Re: [Ohrrpgce] New feature: Achievements
> there are still lots more things that can be done after this is merged (you > didn't really list them so I will), including our own non-Steam display of > achievements, an editor, script commands (as an alternative to toggling > tags), and debug facilities. Yes, you are correct. As it is right now, you have to manually construct the achievement lump to get it to work. I figured that this is the absolute minimum amount of work to make it viable to merge. > but I did notice that Steam.run_frame still contains ACH_WIN_ONE_GAME test > code Whoops! > and a lot of commented out C++ that probably doesn't need to be there anymore That was left on purpose, but I suppose it doesn’t really need to be there now that we’re sure the signatures are correct. > and also run_frame is not really being called from the right place: it will > be called from 0 to 200 times a second depending on how much idle time is > available. Should it be called more often than whatever the game's framerate > (likely 18fps) is? Really? I put it there since that’s when we invoke the graphics backend. Is there a better place to put it? (I don’t think it needs to be called all that often, but this is the closest thing we have to a main loop, as far as I can tell) -- Mike Caron From: Ralph Versteegen<mailto:teeem...@gmail.com> Sent: November 18, 2021 8:58 To: ohrrpgce<mailto:ohrrpgce@lists.motherhamster.org> Subject: Re: [Ohrrpgce] New feature: Achievements Great, at last! I can't believe that I thought (long ago) that this would be a straightforward thing to add since it's "just" calling an external API... there are still lots more things that can be done after this is merged (you didn't really list them so I will), including our own non-Steam display of achievements, an editor, script commands (as an alternative to toggling tags), and debug facilities. One thing it would be good to get other people's opinion on: there's currently a global flag to switch between permanent achievements (the only sort supported by Steam) or per-game/save ones. Would anyone ever want to mix the two? Consider that non-permanent ones won't be shown with the steam overlay. I didn't have time tonight to look over and write comments on the code, but I did notice that Steam.run_frame still contains ACH_WIN_ONE_GAME test code, and a lot of commented out C++ that probably doesn't need to be there anymore, and also run_frame is not really being called from the right place: it will be called from 0 to 200 times a second depending on how much idle time is available. Should it be called more often than whatever the game's framerate (likely 18fps) is? On Thu, 18 Nov 2021 at 18:35, Mike Caron mailto:caron.m...@outlook.com>> wrote: Naturally, I forget to include a link to the Pull Request. I have been a professional for a decade now https://github.com/ohrrpgce/ohrrpgce/pull/1229 Mike Caron From: Ohrrpgce mailto:ohrrpgce-boun...@lists.motherhamster.org>> on behalf of Mike Caron mailto:caron.m...@outlook.com>> Sent: November 18, 2021 5:33 AM To: ohrrpgce@lists.motherhamster.org<mailto:ohrrpgce@lists.motherhamster.org> mailto:ohrrpgce@lists.motherhamster.org>> Subject: [Ohrrpgce] New feature: Achievements Hello friends, As I mentioned a few weeks ago, I was working on support for Steam achievements in the OHRRPGCE. Well, the fateful day has come when I think it's ready to release to the wild. But, that said, I haven't contributed anything for a while, so I wanted to give you guys a chance to review my work before merging it in. So, I've opened up a Pull Request on GitHub and I would welcome any feedback that anyone would like to provide. Don't hold back! I can take it! Cheers! Mike Caron ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org<mailto:ohrrpgce@lists.motherhamster.org> http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] New feature: Achievements
Naturally, I forget to include a link to the Pull Request. I have been a professional for a decade now https://github.com/ohrrpgce/ohrrpgce/pull/1229 Mike Caron From: Ohrrpgce on behalf of Mike Caron Sent: November 18, 2021 5:33 AM To: ohrrpgce@lists.motherhamster.org Subject: [Ohrrpgce] New feature: Achievements Hello friends, As I mentioned a few weeks ago, I was working on support for Steam achievements in the OHRRPGCE. Well, the fateful day has come when I think it's ready to release to the wild. But, that said, I haven't contributed anything for a while, so I wanted to give you guys a chance to review my work before merging it in. So, I've opened up a Pull Request on GitHub and I would welcome any feedback that anyone would like to provide. Don't hold back! I can take it! Cheers! Mike Caron ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
[Ohrrpgce] New feature: Achievements
Hello friends, As I mentioned a few weeks ago, I was working on support for Steam achievements in the OHRRPGCE. Well, the fateful day has come when I think it's ready to release to the wild. But, that said, I haven't contributed anything for a while, so I wanted to give you guys a chance to review my work before merging it in. So, I've opened up a Pull Request on GitHub and I would welcome any feedback that anyone would like to provide. Don't hold back! I can take it! Cheers! Mike Caron ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] Achievement implementations
It definitely supports all those things, except for the automatic reward part.
Unless I’m missing it in their documentation:
https://partner.steamgames.com/doc/features/achievements#4
These points in particular:
* For achievements that have progress bars use
ISteamUserStats::IndicateAchievementProgress at significant points to show a
pop-up with the progress. For instance, if you need 20 wins you may want to
call this at 10 wins to show that the user has made it halfway.
* Whenever one or more achievements are unlocked, call
ISteamUserStats::SetAchievement for each unlocked achievement and then
ISteamUserStats::StoreStats to immediately upload them. Your game will see a
ISteamUserStats::UserStatsStored_t callback, plus one
ISteamUserStats::UserAchievementStored_t callback for each unlocked
achievement. The Steam Game Overlay will display a notification panel to the
user.
It does appear that you could use these stats to persist between saves, but we
could do that already anyway, and I think a major goal is to be portable, even
if the user doesn’t have Steam, so I think it makes more sense to keep track
ourselves. If the stats are visible in Steam (which, if they are, then I’m not
sure where), the we might want to try and sync them, but that’s a bigger deal,
and maybe something we do later.
(Also hi Simon!)
--
Mike Caron
From: Simon Bradley<mailto:neworigi...@gmail.com>
Sent: October 28, 2021 12:01
To: ohrrpgce@lists.motherhamster.org<mailto:ohrrpgce@lists.motherhamster.org>
Subject: Re: [Ohrrpgce] Achievement implementations
I thought Steam had a system where you could update user statistics on Stream
and it could automatically award achievements when they reach a specific value?
I might be imagining that, or getting it confused with some other achievement
system. I don't know if you'd need to save anything locally for cross-save
counters, I think you could just increment a count on steam.
On Thu, 28 Oct 2021 at 15:08, Mike Caron
mailto:caron.m...@outlook.com>> wrote:
The tag_updates function is the place to put hooks for when a tag changes.
I assumed there might be a place like that. Thanks!
If achievements are displayed through steam, how does this work? You
can cause Steam to popup a progress notification with text of your
choice? Or does it have to be numerical, or predefined in Steam's
achievement editor?
You can just tell it to display progress whenever you want, I believe. I was
thinking that we would just do this whenever "value % interval == 0".
What about achievements that can be progressed towards in separate
saved games? The example you put on the wiki of finding all the ways
to die is one where progress needs to be shared rather than
save-specific.
I was envisioning a game where these deaths were not
game-over-and-reloading-your-last-save inducing, but I suppose that's not
clear. I am not sure how we might want to handle cross-save achievements. Part
of me wants to not handle it at all and let game creators do that themselves
(if or when custom IO is a thing), but it also wouldn't be Too hard to add it
now either. On the third hand, I suspect this is a rather more niche feature,
so maybe we can just punt the issue down the road a bit.
Creating an editor for RELOAD-based data is not yet simple, but I'm
going to make a start on generalising the Purchases editor (most
likely) into a class for defining editable RELOAD trees.
Neat. I haven't actually looked too hard into what editors look like these days
so I'm not sure how bad "not yet simple" is, but if there's anything I can do
to help, please let me know.
Pleased that documentation already exists on the wiki, documenting
everything is a big task!
If I had to pick one lesson that I've learned over the years, it's that
documenting things first makes implementation easier. It's keeping that
documentation up to date that is the challenge
From: Ohrrpgce
mailto:ohrrpgce-boun...@lists.motherhamster.org>>
on behalf of Ralph Versteegen mailto:teeem...@gmail.com>>
Sent: October 28, 2021 6:24 AM
To: ohrrpgce
mailto:ohrrpgce@lists.motherhamster.org>>
Subject: Re: [Ohrrpgce] Achievement implementations
This is going quick!
The permanent data storage you're looking for is
https://rpg.hamsterrepublic.com/ohrrpgce/Persist.reld
The tag_updates function is the place to put hooks for when a tag changes.
> A game may wish to expose the current progress at intervals ("defeat 100
> goblins: 20/100") to motivate the player to continue progresing the
> achievement.
If achievements are displayed through steam, how does this work? You
can cause Steam to popup a progress notification with text of your
choice? Or does it have to be numerical, or predefined in Steam's
achievement editor?
What about achievements that can be progressed towards in separate
saved games? The example you put on the wiki of finding all the way
Re: [Ohrrpgce] Achievement implementations
The tag_updates function is the place to put hooks for when a tag changes.
I assumed there might be a place like that. Thanks!
If achievements are displayed through steam, how does this work? You
can cause Steam to popup a progress notification with text of your
choice? Or does it have to be numerical, or predefined in Steam's
achievement editor?
You can just tell it to display progress whenever you want, I believe. I was
thinking that we would just do this whenever "value % interval == 0".
What about achievements that can be progressed towards in separate
saved games? The example you put on the wiki of finding all the ways
to die is one where progress needs to be shared rather than
save-specific.
I was envisioning a game where these deaths were not
game-over-and-reloading-your-last-save inducing, but I suppose that's not
clear. I am not sure how we might want to handle cross-save achievements. Part
of me wants to not handle it at all and let game creators do that themselves
(if or when custom IO is a thing), but it also wouldn't be Too hard to add it
now either. On the third hand, I suspect this is a rather more niche feature,
so maybe we can just punt the issue down the road a bit.
Creating an editor for RELOAD-based data is not yet simple, but I'm
going to make a start on generalising the Purchases editor (most
likely) into a class for defining editable RELOAD trees.
Neat. I haven't actually looked too hard into what editors look like these days
so I'm not sure how bad "not yet simple" is, but if there's anything I can do
to help, please let me know.
Pleased that documentation already exists on the wiki, documenting
everything is a big task!
If I had to pick one lesson that I've learned over the years, it's that
documenting things first makes implementation easier. It's keeping that
documentation up to date that is the challenge
From: Ohrrpgce on behalf of Ralph
Versteegen
Sent: October 28, 2021 6:24 AM
To: ohrrpgce
Subject: Re: [Ohrrpgce] Achievement implementations
This is going quick!
The permanent data storage you're looking for is
https://rpg.hamsterrepublic.com/ohrrpgce/Persist.reld
The tag_updates function is the place to put hooks for when a tag changes.
> A game may wish to expose the current progress at intervals ("defeat 100
> goblins: 20/100") to motivate the player to continue progresing the
> achievement.
If achievements are displayed through steam, how does this work? You
can cause Steam to popup a progress notification with text of your
choice? Or does it have to be numerical, or predefined in Steam's
achievement editor?
What about achievements that can be progressed towards in separate
saved games? The example you put on the wiki of finding all the ways
to die is one where progress needs to be shared rather than
save-specific.
Creating an editor for RELOAD-based data is not yet simple, but I'm
going to make a start on generalising the Purchases editor (most
likely) into a class for defining editable RELOAD trees.
Pleased that documentation already exists on the wiki, documenting
everything is a big task!
On Thu, 28 Oct 2021 at 14:03, Mike Caron wrote:
>
> Hello friends, sorry I’m late. Got sidetracked with a few things
>
>
>
> As I’ve been discussing with TMC on the Slime Salad discord, I’m working on
> adding achievements and Steamworks support to the OHRRPGCE. This is based on
> a post TMC made here just about three years ago:
> http://lists.motherhamster.org/pipermail/ohrrpgce-motherhamster.org/2018-October/004308.html
>
>
>
> I’ve currently made progress on integrating Steamworks itself, thanks in no
> small part to TMC helping me get my environment set up, and now I’m thinking
> about how this stuff will work in the engine itself. I’ve been documenting
> stuff on the wiki, and I would encourage everyone to take a peek and give any
> feedback they have. Note that this page is not complete, and suggestions to
> fill in blanks would be appreciated.
>
>
>
> https://rpg.hamsterrepublic.com/ohrrpgce/User:Pkmnfrk/Achievements
>
>
>
> I’m hopeful to get this implemented soon, so I can get back to my next 13
> year hiatus.
>
>
>
> Cheers!
>
> --
>
> Mike Caron
>
>
>
> ___
> Ohrrpgce mailing list
> ohrrpgce@lists.motherhamster.org
> http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
___
Ohrrpgce mailing list
ohrrpgce@lists.motherhamster.org
http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
___
Ohrrpgce mailing list
ohrrpgce@lists.motherhamster.org
http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
[Ohrrpgce] Achievement implementations
Hello friends, sorry I’m late. Got sidetracked with a few things As I’ve been discussing with TMC on the Slime Salad discord, I’m working on adding achievements and Steamworks support to the OHRRPGCE. This is based on a post TMC made here just about three years ago: http://lists.motherhamster.org/pipermail/ohrrpgce-motherhamster.org/2018-October/004308.html I’ve currently made progress on integrating Steamworks itself, thanks in no small part to TMC helping me get my environment set up, and now I’m thinking about how this stuff will work in the engine itself. I’ve been documenting stuff on the wiki, and I would encourage everyone to take a peek and give any feedback they have. Note that this page is not complete, and suggestions to fill in blanks would be appreciated. https://rpg.hamsterrepublic.com/ohrrpgce/User:Pkmnfrk/Achievements I’m hopeful to get this implemented soon, so I can get back to my next 13 year hiatus. Cheers! -- Mike Caron ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] Mac app bundles build on Windows crash on Mac
On Thu, Apr 11, 2013 at 6:03 AM, Ralph Versteegen teeem...@gmail.comwrote: Popped in to celebrate with us? :) Perhaps! On 11 April 2013 13:44, James Paige b...@hamsterrepublic.com wrote: Hello, Mike! Actually, maybe we can avoid tar. I know we had reasons in the past, but they might have all evaporated by now. Originally I remember only wanting to use tools likely to be already installed on Linux/Mac, but Windows is the trouble-platform, and we have to bundle/download tools anyway. It is worth thinking over. I think the two reasons for using tar were to preserve executable file bits and to preserve symlinks. Funny enough, tar.exe doesn't do either properly. I looked through the commandline options that 7za supports and saw nothing for file attributes or symlinks. So we're seemingly still better off using tar.exe, at least for creating tars. Which maybe we don't want to do. Wait, you're using a Windows binary to create an archive with symlinks and permission bits on Windows? Where do you get the symlinks in the first place?? If you're going to do that, you would need cygwin or something, whose tar command is presumably not broken. Or, alternately the tar format doesn't seem that complicated. Maybe a highly specialized version is in order? These sound like the words of a man who hasn't read any tar file format docs... or maybe I should say those sound like the words *I* spoke before I had read any tar file format docs :) Yes, just look at what happened to tar.exe when it tried to read BSD tar files! This is an ancient file format seemingly with some weird quirks. I realized, too late of course, that I forgot the ;) I meant to add. I would never seriously advocate trying to re-implement a program with such... let's call it legacy. --- James On Wed, Apr 10, 2013 at 08:22:24PM -0400, Mike Caron wrote: Hi guys, long time no see. Is there some reason you're still using tar? Wouldn't it be easier to just bundle 7zip or something? I understand that it supports tar. Probably a few other formats too. You can obtain the console version here. Or, alternately the tar format doesn't seem that complicated. Maybe a highly specialized version is in order? On Wed, Apr 10, 2013 at 1:29 PM, James Paige b...@hamsterrepublic.com wrote: On Wed, Apr 10, 2013 at 09:52:23AM -0700, James Paige wrote: I was almost finished releasing beelzebufo, when I was puzzled and displayed to discover that Mac App game bundles build on Windows do not work. If you build them on mac, they work fine. Here is the crash message: http://pastebin.com/jhu5H4b8 It looks like the problem has something to do with the SDL framework, but I don't understand how this would fail on windows because there are no more symlinks. dismayed not displayed I took the working app build on a Mac and compared it to the broken app built on Windows. I discovered that Contents/FrameWorks/SDL_mixer.framework/Versions/A/Frameworks/mikmod.framework is empty in the broken app. I tested tar.exe from the command line, and discovered that when it unpacks ohrrpgce-mac.minimal.tar it incorrectly unpacks several of the subfolders into the current directory, instead of where they belong under the OHRRPGCE-Game.app folder heirarchy. So the short answer is that this bug is, yet again, another instance of the windows port of tar.exe being a worthless piece of crap :( --- James ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org -- Mike Caron ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org -- Mike Caron ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] Mac app bundles build on Windows crash on Mac
Hi guys, long time no see. Is there some reason you're still using tar? Wouldn't it be easier to just bundle 7zip or something? I understand that it supports tar. Probably a few other formats too. You can obtain the console version here.http://sourceforge.net/projects/sevenzip/files/7-Zip/9.20/7za920.zip/download Or, alternately the tar formathttp://en.wikipedia.org/wiki/Tar_(computing) doesn't seem that complicated. Maybe a highly specialized version is in order? On Wed, Apr 10, 2013 at 1:29 PM, James Paige b...@hamsterrepublic.comwrote: On Wed, Apr 10, 2013 at 09:52:23AM -0700, James Paige wrote: I was almost finished releasing beelzebufo, when I was puzzled and displayed to discover that Mac App game bundles build on Windows do not work. If you build them on mac, they work fine. Here is the crash message: http://pastebin.com/jhu5H4b8 It looks like the problem has something to do with the SDL framework, but I don't understand how this would fail on windows because there are no more symlinks. dismayed not displayed I took the working app build on a Mac and compared it to the broken app built on Windows. I discovered that Contents/FrameWorks/SDL_mixer.framework/Versions/A/Frameworks/mikmod.framework is empty in the broken app. I tested tar.exe from the command line, and discovered that when it unpacks ohrrpgce-mac.minimal.tar it incorrectly unpacks several of the subfolders into the current directory, instead of where they belong under the OHRRPGCE-Game.app folder heirarchy. So the short answer is that this bug is, yet again, another instance of the windows port of tar.exe being a worthless piece of crap :( --- James ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org -- Mike Caron ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] SVN: james/5386 Fix detection of absolute paths to allow unc paths to RPG files
Building on this, you should add support for running games over ssh: game.exe mcaron@workpc:/home/mcaron/games/gamename.rpg ;) On Mon, Sep 17, 2012 at 5:05 PM, James Paige b...@hamsterrepublic.com wrote: On Mon, Sep 17, 2012 at 02:02:34PM -0700, subvers...@hamsterrepublic.com wrote: james 2012-09-17 14:02:34 -0700 (Mon, 17 Sep 2012) 64 Fix detection of absolute paths to allow unc paths to RPG files You can now use the command line: game.exe \\servername\sharename\foldername\gamename.rpg and it will just work. This should also fix double-clicking on rpg files located at unc paths. --- James ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org -- Mike Caron ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] npc copy count
On Tue, May 22, 2012 at 1:39 PM, Ralph Versteegen teeem...@gmail.com wrote: On 23 May 2012 05:27, Adam Perry arpe...@gmail.com wrote: The documentation for this really needs to change. The description says that it returns the number of copies of an NPC, but the example script shows that it returns the number of copies minus one, No it doesn't, the example is wrong. For that matter part of a sentence is missing from the description and the rest is full of typos! No wonder you decided to trust the example instead. Seems the part sentence is my fault. I like to stop mid word and do som Having npc copy count return the number of /copies/ (that is, all the duplicates) would be silly. It would raise moral issues about which NPC is the original, and whether the copies are actually NPCs at all. I mean, do they have the same rights as the original? They share palettes, appearance and behaviour, sure, but they can move independently. What measure is a copy NPC? which I believe is the correct behavior. I can never remember, though, and every time I look it up, I am disappointed anew. ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org -- Mike Caron Tale of the Cave http://taleotc.com ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] SVN: teeemcee/5135 Add Jay's suggested workaround for bug 956 Mangled graphics on startup
Files (x86)\Microsoft Visual Studio 9.0\Common7\IDE\ Then I can type cl at the command line and see the usage, or cl -help and see the help screen. However, scons still says it can't find cl.exe --- James OK, I was wrong, cl.exe isn't meant to be in your path (delete all that). Instead, scons uses the VS90COMNTOOLS environmental variable to see that VS is installed, and where. I certainly didn't set this variable; it should have been added when VS was installed. On my system it's VS90COMNTOOLS=C:\Program Files\Microsoft Visual Studio 9.0\Common7\Tools\ That environment variable is meant to be set by running the Microsoft Visual Studio - Visual Studio Tools - Visual Studio Command Prompt tool, which opens up cmd.exe, but with a bunch of extra stuff set up to let you run the command line tools. It's actually just a batch file that lives, on my system, at C:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\vcvarsall.bat So, basically, you just need to run this batch file before running scons, and you should be golden. -- Mike Caron Tale of the Cave http://taleotc.com ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] SVN: james/5146 Prompt to download ar+tar+gzip the first time you try to distribute as a
On Thu, Apr 26, 2012 at 6:55 PM, James Paige b...@hamsterrepublic.com wrote: On Thu, Apr 26, 2012 at 3:26 PM, James Paige b...@hamsterrepublic.com wrote: On Thu, Apr 26, 2012 at 03:17:47PM -0700, subvers...@hamsterrepublic.com wrote: james 2012-04-26 15:17:47 -0700 (Thu, 26 Apr 2012) 210 Prompt to download ar+tar+gzip the first time you try to distribute as a debian linux package on Windows (I am reasonably confident that tar.exe and gzip.exe work standalone, but I have my doubts about ar.exe) My reason for thinking so is that tar.exe and gzip.exe are both taken from UnxUtils which all seems to be designed with standalone usage in mind, but ar.exe is taken from MinGW, so I don't know yet whether it depends on any MinGW dll files, and if so, which ones. Okay! I figured it out. There were indeed three mingw .dll files that ar.exe required. Yay for Dependency Walker! I tested Export Debian Linux Package on a pristine Windows XP virtualbox, and it all seems to work great (aside from my previously mentioned file permission lamentations) I am almost feeling bold enough to tackle Export Mac OS X app bundle but I guess there isn't a whole lot of point to doing that one until we figure out a nightly build method for the Mac port. I was trying to get Mac OS X running in a virtualbox, but I ended up frustrated pretty quick :( I would advise against further attempts at trying to get OSX running in a VM. That way list the sort of madness that ends with trying to summon an eldritch abomination an an attempt to cleanse the world of people who would invent hardware locks. Not fun for anyone involved. -- Mike Caron Tale of the Cave http://taleotc.com ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] Teach me some git
The thing to remember about git is that it's all about branches. You are always on a branch, usually called master (but that varies, of course). So, I will use master as the branch name below. To update your repo, you need to do this: git fetch #pulls changes from origin git merge origin/master #updates working copy To commit changes, you just need to do this: git add path/to/file #adds a file git add -u #adds all changed files git add -a #adds all non-ignored files git commit -m your message goes here (obviously, pick whatever add command is most appropriate) Lastly, to push your changes back to origin, just do this: git push origin master (if your branch is not up to date, you will need to refresh it as above) That's all you need to know to get started. However, I strongly suggest you learn git's branching mechanics, because they are really powerful and really useful. At my work, we use a branch per ticket/bugfix, and it works beautifully. Sent from my iPad On 2012-04-18, at 12:56 PM, James Paige b...@hamsterrepublic.com wrote: I figure if I am ever going to learn to use git, I need to learn how to replicate my existing svn usgae patterns with git. I have been told many times that git's functionality is a superset of svn's so this should be no problem. My trouble is that every tutorial I can find, even the ones that are supposedly intended for recent converts from svn seem to want me to jump right in to gits more advanced features, and discard my old workflows. I would much rather first get used to using git exactly thew same way I use svn, and then when I am confortable with that, I can start learning the new fancy things that make git special. I have two different workflows for svn repositories 1) I have read only access, and I am not changing anything. All I want to do is keep my local copy up-to-date with upstream. With svn I just do: svn cleanup svn update in a cron job, and all is happy. However, I am not clear on how to do this in git. I assume it is simple, but I get confused when I search for git equivalent for svn update and every single search result tells me something different, and none of them are trying to do the uber-simple case that I care about anyway. 2) Committing my local changes back to a remote server. This is what I would want to do if I was working on the ohrrpgce using git. I know I that local branches are a big deal with git, and that disconnected operation without an internet connection is a big deal with git, but for my purposes I want to treat a centralized repsitory on a server as the master, and use it to sync any assorted workstations I might happen to be sitting in front of at the moment. --- James, Curious ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] Teach me some git
Whoops, I am remiss, and forgot a few things: - origin is what is known as a remote. This might be a colleague's computer, or it might be a central server (I only have experience with the latter). By default, the remote is named origin. So, when you're dealing with remote branches that live on origin, they will be prefixed with origin/ - a branch in your local repo (master) is independant of the branch on origin (origin/master). This makes sense, because you can commit to your local branch without affecting others, and vice-versa. But, it was non-obvious for me. - because of this, when you update your local repo (git merge), you're performing the exact same operation as merging a branch into your branch (because you are). This includes nastiness like merge conflicts, but also good things like updating history. - adding a file (git add) does not have the same semantics as it does in subversion. Specifically, adding a file only marks it for staging. That is, it says next time I commit, include this file. Already existing files still need to be added when you want to commit their changes. Conversely, if you don't want to commit certain files, you can just not add them. - Other useful git commands: git diff #show uncommitted changes git diff --cached #show staged, but uncommitted changes (i.e., only what you've added) git blame path/to/file #see who done what when git log #review older commits git reset path/to/file #un-add a file git reset --hard #nuke your changes from orbit (make everything clean again) On Wed, Apr 18, 2012 at 2:00 PM, Mike Caron caron.m...@gmail.com wrote: The thing to remember about git is that it's all about branches. You are always on a branch, usually called master (but that varies, of course). So, I will use master as the branch name below. To update your repo, you need to do this: git fetch #pulls changes from origin git merge origin/master #updates working copy To commit changes, you just need to do this: git add path/to/file #adds a file git add -u #adds all changed files git add -a #adds all non-ignored files git commit -m your message goes here (obviously, pick whatever add command is most appropriate) Lastly, to push your changes back to origin, just do this: git push origin master (if your branch is not up to date, you will need to refresh it as above) That's all you need to know to get started. However, I strongly suggest you learn git's branching mechanics, because they are really powerful and really useful. At my work, we use a branch per ticket/bugfix, and it works beautifully. Sent from my iPad On 2012-04-18, at 12:56 PM, James Paige b...@hamsterrepublic.com wrote: I figure if I am ever going to learn to use git, I need to learn how to replicate my existing svn usgae patterns with git. I have been told many times that git's functionality is a superset of svn's so this should be no problem. My trouble is that every tutorial I can find, even the ones that are supposedly intended for recent converts from svn seem to want me to jump right in to gits more advanced features, and discard my old workflows. I would much rather first get used to using git exactly thew same way I use svn, and then when I am confortable with that, I can start learning the new fancy things that make git special. I have two different workflows for svn repositories 1) I have read only access, and I am not changing anything. All I want to do is keep my local copy up-to-date with upstream. With svn I just do: svn cleanup svn update in a cron job, and all is happy. However, I am not clear on how to do this in git. I assume it is simple, but I get confused when I search for git equivalent for svn update and every single search result tells me something different, and none of them are trying to do the uber-simple case that I care about anyway. 2) Committing my local changes back to a remote server. This is what I would want to do if I was working on the ohrrpgce using git. I know I that local branches are a big deal with git, and that disconnected operation without an internet connection is a big deal with git, but for my purposes I want to treat a centralized repsitory on a server as the master, and use it to sync any assorted workstations I might happen to be sitting in front of at the moment. --- James, Curious ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org -- Mike Caron Tale of the Cave http://taleotc.com ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] SVN: teeemcee/5109 gfx_console, the curses-based graphics backend!
