Filed enhancement request 8133
I filed this to ask for 'replace' to have an additional option - to work like the Sed /s command. If you aren't familiar with this relic s/oldtext/newtext/ will replace only the first occurrence of oldtext in a line. In all lines of the file, or the specified subset of the file. s/oldtext/newtext/g will replace all occurrences of oldtext in the file, or the specified subset. Probably the ability to specify subsets is not important any more, there are easy ways to get the same effect. But the ability to hit the first occurrence only in a line would be nice. I'm sure this is not a super high priority for anyone, but it would be a nice addition to the string manipulation abilities, save having to roll one's own, and bring Rev closer to being a real Swiss Army knife. Peter ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution
Re: [On-Rev] Photo Gallery problem
In your code, on the page see script http://kenjikojima.com/Photo/GreatLawn/thumbnail.irev?file=thumbnail.ir ev in the lines 22-23 you write: replace with in tScript replace with in tScript is there a typo? Nonetheless, that works -- and generating on the fly a code page is a great idea :-) Those lines of code will be replacingwith their HTML entities lt; and gt; but the code display translates them back. I had the same problem. Andre has managed to get his script lister to show the entities, but I haven't managed to work out how yet. And yes, having each page include the ability to list it's own code is a very neat idea. Cheers, Sarah ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution
Re: [On-Rev] Photo Gallery problem
On Fri, Jun 26, 2009 at 5:35 PM, Sarah Reicheltsarah.reich...@gmail.com wrote: In your code, on the page see script http://kenjikojima.com/Photo/GreatLawn/thumbnail.irev?file=thumbnail.ir ev in the lines 22-23 you write: replace with in tScript replace with in tScript is there a typo? Nonetheless, that works -- and generating on the fly a code page is a great idea :-) Those lines of code will be replacingwith their HTML entities lt; and gt; but the code display translates them back. I had the same problem. Andre has managed to get his script lister to show the entities, but I haven't managed to work out how yet. Worked it out... If you replace ampersands too, then the lt; and gt; entities will display correctly inside a pre tag. Not quite sure why this works, but it does, so instead of 2 replace lines, use these 3: replace with amp; in tScript replace with lt; in tScript replace with gt; in tScript Cheers, Sarah ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution
Re: Q: dynamic usage of regex in a repeat loop
I don't have an answer, and was hoping someone else might.
The only thing that might work would be to dig deeper into regex themselves.
I believe that there are ways of constructing regex queries so that
one can subsequently pick out the occurrences where a hit occurred.
Bernard
2009/6/25 Franz Böhmisch boehmi...@animabit.de:
Hello I need sth like the following
repeat for each line regexline in field regexlist
try
put matchText(einzelinserat,regexline,region,rooms,ground,price,remarks) into
check
end try
if check = true then
...
end if
end repeat
with regex like
([a-z]{1}) ([0-9,]{1,2}) ([0-9,.]{2,5}) ([0-9,.]+)(.*)
()()([0-9,.]{2,5}) ([0-9,.]+)(.*)
()()(^[0-9,]{1,3}) (\d+) ([0-9.,]+)(.*)
([A-ZÄÖÜ]{1,3}) ([0-9,]{1,2}) ([0-9]{1,5}) ([0-9,.]{1,5})
in the list.
Is it possible to make the matchtext part more flexible for allowing to find
items not in the focus at programming time (without using evaluate ...).
in a kind of
put matchText(einzelinserat,regexline,variablelist) into check
Any ideas?
Regards, Franz
___
use-revolution mailing list
use-revolution@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription
preferences:
http://lists.runrev.com/mailman/listinfo/use-revolution
___
use-revolution mailing list
use-revolution@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription
preferences:
http://lists.runrev.com/mailman/listinfo/use-revolution
Re: Q: dynamic usage of regex in a repeat loop using do and dynamic
hi, thanks for thinking!
Q: dynamic usage of regex in a repeat loop
My current status is to use the do command and the construction of the runrev
command at runtime, which might slow down a little bit because it has to be
done anytime.