When I clicked on this email, I was really expecting something like... well, see the attachement ;) On Fri, Mar 23, 2012 at 6:03 AM, Seth Hetu seth.h...@gmail.com wrote: Modern technology is amazing. --Seth On Fri, Mar 23, 2012 at 4:44 PM, Ralph Versteegen teeem...@gmail.com wrote: On 23 March 2012 21:27, subvers...@hamsterrepublic.com wrote: teeemcee 2012-03-23 01:27:38 -0700 (Fri, 23 Mar 2012) 273 gfx_console, the curses-based graphics backend! Only text is displayed. Keyboard support is pretty bad (Ctrl, Shift, and Alt ignored), no mouse or joystick support. Also, using the -d option causes ?_debug.txt to be printed instead, at the cost of display and user input. --- U wip/SConscript U wip/allmodex.bas U wip/backends.bas U wip/common.bas U wip/common.bi U wip/gfx.bi A wip/gfx_console.bas U wip/gfx_sdl.bas U wip/ohrbuild.py Here are some samples: Editing a textbox! Return to Main Menu + to copy Text Box 2 ALT+C copy style Edit Text Edit Conditionals Edit Choice Box Appearance Next: None Selected Text Search: Connected Boxes... Export text boxes... Import text boxes... Hello, World! Text input working nicely Here's me drawing some graphics! Pal0 Enemy (facing right) Tool:Draw UNDO x=0 y=1 F ROAS MC IE ROTATE Playing autotest.rpg! Freki 10/10 SelfStab Helga 10/10 Item Olaf 10/10 Frumpy 10/10 9 ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org -- Mike Caron Tale of the Cave http://taleotc.com attachment: soj.png___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] Slice based menus
On 2012-02-21, at 12:36 PM, James Paige b...@hamsterrepublic.com wrote: On Tue, Feb 21, 2012 at 07:47:06PM +1300, Ralph Versteegen wrote: On 21 February 2012 06:46, Jay Tennant hierand...@crazyleafgames.com wrote: From: James Paige b...@hamsterrepublic.com Sent: Monday, February 20, 2012 10:00 AM I have been reading this whole thread, and I have to admit, I am totally confused. I am not sure at all what you are working on. Would you mind explaining this project to me? --- James Of course. I'll explain the problem, goal, and implementation of this project. I'll chip in as well... 1. Problem: the OHRRPGCE GUI lacks a uniform implementation I like the OHRRPGCE's interface, but it is difficult to upgrade it to be point-and-clickable. Additional monitoring would be necessary to capture the mouse coordinates and button presses. While that can be done for one menu, the lack of uniform implementation makes it troubling to monitor each menu. And if there was interest to adding more buttons later on, understanding context and rewriting the monitoring code could become daunting. It would be easier if the menu items were objects that could respond to operations performed on them--encapsulation. It would be easier if all menu objects interfaced the same way. It would be easier if there was a manager that uniformly passed messages to the appropriate menu objects that are the focus of the user's input. 2. Goal: create a GUI system that brings uniformity and flexibility to the rendering, interfacing, and management of menu's in the engine. This is for the graphical menus in Custom, of which there aren't many currently: font, sprite, tile, and map editors. The others are just lists of menu items and aren't included in this. The map and sprite editors have lots of different widgets visible onscreen, and the code for them is a mess because we have these monolithic functions that handle everything instead of splitting things up. The map editor and sprite editor code is terrible because it is some of the oldest code in the engine, and because I knew nothing about good design when I wrote it :) There is nothing inherently difficult about creating a complex irregular screen layout that prevents the code from being well organized-- it just doesn't happen to be that way right now :) I love the idea of being able to add callbacks to slices. Consider the sprite editor. I like the idea of having button slices that run a callback if you click on them. They could also have a key shortcut callback so that the existing keyboard shortcuts for them could continue to work. I also like the idea of mouse-down and mouse-up callbacks for the sprite drawing area. I was originally thinking that we would just use a slice to specify the position where the drawing area is located on the screen, and hand-code the rest, but it would be much better if the slice could recieve mouse and keyboard events and pass them on to appropriate sprite-drawing handlers. This is a non-OS specific goal. Everything is rendered inside the engine, as it is now. The objective is to make widgets that are intuitive for users. It is a native implementation. native implementation? What does native mean anyway? I always thought it meant provided by/tightly dependent on the platform (OS/hardware). There was also another goal: look into whether it would be possible to abstract different GUI frameworks. Maybe we could draw widgets using either slices, or something like wxWidgets (which I was especially interested in, due to my gfx_wx attempt) A GUI manager can create, destroy, and pass messages to widgets. A widget can optionally process the messages, or delegate the message processing to another widget. All widgets have a message procedure. Functionally, this is very similar to Window's design. Messages generated by the GUI manager are added to a queue, then processed in the order they are added. Why queue messages instead of dispatching them immediately? So that widgets can peek ahead? 3. Implementation: GuiManager, GuiObject, GUI_* messages, input synthesis, and rendering I've already been working on this since October of last year, though I stopped somewhere in late December (work, etc.). I developed a GuiManager object, which provides methods for sending and posting messages, changing focus, and synthesizing input to appropriate GuiObject's. A GuiObject is an implementation of a widget. I successfully created a Button and Text widget, and a custom widget to test the button and text on. Jay mentioned that he was working on a GUI framework written in C++; not directly OHR-related at the time, but I believe built on gfx_directx? I said that the OHR needed such a thing to clean up the map and sprite editors, so I'd like to see whether we could reuse any of his code. Jay's uses message passing instead of callbacks/methods like GTK+ and wxWidgets do (actually I think
Re: [Ohrrpgce] Was anybody using websvn?
On 15/11/2011 1:39 PM, Adam Perry wrote: On Tue, Nov 15, 2011 at 10:31 AM, James Paige b...@hamsterrepublic.com mailto:b...@hamsterrepublic.com wrote: Was anybody using the websvn interface to the ohrrpgce repository on gilgamesh? I only occasionally used it myself because it was nice to be able to use the blame feature to figure out when a specific line was last changed. I use it occasionally. I don't suppose you can just disable blame? Better yet, why not just add it to robots.txt so MSN bot doesn't even try to go there? -- - Mike Caron ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] Was anybody using websvn?
On 15/11/2011 1:58 PM, Ralph Versteegen wrote: On 16 November 2011 07:42, James Paigeb...@hamsterrepublic.com wrote: On Tue, Nov 15, 2011 at 10:39:25AM -0800, Adam Perry wrote: On Tue, Nov 15, 2011 at 10:31 AM, James Paigeb...@hamsterrepublic.com wrote: Was anybody using the websvn interface to the ohrrpgce repository on gilgamesh? I only occasionally used it myself because it was nice to be able to use the blame feature to figure out when a specific line was last changed. I use it occasionally. I don't suppose you can just disable blame? But that is the best feature! I updated my robots.txt file to tell msnbot to chill out, but it might take a day or two before it re-fetches my robots.txt. After that, I will re-enable websvn and hopefully it will be okay after that. --- James Doesn't the commandline svn tool have a blame feature? It's been so long since I used SVN. Back when I used to, I used TortoiseSVN's blame feature a lot and loved the GUI. However, now that I use Git, I just read blames on the commandline. The difference between blame in svn and in git is that building a blame in svn used to take a minute, while in git it's milliseconds, and there's a vast collection of options for tweaking it! And I find the convenience of using a pager makes up for not using a GUI, so I haven't even bothered looking for one. I sometimes use websvn when I want to look at something in the games or tools portions of the repository when on a machine where I don't have those checked out. ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org Warning: Not necessarily typical results. We use git at work, and running a git blame command usually takes about 10 seconds. Now, granted, we have lots of history, branches and altogether nastiness. I assume that runtime is directly proportional to: 1. How recently the oldest line was touched, and 2. How complex the history is between now and that point. (Of course, adding a graphical tool on top won't speed any of that up, but that's beside the point) -- - Mike Caron ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] SVN: teeemcee/4729 linkgcc is now the default on Windows, and it now links libgcc_s, libstd
On 05/09/2011 2:06 PM, Ralph Versteegen wrote: (Also, I never realised that Mike created music_native_subs.bas by porting MIDI code from SDL_Mixer to FB... I plan to replace that with the original C (hopefully containing new bugfixes)) I did what now? On forever ago, music_native_subs.bas wrote: ' Many subs in this file are ported from SDL_Mixer (used under the GPL): ' /* ' native_midi: Hardware Midi support for the SDL_mixer library ' Copyright (C) 2000,2001 Florian 'Proff' Schulze ' */ ' ' Ported to FreeBasic by Mike Caron Oh. You know, I have absolutely no recollection of this. Why in the heck would I use SDL_mixer as a base?? I remember looking at allegro, but not SDL... -- - Mike Caron ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] SVN: james/4555 Timer triggers were NEVER being triggered inside battle
On 7/7/2011 20:45, subvers...@hamsterrepublic.com wrote: james 2011-07-07 17:45:37 -0700 (Thu, 07 Jul 2011) 779 Timer triggers were NEVER being triggered inside battle (in spite of the fact that timers were always running in battle even if you didn't want them to be) The only reason timers in battle actually got triggered in random battles was because the timer code would get re-called for one more tick after the battle ended. I have no idea why this was happening for random battles but not for fight formation battles. Now timers that run out in battle call their trigger immediately, it just wont execute until after battle because scripts don't run in battle Hmm, I don't remember exactly, but I think that at the time, that was not a viable solution (don't quote me on that). I suspect that what I would have done would be to set an extra flag on the timer to indicate that hey, this timer expired at a time we couldn't execute scripts, please run me. This may be the one more time behaviour you were experiencing. I suspect that this was still the case, but when you use fight formation, the script interpreter is not actually in a wait state. Thus, the timer script would not have an opportunity to run right then, and would subsequently be forgotten. This theory may be jossed by the presence of after battle scripts, which I assume still run after fight formation battles. I might possibly have broken instead-of-battle and after-battle scripts by doing this... more tests coming soon to verify that I didn't. autotest.rpg now checks six different combinations of timer/battle interactions. I think the edge cases for this particular interaction would be: 1. Timer fires on same tick as battle 2. Timer fires during battle start (if there was any animation at the start of battle...) 3. Timer fires while waiting in battle 4. Timer fires while animating 5. Timer fires during victory dance 6. Timer fires tick after battle And, of course, all that again with the timer aborts battle flag. --- U wip/game.bas U wip/testgame/autotest.hss U wip/testgame/autotest.rpg ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org -- - Mike Caron ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] Strange bug in Wandering Hamster
On 14/06/2011 4:10 PM, James Paige wrote: On Mon, Jun 13, 2011 at 04:52:14PM -0500, Mike Caron wrote: The other day, I fired up Wandering Hamster to check out the parallaxing effect on Troll Mountain (very neat, incidentally). Afterwards, my party was a bit beat up, so I went to the nearest inn, which was Broaste's castle. There was a bug in the parallax script that should have caused the world map to be broken when you exited the mountain. I just fixed it today. I have no idea how you avoided it. Hmm, I may have gone to the inn first, and THEN the mountains. Actually, that's almost certainly the case. So, had I not simply exited from the mountain, then it would likely have triggered. As soon as I entered the interior of the castle, however, the menu beeping sound started looping continuously, and continued to play forever, even when I went out side. Is this repeatable? Can you get it to happen every time you visit that map? Yes. I tried it several times, and it happened every time. I did not try restarting the program, but I will try tonight. I was able to make it stop by opening the real menu, and moving the cursor a few times until I managed to preempt the sound. I did not test any other interior maps, but since there is no such distinction in the OHR, I doubt that means anything. It did not occur on the World Map, the castle exterior or the mountain. (incidentally, you should add a category to bugzilla for Wandering Hamster and other official games) I have a page on the Wiki for Wandering Hamster bugs... and I was about to link to it when I saw you had already commented on it before, so i guess you know where it is :) This cannot have been any time in the last 3 years _ That said, I probably should have checked there first (though, how quickly would it have been spotted?) I would rather not put non-engine game bugs on bugzilla. (although what you describe does sound more like an engine bug) Fair enough. --- James ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org -- - Mike Caron ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
[Ohrrpgce] Strange bug in Wandering Hamster
The other day, I fired up Wandering Hamster to check out the parallaxing effect on Troll Mountain (very neat, incidentally). Afterwards, my party was a bit beat up, so I went to the nearest inn, which was Broaste's castle. As soon as I entered the interior of the castle, however, the menu beeping sound started looping continuously, and continued to play forever, even when I went out side. I was able to make it stop by opening the real menu, and moving the cursor a few times until I managed to preempt the sound. I did not test any other interior maps, but since there is no such distinction in the OHR, I doubt that means anything. It did not occur on the World Map, the castle exterior or the mountain. (incidentally, you should add a category to bugzilla for Wandering Hamster and other official games) -- - Mike Caron ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] Pixel Shaders
Ugh, why do these threads get so long. Selective quoting, go! On 6/11/2011 1:22, Ralph Versteegen wrote: Pixel shaders 101: A pixel shader is a function applied to every pixel in a scene that can modify pixels. Some of these can be done purely with pixel shaders (bloom, colour inversion, etc), others need the assistance of vertex shaders (outlines, glowing, etc). Not to be ungrateful for the explanation... but a pixel shader is something that can often be accomplished with just a pixel shader? There are two completely different meanings of the word? No no, let me rephrase: A pixel shader is a function applied to every pixel in a scene that can modify pixels. **Certain effects are possible purely** with pixel shaders (bloom, colour inversion, etc), **while other effects require** the assistance of vertex **or other geometry shaders (outlines, glowing, etc). What are the general thoughts on using shaders? Well, they must be portable between different backends, but exact result doesn't matter. It sounds like these would be all implemented in a not-backend-specific way (unless a backend wanted to go the extra mile to provide a D3D shader language implementation, I guess). Or, just use the standard HLSL that compiles to both major chipset shader implementations? Didn't know there was such a thing: I've heard of HLSL but thought it was just the D3D shader language. Oh, you know what, my mistake. I had assumed for some reason that the schism between shader languages was between chipsets, and not APIs, and as a result, my quick glance completely parsed this line from wikipedia wrong: Popular high-level shader languages include DirectX's HLSL and OpenGL's GLSL, along with Cg, which can compile to both OpenGL and DirectX shader instructions. I am suitably embarrassed. I think the most critical argument here (and the most restrictive) is the shaders will exponentially increase rendering time. So it's a feature that should be carefully considered by the user. An abuse of the term exponential. Shader effects are always linear time in the number of pixels, right? But probably want to stay away from really slow effects, as these days the speed difference between a brand new fast CPU and a slow one is more than an order of magnitude; many people won't notice if they're doing something that kills performance for other people. To be perfectly pedantic, in modern cpus, the speed difference is almost completely irrelevant. Modern cpus are faster by incorporating multiple cores. Of course, if your software doesn't take advantage of that... (The OHR does, but only in the backends. Eg, SDL and DirectX use threads internally. I don't know if or how the OHR would use threads in the front end.) -- - Mike Caron ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] Pixel Shaders
On 10/06/2011 2:21 AM, Ralph Versteegen wrote: On 10 June 2011 17:31, Jay Tennanthierand...@crazyleafgames.com wrote: As most everyone is already familiar, a pixel shader is a piece of code that calculates a color of a pixel on a surface that is being drawn to. Then I didn't know what a pixel shader is. I thought they were programmable parts of GPU hardware pipelines, but apparently we're talking about software here. At what point did he say that a shader doesn't run on GPU? The pixel shaders on the rasterizer right now are simply pass-through shaders, doing no calculation besides possibly sampling a texture. So normal texture sampling counts as a pixel shader, but there's only one such sampler currently. I don't know why you're pluralising. There are many cool effects that can be achieved with shaders, such as negative, blur, bloom, fire, water, frosted glass, glow, etc. I don't see how many of these fit your definition of pixel shaders, as they seem to be functions for drawing something onto a surface, in a possibly arbitrary shape (eg. a blurred triangle), rather than calculating the colour of each pixel of a polygon. (OK, I'm just being difficult.) What do you mean by fire and water effects? Are you serious? I can't tell. Pixel shaders 101: A pixel shader is a function applied to every pixel in a scene that can modify pixels. Some of these can be done purely with pixel shaders (bloom, colour inversion, etc), others need the assistance of vertex shaders (outlines, glowing, etc). By fire and water effects, I assume he's talking about wavy textures, like water and fire in many video games. Don't be obtuse. Can you think of more effects? Do the enemy dissolve effects count as shaders? They aren't necessarily computed per-destination pixel, which I'm guessing is a requirement. But anyway, there are many more enemy death/dissolve effects I'd love to implement: some specifically for enemy death animations, others more generally useful effects. Transparency, tinting, and transformed quads are going to enable quite a few death animations I had in mind. But I've also been thinking of a few disintegration-type effects, like turning to dust from the top down or bottom up, or being blown away by the wind, or splintering. It would be a good idea to move the code for these out of allmodex.bas and into a C file, because the melt effect for one is really expensive to compute. In general, it seems it doesn't make sense to try to stick effects like melt in the rasteriser, because they're not rotation-invariant. Sephy requested being able to use dissolve effects on sprite slices, and I've been meaning to get around to implementing that. So it sounds like you're proposing a generalisation of dissolve effects. You could do this with pixel shaders, although I don't know exactly how. But, dissolve effects are hardly the only use for them, so I'm not sure why you would claim that. What are the general thoughts on using shaders? Well, they must be portable between different backends, but exact result doesn't matter. It sounds like these would be all implemented in a not-backend-specific way (unless a backend wanted to go the extra mile to provide a D3D shader language implementation, I guess). Or, just use the standard HLSL that compiles to both major chipset shader implementations? I imagine that some of the effects you mentioned won't work well for low resolution pixel art. If so, it might be a bad idea to provide (and thus encourage people to use) such shaders rather than making them draw the effects themselves better than an algorithm could hope to accomplish. This is perhaps the most critical thing you've said thus far. Unless, of course, there was a way to specify that the shader should run on the target screen buffer, rather than on the source textures! -- - Mike Caron ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] how to force a gdb traceback?
On 10/06/2011 1:18 PM, James Paige wrote: I wondered if, for debugging purposes, there was an easy way to make gdb stop and produce a trackeback on a particular line? Back in the bad old days of ... some other language, don't even remember which one now, I remember I would type 1/0 to force a devision by zero error, that would give me a stack trace. That doesn't work here because division by zero is a compile time error, not a runtime error. Is there a nice elegant way to force a crash on a specific line? Right now my workaround is this klunky bit of code: DIM foo(0) AS INTEGER DIM bar AS INTEGER bar = 1 debug crash here foo(bar) --- James Sub Crash() dim as integer i = 0; 'this will almost certainly fool the compiler. i = 1 / i End Sub Crash() 'boom -- - Mike Caron ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] The parallax problem
On 6/7/2011 19:49, Ralph Versteegen wrote: On 8 June 2011 07:34, Mike Caroncaron.m...@gmail.com wrote: On 07/06/2011 10:44 AM, James Paige wrote: Well, solved in the sense that I can add six more rows and 10 more columns to my background layer and it will look okay, but I still really want to understand the math needed to make the parallax work with the 1/3 size layer, since that is the same math that will be required when parallax scrolling is a built-in feature on different-sized map layers in the future. I meant solved in the sense that Mike (Willis) already gave the formula for that. First, let us suppose a few things: MapWidth := 3000 LayerWidth := 1000 ScreenWidth := 320 You want the layer to butt up against the left side of the map when the camera is at X = 0, and to butt up against the right side of the map when the camera is as X = (MapWidth - ScreenWidth). When on the left side of the map, the camera's X, the map's X and the layer's X are all 0. This case is easy. On the right side of the map, however, it's not so easy. The camera stop at 2680 (MapWidth - ScreenWidth), while the layer needs to stop at 2000 (MapWidth - LayerWidth). You mean the layer needs to stop at LayerWidth - ScreenWidth = 680. You get marks for correcting working-through, but this makes the rest of your post wrong. Huh? If the layer stops at 680, that means its right edge is at 1680, which puts it squarely off the screen. I thought the idea was to have the right edge stop at 3000, like the other right edges. To get a smooth interpolation, you simply need to map the camera's X% (defined as CameraX / (MapWidth - ScreenWidth)) to the layer's X% (defined as LayerX / (MapWidth - LayerWidth)). This formula is most easily expressed as: LayerX = (CameraX / (MapWidth - ScreenWidth)) * (MapWidth - LayerWidth) Correction (what msw already said): LayerX = (CameraX / (MapWidth - ScreenWidth)) * (LayerWidth - ScreenWidth) We know that when the CameraX% is 50% (1340), the LayerX should be at 1000. Let's verify that: LayerX = (1340 / (3000 - 320)) * (3000 - 1000) LayerX = (1340 / 2680) * 2000 LayerX = 0.5 * 2000 LayerX = 1000 I can't seem to find the original email in which I presume you listed the script, but if you plug that formula in, it should work. -- - Mike Caron ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org -- - Mike Caron ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] The parallax problem
On 6/7/2011 20:46, James Paige wrote: On Wed, Jun 08, 2011 at 12:43:12PM +1200, Ralph Versteegen wrote: On 8 June 2011 11:52, Mike Caroncaron.m...@gmail.com wrote: On 6/7/2011 19:49, Ralph Versteegen wrote: On 8 June 2011 07:34, Mike Caroncaron.m...@gmail.comwrote: On 07/06/2011 10:44 AM, James Paige wrote: Well, solved in the sense that I can add six more rows and 10 more columns to my background layer and it will look okay, but I still really want to understand the math needed to make the parallax work with the 1/3 size layer, since that is the same math that will be required when parallax scrolling is a built-in feature on different-sized map layers in the future. I meant solved in the sense that Mike (Willis) already gave the formula for that. First, let us suppose a few things: MapWidth := 3000 LayerWidth := 1000 ScreenWidth := 320 You want the layer to butt up against the left side of the map when the camera is at X = 0, and to butt up against the right side of the map when the camera is as X = (MapWidth - ScreenWidth). When on the left side of the map, the camera's X, the map's X and the layer's X are all 0. This case is easy. On the right side of the map, however, it's not so easy. The camera stop at 2680 (MapWidth - ScreenWidth), while the layer needs to stop at 2000 (MapWidth - LayerWidth). You mean the layer needs to stop at LayerWidth - ScreenWidth = 680. You get marks for correcting working-through, but this makes the rest of your post wrong. Huh? If the layer stops at 680, that means its right edge is at 1680, which puts it squarely off the screen. I thought the idea was to have the right edge stop at 3000, like the other right edges. It looks like we were talking about completely different things: I was talking about the camera position on the parallax map layer, and you about the offset of the parallax map layer to the map root slice. Either way, I am still completely confused. My attempt to use Mike's suggested formula resulted in the parallax layer being correctly placed at the bottom and right edges of the map, and being at 0,0 for all other camera positions :( set slice x(sl, (camera pixel x / (mapw -- screenw)) * (mapw -- layerw)) set slice y(sl, (camera pixel y / (maph -- screenh)) * (maph -- layerh)) Hmm, this is a thing that I didn't think about until just now. At which points does the script engine round off decimals? You may need to add a few (x * 10) and (y / 10)s in there... --- James ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org -- - Mike Caron ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] SVN: jay/4381 Getting it closer to actually working (doesn't crash anymore, but it isn
On 03/06/2011 2:34 PM, Jay Tennant wrote:
From: Ralph Versteegenteeem...@gmail.com
Sent: Friday, June 03, 2011 11:47 AM
On 3 June 2011 06:00, Jay Tennanthierand...@crazyleafgames.com wrote:
From: subvers...@hamsterrepublic.com
Sent: Thursday, June 02, 2011 12:45 PM
To: ohrrpgce@lists.motherhamster.org
Subject: [Ohrrpgce] SVN: jay/4381 Getting it closer to actually working
(doesn't crash anymore, but it isn
jay
2011-06-02 10:45:09 -0700 (Thu, 02 Jun 2011)
94
Getting it closer to actually working (doesn't crash anymore, but it isn't
drawing right yet).
Oh, it's drawing *mostly* correct. That's encouraging! :)
Sounds good! So do you have a test program? Actually, I'll add a
function to allmodex.bas to draw transformed rects, and can stick a
test of it in the 'secret_menu' in custom.
I'm running a test program of it right now, but it's so disappointing: the
overflow problem. If I have a fixed point number divided by a very small
fraction (say 1/65536th) I get a terribly inaccurate number. I don't want
to switch to floating point, but I don't think I have much choice. Is there
a solution to this that I'm not getting?
Where the issue arises: the rasterizer evaluates a surface to be drawn to
one line at a time. The first pass is simply to figure what range of pixels
on each line is to be evaluated. The second pass (after all line's pixel
ranges are figured) is to interpolate the edge values across the line for
each pixel. The error is occurring during the first pass, in figuring the
left-most and right-most boundary of a line of pixels to be rasterized.
For each line, I take the slope of the edges of the triangle and find all
x-intercepts. The left-most intercept and the right-most intercept
(as long as they're within range of the triangle) are the boundaries of
the pixels.
I realize they're some optimizations I could perform (like calculating the
edge slopes once), but the error occurs when an edge is almost completely
vertical. The fixed point integer can't represent +/- infinite, so an
overflow/underflow occurs in determining slope (dy / almost 0).