--
content of field auftrag:
put matchText(field original,$regex,$variablelist) into decision
content of field eval:
([A-ZÄÖÜ]{1,3}) ([0-9,]{1,5}) ([0-9,.]+)(.*)
region, zimmerzahl,wohnfläche,preis,bemerkung
script:
---
repeat
..
try
put field auftrag into befehl;
replace $regex with line 1 of field eval in befehl;
replace $variablelist with line 2 of field eval in befehl;
do befehl
end try
..
end repeat
Suggestion for improvement in next versions might be the possibility to use an
array variable to get the values of the () variables in array[1] etc.
matchtext(text,regex,array)
matchtext(field orig,regexline,resultarray)
Regards, Franz
___
use-revolution mailing list
use-revolution@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription
preferences:
http://lists.runrev.com/mailman/listinfo/use-revolution
Re: my sqlite code is slow :(
I'll add this to the things I need to try, but that'd need quite a fancy values management, as to not insert the wrong thing into the wrong insert. Also I'm currently on a different project ;) Thanks Björnke On 26 Jun 2009, at 07:35, Bill Marriott wrote: Does SQLite support multi-line SQL statements via revExecuteSQL? You might try assembling the whole operation in a variable, BEGIN INSERT INSERT INSERT ... COMMIT and see how that flies ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution
Re: Wonky revTools on Linux
And version 5 has just been released. I'm downloading it now, looking forward to trialing the new features. For info too, they have lots and lots of video tutorials and a thriving community of helpers who can help you learn and get the best from it. It can also import and export Illustrator format (subject to certain limitations), and works with many Photoshop Plugins. Bill Marriott wrote: We use the Windows version of Xara Xtreme for probably half of our marketing production. It's a lot of fun, an easy UI, and quite a bit faster than Inkscape on Linux. It has capabilities that rival or exceed Illustrator (and is also an order of magnitude faster). And the full commercial version for Windows is just $89. I use it for a lot of my UI work and highly recommend it. -- View this message in context: http://www.nabble.com/Wonky-revTools-on-Linux-tp24182796p24220060.html Sent from the Revolution - User mailing list archive at Nabble.com. ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution
Re: Error on shutdown
I use this script a lot. I have also found it necessary to add revStartCentre to the list of stacks put into myDontClose, otherwise it opens and shuts itself again when your project exits, in the IDE. Mark Schonewille-3 wrote: Hi Scott, I'm not sure about this, because usually I see a different type of error on shutdown, but the following script might help: http://runrev.info/Quit%20your%20standalone.htm . Just adjust it to your needs (e.g. by removing the MySQL part or by adding code for custom externals if necessary). -- Best regards, Mark Schonewille -- View this message in context: http://www.nabble.com/Error-on-shutdown-tp24042255p24221552.html Sent from the Revolution - User mailing list archive at Nabble.com. ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution
Re: Unwanted characters from pasted-in text.
Thanks Björnke! that indeed is the magic incantation thanks again - Stephen Barncard San Francisco http://barncard.com 2009/6/25 Björnke von Gierke b...@mac.com I'm not sure, but those could be utf-8 chars, did you try to convert from utf8? function revText theText return unidecode(uniencode(theText,utf8)) end revText ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution
[OT] Shift-Lock
I really miss the old SHIFT-LOCK key: http://mathewson.110mb.com/kp.html does anybody know how to convert the Caps-Lock key to a Shift-Lock key? ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution
Inefficient code
I have written this code a few different ways. Each seems to be equally inefficient. I am attempting to compare to images, pixel by pixel, and record the differences. ON mouseUp pMouseBtnNo put 0 into c set the startvalue of scrollbar Progress to 0 set the endvalue of scrollbar Progress to the length of imagedata of image Alpha put the imagedata of image Alpha into idataalpha put the imagedata of image Beta into idatabeta REPEAT FOR each char myChar in idataalpha set the thumbposition of scrollbar Progress to c IF myChar is not char (c) of idatabeta THEN put c char (c) of idatabeta return after hAll END IF put c+1 into c END repeat put hAll into field diff END mouseUp It takes many minutes to process a 200x200 image. I want this code to eventually compare full screen-captures, so if there is any way to speed it up, I am open for advice. Thanks Bert ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution
Re: Rev 4 Preview Webinar - Video Link
Bill: Be my guest. We've already spent dozens of hours trying to work around this. Wow. This took several minutes: http://imp.on-rev.com/webinars/index.html Plays on iPods too. (And Zunes. And anything else that plays H.264.) The GoToWebinar software we use for our webinars records *only* in WMV - So I can suggest you use something without that serious limitation? file size ... loss in quality My priority would be compatibility, especially since it's a web-inar. (BTW, Web Windows.) Note also that you said you prefer mp4, as if this were a universal format. It is, for all intents and purposes. It's the new MPG-1. It is not owned by Microsoft, and it is not owned by Apple or any other OS vendor. Lots of people on the web think that's important. But most Windows users cannot view that format, without having QuickTime or other additional software installed Software has to be installed to do anything on a computer. The Rev 4 plug-in will have to be installed, for example. So, if you've got an idea that doesn't have these downsides, let me know. Your own upsides require Mac users to install technology that has been officially discontinued for them by Microsoft, in addition to the problem of OS vendor ownership of the video format. And I thought QT, which natively supports but does not own mp4, was necessary for Revolution video on Windows anyway. I admit I'm perhaps overly sensitive to video formats on the web, as well as the continuous onslaught of (clumsy and technologically inferior) Windows-only lock-ins masquerading as solutions on the internet. I do understand that there are trade-offs among compatibility, efficiency, practicality and commerce, but this particular trade-off didn't seem necessary to me. Thanks, m ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution
Re: Inefficient code
Bert Shuler wrote: I have written this code a few different ways. Each seems to be equally inefficient. I am attempting to compare to images, pixel by pixel, and record the differences. ON mouseUp pMouseBtnNo put 0 into c set the startvalue of scrollbar Progress to 0 set the endvalue of scrollbar Progress to the length of imagedata of image Alpha put the imagedata of image Alpha into idataalpha put the imagedata of image Beta into idatabeta REPEAT FOR each char myChar in idataalpha set the thumbposition of scrollbar Progress to c IF myChar is not char (c) of idatabeta THEN put c char (c) of idatabeta return after hAll END IF put c+1 into c END repeat put hAll into field diff END mouseUp It takes many minutes to process a 200x200 image. I want this code to eventually compare full screen-captures, so if there is any way to speed it up, I am open for advice. Take out the progress bar and then see how it does. Setting a progress indicator on every repeat often takes longer than running the entire rest of the handler. If you do need a progress bar, only update it occasionally; for example, every few hundred iterations: if c mod 500 = 0 then set the thumbpos of sb progress to c -- Jacqueline Landman Gay | jac...@hyperactivesw.com HyperActive Software | http://www.hyperactivesw.com ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution
Re: Inefficient code
Bert- Friday, June 26, 2009, 12:50:50 PM, you wrote: It takes many minutes to process a 200x200 image. I want this code to eventually compare full screen-captures, so if there is any way to speed it up, I am open for advice. Obvious statement #1: this will run *much* faster if you don't update the scrollbar (comment out that line and you'll see what I mean)... OK - so how to deal with this and still show progress? Do you really need to update the scrollbar for *every* byte in the imagedata? A 200x200 image has 40,000 pixels - that's 40,000 scrollbar updates. How about something like put c+1 into c if c mod 100 = 0 then set the thumbposition... end if -- -Mark Wieder mwie...@ahsoftware.net ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution
Re: Rev 4 Preview Webinar - Video Link
On Jun 26, 2009, at 4:04 PM, Michael wrote: Wow. This took several minutes: http://imp.on-rev.com/webinars/index.html Unfortunately, although you're right in all of your arguments, your example file makes some of Bill's points for him. The linked file has to be fully downloaded before it will play. It's also iPod sized, which makes it a little hard to read the text. I too did a test, at full size, and mine was about as good as the WMV, and less than half the file size (it was about 2/3rd the size of your small dimensions file too). So it is possible to give a good example MP4 file. Ideally it would be nice to compress straight to an MP4, and not be recompressing a WMV. I guess that the GoToMeeting software is recording straight to compressed WMV. If it recorded to uncompressed, that would be the right file to make other formats from. I may post my MP4 recompressed example, just for interest. ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution
Re: Rev 4 Preview Webinar - Video Link
Bill, I am sure you guys probably thought about this before. You can use external monitor adapters on both computers hooked to dv cameras to record both screens at the same time. Then you can pick these high quality sources and edit to recreate the cuts you made live. it's a ton of work and needs two cameras but it has no quality downside. I am fine with the WMVs you put though, they play fine on my mac with vlc cheers Andre On Wed, Jun 24, 2009 at 11:25 PM, Bill Marriott w...@wjm.org wrote: Michael, Really? Really. I think you should forego foisting MS POS formats on everyone and use mp4 for distribution instead. I can explain how if necessary. Be my guest. We've already spent dozens of hours trying to work around this. The GoToWebinar software we use for our webinars records *only* in WMV -- our choices are WMV that uses a special GoToWebinar codec that works only on Windows, and requires installation of the GoToWebinar software; or a cross-platform, portable WMV. Every option we have tried to transcode such files has resulted in, at the minimum, a file size of twice as large, and a loss in quality (such as what you would get converting a WMA file to MP3). For example, we had to recode an earlier webinar because of a glitch and the file size went from 50MB to 140MB after our best efforts. Some people have suggested we use a second computer to capture the video using different screen recording software, but this is also sub-optimal as a) the second computer is capturing compressed audio/video as a guest of the conference, not capturing the source screens; b) the GoToWebinar software captures the screens at a consistent resolution, without needing to worry about resetting the capture rect as we change presenters and other details; c) the other programs often require installation of a special codec, themselves. Note also that you said you prefer mp4, as if this were a universal format. But most Windows users cannot view that format, without having QuickTime or other additional software installed, and we wouldn't want to foist anything on them. Plus WMV is a progressive download that starts to play fairly quickly, as opposed to requiring the whole file to be downloaded. So, if you've got an idea that doesn't have these downsides, let me know. - Bill ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution -- http://www.andregarzia.com All We Do Is Code. ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution
Re: Inefficient code
I just popped together a stack using your code and 2 versions of an image. I made 2 buttons: 'COMPARE' contained your code 'COMPARE 2' contained your code with all references to your progress bar removed I also introduced 2 extra flds; 'Stime' and 'Etime' In both COMPARE and COMPARE 2 I inserted the following lines: COMPARE ON mouseUp put the time into fld Stime put 0 into c set the startvalue of scrollbar Progress to 0 set the endvalue of scrollbar Progress to the length of imagedata of image Alpha put the imagedata of image Alpha into idataalpha put the imagedata of image Beta into idatabeta REPEAT FOR each char myChar in idataalpha set the thumbposition of scrollbar Progress to c IF myChar is not char (c) of idatabeta THEN put c char (c) of idatabeta return after hAll END IF put c+1 into c END repeat put hAll into field diff put the time into fld Etime END mouseUp COMPARE 2 ON mouseUp put the time into fld Stime put 0 into c put the imagedata of image Alpha into idataalpha put the imagedata of image Beta into idatabeta REPEAT FOR each char myChar in idataalpha IF myChar is not char (c) of idatabeta THEN put c char (c) of idatabeta return after hAll END IF put c+1 into c END repeat put hAll into field diff put the time into fld Etime END mouseUp running the script in btn COMPARE 2 the script took less than a minute running the script in btn COMPARE is taking so long while I am writing this that I am getting cheesed-off and I am going to abort the thing. The problem is your progress bar: get rid of it! ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution
Re: Rev 4 Preview Webinar - Video Link
On Jun 26, 2009, at 4:28 PM, Colin Holgate wrote: I too did a test, at full size, and mine was about as good as the WMV, and less than half the file size (it was about 2/3rd the size of your small dimensions file too). So it is possible to give a good example MP4 file. Ideally it would be nice to compress straight to an MP4, and not be recompressing a WMV. I guess that the GoToMeeting software is recording straight to compressed WMV. If it recorded to uncompressed, that would be the right file to make other formats from. Yes, GoToMeeting compresses to WMV automatically. For the webinars my company does we have settled on using a 3rd party recorder. We set up Screenflow on a laptop and record the webinar. The audio/video quality is MUCH better than the WMV and you have more options for output. Regards, -- Trevor DeVore Blue Mango Learning Systems www.bluemangolearning.com-www.screensteps.com ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution
Re: Rev 4 Preview Webinar - Video Link
On Jun 26, 2009, at 4:37 PM, Andre Garzia wrote: Then you can pick these high quality sources and edit to recreate the cuts you made live. it's a ton of work and needs two cameras but it has no quality downside. There are problems with your suggestion. It might be slightly harder to do the cutting between the cameras when the two machines are in different continents, and the resolution of Kevin's camera would be different to Bill's, and both would be a lot lower resolution than the webinar (NTSC is effectively 640x480, PAL 768x576, and the webinar was 1024x768). I am fine with the WMVs you put though, they play fine on my mac with vlc It plays fine when using Flip4Mac too. ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution
Re: Rev 4 Preview Webinar - Video Link
On Jun 26, 2009, at 4:28 PM, Colin Holgate wrote: I may post my MP4 recompressed example, just for interest. I don't have the right license to export WMV from QuickTime Player, so this has a watermark in it, but you get the idea. The link is a MOV reference movie, which helps solve the need for the whole file to download (I hope): http://xfiles.funnygarbage.com/~colinholgate/video/rev4big.mov Heres the small dimension one, as a reference movie too: http://xfiles.funnygarbage.com/~colinholgate/video/rev4tiny.mov ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution
Re: Rev 4 Preview Webinar - Video Link
I think one can get a bit too hoity-toity about commercial software; and I am not fan number 1 of Microsoft either; BUT, as VLC (which is Open Source, Hurrah, Hurrah, Hurrah) plays WMV files just like that we should not complain. I know, Andre, that if somebody suggested I ran out and bought 2 dv cameras and then spent ages fiddling around reimporting the recorded video files and so on I would get a bit miffed. Obviously the chaps on the Webinar did what came naturally with the broadcast software; I know I would. Andre Garzia wrote: Bill, I am sure you guys probably thought about this before. You can use external monitor adapters on both computers hooked to dv cameras to record both screens at the same time. Then you can pick these high quality sources and edit to recreate the cuts you made live. it's a ton of work and needs two cameras but it has no quality downside. I am fine with the WMVs you put though, they play fine on my mac with vlc cheers Andre On Wed, Jun 24, 2009 at 11:25 PM, Bill Marriott w...@wjm.org wrote: Michael, Really? Really. I think you should forego foisting MS POS formats on everyone and use mp4 for distribution instead. I can explain how if necessary. Be my guest. We've already spent dozens of hours trying to work around this. The GoToWebinar software we use for our webinars records *only* in WMV -- our choices are WMV that uses a special GoToWebinar codec that works only on Windows, and requires installation of the GoToWebinar software; or a cross-platform, portable WMV. Every option we have tried to transcode such files has resulted in, at the minimum, a file size of twice as large, and a loss in quality (such as what you would get converting a WMA file to MP3). For example, we had to recode an earlier webinar because of a glitch and the file size went from 50MB to 140MB after our best efforts. Some people have suggested we use a second computer to capture the video using different screen recording software, but this is also sub-optimal as a) the second computer is capturing compressed audio/video as a guest of the conference, not capturing the source screens; b) the GoToWebinar software captures the screens at a consistent resolution, without needing to worry about resetting the capture rect as we change presenters and other details; c) the other programs often require installation of a special codec, themselves. Note also that you said you prefer mp4, as if this were a universal format. But most Windows users cannot view that format, without having QuickTime or other additional software installed, and we wouldn't want to foist anything on them. Plus WMV is a progressive download that starts to play fairly quickly, as opposed to requiring the whole file to be downloaded. So, if you've got an idea that doesn't have these downsides, let me know. - Bill ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution
Re: Rev 4 Preview Webinar - Video Link
On Jun 26, 2009, at 4:46 PM, Trevor DeVore wrote: We set up Screenflow on a laptop and record the webinar. The audio/ video quality is MUCH better than the WMV and you have more options for output. Now you mention it, any of us could volunteer to record the webinar, with whatever recorder we want. I use SnapzPro X v2, and that does a very nice job. It would need a good amount of drive space, and then a few hours to compress it later. ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution
Re: Inefficient code
Bert Shuler wrote: ON mouseUp pMouseBtnNo put 0 into c set the startvalue of scrollbar Progress to 0 set the endvalue of scrollbar Progress to the length of imagedata of image Alpha put the imagedata of image Alpha into idataalpha put the imagedata of image Beta into idatabeta REPEAT FOR each char myChar in idataalpha set the thumbposition of scrollbar Progress to c IF myChar is not char (c) of idatabeta THEN put c char (c) of idatabeta return after hAll END IF put c+1 into c END repeat put hAll into field diff END mouseUp Ditto what the others have said about the progress bar. Those updates prompt so many layers of OS rendering code that they take quite a toll. I went to a mod solution with one of my projects and it gave me an order of magnitude speed boost. But I have a question about the algorithm's logic: if c is initialized to 0 but the data in idatalpha is traversed starting with the first character, wouldn't the comparison always be one character off? -- Richard Gaskin Fourth World Revolution training and consulting: http://www.fourthworld.com Webzine for Rev developers: http://www.revjournal.com ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution
Re: Inefficient code
On 26 Jun 2009, at 23:07, Richard Gaskin wrote: Ditto what the others have said about the progress bar. Those updates prompt so many layers of OS rendering code that they take quite a toll. I went to a mod solution with one of my projects and it gave me an order of magnitude speed boost. But I have a question about the algorithm's logic: if c is initialized to 0 but the data in idatalpha is traversed starting with the first character, wouldn't the comparison always be one character off? -- I think you're right - the first comparison would be char 1 of idataalpha to char 0 (non-existent) f idatabeta, then 2 to 1 etc. And for any benchmark junkies out there, add 1 to c is nearly twice as fast as put c + 1 into c on my machine, though it would only make a small difference even with hundreds of thousands of iterations (maybe in this case with large images). Best, Mark ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution
Re: Rev 4 Preview Webinar - Video Link
Michael, Wow. This took several minutes: http://imp.on-rev.com/webinars/index.html Yes, we know how to convert to mp4, that isn't the problem. Your version of the video is not acceptable: - It's more than 3x the file size: 151 Mb; the original WMV is 46 Mb - It's half the dimensions of the original, the text of scripts is unreadable. We simply cannot demonstrate a programming product without the code being crisp and clear. - It does not stream -- must be downloaded completely before viewing, which makes the file size even more of an issue - It's in mp4, which simply transfers the burden of installing software from Mac to Windows users For all its flaws, WMV has a built-in codec suitable for screen captures (in which most of the pixels from moment-to-moment are unchanged) as opposed to H.264, which is optimized for video. While you balk at the format for political reasons, Flip4Mac a free and quick download for end users, and WMA/WMV is common enough that many people have it already. It's not like we're forcing anyone to convert their library of videos or author using WMV. The videos even play back using the QuickTime Player, or within the browser window itself. GoToWebinar has its issues, but it's the best value for web conferences we've found for our needs. Webex, for example, accommodates only 25 users per meeting without special arrangement. Our Rev 4.0 webinar had almost 350 attendees. And the video is created automatically at the conclusion of the event, which means we have to spend a whole five minutes of post-production (to trim the start and end) and we're done. That enables us to upload the result quicker, and leaves me more time to participate in the use-list. :) - Bill ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution
Re: Inefficient code