I think the solution is using floating point math for this section. Can
you think of another solution? Other than that, the rasterizer's working
pretty nicely. I'll commit the code when I'm able to.
Does you algorithm allow for swapping axes? Eg,
if(dy dx) {
//swap values, and flip the result turn-ways
}
I don't know if this is possible in your case, however.
--
- Mike Caron
___
Ohrrpgce mailing list
ohrrpgce@lists.motherhamster.org
http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] Proposals: Vector and Matrix math, Per-vertex computations
can rotate an object, what stops you from rotating a clipping rect? I guess that's more or less like a mask. I thought you said that DirectX could only clip to a non-rotated rectangle on the render surface? That's what I was referring to. Ah, I see. It is possible to clip other shaped objects using another surface as the clip mask. I don't have time to think about it right now, though. It's not possible to clip a rotated object, then obviously you just need to clip first, THEN rotate! Frames, which need to be changed to Surface's and Frame views, will need new surface interfaces to access from the backend. The Plan_for_gfx_directx, sketched interfaces, listed a suggested set. To write those into loading, termination, copying, updating, and other calls, I estimate about 15-25 hours. Implementing the backends will take about 10 hours each. I think the first goal should be the SDL backend, and I'd like it in C/C++. OpenGL will be used to accelerate the rendering. Would like to stress again that we should retrofit the existing backends... or actually, probably just gfx_sdl, I'm not sure anyone actually uses gfx_fb or gfx_alleg. Speaking of which, would it be possible to get download stats of the alternative backend nightlies? It'd be fine to do that, but it's also very easy to convert it to c/c++. I guess it's just my preference, but it doesn't need to be c/c++. However, I could guarantee I'd get OpenGL working in c/c++, can't say the same for fb. I was thinking about retrofitting gfx_fb/gfx_sdl without using OpenGL, keeping them very close to their current state, meaning their behaviour is unlikely to change. If you switch them to OpenGL, they'll surely start behaving different for some people -- I wanted gfx_sdl around as a backup for people with OpenGL problems. By all means, write the new SDL 1.3 backend in C++. Actually, using external libraries from FB is problematic, because you have to use headers that have been translated from C to FB either automatically or by hand. Often there are differences (macros missing, type names changed), or the headers are out of date (SDL_mixer.bi for example), and the system headers in particular are terrible because they're system specific (and anyway the translation is mostly broken). ___ Unlimited Disk, Data Transfer, PHP/MySQL Domain Hosting http://www.doteasy.com ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org -- Mike Caron Tale of the Cave http://taleotc.com ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] 16-color Palette manipulation with scripts
On 5/27/2011 20:20, David Gowers (kampu) wrote: On Sat, May 28, 2011 at 1:17 AM, James Paige b...@hamsterrepublic.com mailto:b...@hamsterrepublic.com wrote: This makes a lot of sense to me now. I wrote up a summary of it here: http://rpg.hamsterrepublic.com/ohrrpgce/Types_of_Palettes There are four types of palettes.. (describes three) Are you trying to make a Monty Python reference? ;) 'Type four, there IS NO type four..' First thou shalt take out the holy colours, then thou shalt count to three, no more, no less. Three is the palette thou shalt count, and the number of thy palettes shall be three. Four thou shalt not count, neither count thou two, excepting that thou then proceed to three. Five is right out. Once the palette three, being the third palette, be reached, then lobbest thou thy Master palette towards thy game, which, being made of sprites and such, shall be coloured. -- - Mike Caron ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] 16-color Palette manipulation with scripts
On Thu, May 26, 2011 at 12:47 PM, Jay Tennant hierand...@crazyleafgames.com wrote: From: Ralph Versteegen teeem...@gmail.com Sent: Thursday, May 26, 2011 11:32 AM My concern here, is what kind of problems a feature like this could cause in the future? Actually, wait, I'm not sure that a change to a pallete should affect all other sprites. I think it makes sense to only allow changing the palette for a single sprite (slice): use copy-on-write. Wait, what? Create a new 16-color palette just for one sprite? I think a 16-color palette change should affect all objects that use the 16-color palette. We already have to keep supporting 16 color palette changes on sprites, even when 256 color sprites are enabled in the future-- I wouldn't want to treat them separately at all. Neither would I, but one is like a pointer, and the other is a pointer to a pointer. The 16-color will have to be dereferenced to its 256-color palette. I haven't heard a complete solution to the 16-to-256 palette problem yet, nor have I solved it. Don't use indices, but use actual colours instead? set pal color ( 123, make rgb(255, 0, 0)) -- Mike Caron Tale of the Cave http://taleotc.com ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] 16-color Palette manipulation with scripts
On Thu, May 26, 2011 at 1:01 PM, Jay Tennant hierand...@crazyleafgames.com wrote: We already have to keep supporting 16 color palette changes on sprites, even when 256 color sprites are enabled in the future-- I wouldn't want to treat them separately at all. Neither would I, but one is like a pointer, and the other is a pointer to a pointer. The 16-color will have to be dereferenced to its 256-color palette. I haven't heard a complete solution to the 16-to-256 palette problem yet, nor have I solved it. Don't use indices, but use actual colours instead? set pal color ( 123, make rgb(255, 0, 0)) Yes, but with any changes to the master palette, all the 16-color palettes will have to be changed as well. So the distinction between a master (256-color) palette and a 16-color is still there. Wait a second... are master palettes ever altered during runtime? Anytime there's a fade, for example. However, I wasn't suggesting changing 16-colour palette to use 24-bit colours, I meant having the set pal color command look up the colour in the master palette, and using that. Of course, if you change Red to Blue because the party switched sides or something, this would still break. I guess the other alternative is to let 256-colour sprites use 256-colour palettes, which index into the master palette. (There is no problem that can't be solved by adding another level of indirection ;) -- Mike Caron Tale of the Cave http://taleotc.com ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] No, THIS is how Parallax should work
On Tue, May 24, 2011 at 12:46 PM, James Paige b...@hamsterrepublic.com wrote: On Tue, May 24, 2011 at 01:06:26AM -0400, Mike Caron wrote: On 5/24/2011 0:21, James Paige wrote: Okay, so eventually, it would be cool if we could allow certain map layers to have a size that is different than the map size, and they would automatically parallax scroll. Completely implementing that will be pretty tricky, because of the map file format changes that would have to be implemented, and the complexity of making the map editor handle layers of different sizes. However, on the map display side of things in game, the implementation would actually be pretty simple and straightforward. The plotscript interface for parallaxing map layer 0 could be as simple as: parallax map layer(layer number, width, height) And the rest could happen internally through fairly simple magic. This would assume that the layer you want to parallax has tiles only in the upper left width x height tiles, and that tiles outside that range wouldn't show up through normal scrolling. Of course this wouldn't automatically handle wrapping... actually I am not sure what to think about parallax on a wrapping map... The way I see it, there are two kinds of parallaxing effects one might want to use: 1. The treetop kind. The parallaxing layers (hence forth, the paralayer) are relatively small, but tile to fill up the screen. Some times, they have a constant movement vector (clouds), sometimes they scroll with the camera (leaves), sometimes they do both (rain). However, in all cases, they don't need more than a screen's worth of graphics. Wrapping is irrelevant, as the paralayer's location is not tied the camera's location. Ah, interesting. I had only been thinking of type 2, I had neglected thinking about type 1. 2. The scenic background kind. The paralayer never scrolls automatically, and always scrolls at a fixed multiplier to the camera speed. This is when the graphics on the paralayer are supposed to correspond to graphics on other layers. A classic example is a cliff with a castle or something in the background that only reveals itself when the camera pans up. The size of the paralayer is directly proportional to the movement speed multiplier. If the paralayer moves twice as fast as the camera, then it needs to be twice as big (along each axis). I was thinking you would only specify the size of the paralayer, and then it would automatically figure out the correct movement speed... but this would only work for a background layer that is smaller than the map size. It would not work for a foreground layer that is smaller than the map size, like the type 1 forest layer you described above. Type 1 isn't necessarily foreground. A Link to the Past, for example, uses such a layer on Death Mountain in the Dark World to do lavafloes in certain places by having the lava in the background scroll, and having rocks and such in the foreground with transparent areas, making it look like the lava is coming from the rocks. The defining feature of Type 1 is that it doesn't really have a size -- for all intents and purposes, it's unbounded. Anyway, as long as you allow for speeds 1.0, then there shouldn't be a problem supporting layers of any size: Consider a base map of size 64x64. A paralayer of this size will have a scroll factor of 1. Now, consider a paralayer that's 32x32. This layer will scroll at a speed of 0.5 (slower). A third paralayer, 128x128, would scroll at a speed of 2 (faster). Essentially, as long as the layers scroll such that each edge lines up when the camera is 1/2 screen distance away from that side of the map, it should be okay. Wrapping can be handled the same way as normal map wrapping, but the game maker needs to ensure that it wraps properly (just like with any other map layer). Yeah... that seems sensible. I think the map drawing code would already be capable of drawing some layers wrapping even if the rest of the map was not wrapping. -- While I think that it would be possible to unify the two kinds, I wonder if it is necessary? I remember way back (circa 2006ish) having a discussion about the first kind of parallaxing, using Backdrops as the graphics. I guess that it would be pretty easy to simulate it with a slice and some scripting, so I guess we can ignore it. The second kind would be a bit more difficult, so it should definitely get engine support. I don't think that the editor should need to support editing multiple layer sizes: It would get very messy and confusing (for everyone involved) if two layers were not tile-aligned, for example. I was thinking the editor could just align itself with the layer you were currently working on. Right, but if the scroll factor is not a whole number, then the layers will not be tile aligned, creating confusion (especially since the map editor does not scroll smoothly!) Instead, editing a paralayer should
Re: [Ohrrpgce] No, THIS is how Parallax should work
On 5/24/2011 0:21, James Paige wrote: Okay, so eventually, it would be cool if we could allow certain map layers to have a size that is different than the map size, and they would automatically parallax scroll. Completely implementing that will be pretty tricky, because of the map file format changes that would have to be implemented, and the complexity of making the map editor handle layers of different sizes. However, on the map display side of things in game, the implementation would actually be pretty simple and straightforward. The plotscript interface for parallaxing map layer 0 could be as simple as: parallax map layer(layer number, width, height) And the rest could happen internally through fairly simple magic. This would assume that the layer you want to parallax has tiles only in the upper left width x height tiles, and that tiles outside that range wouldn't show up through normal scrolling. Of course this wouldn't automatically handle wrapping... actually I am not sure what to think about parallax on a wrapping map... The way I see it, there are two kinds of parallaxing effects one might want to use: 1. The treetop kind. The parallaxing layers (hence forth, the paralayer) are relatively small, but tile to fill up the screen. Some times, they have a constant movement vector (clouds), sometimes they scroll with the camera (leaves), sometimes they do both (rain). However, in all cases, they don't need more than a screen's worth of graphics. Wrapping is irrelevant, as the paralayer's location is not tied the camera's location. 2. The scenic background kind. The paralayer never scrolls automatically, and always scrolls at a fixed multiplier to the camera speed. This is when the graphics on the paralayer are supposed to correspond to graphics on other layers. A classic example is a cliff with a castle or something in the background that only reveals itself when the camera pans up. The size of the paralayer is directly proportional to the movement speed multiplier. If the paralayer moves twice as fast as the camera, then it needs to be twice as big (along each axis). Wrapping can be handled the same way as normal map wrapping, but the game maker needs to ensure that it wraps properly (just like with any other map layer). -- While I think that it would be possible to unify the two kinds, I wonder if it is necessary? I remember way back (circa 2006ish) having a discussion about the first kind of parallaxing, using Backdrops as the graphics. I guess that it would be pretty easy to simulate it with a slice and some scripting, so I guess we can ignore it. The second kind would be a bit more difficult, so it should definitely get engine support. I don't think that the editor should need to support editing multiple layer sizes: It would get very messy and confusing (for everyone involved) if two layers were not tile-aligned, for example. Instead, editing a paralayer should take place in its own map editor with only the paralayer visible. To make editing easier, there could be a paralayer preview that /did/ allow the maker to scroll around the map and make sure everything was okay. Other random thoughts that I'm not sure how to put in prose format since it's 1:00 am and I have to be up at 6:30 am: - Since the scroll speed and the required paralayer size are directly related, setting one should also set the other. - Because of this, I don't think allowing a script to parallaxize an existing layer would be useful. It would, in effect, either create a paralayer with the wrong dimensions (either it will need to draw the blank area past the edge of the map, or some will never be seen [which would wreak havoc with wrapping!]), or create paralayer with a 1:1 scroll ratio, which is pointless - Maybe allow other maps to be used as paralayers? That would handily solve the editor problem, at the expense of having to wrangle multiple maps at once. Also, would the NPCs on the parallaxed map be meaningful, etc? - Otherwise, the paralayers can be stored as additional blobs of tilemap data. - etc -- - Mike Caron ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] Frame type
On 5/20/2011 23:51, Ralph Versteegen wrote: On 21 May 2011 14:29, Jay Tennanthierand...@crazyleafgames.com wrote: Then the backend will enqueue a render order (though will probably render immediately right now for testing purposes). Any time a frame needs to be drawn to, the procedure that handles that will finish by copying the data to the backend's surface. Ah, somehow this seems too simple :) You didn't mention 16-colour palettes! It's funny you should mention this. Here's a suggestion to mitigate the effects of 16-colour palettes: Why not have anything that might have a 16-colour sprite keep a handle to a second sprite, which is a 256-colour version of the 16-colour sprite? Then, whenever the whatever changes palettes, just create a new sprite with the updated palette. For normal usage (i.e., palette changes happening once at creation and then never again), this would be a great memory vs speed trade off. The downside would be that if someone decides to emulate the Mario starman and cycle palettes rapidly, this would be an equal slowdown. This might be mitigated by using _another_ cache that remembers [sprite, palette] pairs. It would also be possible to take this a step further, and do the same thing at the master palette level, 'compiling' everything into desktop colour depth-bit sprites, but this would make fades and other (very common) master palette changes very messy. ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org -- - Mike Caron ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] SVN: james/4242 Remove Streaming from the sound effects import menu, because it doesn'
On 4/27/2011 12:46, subvers...@hamsterrepublic.com wrote: james 2011-04-27 09:46:05 -0700 (Wed, 27 Apr 2011) 135 Remove Streaming from the sound effects import menu, because it doesn't do anything, and I can't figure out what it was meant to do. --- U wip/menus.bas Odds are, it was meant to allow one to select between Streaming (where the sound is streamed from disk) and Static (where the sound is loaded into memory completely). The former is more appropriate for large sounds (eg, Music) while the latter is more appropriate for short sounds. -- - Mike Caron ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] SVN: james/4242 Remove Streaming from the sound effects import menu, because it doesn'
On 4/27/2011 13:02, James Paige wrote: On Wed, Apr 27, 2011 at 12:52:26PM -0400, Mike Caron wrote: On 4/27/2011 12:46, subvers...@hamsterrepublic.com wrote: james 2011-04-27 09:46:05 -0700 (Wed, 27 Apr 2011) 135 Remove Streaming from the sound effects import menu, because it doesn't do anything, and I can't figure out what it was meant to do. --- U wip/menus.bas Odds are, it was meant to allow one to select between Streaming (where the sound is streamed from disk) and Static (where the sound is loaded into memory completely). The former is more appropriate for large sounds (eg, Music) while the latter is more appropriate for short sounds. I wasn't aware of the backends being able to handle that distinction. Maybe someone was planning to try and add it but never got around to it? Anyway, if we need a streaming option later, it can easily be added back. They probably can't. Fortunately, it's not a big deal these days. After all, what's 10 Megs of Ram between you or me? :P -- - Mike Caron ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] Build system pruning
On 4/25/2011 1:19, James Paige wrote: On Sat, Apr 23, 2011 at 01:22:58PM +1200, Ralph Versteegen wrote: We have 15 build scripts, and certain changes require updating almost all of them. It's time to get rid of some of this ridiculous redundancy. I like the fact that the *.bat scripts and Makefile don't require gcc, python, and scons on top of freebasic, because that's a lot to install. (These use pre-compiled objects at win32/*.o, though that could be added to scons of course) However, it looks like make actually requires mingw or msys anyway, make.exe won't run by itself. Should we get rid of Makefile (once I add Mac support to scons), or the *.sh and *.bat build scripts, or both? I vote for starting by removing most of the *.sh, *.bat scripts, since it's too many files to maintain and test. makegame.bat, makeedit.bat are especially nasty. I am trhilled with scons, and have been using it exclusively for a while. I would be happy to say goodbye to the .bat and .sh scripts. It's probably about time to make mingw a requirement on Windows as the number of C/C++ source files grows, also the script interpreter is written in C. I concur. With two alternatives, makegame.sh, makeedit.sh, makeutil.sh, makereload.sh may be safely deleted. (Though I notice there's some FreeBSD-specific code in there) What's check.bat? That doesn't seem to be used. No need for mid2bam.bat or mid2bam.sh since you hardly ever want to build that: mid2bam.sh hasn't even worked since FB 0.16! Judging by the svn log there is no need to keep check.bat --- James ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org I am probably the last person to offer advice on build scripts, but has anyone looked into using CMake (http://www.cmake.org/)? The deal with CMake is that it's sort of like automake, but it generates build scripts for several different platforms and compilers. For example, it can generate Visual Studio projects, Makefiles for mingw, etc. Personally, I would prefer to have the option to build the C/++ stuff with Visual Studio, since I have never been able to install MinGW for some reason. Plus, Visual Studio Express is free (http://www.microsoft.com/express/Windows/). Of course, I don't know how hard it would be to teach it about FreeBasic, but I imagine it's not that difficult. -- - Mike Caron ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] Zenzizenzic nightly crash - Animation test
On 4/17/2011 4:06, Matthew Boyd wrote: I downloaded the latest nightly on April 17, 2011, and it's been doing okay so far. When I went to check out the new enemy dissolve animations in the enemy editor, I got this error message; it occurs every time I play a dissolve animation, even the old ones. I haven't tested this in an actual battle yet. After backing up my game and pressing random keys to get out of the error message, my palette got screwed up. The error also kept re-appearing on every frame of the animation till it was done. On a whim, I went to the master palette editor - the visit alone seems to have fixed the palette problem. Baconlabs from the forums Could I point out that that's a terrible error message? It took me three reads to figure out what it's asking the user to do. In my opinion, it would be worded better like this: An error has occurred! Press Escape to quit, or any other key to try and continue. If you quit now, when you run Custom again, it will ask if you want to recover your game. Please back up your Game before you do this, as the internal state may be corrupt. If this error is unexpected, please send an email to ohrrpgce-cr...@hamsterrepublic.com. That reads a lot cleaner to me, and has less redundant technical information. -- - Mike Caron ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] SVN: james/4171 compile hspeak using Euphoria4.
On 2/25/2011 17:15, subvers...@hamsterrepublic.com wrote: james 2011-02-25 14:15:32 -0800 (Fri, 25 Feb 2011) 233 compile hspeak using Euphoria4. WARNING: this only works if hspeak.exw is located in a folder that has no spaces. If you try to compile from C:\Documents And Settings\Homer J Simpson\My Documents\src\ohr\wip\ you will be disapointed --- U wip/makehspeak.bat I looked at makehspeak.bat, expecting to find Oh, maybe he just needs a pair of quotes. I was wrong. WTF? Why would euphoria bomb if a file contains spaces in its path? It would be different if you called euc -gcc %CWD%\hspeak.exw Since that might expand to something like euc -gcc c:\ohr source\hspeak.exw which rightfully would be parsed as two arguments. I can't conceive of what insane things they must be doing with their command line processing. Expanding the file path inline? That's so much more work than the alternative! Argh, this broke my brain. -- - Mike Caron ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] Nightly Missing gver.txt
gver.txt (and cver.txt) are generated by verprint.exe, and its sole purpose is to provide the version string in custom's main menu. It doesn't do anything at runtime. If you're compiling the OHR from source, using makegame/edit.bat should compile verprint.exe for you, so that it can generate these files. On 2/22/2011 23:10, Tim Bell wrote: About 30 minutes ago, I downloaded the nightly build zipfile from the wiki (it's currently 11:07 PM EST), and when I went to start compiling what I've changed, I discovered that I'm missing gver.txt (called in compat.bi). While I do have one that I could use as a backup, it's OHRRPGCE wip 20080308 fb/sdl. Since this particular build I'm making is just for personal amusement, will anything be horribly broken if I don't have the official gver.txt? And why was it missing from the nightly in the first place? ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] SVN: james changed revision 25 log comment by james
On 1/20/2011 20:22, subvers...@hamsterrepublic.com wrote: --- tmp/james-25-old2011-01-20 17:22:19.0 -0800 +++ tmp/james-25-new2011-01-20 17:22:19.0 -0800 @@ -1 +1 @@ -Journaled about trip to New York. +James Mangled this commit log message by accident and doesn't know what it used to say :( \ No newline at end of file --- ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org Hmm, 25 was before you set up the emailing thing. However, according to your first log mangling message: -applied Simon Bradley's patch to move assembly declares into allmodex.bi +Journaled about trip to New York. ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] SVN: teeemcee/4101 diriswriteable wasn't really working properly on Windows when a director
On 1/12/2011 11:55, subvers...@hamsterrepublic.com wrote: teeemcee 2011-01-12 08:55:46 -0800 (Wed, 12 Jan 2011) 371 diriswriteable wasn't really working properly on Windows when a directory on writeable media is made read-only, so I added a kludge for that. Doesn't Windows have a proper concept of unwriteable directories? Not really. The best way to check is to try writing to a file, and see if it fails. You could try explicitly checking security, media, etc., but that would be a lot of code and you'd definitely miss stuff. (Also, I'm aware that Game will immediately crash if trying to play a game from read-only media, due to trying to write g_debug.txt. I have a branch for that...) --- U wip/util.bas ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] SVN: teeemcee/4077 Rewrite inefficient looping with halirious comment in guessdefaultpals
On 1/3/2011 12:56, subvers...@hamsterrepublic.com wrote: teeemcee 2011-01-03 09:56:57 -0800 (Mon, 03 Jan 2011) 299 Rewrite inefficient looping with halirious comment in guessdefaultpals Here is the amazing comment: FOR j = 0 TO gen(genMaxHero) 'I reversed the loops, because it's more efficient FOR i = 0 TO sets 'to do the file I/O in the outer loop loadherodata @her, j ... NEXT NEXT Whoops! Judging by the diff between 1188 and 1189 (when that change was made), I did that at the same time I rewrote everything to use the hero UDT instead of arrays. To be fair, the comment /is/ accurate. It simply didn't reflect reality. :) --- U wip/common.bas ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] SVN: teeemcee/4077 Rewrite inefficient looping with halirious comment in guessdefaultpals
On 1/3/2011 14:01, Mike Caron wrote: On 1/3/2011 12:56, subvers...@hamsterrepublic.com wrote: teeemcee 2011-01-03 09:56:57 -0800 (Mon, 03 Jan 2011) 299 Rewrite inefficient looping with halirious comment in guessdefaultpals Here is the amazing comment: FOR j = 0 TO gen(genMaxHero) 'I reversed the loops, because it's more efficient FOR i = 0 TO sets 'to do the file I/O in the outer loop loadherodata @her, j ... NEXT NEXT Whoops! Judging by the diff between 1188 and 1189 (when that change was made), I did that at the same time I rewrote everything to use the hero UDT instead of arrays. To be fair, the comment /is/ accurate. It simply didn't reflect reality. :) And, actually, just looking at the code now, I see the inner loop actually isn't needed at all (as you rewrote) --- U wip/common.bas ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] SVN: teeemcee/4072 Updated DT0 and HeroDef for percentage based elementals.
On 1/1/2011 14:30, Ralph Versteegen wrote: On 2 January 2011 08:25, Mike Caroncaron.m...@gmail.com wrote: On 1/1/2011 14:16, Ralph Versteegen wrote: On 2 January 2011 08:09,subvers...@hamsterrepublic.comwrote: teeemcee 2011-01-01 11:09:07 -0800 (Sat, 01 Jan 2011) 367 Updated DT0 and HeroDef for percentage based elementals. -DT0 is binsize-sized now -Changed return value from percent_grabber -Added SINGLE overloads for percent_grabberpercent_format, and changed their handling of sig. fig.s -Changed Yorgo's resists in test.rpg. Note that interference from the default weapon prevents values higher than 200% on the Status screen The default weapon is a nuisance, but I think I'll fix it by adding a popup warning to the hero elemental resists menu warning the user to change to a different equip merging scheme, once they're available. However, the default weapon would still be quite a nuisance for some merging schemes, like taking the average difference from 100% of all equipment and the native values. Maybe a per-hero bit ignore elemental resistances of default weapon It's possible that I am missing something, but what about the default weapon, exactly, is causing problems? Or, maybe a better question is, how are you figuring out effective resistance? The intuitive method would be to multiply all equipment together. That way, if you're equipped with the paper armour and oil sword, each with a weakness to fire (200% damage), you take 400% damage from fire. Conversely, if you equip the asbestos armour (with 50% fire damage) instead of the paper armour, it would cancel out the oil sword. Immunity would work the same (* 0 is always 0), and so would absorbtion. You missed it, but we have to provide compatibility with older versions. And in older versions, two items with 200% damage from fire (weak bits set) equals a total of 200% damage from fire (weak bit set). I will definitely add multiplication of values as an alternative, but I thinking about instead using sum-of-differences-from-100% as the new default (so 200% and 200% becomes 300%). That way an immunity item plus an absorption item give you absorption. Or perhaps geometric-sum-of-differences, so that 50% and 200% still cancel each other out... but that's difficult to understand. Right, I read that. But, the way you phrased it above, it sounds like the default weapon is somehow interfering with the formula. Either way, good luck with that! --- U wip/bmod.bas U wip/bmodsubs.bas U wip/common.bas U wip/common.bi U wip/const.bi U wip/customsubs.bas U wip/customsubs.bi U wip/loading.bas U wip/menustuf.bas A wip/ohrhelp/hero_elementals.txt U wip/subs.bas U wip/testgame/test.rpg U wip/udts.bi ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] SVN: teeemcee/4052 Add developer menu (type 'spam' at Custom's top level menu), replacing i
Perhaps I am a bit late on this, but... Why not just add it normally, but wrap it with something like #ifdef SHOWDEVMENU and add -D SHOWDEVMENU to your personal build script? That way, it's kept out of the nightly, but present in a test build. Or, if you want it accessible for other people to test (but, only if they know the danger), make it a command line parameter. On 12/29/2010 23:01, subvers...@hamsterrepublic.com wrote: teeemcee 2010-12-29 20:01:12 -0800 (Wed, 29 Dec 2010) 227 Add developer menu (type 'spam' at Custom's top level menu), replacing individual key combinations for different menus (eg. Ctrl+R for the RELOAD editor). My initial passphrase choice was 'spiffy', but that was too much typing --- U wip/custom.bas ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] FB arrays: extend or scrap?
On 11/26/2010 5:34, Ralph Versteegen wrote: QB/FB arrays are the worst designed feature in any programming language I know of (except for *possibly* some of the stuff in MUMPS http://thedailywtf.com/Articles/A_Case_of_the_MUMPS.aspx) (A quick rant: -not 1st class: you can't return them or even place them in types! -uninitialised dynamic arrays are indistinguishable and will crash your program. You must initialise them. -to figure out whether an array is static or dynamic you have to check the dialect, the OPTION statements in effect, and whether the index in the DIM statement is a variable or a constant -unlike the rest of the language, arrays aren't properly statically typed: the compiler will let you REDIM a static array and crash your program -zero-length arrays not supported -almost zero support for arrays in the standard library: just REDIM, LBOUND, UBOUND, ERASE. You can't even copy an array, or get a pointer to the array descriptor. ) A couple days ago I had finally had enough, and started on a library of array utility functions written in C, effectively extending libfb (addressing just the last 2 complaints above). The idea is they act on 1D arrays starting at -1, with the -1 element ignored so we can have zero-length arrays. Here are a few prototypes (you use a macro to declare overloaded versions for each type T you want to support): declare sub array_copy overload (array() as T, source() as T) 'Append an element or an array declare sub array_append overload (array() as T, value as T) declare sub array_append overload (array() as T, appendlist() as T) 'Removes index n from array declare sub array_delete overload (array() as T, byval n as integer) 'Insert a value/array at an index. Negative indices count from end (-1 appends) declare sub array_insert overload (array() as T, byval insertbefore as integer, value as T) declare sub array_insert overload (array() as T, byval insertbefore as integer, insertlist() as T) declare sub array_sort overload (array() as T) A couple nights into implementing them (expected to be done by now, but the libfb array code is horribly over-complex, eg. there are about 8 different implementations of REDIM) I suddenly realised: if FB arrays are so horrible, why am I using them at all? It would be FAR easier to create a new array structure and write an extensive library for it in C. The code would be much simple, faster, there'd be no worry about keeping up with changes to libfb, and I can fix more of the problems with FB arrays. The disadvantages would be having to manually allocate and free them, and no bounds checking when compiled with -exx. However when we finally switch to -lang fb we'll get those back. I would implement these arrays like this: array objects would just be pointers of the desired type, which can be indexed like a normal pointer. The array descriptor (containing length and other stuff) would be stored before the 0th element. These pointers can of course be embedded in UDTs or returned from functions. Because code should be self documenting, to avoid confusion with normal pointers I'd do this: #define array ptr Example code: FUNCTION generate_test_array() as string array DIM arr as string array array_new arr' allocate zero-length array array_append arr, test array_append arr, array_of(foo, bar) ' I think I'll manage this array_sort arr RETURN arr END FUNCTION DIM arr as string array arr = generate_test_array ' Can only do this as arr is a NULL pointer array_set arr, generate_test_array ' Safe way to assign an array print arr[length(arr) - 1] array_free arr So the question to the other developers is: would you rather have a library for FB arrays, or use these new arrays? I am definitely in favour of the new arrays, and would start using them and, where beneficial, converting code I'm working on to use them as soon as possible (tomorrow?). I'm curious about one thing. Would these new arrays be dynamic or statically sized? I assume the former. In which case, you do you handle the realloc()/Reallocate() semantics? Specifically, when you use realloc(), you have to use the return value as the new pointer (if it's not null), since the memory block may have moved to accommodate the new size. However, the syntax you're proposing does not allow for the re-assignment of the variable containing the array. (Unless you intend for some preprocessor trickery, in which case foo on you.) The best way to replace FB's arrays would be to create an array object which overloads the [] operator, and keeps the memory buffer internal. Ideally, it supports templates for maximum efficiency. Class ArrayT length as uinteger data as T ptr Public Constructor Array(size as uinteger) length = size data = new T[size] End Constructor Public Operator[](index as uinteger) as T ref if index 0 or index = length throw new ArgumentException() return
Re: [Ohrrpgce] FB arrays: extend or scrap?
On 11/26/2010 7:49, Ralph Versteegen wrote: On 26 November 2010 23:59, Mike Caroncaron.m...@gmail.com wrote: On 11/26/2010 5:34, Ralph Versteegen wrote: QB/FB arrays are the worst designed feature in any programming language I know of (except for *possibly* some of the stuff in MUMPS http://thedailywtf.com/Articles/A_Case_of_the_MUMPS.aspx) (A quick rant: -not 1st class: you can't return them or even place them in types! -uninitialised dynamic arrays are indistinguishable and will crash your program. You must initialise them. -to figure out whether an array is static or dynamic you have to check the dialect, the OPTION statements in effect, and whether the index in the DIM statement is a variable or a constant -unlike the rest of the language, arrays aren't properly statically typed: the compiler will let you REDIM a static array and crash your program -zero-length arrays not supported -almost zero support for arrays in the standard library: just REDIM, LBOUND, UBOUND, ERASE. You can't even copy an array, or get a pointer to the array descriptor. ) A couple days ago I had finally had enough, and started on a library of array utility functions written in C, effectively extending libfb (addressing just the last 2 complaints above). The idea is they act on 1D arrays starting at -1, with the -1 element ignored so we can have zero-length arrays. Here are a few prototypes (you use a macro to declare overloaded versions for each type T you want to support): declare sub array_copy overload (array() as T, source() as T) 'Append an element or an array declare sub array_append overload (array() as T, value as T) declare sub array_append overload (array() as T, appendlist() as T) 'Removes index n from array declare sub array_delete overload (array() as T, byval n as integer) 'Insert a value/array at an index. Negative indices count from end (-1 appends) declare sub array_insert overload (array() as T, byval insertbefore as integer, value as T) declare sub array_insert overload (array() as T, byval insertbefore as integer, insertlist() as T) declare sub array_sort overload (array() as T) A couple nights into implementing them (expected to be done by now, but the libfb array code is horribly over-complex, eg. there are about 8 different implementations of REDIM) I suddenly realised: if FB arrays are so horrible, why am I using them at all? It would be FAR easier to create a new array structure and write an extensive library for it in C. The code would be much simple, faster, there'd be no worry about keeping up with changes to libfb, and I can fix more of the problems with FB arrays. The disadvantages would be having to manually allocate and free them, and no bounds checking when compiled with -exx. However when we finally switch to -lang fb we'll get those back. I would implement these arrays like this: array objects would just be pointers of the desired type, which can be indexed like a normal pointer. The array descriptor (containing length and other stuff) would be stored before the 0th element. These pointers can of course be embedded in UDTs or returned from functions. Because code should be self documenting, to avoid confusion with normal pointers I'd do this: #define array ptr Example code: FUNCTION generate_test_array() as string array DIM arr as string array array_new arr' allocate zero-length array array_append arr, test array_append arr, array_of(foo, bar) ' I think I'll manage this array_sort arr RETURN arr END FUNCTION DIM arr as string array arr = generate_test_array ' Can only do this as arr is a NULL pointer array_set arr, generate_test_array ' Safe way to assign an array print arr[length(arr) - 1] array_free arr So the question to the other developers is: would you rather have a library for FB arrays, or use these new arrays? I am definitely in favour of the new arrays, and would start using them and, where beneficial, converting code I'm working on to use them as soon as possible (tomorrow?). I'm curious about one thing. Would these new arrays be dynamic or statically sized? I assume the former. In which case, you do you handle the realloc()/Reallocate() semantics? I forgot to mention? Yeah, always dynamic. Specifically, when you use realloc(), you have to use the return value as the new pointer (if it's not null), since the memory block may have moved to accommodate the new size. However, the syntax you're proposing does not allow for the re-assignment of the variable containing the array. (Unless you intend for some preprocessor trickery, in which case foo on you.) BYREF BYREF BYREF Well, *expletive*. I guess I've been using other languages too long. Don't worry, there's preprocessor trickery involved. However there's less than required for extending FB arrays. Also, I am aware that my attempts at inheritance, hash tables, and lists using macros turned into a disgusting mess; I promise I won't do that again. In fact I
Re: [Ohrrpgce] plotscript dictionary down
On 11/20/2010 13:35, Jeremy Bursey wrote: Just tried to use the nightly plotscript dictionary and I got this error: The XML page cannot be displayed Cannot view XML input using XSL style sheet. Please correct the error and then click the Refresh javascript:location.reload() button, or try again later. End tag 'li' does not match the start tag 'ul'. Error processing resource 'http://hamsterrepublic.com/ohrrpgce/docs/plotdic... likey:Left Win Logo # equal to key:Left Command/li/li ... All fixed. It was simply an extra /li tag... ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] plotscript dictionary down
On 11/20/2010 20:18, Jeremy Bursey wrote: /All fixed. It was simply an extra /li tag.../ You may want to double-check that fix because the error still shows up. I closed out of IE8 just to make sure, and it still showed an error. Did you update your working copy? Or, if you're trying to test it in a nightly build, you need to wait until tomorrow, since I only committed it this afternoon. ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] Grievances with gfx_directx?
On 11/11/2010 15:06, Jay Tennant wrote:
From: Ralph Versteegenteeem...@gmail.com
I'd like the hotswapping-of-displays crashes to be
resolved.
This seems to be impossible to detect. I've searched all over
the docs and the internet, but I only found that Windows
does not send notifications about monitor connection state. It
seems to know about it, but will not raise an event for the
rest of the system.
http://social.msdn.microsoft.com/Forums/en/etw/thread/01373322-30ea-4abc-b7ca-0e698febe39d
Maybe by checking for connected monitors once every 3 seconds?
I've done a bit of playing around, and discovered that it is possible to
get monitor notifications.
WARNING: WIN32 CODE AHEAD, LINUX USERS AVERT YOUR EYES
#include windows.h
#include dbt.h
//this is normally defined in the Driver Development Kit, which I don't have
const GUID GUID_MONITORS = { 0xe6f07b5f, 0xee97, 0x4a90, {0xb0, 0x76,
0x33, 0xF5, 0x7B, 0xF4, 0xEA, 0xA7} };
void RegisterNotification(HWND hWnd) {
DEV_BROADCAST_DEVICEINTERFACE flt;
ZeroMemory(flt, sizeof(flt));
flt.dbcc_size = sizeof(flt);
flt.dbcc_devicetype = DBT_DEVTYP_DEVICEINTERFACE;
flt.dbcc_classguid = GUID_MONITORS;
RegisterDeviceNotification(hWnd, flt, DEVICE_NOTIFY_WINDOW_HANDLE);
}
LRESULT CALLBACK WndProc(HWND hWnd, UINT message, WPARAM wParam, LPARAM
lParam)
{
switch (message) {
//other messages, etc
case WM_CREATE:
RegisterNotification(hWnd);
case WM_DEVICECHANGE:
{
PDEV_BROADCAST_DEVICEINTERFACE b =
(PDEV_BROADCAST_DEVICEINTERFACE) lParam;
switch(wParam) {
case DBT_DEVICEARRIVAL:
LogMsg(Device arrival: %S, b-dbcc_name);
break;
case DBT_DEVICEREMOVECOMPLETE:
LogMsg(Device removal: %S, b-dbcc_name);
break;
}
return DefWindowProc(hWnd, message, wParam, lParam);
}
}
}
END OF WIN32 CODE
With this code (and, obviously, the glue I omitted), I get this when
unplugging and replugging my monitor:
Device removal:
\\?\DISPLAY#HSD0019#5b07f9880UID1048833#{e6f07b5f-ee97-4a90-b076-33f57bf4eaa7}
Device arrival:
\\?\DISPLAY#HSD0019#5b07f9880UID1048833#{e6f07b5f-ee97-4a90-b076-33f57bf4eaa7}
Obviously, the name is irrelevant, but this is a good place to enumerate
displays, etc.
___
Ohrrpgce mailing list
ohrrpgce@lists.motherhamster.org
http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] SVN: james/3886 a work-in-progress editor definition file
On 10/14/2010 0:07, James Paige wrote: On Wed, Oct 13, 2010 at 08:27:34PM -0700, Jay Tennant wrote: From: James Paigeb...@hamsterrepublic.com Sent: Wednesday, October 13, 2010 9:36 PM On Wed, Oct 13, 2010 at 07:23:23PM -0700, subvers...@hamsterrepublic.com wrote: james 2010-10-13 19:23:23 -0700 (Wed, 13 Oct 2010) 42 a work-in-progress editor definition file I am pretty happy with the RELOAD Editor. Has anybody else tried it yet? (CTRL+R from the main menu in custom) However, I am seriously considering making an actual editor-editor-editor. It would be similar to the RELOAD editor except it would strictly enforce the structure of a .editor file, and would automate the creation of each widget type. Forgive my nescience, but why do you need or want an editor-editor-editor? Isn't that like a pointer to a pointer to a pointer? Thank you for making me google nescience. I am now that much less nescient. I know the idea of an editor-editor-editor *sounds* crazy, especially since I have not actually started the editor-editor yet, but in the course of planning and preparing the editor-editor, I have come to believe that an editor-editor-editor might be exactly the right amount of indirection to trigger a cascade of awesomeness. Although maybe I am confusaing things by calling it the editor-editor-editor. In my mind I have been calling the interpreter for the editor definition files the editor-editor so a tool to create those files is the editor-editor-editor, but really I should drop a level of the indirection in my *name* of this thing. just simply editor-editor is a better description of what I have been talking about when I say editor-editor-editor I was thinking about the RELOAD format today, and I realized that it could be made more efficient, if we produced a manifest of sorts, describing the valid contents of a RELOAD document. We could use it to optimize and validate certain operations. I think we could use JSON to describe it... Actually, no. It would be more efficient if we created a new binary format for this. Maybe... Markup Extra Transcription of Additions. Yes, that should work. And, of course, you can add a new editor-editor-editor editor to edit it. Oh, but how are we going to mark that up? Hmm... Editor Language DRaft Information Transcription CHannel, edited in the editor-editor-editor-editor editor of course! And yes, the idea of using the editor-editor to create the editor-editor-editor has occured to me, but then an 80-year-old version of Ralph with a long silver beard and robotic arms appeared in a time machine and beat me senseless before disappearing in a flash of light. ...S I decided not to attempt that in this particular timeline. Hmm. Have you tried installing magnets inside the walls? I have nagging relatives, and to avoid them I've installed neodymium magnets all over the house. Because they have metal fillings, their faces are naturally drawn to the walls when they pass by them. Of course, the side effects include rainbow colored crt's, the wall gnomes have become magnetized (or at least that's their excuse for taking all my change!), and I occasionally wake up pointing northwards. I would try that, but I am concerned about how it might interact with my tinfoil hat. Don't listen to this hippie crap. All you need is a spell of warding against extra-planar travellers. Man, doesn't anyone take levels in sorcerer these days? --- James ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] SVN: james/3886 a work-in-progress editor definition file
On 10/14/2010 22:13, James Paige wrote: On Thu, Oct 14, 2010 at 10:07:03PM -0400, Mike Caron wrote: On 10/14/2010 0:07, James Paige wrote: On Wed, Oct 13, 2010 at 08:27:34PM -0700, Jay Tennant wrote: From: James Paigeb...@hamsterrepublic.com Sent: Wednesday, October 13, 2010 9:36 PM On Wed, Oct 13, 2010 at 07:23:23PM -0700, subvers...@hamsterrepublic.com wrote: james 2010-10-13 19:23:23 -0700 (Wed, 13 Oct 2010) 42 a work-in-progress editor definition file I am pretty happy with the RELOAD Editor. Has anybody else tried it yet? (CTRL+R from the main menu in custom) However, I am seriously considering making an actual editor-editor-editor. It would be similar to the RELOAD editor except it would strictly enforce the structure of a .editor file, and would automate the creation of each widget type. Forgive my nescience, but why do you need or want an editor-editor-editor? Isn't that like a pointer to a pointer to a pointer? Thank you for making me google nescience. I am now that much less nescient. I know the idea of an editor-editor-editor *sounds* crazy, especially since I have not actually started the editor-editor yet, but in the course of planning and preparing the editor-editor, I have come to believe that an editor-editor-editor might be exactly the right amount of indirection to trigger a cascade of awesomeness. Although maybe I am confusaing things by calling it the editor-editor-editor. In my mind I have been calling the interpreter for the editor definition files the editor-editor so a tool to create those files is the editor-editor-editor, but really I should drop a level of the indirection in my *name* of this thing. just simply editor-editor is a better description of what I have been talking about when I say editor-editor-editor I was thinking about the RELOAD format today, and I realized that it could be made more efficient, if we produced a manifest of sorts, describing the valid contents of a RELOAD document. We could use it to optimize and validate certain operations. I think we could use JSON to describe it... Actually, no. It would be more efficient if we created a new binary format for this. Maybe... Markup Extra Transcription of Additions. Yes, that should work. And, of course, you can add a new editor-editor-editor editor to edit it. Oh, but how are we going to mark that up? Hmm... Editor Language DRaft Information Transcription CHannel, edited in the editor-editor-editor-editor editor of course! I... can't... tel if you are teasing me or not!... I'm teasing both of us :D And yes, the idea of using the editor-editor to create the editor-editor-editor has occured to me, but then an 80-year-old version of Ralph with a long silver beard and robotic arms appeared in a time machine and beat me senseless before disappearing in a flash of light. ...S I decided not to attempt that in this particular timeline. Hmm. Have you tried installing magnets inside the walls? I have nagging relatives, and to avoid them I've installed neodymium magnets all over the house. Because they have metal fillings, their faces are naturally drawn to the walls when they pass by them. Of course, the side effects include rainbow colored crt's, the wall gnomes have become magnetized (or at least that's their excuse for taking all my change!), and I occasionally wake up pointing northwards. I would try that, but I am concerned about how it might interact with my tinfoil hat. Don't listen to this hippie crap. All you need is a spell of warding against extra-planar travellers. Man, doesn't anyone take levels in sorcerer these days? I'm still single-classing Barbarian, sorry. Fair enough. You need it to wrangle with all those Pythons. --- James ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] SVN: james/3886 a work-in-progress editor definition file
On 10/14/2010 22:23, James Paige wrote: On Thu, Oct 14, 2010 at 10:15:30PM -0400, Mike Caron wrote: On 10/14/2010 22:13, James Paige wrote: On Thu, Oct 14, 2010 at 10:07:03PM -0400, Mike Caron wrote: On 10/14/2010 0:07, James Paige wrote: On Wed, Oct 13, 2010 at 08:27:34PM -0700, Jay Tennant wrote: From: James Paigeb...@hamsterrepublic.com Sent: Wednesday, October 13, 2010 9:36 PM On Wed, Oct 13, 2010 at 07:23:23PM -0700, subvers...@hamsterrepublic.com wrote: james 2010-10-13 19:23:23 -0700 (Wed, 13 Oct 2010) 42 a work-in-progress editor definition file I am pretty happy with the RELOAD Editor. Has anybody else tried it yet? (CTRL+R from the main menu in custom) However, I am seriously considering making an actual editor-editor-editor. It would be similar to the RELOAD editor except it would strictly enforce the structure of a .editor file, and would automate the creation of each widget type. Forgive my nescience, but why do you need or want an editor-editor-editor? Isn't that like a pointer to a pointer to a pointer? Thank you for making me google nescience. I am now that much less nescient. I know the idea of an editor-editor-editor *sounds* crazy, especially since I have not actually started the editor-editor yet, but in the course of planning and preparing the editor-editor, I have come to believe that an editor-editor-editor might be exactly the right amount of indirection to trigger a cascade of awesomeness. Although maybe I am confusaing things by calling it the editor-editor-editor. In my mind I have been calling the interpreter for the editor definition files the editor-editor so a tool to create those files is the editor-editor-editor, but really I should drop a level of the indirection in my *name* of this thing. just simply editor-editor is a better description of what I have been talking about when I say editor-editor-editor I was thinking about the RELOAD format today, and I realized that it could be made more efficient, if we produced a manifest of sorts, describing the valid contents of a RELOAD document. We could use it to optimize and validate certain operations. I think we could use JSON to describe it... Actually, no. It would be more efficient if we created a new binary format for this. Maybe... Markup Extra Transcription of Additions. Yes, that should work. And, of course, you can add a new editor-editor-editor editor to edit it. Oh, but how are we going to mark that up? Hmm... Editor Language DRaft Information Transcription CHannel, edited in the editor-editor-editor-editor editor of course! I... can't... tel if you are teasing me or not!... I'm teasing both of us :D And the fact that I had to ask means you did it well :) The sad part is that it only took me a few moments to come up with the acronyms... Maybe my destiny is to make meta-file formats?! Nooo --- James ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] better palette options
On 10/10/2010 7:04, Ralph Versteegen wrote: On 9 October 2010 14:39, David Gowers (kampu)00a...@gmail.com wrote: On Sat, Oct 9, 2010 at 5:07 AM, Ralph Versteegenteeem...@gmail.com wrote: On 9 October 2010 02:48, David Gowers (kampu)00a...@gmail.com wrote: On Fri, Oct 8, 2010 at 11:42 PM, David Gowers (kampu)00a...@gmail.com wrote: http://bpaste.net/show/10001/ ^ The format proposition Generally this looks good. Aside from the following long string of criticisms ;) I'm not sure, but I believe you're suggesting using a plain binary lump instead of Reload for animation? If Reload is actually part of it, I don't see much point in a separate lump per animation. RELOAD was not part of it (well, honestly RELOAD vaguely annoys me, the way it replicates XML's weird data model. So it's not the first thing I would think of.) What about RELOAD (or, I guess, XML's) data model don't you like? I'm not trying to defend it, I honestly want to know what you don't like. Perhaps something could be improved? But it is now. I wasn't actually requesting it. RELOAD as container for animation data, as in your proposal, seems like a fairly good idea. RELOAD as container for pixel data may be overly heavy. I agree that pixel data should not be stored in a RELOAD container. Rather, all the animation data should be stored in a single RELOAD document, and the images separate from that. RELOAD annoys me a little too, due to much increased complexity of read/writing. But, surely, you agree that the benefits of having a strongly typed and named data format outweigh the complexity, right? Honestly, I think the lump resizing scheme for older lumps is a lot more complex than RELOAD... I notice that separate image size and bounding box size/offset are missing. Intentionally? Missing? I understood the base bounding box size and location as being immutable (and based off say, a manually definable box - see slice editor). While this was a bit annoying, it made some amount of sense. I'm not sure how to avoid confusing the user if they are not immutable. Please, explain further how you conceive of the interaction of bounding boxes and image offsets? It's a conflation of two (or three) different ideas. Perhaps they should be separated. A srcrect is just a reference point (offset) plus width and height equal to the existing frame size limits. I called these bounding boxes because in the current tile-based movement system (and also somewhat in in-battle movement and positioning) the frame size is effectively a bounding box. But if/when we add pixel-based movement one may well want the bounding box to depend on the current frame. I think that's a really bad idea. The point of bounding boxes is that they are independent of the graphic on screen. Case in point: What if, on a frame with a small bounding box, a character moves up against a wall. Then, on the next frame, he has a large bounding box, which now intersects with the wall. Which of the following happens? 1. The character can now move into the wall? 2. The character is ejected (a la super mario bros)? 3. The character can only move out from the wall? 4. The character can't move at all? 5. The character is disintegrated since two particles cannot share the same quantum state? If the bounding boxes are fixed (or, at least, independent of sprites), I don't understand the difference between the 'animset' command, and chaining to another animation using 'end'. animset != anim. it doesn't set the anim, it sets the animset. Anyway I removed that by allowing varlength tuples. What varlength tuples? You're also missing animation names. I think naming animations is better than numbering them. Then you can add a attackee flinch animation to an enemy, say, just by adding an animation and naming it hurt. +1. I was thinking about this problem and couldn't figure out how to do it nicely, but this is excellent. But there's a complication, referring to animations by name may not always be appropriate. Quoting from the 'text tags' thread, where I proposed embedding tags in textboxes which trigger portrait animations: Maybe animations should get both a name and a number. The name can be used everywhere that it makes sense to use a name, and the number can be used in whatever small number of places a name isn't practical (like here) OK, sounds practical. But I wonder whether the ID number should be an index into the big global table of animation names (shared by all spritesets of any sort), or refer to the n-th animation currently defined for this spriteset. I'd prefer the first, since it might make sense to sort animations, say alphabetically. The check/setctr codes seem to be very inefficient for creating loops: setctr ctr0 = x label 1 ...loop contents setctr ctr0 -= 1 checkctr ctr0 == 0 goto label 2 goto label 1 label 2 Those last 5 instructions could be replaced with a loop(ctr,label) code: decrement ctr and goto label if
[Ohrrpgce] Sprites, animation food for thought
First off, a disclaimer: This is not an actual proposal for a data format for the OHR. This is something I've cooked up for another project which may or may not ever see the light of day (hopefully, it will, but it's immaterial to this discussion). This is how I've implemented sprites and animations in this other project: First, all pixel data is stored in sprite sheets. These are PNGs that contain all the sprites for a given on-screen entity. Technically, as I am writing it in .NET, nothing is stopping them from being BMP or JPG or something, but I prefer PNG for its true-colour and transparency support. Then, with each sprite sheet, there is an XML document that describes it: spriteset xmlns=http://mike-caron.com/za; name=link1 file=link.png frame x=33 y=1 width=16 height=22 ox=8 oy=19 name=stand-down / frame x=3 y=31 width=16 height=22 ox=8 oy=19 name=walk1-down / frame x=33 y=31 width=16 height=22 ox=8 oy=19 name=walk2-down / !-- etc. -- Each frame has a name, a position and size on the sheet, and an origin. The origin is aligned up with the actual entity's origin (which, ATM is always their (x,y) position, but that may be customizable later on) Then, after the frame definitions, we have the animation definitions: animation name=stand deftime=1 loop=no dir dir=down framestand-down/frame /dir dir dir=up framestand-up/frame /dir dir dir=right framestand-right/frame /dir dir dir=left framestand-left/frame /dir /animation This is mostly self explanitory. Every frame has either one or four directions defined, depending on if the sprite has meaningful directions (an item on the ground, for example, probably only has one direction. An NPC, on the other hand, will have all four). Each frame in a direction can have a time parameter, which describes how many ticks the frame is displayed for. deftime is just the default such value. Obviously, as well, if loop is yes, the animation will start from the beginning. If no, it will just stop on the last frame forever. Finally, every entity in the game references both a sprite sheet and an animation. The idea is that you could change one or the other. For example, if you get a better shield, the sprite sheet changes from link1 to link2. And, obviously, the animation changes based on what is happening in the game. A few notes on this structure: - There is no accounting for palettes. If you wanted to use a paletted sprite, then by all means you are able, but don't expect to be able to change that palette in game. The canonical way to add palette swaps is to have a different sprite sheet with the same animations (and, I may add some way to automate this. Eg, this sprite sheet is exactly the same as this other sprite sheet, so use those definitions) - There is also no way to get any fancier animations. I.e, there's no way to automatically oscillate or go backwards or whatever. The way to achieve these effects would be to duplicate the frames in the animation. - I may or may not add sprite flipping. - The way the images are laid out on the sheet has absolutely nothing to do with how they are animated or displayed. If you wanted the same frame, but offset by 5 pixels or whatever, you just modify the origin. No muss, no fuss. - Also, there are no inherent limitations on how large or small the sprite is. If you want to make a 2d version of Shadow of the Colossus, then be my guest! (just, don't expect the video card to necessarily play along ;) So... yeah. Make of that what you will, I'm just adding ideas to the pot. I am aware that this format, as is, is unsuitable for the OHR. It's just my two cents. ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
[Ohrrpgce] Sprites, animation food for thought
First off, a disclaimer: This is not an actual proposal for a data format for the OHR. This is something I've cooked up for another project which may or may not ever see the light of day (hopefully, it will, but it's immaterial to this discussion). This is how I've implemented sprites and animations in this other project: First, all pixel data is stored in sprite sheets. These are PNGs that contain all the sprites for a given on-screen entity. Technically, as I am writing it in .NET, nothing is stopping them from being BMP or JPG or something, but I prefer PNG for its true-colour and transparency support. Then, with each sprite sheet, there is an XML document that describes it: spriteset xmlns=http://mike-caron.com/za; name=link1 file=link.png frame x=33 y=1 width=16 height=22 ox=8 oy=19 name=stand-down / frame x=3 y=31 width=16 height=22 ox=8 oy=19 name=walk1-down / frame x=33 y=31 width=16 height=22 ox=8 oy=19 name=walk2-down / !-- etc. -- Each frame has a name, a position and size on the sheet, and an origin. The origin is aligned up with the actual entity's origin (which, ATM is always their (x,y) position, but that may be customizable later on) Then, after the frame definitions, we have the animation definitions: animation name=stand deftime=1 loop=no dir dir=down framestand-down/frame /dir dir dir=up framestand-up/frame /dir dir dir=right framestand-right/frame /dir dir dir=left framestand-left/frame /dir /animation This is mostly self explanitory. Every frame has either one or four directions defined, depending on if the sprite has meaningful directions (an item on the ground, for example, probably only has one direction. An NPC, on the other hand, will have all four). Each frame in a direction can have a time parameter, which describes how many ticks the frame is displayed for. deftime is just the default such value. Obviously, as well, if loop is yes, the animation will start from the beginning. If no, it will just stop on the last frame forever. Finally, every entity in the game references both a sprite sheet and an animation. The idea is that you could change one or the other. For example, if you get a better shield, the sprite sheet changes from link1 to link2. And, obviously, the animation changes based on what is happening in the game. A few notes on this structure: - There is no accounting for palettes. If you wanted to use a paletted sprite, then by all means you are able, but don't expect to be able to change that palette in game. The canonical way to add palette swaps is to have a different sprite sheet with the same animations (and, I may add some way to automate this. Eg, this sprite sheet is exactly the same as this other sprite sheet, so use those definitions) - There is also no way to get any fancier animations. I.e, there's no way to automatically oscillate or go backwards or whatever. The way to achieve these effects would be to duplicate the frames in the animation. - I may or may not add sprite flipping. - The way the images are laid out on the sheet has absolutely nothing to do with how they are animated or displayed. If you wanted the same frame, but offset by 5 pixels or whatever, you just modify the origin. No muss, no fuss. - Also, there are no inherent limitations on how large or small the sprite is. If you want to make a 2d version of Shadow of the Colossus, then be my guest! (just, don't expect the video card to necessarily play along ;) So... yeah. Make of that what you will, I'm just adding ideas to the pot. I am aware that this format, as is, is unsuitable for the OHR. It's just my two cents. ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
[Ohrrpgce] Sprites, animation food for thought
First off, a disclaimer: This is not an actual proposal for a data format for the OHR. This is something I've cooked up for another project which may or may not ever see the light of day (hopefully, it will, but it's immaterial to this discussion). This is how I've implemented sprites and animations in this other project: First, all pixel data is stored in sprite sheets. These are PNGs that contain all the sprites for a given on-screen entity. Technically, as I am writing it in .NET, nothing is stopping them from being BMP or JPG or something, but I prefer PNG for its true-colour and transparency support. Then, with each sprite sheet, there is an XML document that describes it: spriteset xmlns=http://mike-caron.com/za; name=link1 file=link.png frame x=33 y=1 width=16 height=22 ox=8 oy=19 name=stand-down / frame x=3 y=31 width=16 height=22 ox=8 oy=19 name=walk1-down / frame x=33 y=31 width=16 height=22 ox=8 oy=19 name=walk2-down / !-- etc. -- Each frame has a name, a position and size on the sheet, and an origin. The origin is aligned up with the actual entity's origin (which, ATM is always their (x,y) position, but that may be customizable later on) Then, after the frame definitions, we have the animation definitions: animation name=stand deftime=1 loop=no dir dir=down framestand-down/frame /dir dir dir=up framestand-up/frame /dir dir dir=right framestand-right/frame /dir dir dir=left framestand-left/frame /dir /animation This is mostly self explanitory. Every frame has either one or four directions defined, depending on if the sprite has meaningful directions (an item on the ground, for example, probably only has one direction. An NPC, on the other hand, will have all four). Each frame in a direction can have a time parameter, which describes how many ticks the frame is displayed for. deftime is just the default such value. Obviously, as well, if loop is yes, the animation will start from the beginning. If no, it will just stop on the last frame forever. Finally, every entity in the game references both a sprite sheet and an animation. The idea is that you could change one or the other. For example, if you get a better shield, the sprite sheet changes from link1 to link2. And, obviously, the animation changes based on what is happening in the game. A few notes on this structure: - There is no accounting for palettes. If you wanted to use a paletted sprite, then by all means you are able, but don't expect to be able to change that palette in game. The canonical way to add palette swaps is to have a different sprite sheet with the same animations (and, I may add some way to automate this. Eg, this sprite sheet is exactly the same as this other sprite sheet, so use those definitions) - There is also no way to get any fancier animations. I.e, there's no way to automatically oscillate or go backwards or whatever. The way to achieve these effects would be to duplicate the frames in the animation. - I may or may not add sprite flipping. - The way the images are laid out on the sheet has absolutely nothing to do with how they are animated or displayed. If you wanted the same frame, but offset by 5 pixels or whatever, you just modify the origin. No muss, no fuss. - Also, there are no inherent limitations on how large or small the sprite is. If you want to make a 2d version of Shadow of the Colossus, then be my guest! (just, don't expect the video card to necessarily play along ;) So... yeah. Make of that what you will, I'm just adding ideas to the pot. I am aware that this format, as is, is unsuitable for the OHR. It's just my two cents. ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] better palette options
On Sun, Oct 10, 2010 at 6:20 PM, David Gowers (kampu) 00a...@gmail.com wrote: On Mon, Oct 11, 2010 at 3:36 AM, Mike Caron caron.m...@gmail.com wrote: What about RELOAD (or, I guess, XML's) data model don't you like? That nodes with children can also have values. That's a pretty extraordinary behaviour for a tree data model, IMO. Unless you expect people to XSLT their XML source data, I think a model like YAML or JSON is much simpler while still supporting the basic data model. (you can currently convert YAML or JSON to RELOAD with a bit of work. I'm just arguing that the nodes with children can also have values is a specific, and probably unneeded, accommodation for XML's strange behaviour.) Oh. I see. Originally, it was such that nodes could have either children or a value. However, TMC suggested we do it this way as an optimization. I'm not trying to defend it, I honestly want to know what you don't like. Perhaps something could be improved? But it is now. I wasn't actually requesting it. RELOAD as container for animation data, as in your proposal, seems like a fairly good idea. RELOAD as container for pixel data may be overly heavy. I agree that pixel data should not be stored in a RELOAD container. Rather, all the animation data should be stored in a single RELOAD document, and the images separate from that. RELOAD annoys me a little too, due to much increased complexity of read/writing. But, surely, you agree that the benefits of having a strongly typed and named data format outweigh the complexity, right? Honestly, I think the lump resizing scheme for older lumps is a lot more complex than RELOAD... I think that's a really bad idea. The point of bounding boxes is that they are independent of the graphic on screen. Case in point: What if, on a frame with a small bounding box, a character moves up against a wall. Then, on the next frame, he has a large bounding box, which now intersects with the wall. Which of the following happens? 1. The character can now move into the wall? 2. The character is ejected (a la super mario bros)? 3. The character can only move out from the wall? 4. The character can't move at all? 5. The character is disintegrated since two particles cannot share the same quantum state? If the bounding boxes are fixed (or, at least, independent of sprites), ..?? Hello sentence which is not there. :\ I have no idea how that happened, but I think the rest of the sentence was something like: If the bounding boxes are fixed (or, at least, independent of sprites), then you don't have to worry about this. I don't understand the difference between the 'animset' command, and chaining to another animation using 'end'. animset != anim. it doesn't set the anim, it sets the animset. Anyway I removed that by allowing varlength tuples. What varlength tuples? You're also missing animation names. I think naming animations is better than numbering them. Then you can add a attackee flinch animation to an enemy, say, just by adding an animation and naming it hurt. +1. I was thinking about this problem and couldn't figure out how to do it nicely, but this is excellent. But there's a complication, referring to animations by name may not always be appropriate. Quoting from the 'text tags' thread, where I proposed embedding tags in textboxes which trigger portrait animations: Maybe animations should get both a name and a number. The name can be used everywhere that it makes sense to use a name, and the number can be used in whatever small number of places a name isn't practical (like here) OK, sounds practical. But I wonder whether the ID number should be an index into the big global table of animation names (shared by all spritesets of any sort), or refer to the n-th animation currently defined for this spriteset. I'd prefer the first, since it might make sense to sort animations, say alphabetically. The check/setctr codes seem to be very inefficient for creating loops: setctr ctr0 = x label 1 ...loop contents setctr ctr0 -= 1 checkctr ctr0 == 0 goto label 2 goto label 1 label 2 Those last 5 instructions could be replaced with a loop(ctr,label) code: decrement ctr and goto label if ctr 0. A relative frame change code would also be very useful for loops. oh right, ctr and label can fit into 16 bits easily(3+8) Good idea. (note that I'm leaving in the original ops -- they have a specific use that cannot be readily attained by a loop command : interleaving instructions, like 'wait 1' that only execute eg on= second loop through a particular area of code. Your definition of loop isn't correct. The actual encoding of the bytestring, bits per opcode and argument, isn't specified. Making them all 16 bit signed is easy, but we if want to cram multiple arguments into the same short, maybe it would just be simpler to same that opcodes are 8 bit, and arguments are either 8 or 16 bit
[Ohrrpgce] Fwd: Sprites, animation food for thought
Yikes, I fail at email forever. This was meant to go to the rest of the list. -- Forwarded message -- From: Mike caron.m...@gmail.com Date: Sun, Oct 10, 2010 at 11:35 PM Subject: Re: [Ohrrpgce] Sprites, animation food for thought To: Seth Hetu seth.h...@gmail.com (Oops, I figured it would. Thunderbird claimed the email didn't send the first two times) Anyway, yes. As I mentioned below the relevant snippet, each frame in an animation can have its own delay. deftime simply specified the default delay for that animation as a whole. -- Mike Caron -Original Message- From: Seth Hetu seth.h...@gmail.com Date: Mon, 11 Oct 2010 11:32:41 To: ohrrpgce@lists.motherhamster.org; caron.m...@gmail.com Subject: Re: [Ohrrpgce] Sprites, animation food for thought 3 times, but nevermind that. I've got a question about animations. Is it possible to add a delay value, which delays the next frame in the animation? Like: animation name=stand deftime=1 loop=no dir dir=down framestand-down/frame /dir dir dir=up delay=3ms framestand-up/frame /dir dir dir=right framestand-right/frame /dir dir dir=left framestand-left/frame /dir /animation Then the person remains standing for 3s longer than the other directions? The reason I ask is that I like your idea of defining the animation by wrapping a PNG sheet with XML. So maybe it would be possible/easy to add more tags and options to an animation to increase its power. Then, it's all described in the XML+PNG combo, which makes it really easy to edit external to the OHR. Just throwing out ideas, since your first email seemed to encourage it. Cool system overall. --Seth On Mon, Oct 11, 2010 at 2:01 AM, Mike caron.m...@gmail.com wrote: Shoot, if this email went out multiple times, I apologise. If not, please disregard this. -- Mike Caron -Original Message- From: Mike Caron caron.m...@gmail.com Date: Sun, 10 Oct 2010 13:37:53 To: ohrrpgce@lists.motherhamster.org Subject: Sprites, animation food for thought First off, a disclaimer: This is not an actual proposal for a data format for the OHR. This is something I've cooked up for another project which may or may not ever see the light of day (hopefully, it will, but it's immaterial to this discussion). This is how I've implemented sprites and animations in this other project: First, all pixel data is stored in sprite sheets. These are PNGs that contain all the sprites for a given on-screen entity. Technically, as I am writing it in .NET, nothing is stopping them from being BMP or JPG or something, but I prefer PNG for its true-colour and transparency support. Then, with each sprite sheet, there is an XML document that describes it: spriteset xmlns=http://mike-caron.com/za; name=link1 file=link.png frame x=33 y=1 width=16 height=22 ox=8 oy=19 name=stand-down / frame x=3 y=31 width=16 height=22 ox=8 oy=19 name=walk1-down / frame x=33 y=31 width=16 height=22 ox=8 oy=19 name=walk2-down / !-- etc. -- Each frame has a name, a position and size on the sheet, and an origin. The origin is aligned up with the actual entity's origin (which, ATM is always their (x,y) position, but that may be customizable later on) Then, after the frame definitions, we have the animation definitions: This is mostly self explanitory. Every frame has either one or four directions defined, depending on if the sprite has meaningful directions (an item on the ground, for example, probably only has one direction. An NPC, on the other hand, will have all four). Each frame in a direction can have a time parameter, which describes how many ticks the frame is displayed for. deftime is just the default such value. Obviously, as well, if loop is yes, the animation will start from the beginning. If no, it will just stop on the last frame forever. Finally, every entity in the game references both a sprite sheet and an animation. The idea is that you could change one or the other. For example, if you get a better shield, the sprite sheet changes from link1 to link2. And, obviously, the animation changes based on what is happening in the game. A few notes on this structure: - There is no accounting for palettes. If you wanted to use a paletted sprite, then by all means you are able, but don't expect to be able to change that palette in game. The canonical way to add palette swaps is to have a different sprite sheet with the same animations (and, I may add some way to automate this. Eg, this sprite sheet is exactly the same as this other sprite sheet, so use those definitions) - There is also no way to get any fancier animations. I.e, there's no way to automatically oscillate or go backwards or whatever. The way to achieve these effects would be to duplicate the frames in the animation. - I may or may not add sprite
Re: [Ohrrpgce] SVN: teeemcee/3847 Windows builds were completely broken when using FB 0.21 or later.
On 10/8/2010 2:01, subvers...@hamsterrepublic.com wrote: teeemcee 2010-10-07 23:01:32 -0700 (Thu, 07 Oct 2010) 380 Windows builds were completely broken when using FB 0.21 or later. Turns out that __FB_UNIX__ is *always* defined, to either true or false, unlike all the other __FB_*__ defines. Good work, FB devs. Actually I'd already complained about this on the forum... to the guy who wrote the patch for this when he was asking for feedback. He fixed all the other problems I pointed out :( Dude, you need to get with the times. Don't you know that #ifdef is SO last decade. Modern programs do this: IF __FB_WINDOWS__ THEN #include whatever ELSEIF __FB_UNIX__ THEN #include whateverx END IF Sheesh. ;) --- U wip/compat.bi ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] better palette options
On 10/8/2010 0:23, Ralph Versteegen wrote: On 8 October 2010 14:59, David Gowers (kampu)00a...@gmail.com wrote: On Fri, Oct 8, 2010 at 9:44 AM, Jeremy Burseyzippywi...@hotmail.com wrote: I'm only subscribed to the mailing digest, not the individual alerts, so I have to cut and paste those comments related to the palette discussion I introduced last night. Hopefully this doesn't come out all messy. Did james modify my palette in that way? I know he changed it slightly before installing as the new default. That's a shame if so; I specifically designed it to not waste a single entry. The game I'm updating actually started 10 years ago on the old palette, so I am using most of the original colors. I just had to move all my yellows to the top row to make room for all my sand tiles. So that's the specific issue I'm facing. At the moment, using my updated palette turns all the UI colors yellow. I'd like to use yours at some point, actually, but the time it would take me to convert 160 palettes, 300 backdrops, over 1,000 sprites, and so on to the compatable colors would cause me unthinkable pain, and I'm not ready to take that plunge yet. An update to how palettes are handled would certainly make the effort more desirable. i wrote an app which does that automatically (CHGPAL). have you tried it on a copy of your rpg? http://rpg.hamsterrepublic.com/ohrrpgce/Graphics_Utilities CHGPAL doesn't handle box borders, portraits or UI colours, and it uses the old master palette format instead of the new one, but it shouldn't be much work to update it if desired. It's a simple enough procedure that CUSTOM could conceivably implement it as an option on the master palette screen. Yes, it would be good. I don't know how to best pick the nearest match in the new palette for a colour (and should you let the user override this somehow, as they can when importing backdrops?), but in this case Jeremy just wants to move colours around the palette. That's the part that made me question whether this was worth discussing. have you seen http://rpg.hamsterrepublic.com/ohrrpgce/Plan_for_256_color_sprites I have, actually. But it was my understanding that this was still light years away from becoming reality, if it was ever meant to become reality. I'm working on some formal YAML propositions regarding sprite upgrades. They have the property of being very succint, so hopefully that will help things to happen What is a formal YAML proposition?? A proposal for a new graphics format? I'm keen to have that agreed upon. Just to add my two cents, I think PNG would be a good format for sprites. As a bonus, you can embed whatever metadata you might need directly in the image. ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] SVN: james/3838 Oops! I was using Reload.rli* constants instead of Reload.rlt* constants
On 10/7/2010 0:56, James Paige wrote: On Wed, Oct 06, 2010 at 09:55:07PM -0700, subvers...@hamsterrepublic.com wrote: james 2010-10-06 21:55:07 -0700 (Wed, 06 Oct 2010) 74 Oops! I was using Reload.rli* constants instead of Reload.rlt* constants! Ha ha ha! You feel for my devious trap! ... Honestly, though, I probably could have come up with a better name. And/or not put them in reload.bi, since they don't belong there. Could anybody explain the difference between these to me? I see that the rlt constants only have one integer type, what is the purpose of the rli constants and their small multitude of different integer types? rli (ReLoad Internal) are the types as stored on disk. Those constants are only used during loading/saving. rlt (ReLoad Types) are the logical types as seen by clients (number, string, etc). They correspond to the GetInteger(), etc functions. --- James ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] SVN: james/3838 Oops! I was using Reload.rli* constants instead of Reload.rlt* constants
On 10/7/2010 1:00, Mike Caron wrote: On 10/7/2010 0:56, James Paige wrote: On Wed, Oct 06, 2010 at 09:55:07PM -0700, subvers...@hamsterrepublic.com wrote: james 2010-10-06 21:55:07 -0700 (Wed, 06 Oct 2010) 74 Oops! I was using Reload.rli* constants instead of Reload.rlt* constants! Ha ha ha! You feel for my devious trap! ... Honestly, though, I probably could have come up with a better name. And/or not put them in reload.bi, since they don't belong there. Could anybody explain the difference between these to me? I see that the rlt constants only have one integer type, what is the purpose of the rli constants and their small multitude of different integer types? rli (ReLoad Internal) are the types as stored on disk. Those constants are only used during loading/saving. rlt (ReLoad Types) are the logical types as seen by clients (number, string, etc). They correspond to the GetInteger(), etc functions. Also, just looking at the editor's code, I noticed that you cap integers to be 16-bit signed. Is this a function of the grabber not supporting higher, or something else (since, RELOAD supports up to 64-bit integers)? --- James ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] SVN: james/3790 Attack queue is actually used now. (although non-blocking attacks are no
On 9/27/2010 12:31 PM, James Paige wrote: On Sun, Sep 26, 2010 at 07:45:13PM +1200, Ralph Versteegen wrote: On 26 September 2010 17:53, James Paigeb...@hamsterrepublic.com wrote: On Sat, Sep 25, 2010 at 10:12:11PM -0700, subvers...@hamsterrepublic.com wrote: =The way it WILL work eventually= * An attack will be targetted and confirmed * The attack and targetting and delay state will be stored directly in the queue without ever going in bslot() * When the delay runs out on the queued attack, the attack and targetting stat will be copied directly to BattleState * The attack will animate, using targetting state from BattleState, not bslot() Odd question: no possibility of multiple simultaneous attacks in future? I understand that attack info is copied into BattleState for convenience and because that was the easiest cleanup path, but that precludes simultaneous attacks. I am not ruling out the possibility of multiple simultaneous attacks in the future, but I would consider that to be a separate transition. That is all about cleaning up the attack animation code, whereas the attack queue is all about cleaning up the attack targetting and delaying code. The main reason I suggest supporting them is if people have multitasking battlescripting, and might like have simultaneous attacks, and use real attacks to do so rather than slice hackery. Of course, it would hardly be easy to support; it'd present a whole lot of problems to be solved. I am having difficulty thinking of RPG games that allow simultaneous attacks. Not talking about combo-attacks, or other cases of multi-attacker attack animations, which is certainly something we want to allow eventually. What I am talking about is fully asynchronous simultaneous attacks. What RPG does that? (so I can play it to get a better feel for it ;) The only one that I can think of is SaGa Frontier. Although I've never played it, it seems that you can chain together normal attacks into pseudo combos to execute simultaneously? Or something? Man, the SaGa games are weird :S http://www.youtube.com/watch?v=8Y_n1xcgusM (spoilers if you can read japanese, I guess) --- James Paige ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] SVN: james/3790 Attack queue is actually used now. (although non-blocking attacks are no
On 9/27/2010 1:08 PM, James Paige wrote: On Mon, Sep 27, 2010 at 12:51:40PM -0400, Mike Caron wrote: On 9/27/2010 12:31 PM, James Paige wrote: On Sun, Sep 26, 2010 at 07:45:13PM +1200, Ralph Versteegen wrote: On 26 September 2010 17:53, James Paigeb...@hamsterrepublic.com wrote: On Sat, Sep 25, 2010 at 10:12:11PM -0700, subvers...@hamsterrepublic.com wrote: =The way it WILL work eventually= * An attack will be targetted and confirmed * The attack and targetting and delay state will be stored directly in the queue without ever going in bslot() * When the delay runs out on the queued attack, the attack and targetting stat will be copied directly to BattleState * The attack will animate, using targetting state from BattleState, not bslot() Odd question: no possibility of multiple simultaneous attacks in future? I understand that attack info is copied into BattleState for convenience and because that was the easiest cleanup path, but that precludes simultaneous attacks. I am not ruling out the possibility of multiple simultaneous attacks in the future, but I would consider that to be a separate transition. That is all about cleaning up the attack animation code, whereas the attack queue is all about cleaning up the attack targetting and delaying code. The main reason I suggest supporting them is if people have multitasking battlescripting, and might like have simultaneous attacks, and use real attacks to do so rather than slice hackery. Of course, it would hardly be easy to support; it'd present a whole lot of problems to be solved. I am having difficulty thinking of RPG games that allow simultaneous attacks. Not talking about combo-attacks, or other cases of multi-attacker attack animations, which is certainly something we want to allow eventually. What I am talking about is fully asynchronous simultaneous attacks. What RPG does that? (so I can play it to get a better feel for it ;) The only one that I can think of is SaGa Frontier. Although I've never played it, it seems that you can chain together normal attacks into pseudo combos to execute simultaneously? Or something? Man, the SaGa games are weird :S http://www.youtube.com/watch?v=8Y_n1xcgusM I have that game. I am not 100% certain, but what i think is happening there is just a combo-attack. I should have said, most of it is where one animation begins before the last one has ended. However, I can distinctly recall a game where each party member jumps forward and does their attack in quick succession, so that the animations line up something like: ---A--- ---B ---C ---D ---E I thought it was Suikoden 1 or 2, but I couldn't find any relevant clip... Fakeedit: Yeah, it is Suikoden 2. Although each attack happens serially, the (non-flashy) unrelated animations overlap. http://www.youtube.com/watch?v=UpmAltLjwF8 To get the same effect in the OHR, we would have a system like this: * Attacker A does combo part 1 * Attacker B does combo part 2 before the delay runs out for Combo part 1 * Combo part 2 detects that combo part 1 is already queued by a valid team member, and replaces its queue entry with a FullCombo attack. * FullCombo would animate both attackers At no time would we be simultaneously animating to unrelated attack animations. --- James ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] SVN: kampu/3758 SConstruct based build system. Builds game, custom, and bam2mid currentl
On 9/21/2010 1:56 AM, Seth Hetu wrote: One more bit of info, from the scons wiki: If possible on the platform, the duplication is performed by linking rather than copying; see also the --duplicate command-line option. Moreover, only the files needed for the build are duplicated; files and directories that are not used are not present in variant_dir. So the files are _linked_ under Linux, which is why they compile properly after svn up. Which is why we only see this bug on Windows. The section you quoted seems to suggest that if linking is not possible (i.e., on Windows (actually, you can, but it's a pain)), it should copy, rather than fail. Just throwing that in there. It might be relevant to say that I'm using Windows Vista 32-bit Business Edition. Also, I have never used variant_dir before, so I am unfamiliar with its effects. --Seth On Tue, Sep 21, 2010 at 1:52 PM, Seth Hetuseth.h...@gmail.com wrote: I may have found the source of both problems. First, I updated to 0.21.1. (Just FYI). No noticeable change. Next, I checked. Yes, udts.bi is in build/. HOWEVER, it does not contain the definition of ZoneMap. And even stranger, the udts.bi in the wip/ folder DOES. This was especially odd, as I'd just done a scons -c. As reported before, though, -c only cleans binary files. This _should_ have been ok, though, since the source would have been copied into build/ and then re-compiled. But it obviously was not. So, my guess is that your scons script is failing to copy new source changes to the build/ folder. Everything else works ok, since no source file has changed in the build folder. So something's wrong with the MD5 check on copying everything to the build folder. Sure enough, deleting build and running scons completed the build. I'll have a look at the script; just wanted to give you guys a heads up. --Seth On Tue, Sep 21, 2010 at 8:59 AM, Ralph Versteegenteeem...@gmail.com wrote: On 21 September 2010 12:51, David Gowers (kampu)00a...@gmail.com wrote: On Tue, Sep 21, 2010 at 9:59 AM, Ralph Versteegenteeem...@gmail.com wrote: OK, now the killer problem: it doesn't rebuild anything when I edit a file!! Even if I run scons -c, it doesn't rebuild anything afterwards (but maybe -c is completely unsupported?) -c works fine for me. Did you change the *content*, not just edit something in and out again? Yes, I changed something. To clarify : if I add a comment to game.bas, game.o will be rebuilt, because I altered the MD5 sum. then ohrrpgce-game will *not* be rebuilt, because game.o will have the same MD5 sum as before. If your result differs from this, I agree it's a problem. I should have said: I changed something significant. ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] SVN: teeemcee/3756 Last minute change to the zonemap file format: switch to a span-based fo
On 9/16/2010 11:43 PM, subvers...@hamsterrepublic.com wrote: teeemcee 2010-09-16 20:43:54 -0700 (Thu, 16 Sep 2010) 372 Last minute change to the zonemap file format: switch to a span-based format (as already documented on the Wiki) This format is easier to write, but surprising slower to read and write (about half), and the file size is now mostly RELOAD overhead (wish Mike hadn't rejected my suggestions for better space efficiency). Which ones were those? :S Also finally added zonemap support to map resizing. --- U wip/loading.bas U wip/loading.bi U wip/mapsubs.bas ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] SVN: teeemcee/3756 Last minute change to the zonemap file format: switch to a span-based fo
On 9/17/2010 12:01 AM, Ralph Versteegen wrote: On 17 September 2010 15:46, Mike Caroncaron.m...@gmail.com wrote: On 9/16/2010 11:43 PM, subvers...@hamsterrepublic.com wrote: teeemcee 2010-09-16 20:43:54 -0700 (Thu, 16 Sep 2010) 372 Last minute change to the zonemap file format: switch to a span-based format (as already documented on the Wiki) This format is easier to write, but surprising slower to read and write (about half), and the file size is now mostly RELOAD overhead (wish Mike hadn't rejected my suggestions for better space efficiency). Which ones were those? :S -Use a VLI instead of an INT for node size. Even if you reserve 4 bytes whenever the node has any children, and then write the size later, you can probably use a single byte for leaf nodes. Except, the reason I use four bytes is because: 1. It's very difficult to slide the rest of the file over to get rid of the extra space. Even if that were easy, 2. It would create a cascading effect on the sizes of all other nodes. Both those problems would be solved if we rendered the file into memory first, but I'm not particularly enthused about writing this code. -Add a bit to the type byte that indicates whether or not there are any children, so that you can skip the VLI for number of children for leaves. This is reasonable, since I don't foresee having to add any new types. Those reduce the overhead from 7 bytes per leaf to 3. So if the average size of a .Z 'spans' data node is 3 bytes (guessing), it's a significant saving. I forgot that strings also have a size VLI, so it's actually 8 vs. 4. I agree that it's significant. An alternative, however, might be to forgo the size counts entirely. The reason they exist is to allow efficient scanning of the file in cases where you only care about part of the file. However, this is impossible with the current API, so... Proposal: Bump the version to 2 and omit the size parameters entirely. Thoughts? Also finally added zonemap support to map resizing. --- U wip/loading.bas U wip/loading.bi U wip/mapsubs.bas ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] SVN: teeemcee/3756 Last minute change to the zonemap file format: switch to a span-based fo
On 9/17/2010 12:23 AM, Ralph Versteegen wrote: On 17 September 2010 16:11, Mike Caroncaron.m...@gmail.com wrote: On 9/17/2010 12:01 AM, Ralph Versteegen wrote: On 17 September 2010 15:46, Mike Caroncaron.m...@gmail.comwrote: On 9/16/2010 11:43 PM, subvers...@hamsterrepublic.com wrote: teeemcee 2010-09-16 20:43:54 -0700 (Thu, 16 Sep 2010) 372 Last minute change to the zonemap file format: switch to a span-based format (as already documented on the Wiki) This format is easier to write, but surprising slower to read and write (about half), and the file size is now mostly RELOAD overhead (wish Mike hadn't rejected my suggestions for better space efficiency). Which ones were those? :S -Use a VLI instead of an INT for node size. Even if you reserve 4 bytes whenever the node has any children, and then write the size later, you can probably use a single byte for leaf nodes. Except, the reason I use four bytes is because: 1. It's very difficult to slide the rest of the file over to get rid of the extra space. Even if that were easy, 2. It would create a cascading effect on the sizes of all other nodes. You misunderstood. I meant: if the node has no children, determining the size is easy; otherwise reserve 4 bytes (27 bits: 134MB) and come back and write the VLI later. Both those problems would be solved if we rendered the file into memory first, but I'm not particularly enthused about writing this code. We do already write the file to memory first, but of course it's not designed to do this. The technique I've used in the past (albeit, in C# with MemoryStreams) is to create a buffer for the node's children, render the children (passing this buffer), and then rendering the node itself + the buffer into the parent's buffer. The advantages are that you can do nice optimizations like the VLI thing. The downside is that it's a lot of memory management (especially without a garbage collector working behind the scenes!) -Add a bit to the type byte that indicates whether or not there are any children, so that you can skip the VLI for number of children for leaves. This is reasonable, since I don't foresee having to add any new types. You'd still be able to have up to 128 types. Yup, I was agreeing with you Those reduce the overhead from 7 bytes per leaf to 3. So if the average size of a .Z 'spans' data node is 3 bytes (guessing), it's a significant saving. I forgot that strings also have a size VLI, so it's actually 8 vs. 4. I agree that it's significant. An alternative, however, might be to forgo the size counts entirely. The reason they exist is to allow efficient scanning of the file in cases where you only care about part of the file. However, this is impossible with the current API, so... Huh? The sizes are used to skip over nodes when using delay loading. Wasn't that exactly their intention? And that seems worth keeping. Oh, yeah. Wow it's been a while. But, now I remember... Hmm... Another option would be to only include the sizes if the node has children, but that's getting a bit messy. Unfortunately, you don't know in advance whether a node has children or not. Proposal: Bump the version to 2 and omit the size parameters entirely. Thoughts? I figured we would eventually create a new version, and that I would not let my suggestions be excluded again; however there's no hurry to do so now. I guess it's it wouldn't be much work though. Honestly, though, does it really matter that much that there's 8 bytes of overhead instead of 4? I think the law of diminishing returns is kicking in here... Also finally added zonemap support to map resizing. --- U wip/loading.bas U wip/loading.bi U wip/mapsubs.bas ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] BYREF
On 9/14/2010 3:56 AM, Seth Hetu wrote: I'd definitely avoid this situation with gusto. Consider passing '3' (just the number) as by ref into the parameter 'x' ---what happens if the compiler inlines that variable, copying it to each place the variable is used inside the function? That becomes problematic for, e.g.,: x = 10 which becomes: 3 = 10 That's not how function inlining works. And fbc is miles off doing any optimisations as significant as function inlining anyway. I was under the impression that function inlining (e.g., inline expansion) included constant propagation as one of its steps. That, at least, is the reason it's illegal in C++ ---not sure what the FB compiler says if you try this. I don't think so. C++ can pass other types of temporary values by reference, so I assume the reason it's illegal is that if you're passing 3 by reference, it's probably an unintentional bug. Passing by reference indicates that you want to pass a *variable* instead of a value. And like I said, it works in FB. What you said makes more sense in this case. But then if passing by reference always works in FB, what happens if you pass '3' as a parameter by reference? Does it push a temporary variable to the stack and store '3' in there? Does it declare a temporary variable on the heap and pass its memory address to the function? Why is there so much confusion by this? Variable allocation is trivial to do at compile time. If FB needs a temporary variable to pass to a BYREF argument, it just adds sizeof(thatvariable) to the stack size and calls it a night. Or, looking at it from a different angle, why are you NOT surprised that you can do this in C? myfunction(Literal string: omg no variable!); I'm just curious how FB manages this without corrupting memory. It seems like such a weird design choice. Consider this. In QB, everything was BYREF by default. So, this had to be legal: sub mysub(myparam as integer) REM whatever end sub mysub 1 If it weren't legal, you'd get an error on a seemingly standard BASIC program. --Seth ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] BYREF
On 9/12/2010 10:31 PM, Seth Hetu wrote: I have a feeling that this should be illegal in FB. I could totally see FB forgetting that it stored a variable in its nondeclared variable storage and overwriting it in a later function. I'd definitely avoid this situation with gusto. Consider passing '3' (just the number) as by ref into the parameter 'x' ---what happens if the compiler inlines that variable, copying it to each place the variable is used inside the function? That becomes problematic for, e.g.,: x = 10 which becomes: 3 = 10 That, at least, is the reason it's illegal in C++ ---not sure what the FB compiler says if you try this. (Of course, in early Fortran dialects, side-effecting 3 to 10 was completely possible... but that's a whole different can of worms (and I've been told it doesn't work in newer compilers).) I imagine that any sane compiler would turn that situation into this: dim temp as integer = 3 byrefsub(temp) --Seth ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] BYREF
On 9/13/2010 12:37 AM, Keith Gable wrote: On Mon, 2010-09-13 at 00:33 -0400, Mike Caron wrote: I imagine that any sane compiler would turn that situation into this: dim temp as integer = 3 byrefsub(temp) This is FreeBASIC we're talking about. I'd be worried that it would assume that memory is freed up when the goto/gosub/function/sub finishes. Who was talking about FreeBASIC? I said *sane* compiler. ;) ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] BYREF
On 9/11/2010 4:27 PM, James Paige wrote: I keep forgetting about BYREF. I know when to use it on ints, and I know never to use it on strings, and I know it is utterly and completely irrelvant on arrays But what I can't remember is whether it matters for UDTs You generally want to pass UDTs BYREF, especially if you want to modify the UDT in the sub. You can pass them BYVAL, but: 1. They will be read only (it passes a copy) 2. If the UDT is big and passed around frequently, it will be considerably slower than passing BYREF. --- James ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] BYREF
On 9/11/2010 8:18 PM, Ralph Versteegen wrote: On 12 September 2010 09:16, James Paigeb...@hamsterrepublic.com wrote: On Sat, Sep 11, 2010 at 05:06:22PM -0400, Mike Caron wrote: On 9/11/2010 4:27 PM, James Paige wrote: I keep forgetting about BYREF. I know when to use it on ints, and I know never to use it on strings, and I know it is utterly and completely irrelvant on arrays But what I can't remember is whether it matters for UDTs You generally want to pass UDTs BYREF, especially if you want to modify the UDT in the sub. You can pass them BYVAL, but: 1. They will be read only (it passes a copy) 2. If the UDT is big and passed around frequently, it will be considerably slower than passing BYREF. But doesn't -lang deprecated pass them BYREF by default or something? That would explain why all the many times I have forgotten to specify BYREF on UDTs has not broken anything (that I have noticed yet). --- James UDTs by default are always passed BYREF, in all dialects, so I'd rather we didn't specify BYREF. What I didn't realise was that Option Byval changes UDTs to pass BYVAL by default! Argh! How could the FB devs be so cruel to us! :( You know, in most languages, UDTs are passed BYVAL BYDEF- er, by default. Like, in C for example. Besides, wouldn't you want it to be consistent everywhere? On the whole our code will be cleaner if we explicitly pass all UDTs BYREF, declare Option Byval everywhere, and then remove all the explicit BYVAL passing of integers. But I'm a bit less motivated to do so now. ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] BYREF
On 9/12/2010 1:01 AM, Keith Gable wrote: On Sun, 2010-09-12 at 12:57 +1200, Ralph Versteegen wrote: C didn't have references. I thought that a reference in BASIC languages is the same thing as a pointer in C languages. In VB6, ByVal creates a copy of the object (I'm sure it does the same thing in FB), whereas ByRef is a pointer, even to the extent that C APIs that follow the standard calling convention that require pointers will work if you define the pointers as ByRef. Though that same API could be declared ByVal but then you use AddressOf when calling it. [This site][1] seems to reinforce this, at least for VB6. Are things different in FB? Just curious. [1]: http://www.cpearson.com/excel/byrefbyval.aspx No, you are correct. I will make a table: || |ALMOST EQUIVALENT THINGS| || | BASIC| C| C++ | |--+--+--| | BYREF| Pointers | References * | |--+--+--| | BYREF as integer | int *| int | || - BYVAL parameters are copies of the original, passed via stack. They can be modified with no consequence (unless it's a UDT with pointers, since the pointers will still point at the old data!) - BYREF parameters are pointers, but they cannot be reassigned. - Pointers are pointers (duh), but they CAN be reassigned. - References are aliases to the variables they reference. They cannot be reassigned, and are effectively the same as the original variable. ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] NPC instance script commands
On 9/10/2010 10:58 AM, Ralph Versteegen wrote: On 10 September 2010 03:18, James Paigeb...@hamsterrepublic.com wrote: On Thu, Sep 09, 2010 at 10:18:41PM +1200, Ralph Versteegen wrote: Well, this email was originally going to be part of the one about SAV, but then I remembered that NPC instance data has never been read from .SAV! I'd like start adding of data to NPC instances, like zone restrictions. (I'll have to create an NPC instance editor.) Actually, I already started. A few months ago I wrote these (below), but didn't check them in because I was waiting for .SAV to be dropped. Opps! Sorry Jeremy! But anyway, I just thought I may as well ask about functions for reading/writing NPC instance data. functions: 472,setnpcbit,3,0,0,1 #set NPC instance bit (npcref, bit, value) 473,getnpcbit,2,0,0 #get NPC instance bit (npcref, bit) constants: 0,NPCbit:ignore walls 1,NPCbit:not obstruction 2,NPCbit:suspend use 3,NPCbit:suspend movement (Incidentally, those bits are not actually stored as bits in the UDT.) And then other non-flag data like zone restrictions would each get a pair of getter/setting functions. Would people prefer something more uniform like alter/read npc (below)? Or the opposite extreme of individual read/write functions for each bit? 472,read npc instance,3,0,0,1 #set NPC instance data (npcref, data id, value) 473,write npc instance,2,0,0 #get NPC instance data (npcref, data id) 0,NPCinst:ignore walls 1,NPCinst:not obstruction 2,NPCinst:suspend use 3,NPCinst:suspend movement 4,NPCinst:zone I like individual functions for each bit. I may have advocated generic functions with constants in the past, but I am really starting to hate the heck out of them. it is all about what I want to be writing as an end-user in my script: # this is good set npc ignore walls(npc, true) # this is bad set npc bit(npc, NPCbit:ignore walls, true) Well, I choose the second approach because I hate having too many functions (and having to document them :) ), because it makes the dictionary extremely verbose: consulting a compact table as for alternpc is much faster (I actually prefer read/alternpc to trying to remember over 100 menu functions, for example). And our commands are inconsistent and difficult to remember as a rule -- I can never remember whether to use slice parent or parent slice -- so I end up looking up an awful lot. However a few paragraphs of documentation for every tiny bit is much more fair for people who aren't familiar with engine internals in the first place. However despite all that, I must agree with you: I didn't realise how overly verbose the setnpcbit approach was. But while we're on the topic of too many functions which are impossible to remember, I think I'll sidetrack into adding structures to HS, which I've been meaning to talk about. Having actually used Euphoria a whole a couple months back, I take back what I said about it being a nice language. I think not supporting user defined data structures in the new HamsterSpeak is out of the question, it's just unworkable. And following this train of reasoning, if we have user defined structures, and function objects, then we may as well add some form of object orientation, since you're already nearly there. And if we now present a slice handle, npc reference, menu handle or whatever else as an object with members and/or methods, we can provide an alternative way to interact these objects that's less verbose, highly consistent, easy to remember and compactly documentable: menuitem.caption instead of get menu item caption (menuitem) heroes[who].stats[stat.hp] += x instead of set hero stat(who, stat:hp, get hero stat(who, stat:hp) + x) and so on. The exact details don't matter right now, but this is very similar to the Magic Variables plan. I'm having deja vu; this is a lot like the transition between the mathematical operators being functions and them being actual symbols. (although, this transition occurred long before I had ever used the OHR) Hypothetically speaking, how would you implement these structs? Would you go for a classical model (like how they work in C), or something else? If you're looking for suggestions, I think Lua has the best system. Basically, every variable that isn't a basic type (integer, string, etc) is a table (i.e., a set of key-value pairs). If the indicies are numeric, you call it an array, whereas if the keys are strings, you call it a hash. If the values are functions, then it must be an object. Even better, you can set what's called a metatable, which defines functions that are called when the table is operated on, which allows you to simulate properties. The identity of an object is based on its metatable. If get_metatable(objA) == get_metatable(objB), then they must be instances of the same class. One of Lua's interesting quirks is that it actually has two ways to invoke methods on objects. If you use the dot operator
Re: [Ohrrpgce] Script structures/objects (Was: NPC instance script commands)
On 9/10/2010 12:26 PM, Ralph Versteegen wrote: On 11 September 2010 03:15, Mike Caroncaron.m...@gmail.com wrote: On 9/10/2010 10:58 AM, Ralph Versteegen wrote: On 10 September 2010 03:18, James Paigeb...@hamsterrepublic.comwrote: On Thu, Sep 09, 2010 at 10:18:41PM +1200, Ralph Versteegen wrote: Well, this email was originally going to be part of the one about SAV, but then I remembered that NPC instance data has never been read from .SAV! I'd like start adding of data to NPC instances, like zone restrictions. (I'll have to create an NPC instance editor.) Actually, I already started. A few months ago I wrote these (below), but didn't check them in because I was waiting for .SAV to be dropped. Opps! Sorry Jeremy! But anyway, I just thought I may as well ask about functions for reading/writing NPC instance data. functions: 472,setnpcbit,3,0,0,1 #set NPC instance bit (npcref, bit, value) 473,getnpcbit,2,0,0 #get NPC instance bit (npcref, bit) constants: 0,NPCbit:ignore walls 1,NPCbit:not obstruction 2,NPCbit:suspend use 3,NPCbit:suspend movement (Incidentally, those bits are not actually stored as bits in the UDT.) And then other non-flag data like zone restrictions would each get a pair of getter/setting functions. Would people prefer something more uniform like alter/read npc (below)? Or the opposite extreme of individual read/write functions for each bit? 472,read npc instance,3,0,0,1 #set NPC instance data (npcref, data id, value) 473,write npc instance,2,0,0 #get NPC instance data (npcref, data id) 0,NPCinst:ignore walls 1,NPCinst:not obstruction 2,NPCinst:suspend use 3,NPCinst:suspend movement 4,NPCinst:zone I like individual functions for each bit. I may have advocated generic functions with constants in the past, but I am really starting to hate the heck out of them. it is all about what I want to be writing as an end-user in my script: # this is good set npc ignore walls(npc, true) # this is bad set npc bit(npc, NPCbit:ignore walls, true) Well, I choose the second approach because I hate having too many functions (and having to document them :) ), because it makes the dictionary extremely verbose: consulting a compact table as for alternpc is much faster (I actually prefer read/alternpc to trying to remember over 100 menu functions, for example). And our commands are inconsistent and difficult to remember as a rule -- I can never remember whether to use slice parent or parent slice -- so I end up looking up an awful lot. However a few paragraphs of documentation for every tiny bit is much more fair for people who aren't familiar with engine internals in the first place. However despite all that, I must agree with you: I didn't realise how overly verbose the setnpcbit approach was. But while we're on the topic of too many functions which are impossible to remember, I think I'll sidetrack into adding structures to HS, which I've been meaning to talk about. Having actually used Euphoria a whole a couple months back, I take back what I said about it being a nice language. I think not supporting user defined data structures in the new HamsterSpeak is out of the question, it's just unworkable. And following this train of reasoning, if we have user defined structures, and function objects, then we may as well add some form of object orientation, since you're already nearly there. And if we now present a slice handle, npc reference, menu handle or whatever else as an object with members and/or methods, we can provide an alternative way to interact these objects that's less verbose, highly consistent, easy to remember and compactly documentable: menuitem.caption instead of get menu item caption (menuitem) heroes[who].stats[stat.hp] += x instead of set hero stat(who, stat:hp, get hero stat(who, stat:hp) + x) and so on. The exact details don't matter right now, but this is very similar to the Magic Variables plan. I'm having deja vu; this is a lot like the transition between the mathematical operators being functions and them being actual symbols. (although, this transition occurred long before I had ever used the OHR) Hypothetically speaking, how would you implement these structs? Would you go for a classical model (like how they work in C), or something else? I was thinking something along the lines of Python or Lua, which implement objects in very similar ways. I'm not all that familiar with Lua though. We can afford to be less flexible than these though. For example, should you be able to add new members to builtin data types, like setting npcref.is_hallucinating := true? (Realise that if we do this, it would be handled entirely by the script interpreter, and in fact I think it's quite likely we can add builtin types without having to significantly modify any existing commands beyond what's needed for dynamic types anyway) I think it looks elegant, but isn't there a rule of thumb that composition is preferred to
Re: [Ohrrpgce] Binary data in RELOAD to XML?
On 8/23/2010 8:19 AM, Ralph Versteegen wrote: On 23 August 2010 05:24, Mike Caroncaron.m...@gmail.com wrote: On 8/22/2010 3:23 AM, Ralph Versteegen wrote: On 22 August 2010 08:14, Mike Caroncaron.m...@gmail.comwrote: On 8/21/2010 3:55 PM, Ralph Versteegen wrote: On 21 August 2010 05:22, Mike Caroncaron.m...@gmail.com wrote: On 8/20/2010 1:15 PM, Ralph Versteegen wrote: On 21 August 2010 05:11, James Paigeb...@hamsterrepublic.com wrote: On Fri, Aug 20, 2010 at 12:55:50PM -0400, Mike Caron wrote: On 8/20/2010 12:52 PM, Ralph Versteegen wrote: On 21 August 2010 04:43, Mike Caroncaron.m...@gmail.com wrote: On 8/20/2010 12:41 PM, Ralph Versteegen wrote: On 21 August 2010 03:58, Mike Caroncaron.m...@gmail.com wrote: On 8/20/2010 11:45 AM, Ralph Versteegen wrote: On 21 August 2010 02:24, Mike Caroncaron.m...@gmail.com wrote: On 8/20/2010 9:58 AM, Ralph Versteegen wrote: Currently reload2xml can't properly export binary data stored in strings in reload files. It would be nice to be able to hand edit xml and convert back. What's the preferred way to write binary?#nnn; escape codes, or that Base64 stuff? (Is that actually part of the xml standard?) Strictly speaking, we only really need to escape characters below 32. Everything else should be okay to write out. Follow-up question: what about bytes above 127? Is there a chance of confusing xml parsers into guessing UTF8 encoding? I suppose it doesn't matter too much, since we can force libxml2 to read with ASCII encoding (but it would be nice to know that we have to do so). The best way to handle this, actually, would be to add the XML header, which allows you to specify the encoding. In our case, it should look something like: ?xml version=1.0 encoding=iso-8859-1? In light of that, I would prefer the#; syntax, since it means that no one has to do any extra work to process the resulting XML file. If you used Base64 (which has nothing to do with XML), you'd have to mark it as such in order to distinguish it from a regular string that just so happens to look like Base64. OK, cool. I'll add that then soonish (unless you jumped on the chance to do so). No, go ahead. Change of plan required. ' #0; is not permitted, however, as the null character is one of the control characters excluded from XML, even when using a numeric character reference.[14] An alternative encoding mechanism such as Base64 is needed to represent such characters. ' - Wikipedia And I just confirmed libxml2 spits. Damn. I guess that kind of makes sense, though. Maybe... emit anull/ element instead? Ugh. Or, Base64 it. Sigh. Well, I know next to nothing about XML, what's the idiomatic way to do that? Add a special attribute to nodes containing base64-encoded data, likefoo base=64assdfasdf234/foo ? No, the proper way to do this is something like this: basenode xmlns:reload=http://hamsterrepublic.com/RELOAD; foo reload:encoding=base64.../foo /basenode Note: that namespace URL doesn't have to exist, it just has to be unique. At first I thought you may be pulling my leg and going overboard... but such is XML! Actually, we could just cheese it and say: root xmlns:reload=uri:reload But, that's considered bad form. The only reason I'm suggesting using namespaces is because the structure and the data are intermingled. How is xml2reload supposed to know if base=64 means that the node is encoded, or it represents a strangely-based number, or that all your base count up to 64? reload:base=64 would be perfectly unambiguous if it were agreed upon, though. But I agree it would be pretty weird. It would be, yes. But, then when we add some other encoding for some reason... etc etc Another, more XMLish way, might be to encode documents like this: RELOADDocument node nameroot/name typenull/type children node namemynode/name typestring/type content123/content children.../children /node /children /node /RELOADDocument Now who's pulling whose leg? ;) Ugh I hate XML (and libxml) so much. I volunteered for this when I thought it was going to be a quick fix with#nnn; but now I'm in XML deeper than I ever hoped to get in my life. Should have taken the blue pill... Anyway some more problems I'd likesapproval/s suggestions on: * null-name nodes? We already use these in RELOAD documents. The solution I came up with is to write the name as 'reload:_' and convert it back to a null name in xml2reload (while still being able to distinguish the null-name nodes libxml2 inserts). Was this a good choice? Sure, that seems like a reasonable solution. * strings with leading/trailing whitespace. Whitespace is a bitch in XML. I had quite a bit of trouble, back when I did the plotscripting dictionary, since hssed wanted such particular spacing for the help file! I am quite certain now
Re: [Ohrrpgce] LibXML2 sucks?
On 8/23/2010 8:04 AM, Ralph Versteegen wrote:
On 23 August 2010 10:45, Mike Caroncaron.m...@gmail.com wrote:
The other email thread was getting kinda big, and this is tangentially
related, so here's a separate thread.
TMC, you mentioned that LibXML2 was eating nodes with all whitespace. I am
unable to reproduce this problem.
Using this document:
node
text beforechild text inside/childtext after
blanknode /blanknode
/node
And, this program:
void print_element_names(xmlNode * a_node) {
xmlNode *cur_node = NULL;
for (cur_node = a_node; cur_node; cur_node = cur_node-next) {
for(int i = 0; i indent; i++) printf( );
switch(cur_node-type) {
case XML_ELEMENT_NODE:
printf(node type: Element, name: %s\n, cur_node-name);
break;
case XML_TEXT_NODE:
printf(node type: Text, content: \%s\\n,
xmlNodeGetContent(cur_node));
break;
}
indent++;
print_element_names(cur_node-children);
indent--;
}
}
(set up code omitted, and in C because that's what the sample I tweaked was
in)
I got this output:
D:\projects\LibXml2Testdebug\LibXml2Test.exe ws.xml
node type: Element, name: node
node type: Text, content:
text before
node type: Element, name: child
node type: Text, content: text inside
node type: Text, content: text after
node type: Element, name: blanknode
node type: Text, content:
node type: Text, content:
So, it appears to be preserving whitespace properly.
Ah! I see what's going. xml2reload was using xmlNodeIsEmpty, which I'd
failed to notice. Knowing that, I managed to get everything working
without quotes.
Yay!
Why don't you commit what you have, and I will take a look at it?
OK... you asked for it!
Okay, let me just check my blackberry this mornin- holy crap!
___
Ohrrpgce mailing list
ohrrpgce@lists.motherhamster.org
http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
___
Ohrrpgce mailing list
ohrrpgce@lists.motherhamster.org
http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] SVN: teeemcee/3668 More horror! MenuDefs are meant to be deleted with ClearMenuData, or the
On 8/23/2010 1:39 PM, James Paige wrote: On Tue, Aug 24, 2010 at 05:22:00AM +1200, Ralph Versteegen wrote: On 24 August 2010 04:33, James Paigeb...@hamsterrepublic.com wrote: On Mon, Aug 23, 2010 at 04:56:27AM -0700, subvers...@hamsterrepublic.com wrote: teeemcee 2010-08-23 04:56:27 -0700 (Mon, 23 Aug 2010) 218 More horror! MenuDefs are meant to be deleted with ClearMenuData, or they leak all their items! Fix the numerous places where we were doing that. It would be REALLY great if FB let us use constructors and destructors. Isn't -lang deprecated the thing that is holding us back from using them? --- James Yep. There are a few UDTs now which have to be manually destructed, and I feel really bad every time I introduce one because it makes our code less friendly and more and more C-like. I suspect that constructor and destructor support could be added to -lang deprecated by patching a few lines in the compiler, and I am almost considering trying it. And if I didn't have all these OHR features to work on, I would definitely be working on fb2c++ right now. Have you been following the forum thread in which a couple people have threatened to bribe you to work on it with the goal of an XNA port? What, really? While I would really enjoy a .NET port of free basic, I doubt this would happen any time in the near future. --- James ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] SVN: teeemcee/3668 More horror! MenuDefs are meant to be deleted with ClearMenuData, or the
On 8/23/2010 1:39 PM, James Paige wrote: On Tue, Aug 24, 2010 at 05:22:00AM +1200, Ralph Versteegen wrote: On 24 August 2010 04:33, James Paigeb...@hamsterrepublic.com wrote: On Mon, Aug 23, 2010 at 04:56:27AM -0700, subvers...@hamsterrepublic.com wrote: teeemcee 2010-08-23 04:56:27 -0700 (Mon, 23 Aug 2010) 218 More horror! MenuDefs are meant to be deleted with ClearMenuData, or they leak all their items! Fix the numerous places where we were doing that. It would be REALLY great if FB let us use constructors and destructors. Isn't -lang deprecated the thing that is holding us back from using them? --- James Yep. There are a few UDTs now which have to be manually destructed, and I feel really bad every time I introduce one because it makes our code less friendly and more and more C-like. I suspect that constructor and destructor support could be added to -lang deprecated by patching a few lines in the compiler, and I am almost considering trying it. And if I didn't have all these OHR features to work on, I would definitely be working on fb2c++ right now. Have you been following the forum thread in which a couple people have threatened to bribe you to work on it with the goal of an XNA port? What, really? While I would really enjoy a .NET port of free basic, I doubt this would happen any time in the near future. --- James ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] Binary data in RELOAD to XML?
On 8/22/2010 3:23 AM, Ralph Versteegen wrote: On 22 August 2010 08:14, Mike Caroncaron.m...@gmail.com wrote: On 8/21/2010 3:55 PM, Ralph Versteegen wrote: On 21 August 2010 05:22, Mike Caroncaron.m...@gmail.comwrote: On 8/20/2010 1:15 PM, Ralph Versteegen wrote: On 21 August 2010 05:11, James Paigeb...@hamsterrepublic.com wrote: On Fri, Aug 20, 2010 at 12:55:50PM -0400, Mike Caron wrote: On 8/20/2010 12:52 PM, Ralph Versteegen wrote: On 21 August 2010 04:43, Mike Caroncaron.m...@gmail.comwrote: On 8/20/2010 12:41 PM, Ralph Versteegen wrote: On 21 August 2010 03:58, Mike Caroncaron.m...@gmail.com wrote: On 8/20/2010 11:45 AM, Ralph Versteegen wrote: On 21 August 2010 02:24, Mike Caroncaron.m...@gmail.com wrote: On 8/20/2010 9:58 AM, Ralph Versteegen wrote: Currently reload2xml can't properly export binary data stored in strings in reload files. It would be nice to be able to hand edit xml and convert back. What's the preferred way to write binary?#nnn; escape codes, or that Base64 stuff? (Is that actually part of the xml standard?) Strictly speaking, we only really need to escape characters below 32. Everything else should be okay to write out. Follow-up question: what about bytes above 127? Is there a chance of confusing xml parsers into guessing UTF8 encoding? I suppose it doesn't matter too much, since we can force libxml2 to read with ASCII encoding (but it would be nice to know that we have to do so). The best way to handle this, actually, would be to add the XML header, which allows you to specify the encoding. In our case, it should look something like: ?xml version=1.0 encoding=iso-8859-1? In light of that, I would prefer the#; syntax, since it means that no one has to do any extra work to process the resulting XML file. If you used Base64 (which has nothing to do with XML), you'd have to mark it as such in order to distinguish it from a regular string that just so happens to look like Base64. OK, cool. I'll add that then soonish (unless you jumped on the chance to do so). No, go ahead. Change of plan required. ' #0; is not permitted, however, as the null character is one of the control characters excluded from XML, even when using a numeric character reference.[14] An alternative encoding mechanism such as Base64 is needed to represent such characters. ' - Wikipedia And I just confirmed libxml2 spits. Damn. I guess that kind of makes sense, though. Maybe... emit anull/element instead? Ugh. Or, Base64 it. Sigh. Well, I know next to nothing about XML, what's the idiomatic way to do that? Add a special attribute to nodes containing base64-encoded data, likefoo base=64assdfasdf234/foo? No, the proper way to do this is something like this: basenode xmlns:reload=http://hamsterrepublic.com/RELOAD; foo reload:encoding=base64.../foo /basenode Note: that namespace URL doesn't have to exist, it just has to be unique. At first I thought you may be pulling my leg and going overboard... but such is XML! Actually, we could just cheese it and say: root xmlns:reload=uri:reload But, that's considered bad form. The only reason I'm suggesting using namespaces is because the structure and the data are intermingled. How is xml2reload supposed to know if base=64 means that the node is encoded, or it represents a strangely-based number, or that all your base count up to 64? reload:base=64 would be perfectly unambiguous if it were agreed upon, though. But I agree it would be pretty weird. It would be, yes. But, then when we add some other encoding for some reason... etc etc Another, more XMLish way, might be to encode documents like this: RELOADDocument node nameroot/name typenull/type children node namemynode/name typestring/type content123/content children.../children /node /children /node /RELOADDocument Now who's pulling whose leg? ;) Ugh I hate XML (and libxml) so much. I volunteered for this when I thought it was going to be a quick fix with#nnn; but now I'm in XML deeper than I ever hoped to get in my life. Should have taken the blue pill... Anyway some more problems I'd likesapproval/ssuggestions on: * null-name nodes? We already use these in RELOAD documents. The solution I came up with is to write the name as 'reload:_' and convert it back to a null name in xml2reload (while still being able to distinguish the null-name nodes libxml2 inserts). Was this a good choice? Sure, that seems like a reasonable solution. * strings with leading/trailing whitespace. Whitespace is a bitch in XML. I had quite a bit of trouble, back when I did the plotscripting dictionary, since hssed wanted such particular spacing for the help file! I am quite certain now that using XML to represent RELOAD documents as text was a horrible idea, and we should have used something very
[Ohrrpgce] LibXML2 sucks?
The other email thread was getting kinda big, and this is tangentially
related, so here's a separate thread.
TMC, you mentioned that LibXML2 was eating nodes with all whitespace. I
am unable to reproduce this problem.
Using this document:
node
text beforechild text inside /childtext after
blanknode /blanknode
/node
And, this program:
void print_element_names(xmlNode * a_node) {
xmlNode *cur_node = NULL;
for (cur_node = a_node; cur_node; cur_node = cur_node-next) {
for(int i = 0; i indent; i++) printf( );
switch(cur_node-type) {
case XML_ELEMENT_NODE:
printf(node type: Element, name: %s\n, cur_node-name);
break;
case XML_TEXT_NODE:
printf(node type: Text, content: \%s\\n,
xmlNodeGetContent(cur_node));
break;
}
indent++;
print_element_names(cur_node-children);
indent--;
}
}
(set up code omitted, and in C because that's what the sample I tweaked
was in)
I got this output:
D:\projects\LibXml2Testdebug\LibXml2Test.exe ws.xml
node type: Element, name: node
node type: Text, content:
text before
node type: Element, name: child
node type: Text, content: text inside
node type: Text, content: text after
node type: Element, name: blanknode
node type: Text, content:
node type: Text, content:
So, it appears to be preserving whitespace properly.
Why don't you commit what you have, and I will take a look at it?
___
Ohrrpgce mailing list
ohrrpgce@lists.motherhamster.org
http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] Binary data in RELOAD to XML?
On 8/21/2010 3:55 PM, Ralph Versteegen wrote: On 21 August 2010 05:22, Mike Caroncaron.m...@gmail.com wrote: On 8/20/2010 1:15 PM, Ralph Versteegen wrote: On 21 August 2010 05:11, James Paigeb...@hamsterrepublic.comwrote: On Fri, Aug 20, 2010 at 12:55:50PM -0400, Mike Caron wrote: On 8/20/2010 12:52 PM, Ralph Versteegen wrote: On 21 August 2010 04:43, Mike Caroncaron.m...@gmail.com wrote: On 8/20/2010 12:41 PM, Ralph Versteegen wrote: On 21 August 2010 03:58, Mike Caroncaron.m...@gmail.comwrote: On 8/20/2010 11:45 AM, Ralph Versteegen wrote: On 21 August 2010 02:24, Mike Caroncaron.m...@gmail.com wrote: On 8/20/2010 9:58 AM, Ralph Versteegen wrote: Currently reload2xml can't properly export binary data stored in strings in reload files. It would be nice to be able to hand edit xml and convert back. What's the preferred way to write binary?#nnn; escape codes, or that Base64 stuff? (Is that actually part of the xml standard?) Strictly speaking, we only really need to escape characters below 32. Everything else should be okay to write out. Follow-up question: what about bytes above 127? Is there a chance of confusing xml parsers into guessing UTF8 encoding? I suppose it doesn't matter too much, since we can force libxml2 to read with ASCII encoding (but it would be nice to know that we have to do so). The best way to handle this, actually, would be to add the XML header, which allows you to specify the encoding. In our case, it should look something like: ?xml version=1.0 encoding=iso-8859-1? In light of that, I would prefer the#; syntax, since it means that no one has to do any extra work to process the resulting XML file. If you used Base64 (which has nothing to do with XML), you'd have to mark it as such in order to distinguish it from a regular string that just so happens to look like Base64. OK, cool. I'll add that then soonish (unless you jumped on the chance to do so). No, go ahead. Change of plan required. ' #0; is not permitted, however, as the null character is one of the control characters excluded from XML, even when using a numeric character reference.[14] An alternative encoding mechanism such as Base64 is needed to represent such characters. ' - Wikipedia And I just confirmed libxml2 spits. Damn. I guess that kind of makes sense, though. Maybe... emit anull/ element instead? Ugh. Or, Base64 it. Sigh. Well, I know next to nothing about XML, what's the idiomatic way to do that? Add a special attribute to nodes containing base64-encoded data, likefoo base=64assdfasdf234/foo ? No, the proper way to do this is something like this: basenode xmlns:reload=http://hamsterrepublic.com/RELOAD; foo reload:encoding=base64.../foo /basenode Note: that namespace URL doesn't have to exist, it just has to be unique. At first I thought you may be pulling my leg and going overboard... but such is XML! Actually, we could just cheese it and say: root xmlns:reload=uri:reload But, that's considered bad form. The only reason I'm suggesting using namespaces is because the structure and the data are intermingled. How is xml2reload supposed to know if base=64 means that the node is encoded, or it represents a strangely-based number, or that all your base count up to 64? reload:base=64 would be perfectly unambiguous if it were agreed upon, though. But I agree it would be pretty weird. It would be, yes. But, then when we add some other encoding for some reason... etc etc Another, more XMLish way, might be to encode documents like this: RELOADDocument node nameroot/name typenull/type children node namemynode/name typestring/type content123/content children.../children /node /children /node /RELOADDocument Now who's pulling whose leg? ;) Ugh I hate XML (and libxml) so much. I volunteered for this when I thought it was going to be a quick fix with#nnn; but now I'm in XML deeper than I ever hoped to get in my life. Should have taken the blue pill... Anyway some more problems I'd likesapproval/s suggestions on: * null-name nodes? We already use these in RELOAD documents. The solution I came up with is to write the name as 'reload:_' and convert it back to a null name in xml2reload (while still being able to distinguish the null-name nodes libxml2 inserts). Was this a good choice? Sure, that seems like a reasonable solution. * strings with leading/trailing whitespace. Whitespace is a bitch in XML. I had quite a bit of trouble, back when I did the plotscripting dictionary, since hssed wanted such particular spacing for the help file! xml2reload currently strips it, but it doesn't have to. My first idea was: foo reload:encoding=exact lots of white spacechild.../child child.../child /foo As you can see, the first child if any has to be smack against the end of the string
Re: [Ohrrpgce] Binary data in RELOAD to XML?
On 8/20/2010 11:45 AM, Ralph Versteegen wrote: On 21 August 2010 02:24, Mike Caroncaron.m...@gmail.com wrote: On 8/20/2010 9:58 AM, Ralph Versteegen wrote: Currently reload2xml can't properly export binary data stored in strings in reload files. It would be nice to be able to hand edit xml and convert back. What's the preferred way to write binary?#nnn; escape codes, or that Base64 stuff? (Is that actually part of the xml standard?) Strictly speaking, we only really need to escape characters below 32. Everything else should be okay to write out. Follow-up question: what about bytes above 127? Is there a chance of confusing xml parsers into guessing UTF8 encoding? I suppose it doesn't matter too much, since we can force libxml2 to read with ASCII encoding (but it would be nice to know that we have to do so). The best way to handle this, actually, would be to add the XML header, which allows you to specify the encoding. In our case, it should look something like: ?xml version=1.0 encoding=iso-8859-1? In light of that, I would prefer the#; syntax, since it means that no one has to do any extra work to process the resulting XML file. If you used Base64 (which has nothing to do with XML), you'd have to mark it as such in order to distinguish it from a regular string that just so happens to look like Base64. OK, cool. I'll add that then soonish (unless you jumped on the chance to do so). No, go ahead. I had thought there was some semi-standard for base64 strings in xml: http://www.xml.com/pub/a/98/07/binary/binary.html That article just explores different ways of embedding binary data. There's no standard for this sort of thing. Base64 would be the closest thing to a proper solution, being used in different applications already (eg, embedded resources in HTML). The drawback is a flat 33% size increase, compared to the raw data. In comparison, the entity solution would be a 600% increase for encoded bytes, and 0% for anything else. (plus, no special handling required for parsers) The MIME idea proposed in that article is interesting, but even more complicated than Base64, and not really suited for our purposes. ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] Binary data in RELOAD to XML?
On 8/20/2010 12:41 PM, Ralph Versteegen wrote: On 21 August 2010 03:58, Mike Caroncaron.m...@gmail.com wrote: On 8/20/2010 11:45 AM, Ralph Versteegen wrote: On 21 August 2010 02:24, Mike Caroncaron.m...@gmail.comwrote: On 8/20/2010 9:58 AM, Ralph Versteegen wrote: Currently reload2xml can't properly export binary data stored in strings in reload files. It would be nice to be able to hand edit xml and convert back. What's the preferred way to write binary?#nnn; escape codes, or that Base64 stuff? (Is that actually part of the xml standard?) Strictly speaking, we only really need to escape characters below 32. Everything else should be okay to write out. Follow-up question: what about bytes above 127? Is there a chance of confusing xml parsers into guessing UTF8 encoding? I suppose it doesn't matter too much, since we can force libxml2 to read with ASCII encoding (but it would be nice to know that we have to do so). The best way to handle this, actually, would be to add the XML header, which allows you to specify the encoding. In our case, it should look something like: ?xml version=1.0 encoding=iso-8859-1? In light of that, I would prefer the#; syntax, since it means that no one has to do any extra work to process the resulting XML file. If you used Base64 (which has nothing to do with XML), you'd have to mark it as such in order to distinguish it from a regular string that just so happens to look like Base64. OK, cool. I'll add that then soonish (unless you jumped on the chance to do so). No, go ahead. Change of plan required. ' #0; is not permitted, however, as the null character is one of the control characters excluded from XML, even when using a numeric character reference.[14] An alternative encoding mechanism such as Base64 is needed to represent such characters. ' - Wikipedia And I just confirmed libxml2 spits. Damn. I guess that kind of makes sense, though. Maybe... emit a null/ element instead? Ugh. Or, Base64 it. Sigh. I had thought there was some semi-standard for base64 strings in xml: http://www.xml.com/pub/a/98/07/binary/binary.html That article just explores different ways of embedding binary data. There's no standard for this sort of thing. Base64 would be the closest thing to a proper solution, being used in different applications already (eg, embedded resources in HTML). The drawback is a flat 33% size increase, compared to the raw data. In comparison, the entity solution would be a 600% increase for encoded bytes, and 0% for anything else. (plus, no special handling required for parsers) The MIME idea proposed in that article is interesting, but even more complicated than Base64, and not really suited for our purposes. ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] Binary data in RELOAD to XML?
On 8/20/2010 12:52 PM, Ralph Versteegen wrote: On 21 August 2010 04:43, Mike Caroncaron.m...@gmail.com wrote: On 8/20/2010 12:41 PM, Ralph Versteegen wrote: On 21 August 2010 03:58, Mike Caroncaron.m...@gmail.comwrote: On 8/20/2010 11:45 AM, Ralph Versteegen wrote: On 21 August 2010 02:24, Mike Caroncaron.m...@gmail.com wrote: On 8/20/2010 9:58 AM, Ralph Versteegen wrote: Currently reload2xml can't properly export binary data stored in strings in reload files. It would be nice to be able to hand edit xml and convert back. What's the preferred way to write binary?#nnn; escape codes, or that Base64 stuff? (Is that actually part of the xml standard?) Strictly speaking, we only really need to escape characters below 32. Everything else should be okay to write out. Follow-up question: what about bytes above 127? Is there a chance of confusing xml parsers into guessing UTF8 encoding? I suppose it doesn't matter too much, since we can force libxml2 to read with ASCII encoding (but it would be nice to know that we have to do so). The best way to handle this, actually, would be to add the XML header, which allows you to specify the encoding. In our case, it should look something like: ?xml version=1.0 encoding=iso-8859-1? In light of that, I would prefer the#; syntax, since it means that no one has to do any extra work to process the resulting XML file. If you used Base64 (which has nothing to do with XML), you'd have to mark it as such in order to distinguish it from a regular string that just so happens to look like Base64. OK, cool. I'll add that then soonish (unless you jumped on the chance to do so). No, go ahead. Change of plan required. ' #0; is not permitted, however, as the null character is one of the control characters excluded from XML, even when using a numeric character reference.[14] An alternative encoding mechanism such as Base64 is needed to represent such characters. ' - Wikipedia And I just confirmed libxml2 spits. Damn. I guess that kind of makes sense, though. Maybe... emit anull/ element instead? Ugh. Or, Base64 it. Sigh. Well, I know next to nothing about XML, what's the idiomatic way to do that? Add a special attribute to nodes containing base64-encoded data, likefoo base=64assdfasdf234/foo ? No, the proper way to do this is something like this: basenode xmlns:reload=http://hamsterrepublic.com/RELOAD; foo reload:encoding=base64.../foo /basenode Note: that namespace URL doesn't have to exist, it just has to be unique. I had thought there was some semi-standard for base64 strings in xml: http://www.xml.com/pub/a/98/07/binary/binary.html That article just explores different ways of embedding binary data. There's no standard for this sort of thing. Base64 would be the closest thing to a proper solution, being used in different applications already (eg, embedded resources in HTML). The drawback is a flat 33% size increase, compared to the raw data. In comparison, the entity solution would be a 600% increase for encoded bytes, and 0% for anything else. (plus, no special handling required for parsers) The MIME idea proposed in that article is interesting, but even more complicated than Base64, and not really suited for our purposes. ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] SVN: pkmnfrk/3584 Finally added support for RELOAD delay loading: (see below for details)
On 02/06/2010 2:49 AM, Ralph Versteegen wrote: On 2 June 2010 10:33,subvers...@hamsterrepublic.com wrote: pkmnfrk 2010-06-01 15:33:51 -0700 (Tue, 01 Jun 2010) 1806 Finally added support for RELOAD delay loading: (see below for details) - Added new utility reloadtime.bas, which times exactly how long it takes to load any random RELOAD document. - reload2xml now passes optNoDelay, since it does not benefit from delay loading - reloadTest now includes two document comparison tests: first with no delay loading, and then again with delay loading. - xml2reload now display memory usage (again? I don't know if I ever committed this change), and frees the XML document before optimizing the RELOAD tree (I meant to commit this two days ago) Delay Loading: This means that nodes will only be loaded from disk as they are accessed. Depending on what the tree is being used for this can be a huge time saver. The ideal use-case for this feature would be the hypothetical Textbox lump, whereby only the textboxes that are referenced are actually loaded. This would speed up loading the document initially, with the tradeoff of spreading out the document load to later. Basically, it works like this: 1. When the document is loaded, the root node (including data) is loaded, but its children are not. 2. If its children are requested (eg, by calling GetFirstChild or AddChild or something), it will load all its kids (but not their children), and the requested operation can take place. 3. Repeat for each of those children that are requested. While this takes place, the document holds on to the file handle (so that the document cannot be modified or removed while it still refers to it), and releases it when the document is freed. The old behaviour of loading everything at the beginning can be restored by passing optNoDelay to RELOAD.LoadDocument. I've tested everything pretty thoroughly, but as always, if you find a way to break it (tmc:@), let me know. --- U wip/makereload.bat U wip/makereload.sh U wip/reload.bas U wip/reload.bi U wip/reload2xml.bas U wip/reloadtest.bas A wip/reloadtime.bas U wip/xml2reload.bas This looks great! Not yet subjected to stress testing ;) Two things: you forgot to update AddChild (funny, you actually gave it as an example :P), Hmm, you are correct. I will work on this tomorrow and there's no point marking a node without children as delayed-load. True. Strictly speaking, it doesn't matter at all, but it should be done for completeness. Also, now we should have function to force loading a node and all its descendants: if you want to read a textbox, you don't want to read from file every time you encounter a node with children. Yes, actually I was going to do this tomorrow. I should have mentioned that, but hey. -- Mike ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] SVN: james/3585 New RSAV loadgame support. Not enabled by default yet.
On 02/06/2010 2:43 PM, James Paige wrote: On Wed, Jun 02, 2010 at 02:17:25PM -0400, Mike Caron wrote: On 02/06/2010 5:08 AM, Ralph Versteegen wrote: On 2 June 2010 13:59,subvers...@hamsterrepublic.com wrote: james 2010-06-01 18:59:48 -0700 (Tue, 01 Jun 2010) 153 New RSAV loadgame support. Not enabled by default yet. Saving always saves in both SAV and RSAV format. To load from the RSAV, hold SHIFT when you load. --- U wip/moresubs.bas U wip/savegame.bas We're nearly there! Are we sure we want to store save games in %APP_DATA% on Windows by default instead of the game folder? I generally find it annoying when games do that, even if it might be correct in theory. For example, I was quite annoyed recently when I realised (on linux, but I'm not disputing putting saved games in ~/.ohrrpgce) that all versions of the same game (or even different games with the same filename) share the same save file. Plus, I think very few people will know to look in the APP_DATA directory when they want to access a save file, eg. to copy it to another computer. I try not to go in there, it's a mess of human-unfriendly data stores. For what it's worth, if we're going to store the saves in a central location, I'd rather they either go in a configurable location, or in My Documents\My Games\Ohrrpgce\* My documents can be located by calling SHGetSpecialFolderLocation, but that appears to be rather complicated, so I'll write a wrapper function. I kinda hate to create a new top-level folder in My Documents. Some users are particular about how they organize that folder. Here are the following games that did not share your sympathies, at least when they were installed on my computer: EA Games\The Sims 2 My Games\Age of Mythology My Games\Heroes of Might And Magic V My Games\Oblivion Puzzle Quest Puzzle Quest Galactrix RPG Maker XP RPG Maker VX Sim City 4 Sim City Societies (they didn't even bother to put these under EA Games as above) And, plenty of other non-game programs. I think that in this age, it's a moot point. My suggestion about making it configurable still stands, though. Many other games store save games in %APP_DATA% I do see the problem with different versions sharing the same save data. It means you can no longer put a copy of the game in a different folder with a different copy of game.exe and expect it to get its own save file. I know %APP_DATA% seems like a kinda lousy default location for game authors, but I think it is a great default location for players who only play games and don't make them (which should in theory, (but perhaps not always practice) be the majority of users) What if we put save games in: same\path\as\the\rpg\file\gamename.saves\#.rsav And then fall back to %APP_DATA%\OHRRPGCE\gamename\saves\#.rsav only if the first choice is non-writeable? --- James Paige ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org -- Mike ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] More RELOAD tests
On 01/06/2010 2:23 AM, Ralph Versteegen wrote: On 1 June 2010 16:37, Mike Caroncaron.m...@gmail.com wrote: On 01/06/2010 12:25 AM, Ralph Versteegen wrote: On 1 June 2010 16:15, Mike Caroncaron.m...@gmail.comwrote: On 01/06/2010 12:10 AM, Ralph Versteegen wrote: On 1 June 2010 15:57, Mike Caroncaron.m...@gmail.com wrote: On 31/05/2010 11:44 PM, Ralph Versteegen wrote: On 31 May 2010 11:58, Ralph Versteegenteeem...@gmail.comwrote: On 31 May 2010 10:20, Mike Caroncaron.m...@gmail.comwrote: On 30/05/2010 6:01 PM, Ralph Versteegen wrote: On 31 May 2010 08:57, Mike Caroncaron.m...@gmail.com wrote: D:\ohrrpgcedir gentest.xml 30/05/2010 04:52 AM 237,804,960 gentest.xml D:\ohrrpgcexml2reload gentest.xml gentest.rld Loaded XML document in 20857 ms Memory usage: 301 Parsed XML document in 17049 ms Memory usage: 573142499 Freed XML document in 35378 ms Optimised document in 31149 ms Memory usage: 428328275 Serialized document in 8557 ms Tore down memory in 173 ms Finished in 116510 ms D:\ohrrpgcedir gentest.rld 30/05/2010 04:31 PM 203,271,865 gentest.rld And, here's a profile of CPU and Memory usage while doing the above: http://taleotc.com/Reloadprofile.png Note that on the CPU graph, Green is Usermode, while Red is Kernel mode. And that my CPU thinks it's dual core, so you should consider the 50% mark to be max. Hyperthreading? Aka, Make applications think the CPU can multiple things at once, when in reality it can't!. I guess Hyperthreading is more marketable, though. :) Further, you should care more about the Commit memory, rather than th Physical memory. -- Mike Yikes! Nearly twice as long to free as to load the XML document. Looks like the memory allocator thrashed a lot while freeing memory, which is odd because I would have expected xml nodes to be deleted in about the same order they were created. Maybe libxml deletes attribute nodes after deleting children. (Yes! I checked, and my hunch is correct!) I assume that 3.0GB marks the total virtual memory committed on the machine, and xml2relaod used roughly 1.5GB? Yes, that's correct. What does that xml file look like? It's interesting that the resulting .reld is nearly the same size as the original! That 64MB test case compressed to half the size. I used this program to generate a document: http://www.xml-benchmark.org/downloads.html It seems to have a lot of content in the form of text nodes. Alas, I can't find the place I got the original document, since it was basically more of the same. I suppose we could just cat the 64-meg document to itself, and do a tiny bit of editing :D Serialisation could be a bit faster (you can always increase the buffer size to speed it up, going from 32kb to 64kb caused about a 20% improvement), but mostly I'm happy with that. Honestly, that's the least of my concerns. That's always going to scale linearly with the resulting document size. I'm more concerned, now, with the thing we haven't really been profiling: Loading these stupidly large documents again. I modified reloadtest to load it, and it ain't pretty: Well, I told you to profile them, and have been doing so myself; who cares how long it takes to convert an XML? And speaking of profiling, I tried profiling but got totally useless results missing nearly all running time. Test #19: loadHugeDocument... Pass Took 50 s Test #20: freeHugeDocument... Pass Took 297 ms And, just for completion's sake, here's the 64 Meg-document's RELOAD: Test #19: loadHugeDocument... Pass Took 19 s Test #20: freeHugeDocument... Pass Took 53 ms Ouch. What the devil? Here are my times, after warming up of course, on windows XP this time, with a private heap: reload2reload medline.reld medline2.reld Loaded document in 12743 ms Serialized document in 1062 ms Tore down memory in 17 ms Finished in 13824 ms But! here's what it was in linux (back then the buffer was smaller, I think now the serialization time is about 2.5s): bash-3.1$ time reload2reload medline2.reld medline2_.reld Loaded document in 3724 ms Serialized document in 3506 ms Tore down memory in 597 ms Finished in 7838 ms real0m7.892s user0m5.276s sys 0m0.440s So, it looks like Linux is being way too paranoid about flushing the file to disk as its written, or something like that (there are so many options in the kernel for tuning file systems; looks like they're not set for performance in this case on my machine), and conversely Windows is doing something REALLY stupid when loading it in. I mean, the reload file is read in a linear manner, what on earth could the problem be? Argh! We may have to write a file buffering layer for reading files as well! Or maybe the problem is in allocating nodes? We could test this by loading the file into memory before attempting to load it. OK, I tested and confirmed that file reads are the problem. I loaded the whole file into memory and then read from that (inmem). Then I also
Re: [Ohrrpgce] SVN: teeemcee/3580 Converted Reload.SerializeBin to use BufferedFile instead of stdio FILEs
On 30/05/2010 2:33 AM, subvers...@hamsterrepublic.com wrote: teeemcee 2010-05-29 23:33:46 -0700 (Sat, 29 May 2010) 134 Converted Reload.SerializeBin to use BufferedFile instead of stdio FILEs. It now runs nearly 400 times faster on that 64MB test case. --- U wip/makereload.bat U wip/makereload.sh U wip/reload.bas U wip/util.bas You win! No one will ever be able to match 8 commits in the space of 17 seconds! When I check out what you've done, I will comment on something relevant. -- Mike ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] SVN: teeemcee/3580 Converted Reload.SerializeBin to use BufferedFile instead of stdio FILEs
On 30/05/2010 3:02 AM, Ralph Versteegen wrote: On 30 May 2010 18:37, Mike Caroncaron.m...@gmail.com wrote: On 30/05/2010 2:33 AM, subvers...@hamsterrepublic.com wrote: teeemcee 2010-05-29 23:33:46 -0700 (Sat, 29 May 2010) 134 Converted Reload.SerializeBin to use BufferedFile instead of stdio FILEs. It now runs nearly 400 times faster on that 64MB test case. --- U wip/makereload.bat U wip/makereload.sh U wip/reload.bas U wip/util.bas You win! No one will ever be able to match 8 commits in the space of 17 seconds! When I check out what you've done, I will comment on something relevant. -- Mike Git is great. I have some HSpeak/script debugger work I could have finished off and committed in several parts as well, but I'll leave it for another day. Ah, you lazy git. That explains it. ;) I meant to say somewhere: maybe you want to modify SerializeXML so that it outputs valid XML (eg, print asNULL andfoo =2 as foo value=2, and make xml2reload translate back) or whatever, I left that up to you. I mostly changed it because it was misprinting slicetrees. I'm not sure what to do in the face of null nodes, since and NULL are not the same thing. I'm not even sure than a Node with no name is even legal (it's quite possibly undefined). I don't agree with the change to nodes with both values and children. The exact XML representation of that is: node value children / children / children / /node I don't believe in magic attributes or node names, so if I were going to do that, I'd spit them out as: rootnodewhatever xmlns:RELOAD=http://someurlthatIhaventdecidedyet.com; ... RELOAD:null / /rootnodewhatever I'm not sure that I want to commit to doing this yet, as I'm working on another big project with my spare time... Maybe soon though. If I become inspired. -- Mike ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] Reload.SerializeXML
On 30/05/2010 2:53 AM, Ralph Versteegen wrote: On 28 May 2010 13:04, Mike Caroncaron.m...@gmail.com wrote: On 27/05/2010 8:38 PM, Ralph Versteegen wrote: On 28 May 2010 11:25, James Paigeb...@hamsterrepublic.comwrote: On Thu, May 27, 2010 at 07:07:38PM -0400, Mike Caron wrote: On 27/05/2010 6:38 PM, James Paige wrote: Mike, was there any special reason why Reload.SerializeXML uses print statements rather than writing to a file? --- James I wrote it as a debugging function. If it wrote to a file, then I wouldn't be able to see it on screen in reloadtest! :) -- Mike I guess what I am really looking for is a reload2xml command-line tool so I can easily debug reload files on disk.. and actually it doesn't matter that SerializeXML prints to the console, because I could just do reload2xml somefile.reldsomefile.xml --- James Writing to standard output is the Unix way anyway! Speaking of xml, Mike mentioned a couple weeks ago that the Reload code grinds to a halt when processing some translated 64MB xml document. Since I enjoy optimisation to a rather evil degree, I'd like to look at it sometime. What are some good testcases? Right at this moment, I don't feel like touching any of that stuff, so go nuts. A few tips: - I don't think the private heap has anything to do with potential performance issues. They're the same calls being done by the runtime, just in a different memory block. - One thing I never thought about doing is compiling with the -profile switch. - I never did any formal performance tuning, being as I subscribe to the make it work, then make it fast camp. All that ZString crap was to alleviate my fears of memory corruption caused by having Strings in UDTs. This is the document I mentioned: (warning: ~5 Megs compressed, 64 Megs uncompressed) http://taleotc.com/medline08n0059.zip I haven't looked to closely at the structure of the document, but it seems to be a fairly average, if large, dataset. Other than that, Google isn't being very friendly. Querying xml test documents lists a bunch of XML Tutorials and Unit testing stuff, while large xml test documents seems to focus on *really* big documents (like, 1 Gb), for which I suspect the RELOAD file format would break down :) -- Mike So here are my results (my machine is a 7 year old 3GHz pentium 4 with 1GB of RAM): Beforehand: (I believe that this version of xml2reload did not include the attributes, which add about 30% to the .reld size) bash-3.1$ xml2reload ../medline08n0059.xml medline.reld Loaded XML document in 4839 ms Parsed XML document in 74792 ms Optimised document in 9859 ms Serialized document in 1022907 ms Tore down memory in 3891 ms Finished in 1116305 ms bash-3.1$ time reload2reload plotdict.reld plotdict2.reld Loaded document in 30 ms Serialized document in 2851 ms Tore down memory in 1 ms Finished in 2899 ms real0m2.919s user0m0.112s sys 0m0.188s (where reload2reload is obviously just a 10 liner) Afterwards:=== bash-3.1$ time xml2reload ../medline08n0059.xml medline.reld Loaded XML document in 4207 ms Parsed XML document in 6023 ms Optimised document in 10631 ms Serialized document in 2623 ms Tore down memory in 3097 ms Finished in 26596 ms real0m26.699s user0m24.018s sys 0m1.100s (Also, running reload2reload on medline.reld (which is 31MB) required about 142MB of memory) bash-3.1$ reload2reload plotdict.reld plotdict2.reld Loaded document in 13 ms Serialized document in 21 ms Tore down memory in 8 ms Finished in 60 ms After running a few tests, I discovered the major bottleneck in the private heap implementation: A tiny bit of debugging code which didn't do a whole lot... except enumerate the entire heap on every allocation. When I got rid of it, everything sped up enormously! With private heap: Loading XML document... Loaded XML document in 5139 ms Starting memory usage: 301 Parsing document... Parsed XML document in 6765 ms Memory usage: 237104215 Optimised document in 11849 ms Memory usage: 157105826 Serialized document in 1138 ms Tore down memory in 1974 ms Finished in 28314 ms Without private heap: Loading XML document... Loaded XML document in 4974 ms Starting memory usage: 0 Parsing document... Parsed XML document in 6989 ms Memory usage: 0 Optimised document in 12147 ms Memory usage: 0 Serialized document in 1207 ms Tore down memory in 3257 ms Finished in 28577 ms I ran each a few times, it seems that they're all about the same. the only big difference I can see is the Tore down memory phase, which is indeed faster with the private heap (by 33%!) as intended. I suspect, though, that it's not enough to save it. Oh well, it was fun to get working, now it'll be fun to excise! Anyway, the whole process used ~650 Megs of RAM, which is broken up as such: - Approximately 423 Megs used by libxml2 to load the XML document - Exactly 226.12 Megs (as seen above) used by RELOAD to load the unoptimized RELOAD document I am going
Re: [Ohrrpgce] SVN: teeemcee/3580 Converted Reload.SerializeBin to use BufferedFile instead of stdio FILEs
On 30/05/2010 3:19 AM, Ralph Versteegen wrote: On 30 May 2010 19:08, Mike Caroncaron.m...@gmail.com wrote: On 30/05/2010 3:02 AM, Ralph Versteegen wrote: On 30 May 2010 18:37, Mike Caroncaron.m...@gmail.comwrote: On 30/05/2010 2:33 AM, subvers...@hamsterrepublic.com wrote: teeemcee 2010-05-29 23:33:46 -0700 (Sat, 29 May 2010) 134 Converted Reload.SerializeBin to use BufferedFile instead of stdio FILEs. It now runs nearly 400 times faster on that 64MB test case. --- U wip/makereload.bat U wip/makereload.sh U wip/reload.bas U wip/util.bas You win! No one will ever be able to match 8 commits in the space of 17 seconds! When I check out what you've done, I will comment on something relevant. -- Mike Git is great. I have some HSpeak/script debugger work I could have finished off and committed in several parts as well, but I'll leave it for another day. Ah, you lazy git. That explains it. ;) I meant to say somewhere: maybe you want to modify SerializeXML so that it outputs valid XML (eg, printasNULLandfoo =2as foo value=2, and make xml2reload translate back) or whatever, I left that up to you. I mostly changed it because it was misprinting slicetrees. I'm not sure what to do in the face of null nodes, since andNULL are not the same thing. I'm not even sure than a Node with no name is even legal (it's quite possibly undefined). According to emacs' builtin XML validator (ha) it's not legal in XML, but it's certainly legal in RELOAD. (Not sure which you're wondering.) I meant in RELOAD. I know it's possible in RELOAD, but I never really gave it any thought. And, I see that I optimized this case... I don't agree with the change to nodes with both values and children. The exact XML representation of that is: node value children / children / children / /node I'm not sure that's exact. What is node value children / othervalue children / /node Technically, the structure in XML is: node xml:textvalue/xml:text children / xml:textothervalue/xml:text children / /node If you recall way back, this was the way RELOAD was going to work: A node could have children, or data, not both. I made that change to avoid confusion betweennode =foobbar/b/node and what you get after running nodefoobbar/b/node through xml2reload. Although admittedly the --debug option I added to reload2xml is enough to distinguish those two. I'm not really sure what the difference would be, assuming I'm interpreting your =foo notation correctly. I don't believe in magic attributes or node names, so if I were going to do that, I'd spit them out as: rootnodewhatever xmlns:RELOAD=http://someurlthatIhaventdecidedyet.com; ... RELOAD:null / /rootnodewhatever You lost me. Would this actually be useful (OK, so I suggested it in the first place, so what) I meant, stick any magic attributes and nodes in a namespace to avoid misinterpreting a document which contains those names. I'm not sure that I want to commit to doing this yet, as I'm working on another big project with my spare time... Maybe soon though. If I become inspired. -- Mike And I was meant to be working on term projects and assignments instead of RELOAD. ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org -- Mike ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] Reload.SerializeXML
On 30/05/2010 4:33 AM, Ralph Versteegen wrote: On 30 May 2010 20:23, Mike Caroncaron.m...@gmail.com wrote: On 30/05/2010 4:19 AM, Ralph Versteegen wrote: On 30 May 2010 20:10, Mike Caroncaron.m...@gmail.comwrote: On 30/05/2010 4:03 AM, Ralph Versteegen wrote: On 30 May 2010 19:30, Mike Caroncaron.m...@gmail.com wrote: On 30/05/2010 2:53 AM, Ralph Versteegen wrote: On 28 May 2010 13:04, Mike Caroncaron.m...@gmail.comwrote: On 27/05/2010 8:38 PM, Ralph Versteegen wrote: On 28 May 2010 11:25, James Paigeb...@hamsterrepublic.com wrote: On Thu, May 27, 2010 at 07:07:38PM -0400, Mike Caron wrote: On 27/05/2010 6:38 PM, James Paige wrote: Mike, was there any special reason why Reload.SerializeXML uses print statements rather than writing to a file? --- James I wrote it as a debugging function. If it wrote to a file, then I wouldn't be able to see it on screen in reloadtest! :) -- Mike I guess what I am really looking for is a reload2xml command-line tool so I can easily debug reload files on disk.. and actually it doesn't matter that SerializeXML prints to the console, because I could just do reload2xml somefile.reld somefile.xml --- James Writing to standard output is the Unix way anyway! Speaking of xml, Mike mentioned a couple weeks ago that the Reload code grinds to a halt when processing some translated 64MB xml document. Since I enjoy optimisation to a rather evil degree, I'd like to look at it sometime. What are some good testcases? Right at this moment, I don't feel like touching any of that stuff, so go nuts. A few tips: - I don't think the private heap has anything to do with potential performance issues. They're the same calls being done by the runtime, just in a different memory block. - One thing I never thought about doing is compiling with the -profile switch. - I never did any formal performance tuning, being as I subscribe to the make it work, then make it fast camp. All that ZString crap was to alleviate my fears of memory corruption caused by having Strings in UDTs. This is the document I mentioned: (warning: ~5 Megs compressed, 64 Megs uncompressed) http://taleotc.com/medline08n0059.zip I haven't looked to closely at the structure of the document, but it seems to be a fairly average, if large, dataset. Other than that, Google isn't being very friendly. Querying xml test documents lists a bunch of XML Tutorials and Unit testing stuff, while large xml test documents seems to focus on *really* big documents (like, 1 Gb), for which I suspect the RELOAD file format would break down :) -- Mike So here are my results (my machine is a 7 year old 3GHz pentium 4 with 1GB of RAM): Beforehand: (I believe that this version of xml2reload did not include the attributes, which add about 30% to the .reld size) bash-3.1$ xml2reload ../medline08n0059.xml medline.reld Loaded XML document in 4839 ms Parsed XML document in 74792 ms Optimised document in 9859 ms Serialized document in 1022907 ms Tore down memory in 3891 ms Finished in 1116305 ms bash-3.1$ time reload2reload plotdict.reld plotdict2.reld Loaded document in 30 ms Serialized document in 2851 ms Tore down memory in 1 ms Finished in 2899 ms real0m2.919s user0m0.112s sys 0m0.188s (where reload2reload is obviously just a 10 liner) Afterwards:=== bash-3.1$ time xml2reload ../medline08n0059.xml medline.reld Loaded XML document in 4207 ms Parsed XML document in 6023 ms Optimised document in 10631 ms Serialized document in 2623 ms Tore down memory in 3097 ms Finished in 26596 ms real0m26.699s user0m24.018s sys 0m1.100s (Also, running reload2reload on medline.reld (which is 31MB) required about 142MB of memory) bash-3.1$ reload2reload plotdict.reld plotdict2.reld Loaded document in 13 ms Serialized document in 21 ms Tore down memory in 8 ms Finished in 60 ms After running a few tests, I discovered the major bottleneck in the private heap implementation: A tiny bit of debugging code which didn't do a whole lot... except enumerate the entire heap on every allocation. Ah, I was wondering why you said that in tests it took 20 min to optimise the parsed XML when it only took me 12 seconds. When I got rid of it, everything sped up enormously! With private heap: Loading XML document... Loaded XML document in 5139 ms Starting memory usage: 301 Parsing document... Parsed XML document in 6765 ms Memory usage: 237104215 Optimised document in 11849 ms Memory usage: 157105826 Serialized document in 1138 ms Tore down memory in 1974 ms Finished in 28314 ms Without private heap: Loading XML document... Loaded XML document in 4974 ms Starting memory usage: 0 Parsing document... Parsed XML document in 6989 ms Memory usage: 0 Optimised document in 12147 ms Memory usage: 0 Serialized document in 1207 ms Tore down memory in 3257 ms Finished in 28577 ms Interesting that in those runs it took less than half as long to serialise as on my machine
Re: [Ohrrpgce] Reload.SerializeXML
On 30/05/2010 4:33 AM, Ralph Versteegen wrote: On 30 May 2010 20:23, Mike Caroncaron.m...@gmail.com wrote: On 30/05/2010 4:19 AM, Ralph Versteegen wrote: On 30 May 2010 20:10, Mike Caroncaron.m...@gmail.comwrote: On 30/05/2010 4:03 AM, Ralph Versteegen wrote: On 30 May 2010 19:30, Mike Caroncaron.m...@gmail.com wrote: On 30/05/2010 2:53 AM, Ralph Versteegen wrote: On 28 May 2010 13:04, Mike Caroncaron.m...@gmail.comwrote: On 27/05/2010 8:38 PM, Ralph Versteegen wrote: On 28 May 2010 11:25, James Paigeb...@hamsterrepublic.com wrote: On Thu, May 27, 2010 at 07:07:38PM -0400, Mike Caron wrote: On 27/05/2010 6:38 PM, James Paige wrote: Mike, was there any special reason why Reload.SerializeXML uses print statements rather than writing to a file? --- James I wrote it as a debugging function. If it wrote to a file, then I wouldn't be able to see it on screen in reloadtest! :) -- Mike I guess what I am really looking for is a reload2xml command-line tool so I can easily debug reload files on disk.. and actually it doesn't matter that SerializeXML prints to the console, because I could just do reload2xml somefile.reld somefile.xml --- James Writing to standard output is the Unix way anyway! Speaking of xml, Mike mentioned a couple weeks ago that the Reload code grinds to a halt when processing some translated 64MB xml document. Since I enjoy optimisation to a rather evil degree, I'd like to look at it sometime. What are some good testcases? Right at this moment, I don't feel like touching any of that stuff, so go nuts. A few tips: - I don't think the private heap has anything to do with potential performance issues. They're the same calls being done by the runtime, just in a different memory block. - One thing I never thought about doing is compiling with the -profile switch. - I never did any formal performance tuning, being as I subscribe to the make it work, then make it fast camp. All that ZString crap was to alleviate my fears of memory corruption caused by having Strings in UDTs. This is the document I mentioned: (warning: ~5 Megs compressed, 64 Megs uncompressed) http://taleotc.com/medline08n0059.zip I haven't looked to closely at the structure of the document, but it seems to be a fairly average, if large, dataset. Other than that, Google isn't being very friendly. Querying xml test documents lists a bunch of XML Tutorials and Unit testing stuff, while large xml test documents seems to focus on *really* big documents (like, 1 Gb), for which I suspect the RELOAD file format would break down :) -- Mike So here are my results (my machine is a 7 year old 3GHz pentium 4 with 1GB of RAM): Beforehand: (I believe that this version of xml2reload did not include the attributes, which add about 30% to the .reld size) bash-3.1$ xml2reload ../medline08n0059.xml medline.reld Loaded XML document in 4839 ms Parsed XML document in 74792 ms Optimised document in 9859 ms Serialized document in 1022907 ms Tore down memory in 3891 ms Finished in 1116305 ms bash-3.1$ time reload2reload plotdict.reld plotdict2.reld Loaded document in 30 ms Serialized document in 2851 ms Tore down memory in 1 ms Finished in 2899 ms real0m2.919s user0m0.112s sys 0m0.188s (where reload2reload is obviously just a 10 liner) Afterwards:=== bash-3.1$ time xml2reload ../medline08n0059.xml medline.reld Loaded XML document in 4207 ms Parsed XML document in 6023 ms Optimised document in 10631 ms Serialized document in 2623 ms Tore down memory in 3097 ms Finished in 26596 ms real0m26.699s user0m24.018s sys 0m1.100s (Also, running reload2reload on medline.reld (which is 31MB) required about 142MB of memory) bash-3.1$ reload2reload plotdict.reld plotdict2.reld Loaded document in 13 ms Serialized document in 21 ms Tore down memory in 8 ms Finished in 60 ms After running a few tests, I discovered the major bottleneck in the private heap implementation: A tiny bit of debugging code which didn't do a whole lot... except enumerate the entire heap on every allocation. Ah, I was wondering why you said that in tests it took 20 min to optimise the parsed XML when it only took me 12 seconds. When I got rid of it, everything sped up enormously! With private heap: Loading XML document... Loaded XML document in 5139 ms Starting memory usage: 301 Parsing document... Parsed XML document in 6765 ms Memory usage: 237104215 Optimised document in 11849 ms Memory usage: 157105826 Serialized document in 1138 ms Tore down memory in 1974 ms Finished in 28314 ms Without private heap: Loading XML document... Loaded XML document in 4974 ms Starting memory usage: 0 Parsing document... Parsed XML document in 6989 ms Memory usage: 0 Optimised document in 12147 ms Memory usage: 0 Serialized document in 1207 ms Tore down memory in 3257 ms Finished in 28577 ms Interesting that in those runs it took less than half as long to serialise as on my machine
[Ohrrpgce] More RELOAD tests
D:\ohrrpgcedir gentest.xml 30/05/2010 04:52 AM 237,804,960 gentest.xml D:\ohrrpgcexml2reload gentest.xml gentest.rld Loaded XML document in 20857 ms Memory usage: 301 Parsed XML document in 17049 ms Memory usage: 573142499 Freed XML document in 35378 ms Optimised document in 31149 ms Memory usage: 428328275 Serialized document in 8557 ms Tore down memory in 173 ms Finished in 116510 ms D:\ohrrpgcedir gentest.rld 30/05/2010 04:31 PM 203,271,865 gentest.rld And, here's a profile of CPU and Memory usage while doing the above: http://taleotc.com/Reloadprofile.png Note that on the CPU graph, Green is Usermode, while Red is Kernel mode. And that my CPU thinks it's dual core, so you should consider the 50% mark to be max. Further, you should care more about the Commit memory, rather than th Physical memory. -- Mike ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
Re: [Ohrrpgce] what to do with saved NPC and map state when the map gets edited
On 30/05/2010 5:51 PM, James Paige wrote: So seeing as I will be soon working on NPC and map state saving as a NEW feature, (since the limited support that was already present was broken for 11 or 12 years and nobody noticed (and why should they? Nobody but me knew it was supposed to be a feature)) I digress. Suppose you release your first demo. Everybody plays it, and then saves their game at the end. Then a few weeks later you release your second demo, which among other things includes heavy revision of the NPCs on the same map where the previous demo ends. Everybody who continues from their old save says the NPCs are all crazy! They are in the wrong place! Some of them are missing How do we resolve this situation? For a while I was thinking we should have a version number of some kind on each map that the game author could manually incremment, which would invalidate old saved state for that map. So that is one option. Another option is that each NPC would have a timestamp tracking when it was placed/edited in the map editor. That could be compared to a timestamp on the save game to determine if the NPC's saved position should be restored. And then there is tilemap data... Thoughts? --- James I suggest a bitset that optionally turns this feature on. If it's off, it behaves as it always has (i.e, not saving map state at all). If it's on, then it unconditionally loads the map state on load, and requires that the player leave the map and come back to get the new changes. I don't think it would be a good idea to mix and match NPCs, because then you get into a situation where you're not sure what data to base a reset on. For example, take a nosy guard NPC who is meant to stick near the player while he selects exactly one item from the treasury, and open the door for him when he's done. The player then saves in the treasury. But, the NPC gets moved in the editor, and subsequently gets reset when the game is loaded. Now, hypothetically, the player is stuck, since he can't talk to the NPC to get out! With the existing behaviour, this is still possible, but it's expected, since the guard will always be reset, so you wouldn't do something stupid like let the player save there. Similarly, in the DS Pokémon games, it saves the state of NPCs that are in the same sector (32x32 square) as you. When you load, it loads everything exactly as it was, even if it should be different (eg, if an NPC is supposed to be elsewhere due to it being a different time). This is to prevent the player getting stuck on top of a cut tree or something. That's just my two cents. -- Mike ___ Ohrrpgce mailing list ohrrpgce@lists.motherhamster.org http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