Bert, Others have pointed out the delay introduced by updating a progress bar with every pixel and suggested updating it every 100 or 500 pixels or so. Similarly, comparing byte-by-byte is going to be slow. An immediate, simple improvement will be achieved by comparing *groups* of pixels between the two images. For example, if your image is 10,000 bytes in size, comparing 500 bytes at a time results in 20 comparisons instead of 10,000 comparisons. As you find differences in a block of 500 bytes, you can then down-shift to find the differences within that 500-byte block with more granularity. A refinement on this approach is simply to divide and conquer by constantly dividing the image by half and recursively testing the halves for differences. If the differences between the two images are small, the comparison can be near-instant. One of the classic papers on checking for differences between data sets can be found here: http://xmailserver.org/diff2.pdf Of course, the language in that paper is way beyond my comprehension ;) I'll putter around with expressing these concepts elegantly in Rev, but hopefully this gives you or someone else on the list a starting point for an algorithm that is dramatically faster than byte-for-byte testing. (I'd love to see the O(ND) difference algorithm properly implemented in Rev code.) - Bill Bert Shuler bertshu...@yahoo.com wrote in message news:c226bc83-61ac-4267-a24c-e32c8c7ce...@yahoo.com... I have written this code a few different ways. Each seems to be equally inefficient. I am attempting to compare to images, pixel by pixel, and record the differences. ON mouseUp pMouseBtnNo put 0 into c set the startvalue of scrollbar Progress to 0 set the endvalue of scrollbar Progress to the length of imagedata of image Alpha put the imagedata of image Alpha into idataalpha put the imagedata of image Beta into idatabeta REPEAT FOR each char myChar in idataalpha set the thumbposition of scrollbar Progress to c IF myChar is not char (c) of idatabeta THEN put c char (c) of idatabeta return after hAll END IF put c+1 into c END repeat put hAll into field diff END mouseUp It takes many minutes to process a 200x200 image. I want this code to eventually compare full screen-captures, so if there is any way to speed it up, I am open for advice. Thanks Bert ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution
Re: [OT] Shift-Lock
On Sat, Jun 27, 2009 at 3:35 AM, Richmond Mathewson richmondmathew...@gmail.com wrote: does anybody know how to convert the Caps-Lock key to a Shift-Lock key? I don't know but you might try Ukelele, it's free: http://scripts.sil.org/cms/scripts/page.php?site_id=nrsiitem_id=ukelele If you reassign shift to the caps key, then maybe in the locked state it will be as if you had shift lock. Worth a try I guess. HTH ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution
Re: [OT] Shift-Lock
On Sat, Jun 27, 2009 at 11:20 AM, Kay C Lan lan.kc.macm...@gmail.comwrote: http://scripts.sil.org/cms/scripts/page.php?site_id=nrsiitem_id=ukelele If you reassign shift to the caps key, then maybe in the locked state it will be as if you had shift lock. Worth a try I guess. I note on the above mentioned page, under the first note for the 1.7b1 release, it actually mentions 'copying' a current keyboard layout, then just assigning shift to the caps lock key. Unfortunately it doesn't specify what the exact result is. ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution
Rev cannot open my jpeg !
Hello, I've bought a new camera (Nikon coolpix p6000) and Revolution doesn't want to open my photos (with the set the filename It's just a normal jpeg file, but too big ? If i resize the photos, it's works. Is there a workaround, i use Revolution to choose, name and tag my photos so i need to see it ! Thanks. Rev. 2.91 I've test ith Rev 3.5 it's the same. ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution
Re: Rev cannot open my jpeg !
You don't mention which OS you're on, but I wouldn't be surprised if you've hit a 4096 pixel limitation. If you use the 8 megapixel setting on the camera you should be ok, or resize the 13 megapixel images down to 4095 or less across. ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution
Re: Rev cannot open my jpeg !
From Rev, is your camera listed in the volumes? If not, the camera may be connecting to the computer with Picture Transfer Protocol, which makes the computer treat it differently than a regular USB mass storage device. Phil Davis Ludovic Thébault wrote: Hello, I've bought a new camera (Nikon coolpix p6000) and Revolution doesn't want to open my photos (with the set the filename It's just a normal jpeg file, but too big ? If i resize the photos, it's works. Is there a workaround, i use Revolution to choose, name and tag my photos so i need to see it ! Thanks. Rev. 2.91 I've test ith Rev 3.5 it's the same. -- Phil Davis PDS Labs Professional Software Development http://pdslabs.net ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution
Re: Rev cannot open my jpeg !
Le 27 juin 09 à 07:09, Colin Holgate a écrit : You don't mention which OS you're on, but I wouldn't be surprised if you've hit a 4096 pixel limitation. If you use the 8 megapixel setting on the camera you should be ok, or resize the 13 megapixel images down to 4095 or less across. I'm on MacOS X, and yes my pictures are taken at 13 Mpx. At 8 Mpx it's work, but it's a shame. An we cannot use Rev to resize these pictures !___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution
