Re: Help: Id's can be completly unreliable
Hallo, Sorry for the delay but I have been quite busy. Jacqueline, I found the script it in the button revIDE of the revLibrary stack and it's called addObject private command addObject pObject, pObeyDontSearch, @rList, pScriptsOnly if word 1 of the name of pObject is among the words of card group field and pObeyDontSearch and the dontSearch of pObject then return dontSearch set on object end if local tId put the long id of pObject into tId if word 3 of tId is 0 then # Due to an engine bug, some objects can have IDs of zero. For now, we ignore this... --breakpoint end if ... ... Bob, there are no intercepting handlers. This happend with at least 2 totaly different stacks. mac 10.5.8 livicode 4.5 build 1080 but it happened with a earlyer version aswell. I will try to get a simplifyed stack which more or less 'reliably' exibits this fenomena, but I can't promise anything. I will see if I got some time coming weekend to do this. All the best, Claudi ___ 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
Help: Id's can be completly unreliable
Hallo all, I have a big, and as yet, unsolvable problem. I use quite some tmplt's (groups) I keep in a seperate stack and when I need one I copy this tmplt to the card or group where I need it. So far all works as expected. Now comes the problem: Very often one or more controls of this copyed tmplt or the copyed tmplt itself will get an ID of 0. This is more than a nuisance as I keep discovering more and more problems with controls with an ID of 0. The first problem is the long id the IDE/engine places in the it variable. If there is a ID of 0 in it, it will be invalid. Which means if you do the following it will throw a error saying the control doesn't exist. on mouseUp copy grp myTmplt of cd 1 of this stack to grp myGrp of cd 2 of this stack set the name of it to test -- it: group id 0 of group id 1012 of cd id 1002 of stack myStack end mouseUp I had to change a lot of code because of this. I couldn't be sure the long id in the it var was reliable, because it might contain a ID of 0. This problem can be solved as follows: on MouseUp put the number of controls of grp myGrp of cd 2 of this stack + 1 into tCount copy grp myTmplt of cd 1 of this stack to grp myGrp of cd 2 of this stack set the name of control tCount of grp myGrp of cd 2 of this stack to test end mouseUp Now it doesn't matter if I copy a image/field or a group consisting of many controls, since tCount will point to the next layer within the destination, which will be the group in case a group get's copyed. (you can't use the last control of grp myGrp it will throw a error on compiling and anyway if you copy a grp the last control will point to the control with the highest layer within the copyed grp) Further if you use tmplt's which you copy, always and I mean always, use long names to point to a specific control, as you just can't trust the ID or the long ID. (a major rewrite of my code) Ok this is a nuisance, but there is at least a solution/workaround to it. Now I discovered a new problem with a control which has gotten the ill fated ID of 0. If it happens to be a field you can't get it's formattedWith/Height etc. I simply can't find a way to get at this, it always returns 0 put the width of fld myLongNameFieldReference gives: 220 put the value of yLongNameFieldReference gives: This is some text put the formattedWidth of fld myLongNameFieldReference gives: 0 So the field exists (with a ID of 0) and it has text in it. But I just can't get the formattedWidth of it . I found in a rev lib stack a comment that the engine sometimes returns 0 as ID for controls and that they leave it at the moment. The problem is I now have a card with sometimes 10 or more controls with a ID of 0 which gives all kinds of problems. Does anyone know a trick or workaround to get the formattedWidth of my field? Please. This is a complete stopper for me at the moment since I can't format my field and I use dozens of them which need to be formatted depending on there content. Hoping someone of you bright minds knows/finds something clever to solve my problem. Best wishes, Claudi ___ 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: Help: Id's can be completly unreliable
Hi, After some more tinkering I fortunatly solved the formattedWidth problem. Apparently you have to be on the card containing the field to get the formattedWidth/Height etc, otherwise it will return 0. This is mentioned in the docs, so I could have 'known' this. I am glad I got this working so I can continue. But still be warned and prepared to deal with the ID 0 problems you may encounter. Best wishes, Claudi ___ 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: Help: Id's can be completly unreliable
Richard, Referencing a control with a ID 0 and using ID's to reference this control might result in a error (control doesn't exist) but not always. It depends on what you do. Programaticly putting some text in a field with ID 0 (referencing the control with the long id) didn't generate a error but it also didn't put the text in the field, for instance. Same with properties, no error, but also the propertie didn't get set. This makes using ID's not reliable. So to be save, I don't use ID's anymore. It might go well a hundred times but suddenly you might get an error. Sometimes I get 8 out of 10 times a ID of 0 when copying a grp. This also means the it variable might hold a invalid reference to the newly created grp after copying, so you can't use it reliably and you have to use the work around to reference the newly created grp. This problem with the ID of 0 has happened in different stacks and with different kinds of grp's. It drove me quite nuts for a while, as the IDE/engine gave sometimes a error, but not always. This made it difficult to find what's was going on. Jacqueline, The names of the grp's I copy have names like textPiece for instance (this is the one I have trouble's with right now) I have to say, this grp belongs to a grp called textPieceTmplts. I will change this name to see if that will make a difference. Also like I wrote in my first mail, I found in a rev lib stack a comment that the engine sometimes returns 0 as ID for controls so apparently the 'wizards' in Edinburgh know about this. Best wishes, Claudi ___ 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: ANN: A true scrolling canvas demo
Hi all, Unfortunatly there was a small error in my code, which Thierry found. A local var with the wrong name (sMoveFlag instead of sMoveObj) I have uploaded a new version with the correction. Sorry for this unconvienience. (working on to many scripts at the same time) Best wishes, Claudi On 3 okt 2010, at 23:30, Claudi Cornaz wrote: Hi All, I am glad to anounce the ccScroller group. After many years, on and off, trying to get it to work right, I finaly succeeded in creating a true scrolling canvas like group. I managed to get something to work before but now I found a simple and elegant solution. You can drag objects around in the group and the canvas will always adept to the placement of the objects. It grows and shrinks as neccesary, in the expected and sane way. The final piece I needed was mentioned by Richard quiete a few weeks ago, the boundingRect. The boundingRect alone isn't sufficient but it turned out to be a crucial extra piece I needed to acomplish this canvas group. I tryed to uploaded it to revOnLine but after I logged out I couldn't log in anymore. Apparently the password they send me this afternoon is wrong now. I don't know. So I have no idea if the upload went correctly because I can't find it nowhere anymore. Revonline seems a bit flaky at the moment. I will try again later. Probably the easyest way to get the stack is to execute the following in the messagebox: go stack URL http://www.claudi.on-rev.com/livecode/ccScroller_demo.rev and check it out. I hope (some of) you find it usefull All the best Claudi ___ 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
ANN: A true scrolling canvas demo
Hi All, I am glad to anounce the ccScroller group. After many years, on and off, trying to get it to work right, I finaly succeeded in creating a true scrolling canvas like group. I managed to get something to work before but now I found a simple and elegant solution. You can drag objects around in the group and the canvas will always adept to the placement of the objects. It grows and shrinks as neccesary, in the expected and sane way. The final piece I needed was mentioned by Richard quiete a few weeks ago, the boundingRect. The boundingRect alone isn't sufficient but it turned out to be a crucial extra piece I needed to acomplish this canvas group. I tryed to uploaded it to revOnLine but after I logged out I couldn't log in anymore. Apparently the password they send me this afternoon is wrong now. I don't know. So I have no idea if the upload went correctly because I can't find it nowhere anymore. Revonline seems a bit flaky at the moment. I will try again later. Probably the easyest way to get the stack is to execute the following in the messagebox: go stack URL http://www.claudi.on-rev.com/livecode/ccScroller_demo.rev; and check it out. I hope (some of) you find it usefull All the best Claudi ___ 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: Storing an Array as a Custom Property
Hi Richard, It works if you create it as one var so put gaskin[richard] into tPropSet set the tPropset of tObj to tValue Or you can do on mouseUp put gaskin into tPropset put [richard] after tPropSet set the tPropSet of tObj to tValue end mouseUp I use this the whole time Claudi ___ 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: Storing an Array as a Custom Property
There is another one theat works put richard into tProp put the gasking[tprop] of tObj Apparantly the set name by itself can't be a var, altough the prop name can. So either put the whole thing in a var (set name en property name) or you can only use the property part as a var. Claudi On 2 okt 2010, at 15:18, Richard Gaskin wrote: Claudi Cornaz wrote: It works if you create it as one var so put gaskin[richard] into tPropSet set the tPropset of tObj to tValue Or you can do on mouseUp put gaskin into tPropset put [richard] after tPropSet set the tPropSet of tObj to tValue end mouseUp I use this the whole time Nice! It hadn't occurred to me to put the entire array notation string into a variable. And it avoids the performance penalty of do -- good stuff. Thanks! -- Richard Gaskin Fourth World LiveCode training and consulting: http://www.fourthworld.com Webzine for LiveCode developers: http://www.LiveCodeJournal.com LiveCode Journal blog: http://LiveCodejournal.com/blog.irv ___ 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
controls with ID 0
Hi all, I have a very strange situation where controls get a ID of 0. I had this problem with runrev 4 and now also with the new livecode It has happened a couple of times in different stacks. I create a control in a grp and the control ends up having a ID of 0. It's happened with btn's and now with a fld. It's very frustrating 'cause one of the consequences is that I can't put text in that fld. I managed to put text in the fld using a long id but then I get a formattedWidth of 0 I have the following statements in a script: put the with of tFldId into tWidth (tFldId is a long id. tWidth becomes 129) put the formattedWidth of tFldId into tFormattedWidth (tFormattedWidth becomes 0) This happens every time I create the control. Well actualy I copy a tmplt grp from a stack to a grp in another stack. What should I do? This looks like quite a bug to me. Claudi ___ 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: Which char is NOT in result MD5Digest
Hi Richard and Dave, Thanks for the answers. It works like a charm. Not something I would have found in a long long time. Anyway back to coding again. Claudi On 3 apr 2010, at 17:28, Richard Gaskin wrote: Claudi Cornaz wrote: Is there a character that won't be in the result of the MD5Digest? I try to ceate a list of control id's a comma and the MD5Digest of the script of that control. Unfortunatly the cr character can be part of the resulting MD5 thus throwing of my lines with a part of the MD5. Now of course I can replace the cr but which char to use? I need one that for sure isn't returned by the MD5. Or differently: how can I keep my control id's and corresponding MD5 seperated from each other in a list, to get at each in turn? If you can deal with a string that's a bit longer, you could base64- encode the MD5 digest: put base64Encode(md5Digest(tMyString)) That'll result in a string using only a relatively small subset of ASCII characters, leaving most common delimiters (commas, tabs, etc.) free for you to use however you like. -- Richard Gaskin Fourth World Rev training and consulting: http://www.fourthworld.com Webzine for Rev developers: http://www.revjournal.com revJournal blog: http://revjournal.com/blog.irv ___ 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
Which char is NOT in result MD5Digest
Hi all you wizards, Is there a character that won't be in the result of the MD5Digest? I try to ceate a list of control id's a comma and the MD5Digest of the script of that control. Unfortunatly the cr character can be part of the resulting MD5 thus throwing of my lines with a part of the MD5. Now of course I can replace the cr but which char to use? I need one that for sure isn't returned by the MD5. Or differently: how can I keep my control id's and corresponding MD5 seperated from each other in a list, to get at each in turn? TIA Claudi ___ 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: including a file on on-rev
Hi Thomas, I just started with irev and mysql. I know how you must feel. I can program quite fluently in revtalk and if something doesn't work as expected, which happens fairly often, I always know at least 2 or more ways to tackle the problem or to solve it in a different way. But now I am getting on 'new' ground: irev, connections, ports, mysql, cgi, css, htm, etcl. And all these things need to work together aswell and suddenly I find myself looking all over the place to find 'trivial' things, that is, trivial once you know them. The first hurdle being where to look, where can you find what you need. Copying bits and pieces from code you find, trying to fit it all together. It's like walking through molasses, till you sudenly discover that you're no longer walking but somehow you've started swimming, and now that you look more closely it is not mollasses at all, but it's something like ... water. (and now that you're really taking attention you notice that you have turned into ... a fish) Well anyway here are some pieces of what I tried and failed. include ../lib/ccStats.irev -- failed with FTP error file / testing/..//lib/cc_PageStats.irev not present. the include call was made from a page in the testing folder. Why that folders name is added I don't know include http://claudi.on-rev.com/lib/cc_PageStats.irev; - fails I also tried: include /home/claudi/public_html/lib/ cc_PageStats.irev which for some reason first failed and later on it worked no idea why it first failed but I've been changing and trying all kinds of things which I don't all recall and in which order, so probably I first did something stupid. I also tried some more complicated crazy things to no avail but what those where, I don't know anymore. So it looks like I can use: include /home/claudi/public_html/lib/ cc_PageStats.irev or what I use now: set the defaultFolder to $_Server[DOCUMENT_ROOT] include lib/cc_PageStats.irev I hope this helps Best wishes, Claudi ___ 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: including a file on on-rev
Thanks all for your ideas and responses What a great list this is. I still can't get the ../lib/cc_Pagestats working (strange) Jim the double slash in /testing/..//lib/cc_PageStats.irev was a typo from me in the email By the way thanks for the explanation about linux/apache. It sure helps in getting a overall picture. Setting the defaultFolder as Andre sugested on the other hand works like a charme. So that's the way to go for me. So I am off now to do some more coding and probably finding more problems/learnings on the way. Thanks all Claudi ___ 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
including a file on on-rev
Hi all, I finaly started getting to grips with on-rev and after some initial struggeling I start getting the hang of it. Very powerfull and very nice. However I am now again struggeling with a simple thing but I can't get it to work. I have a folder lib with obviously some irev files in it. this lib folder is at the root level in my public_html folder I need to include one of the files in this lib folder called cc_PageStats from other pages. As long as the calling page is on the root level I can include the file with: include lib/cc_PageStats.irev In that case everything works as expected. Now I have a page in another folder, called testing. This folder is on the root level, so I assumed I need to change the include statement to: include ../lib/cc_PageStats.irev but this doesn't work. I get the error message in my on-Rev desktop app saying: FTP error file /testing/..//lib/cc_PageStats.irev not present. (where does this 'testing' come from. I thought the 2 dots meant to go one level higher in the dir structure and start from there.) Whatever I try I can't get this file included. If anybody has a idea what I am doing wrong please let me know, for I am banging my head now for some time. Even better would be, preferably a one liner, so I can include the file from whatever dir the requested page is, without worring about the number of levels (I have been trying with a 'hard coded adress like http://claudi.on-rev.com/lib/cc_Pagestats and with $_SERVER[document_root] and a couple of others but unfortunatly nothing.) All ideas or suggestions are greatly appreciated Claudi ___ 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: moving domain to on-rev
Hi Stephen, My mistake I didn't mean my ISP but my registar of course. They should change the nameservers . There is a controlpanel but since cuurently they also host the site and unfortunatly I can't change the nameservers. Best wishes, Claudi On 1 jul 2009, at 20:21, stephen barncard wrote: You shouldn't have to ask your ISP to change the nameservers. I haven't had to do that for years. Your domain registrar may be, and usually is, a different entity than your ISP. Registrars usually offer a control panel to do such changes, and anyone with an administrative or technical contact authorization can change it. It's dead-dog simple to change. Three items, copy and paste. - Stephen Barncard San Francisco http://barncard.com 2009/7/1 Claudi Cornaz claud...@fiberworld.nl I will now ask my isp to try again to change the nameservers, hoping it will work this time. ___ 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: moving domain to on-rev
Thanks for the reply Bill, I have now created a subdomain cc-imaginering, which is working. I will now ask my isp to try again to change the nameservers, hoping it will work this time. I have one more question though. Now there is no .com or, .nl in my case, for the subdomains site folder. Do I have to change this for the add on domain, or does the name for the site folder just stay cc-imaginering. Best wishes, Claudi ___ 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
moving domain to on-rev
Hi all, I am trying to move a existing domain to on-rev. The site is on line and I want to transfer it. So as sugested in a couple of posts I started out by adding a domain to my on-rev account via cPanel. I set the home directory of this site at the root level and not within the public_html folder of my home folder. (this was suggested by Bill in one of the posts I've read) cPanel did complain, exactly like I read in the posts but did create the folder for this domain anyway (including a cgi-bin folder), so all looked ok. I transfered some files over and tryed to acces the index file at: www.claudi.on-rev.com/cc-imaginering.nl/index.html (cc-imaginering.nl is the domain I try to move) to see if it works. Unfortunatly I got a 404 errror. (the requested URL /cc-imaginering.nl/ index.html was not found on this server) The same file uploaded to my public_html folder works just fine My first question is how do I get to my site to test it. Is the URL I entred wrong? Is there some setting I overlooked? The second problem I encoutered is the following There is a part of the site which I don't want to be down for a couple of days while the DNS update takes place so I contacted my ISP where the site is currently hosted and asked them to change the nameservers to ns1.on-rev.com and ns2.on-rev.com but leaving the site on there servers till the DNS update has taken place. They where ok with this and tried to do it. Next I got a mail from them telling me that they got the following error from SIDN Errors=1, Warnings=1, Informational=1 ** Summary: REJECTED cc-imaginering.nl. Some problems need to get fixed: - Some of your name servers cannot be used (are broken). ** Full check report: * general reports Warning: all specified name servers are on sub-net 74.54.153.0/24 * primary name server ns1.on-rev.com. [BROKEN] Broken: the name server does not know of domain cc-imaginering.nl. * secondary name server ns2.on-rev.com. Info: problems with the primary, so not all tests run. ** DNScheck 4.2.6, 2009/06/29 15:45:16 CEST+0200 fail I don't know if both things are related, maybe, but I realy like the site to stay working during the DNS update I hope that it is possible. If anyone can help me out I would appriciate it very much Sincerly, Claudi ___ 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: lost features scripteditor 3.0
Thanks Jacqueline, Jim and Jerry. I am glad there are still shortcuts, although a bit awkward. The space bar is so very easy accesable so I can keep my eys on the screen, instead of looking at the keyboard. I' will have a look if I can change them back into the original ones. I think it's a good idea to have a peek at the scripeditor anyway. Jim, nice idea with the globals although I prefer to do it the normal way without the use of globals. I know I can use some chars to mark pieces of code but color is so much easyer to spot when scrolling. It visualy realy stands out much more. (information in a other domain then chars, which are what I am aleady looking at, the script) But alas, it seems to have gone. Jerry, I will certainly give GLX2 another try. My license should be valid till end of oktober, so let's see what you've been up to lately. Anway all of you thanks for the reply's and suggestions, All the best Claudi ___ 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
lost features scripteditor 3.0
Hi all, I didn't have much time to put vers 3 through it's paces yet, but of course I was to curious to see what it would be like. I hope to get some more time for programming again the comming days. (I still have to finish 2 films in the comming weeks but I will only be working a couple of hrs a day on these) I like what I have seen so far, but there where 2 nasty surprises in the script editor. The keyboard shortcuts for step [into] (spacebar) and step over (option spacebar) are not working anymore. This is quite severe 'cause now it's impossible to debug while the mouse is down, like in mouseMove. Because it's of course quite impossible to click on a button in the debugger and to keep the mouse button down at the same time. In almost all my programs you need to drag objects or images around, so it's crucial for me to able to debug during the moseMove handler. I have a trackball and with a match I can lock the mouse button down. This way I can move the mouse around, without getting cramp in my fingers and control the debugger with the keyboard to step into, over or run. ( I always added the command space shortcut to choose run. This way I can realy focus on debugging the specific parts I am interested in, a couple of well placed breakpoints are all that's needed.) Are there new short cuts or are they indeed lost? That would really be a (unnecesary) shame if it wouldn't be restored. The second unpleasant surprise was the loss of color. I don't mean script coloring (which I dislike: to much info for the value. it distracts me) but the ability to define a color to a word or sentence. For instance I use red for debug code. Visualy it's a very vast way of finding parts in my scripts. I need to change the itemdelimiter quite often and I always set the line with that declaration to blue so whith a glance I can see where I set it and to what. I was just planning on adding a popUpMenu to the script editor to be able to work quicker and with some more predifened colors to mark parts in my scripts. Probably this loss of color won't affect a lot of people so I guess I wil have to change my preference of working and find another way to define parts in my scripts. This will mean I will give GLX2 another try. This coloring was actualy my mean reason not to switch to GLX2. I hope and guess that the rest of the vers. 3 experience will be a real pleasure. All the best Claudi ___ 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] Does Sheepshaver work in Leopard?
Hi Jim, I have been using it for one app, Claris Cad and I have not encountered any problems. Don't know about other apps though. Claudi On 1-feb-2008, at 19:51, Jim Carwardine wrote: Hi Folks... I haven't upgraded to Leopard for fear of Sheepshaver not working. Anybody try it?... Jim Jim Carwardine, President CEO OYF Consulting Ph. 902.823.2339 / 866.601.2339 Fx. 902.823-2139 www.StrategicDoing.com StrategicDoing™: Execution depends on employees. Strategic Partner with HiringSmart Canada Ltd. -- ___ 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: core image transitions
Thanks for all the answers. It looks like Scott is right and only the transitions work. Alas I hope the others will become available as well. Well, it looks like I have to figure out some other scheme. Fortunatly the whole concept is still in the making so everything is still open. David, off course you are invited to come and see, in fact all off you are. It will be sometime after he holliday's, I don't know yet the exact date. It will be here in Hal 4 in Rotterdam, the Netherlands. For all those who can't make it to the Netherlands, when I have a nice demo I will try to put something on revOnLine. (depending on the size of the images etc.) Thank you all and best wishes Claudi ___ 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
core image transitions
Hi all, Currently I am working on a arty project. In a kind of dark box images (actually always tiling patterns) will be projected on all 4 sides. In this box will be tables and chairs and people will be sitting there. These patters (hundreds of them) will slowly dissolve one in the other. This part I got working beautifully with blendLevels and different inks. (the patterns will be choosen at random and also the timing will have a random component) The next part is more complicated and I don't know if it can be done in rev. (I hope so) Over these background patterns I need to move spot's of other patterns, a bit like clouds. These clouds should change the undelying pattern in contrast, or color, or luminosity etc. I was hoping I might use the new core transitions to achieve this but I can't get them to work. I manage the CIDissolveTransition to work but alas I can't get any of the blendModes to do anything. I hope there is something obvious I am missing. I have a stack with 2 images and a btn with the script. on mouseUp hide img testImg do show img quote testImg quote with visual effect CISaturationBlendMode slowly -- for instance, I tried all blendModes end mouseUp I also tryed to put the destination img after the blendMode (with visual effect CISaturationBlendMode slowly to bgImg) Alas no dice. After a delay the testImg just pops in front of the bgImg. No transition no combining of the 2 images, nothing. If anyone has a sugestion, please I am in a bit of a fix now, not knowing if it can be done with rev or that I need to find some other way to do this (other solutions are very welcome as well of course) All the best Claudi Ps. I have the newest macBook and of course 10.4 and the latest build of rev. ___ 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
how to revert stack B, C and D from stack A
Hi all, Currently I am working on a updater for my app. I can download the new versions of the stacks, the app is a splash with a lot of seperate lib stacks. In my updater stack I then have a list of stacks that need to be reverted to use the new version, but how to do that? All stacks that need to be reverted are invissible I just can't get it to work. Each time the udater stack itself gets reverted, which is of course the current stack. So I tried to use send in time a myRevert message to each stack on MyRevert set the defaultStack to me revert end myRevert Alas still no dice. How can I do this, if somone has a suggestion please, it would be very helpfull. Thanks a lot, Claudi ___ 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: how to revert stack B, C and D from stack A
Hi Mark, Thanks for the quick reply. All stacks are librarie stacks so they have no substacks. They also are all in use (well most of them) First I download the new versions of the stacks into the bundle with another name then I check if all stacks got downloaded OK. If so then I rename the old ones with a ~ at the end and rename the newly downloaded ones to the name of the old ones. Then I check again if all is well and if so I delete the old ones (the stacks with a ~ at the end) then I like to revert all stacks that have a new version. (I hope this doesn't sound to complicated but I hope this is a safe way to do it. If anywhere in this process something goes wrong I can still backout of the whole update and leave the app as it was before) I found out what I did wrong. The defaultStack is not the same as the current stack. So in this case instead of setting the defaultStack, I can just go invissible to the stack which was renewed in the bundle and then do revert. This works. I got confused between the defaultStack and the current stack. This got me confused the otherway round before when getting the mouseLoc in a librarie routine. In other words I wanted the mouseLoc in my document stack but the routine was in a invissible librarie stack. Without setting the defaultStack I got the mouseLoc relatieve to the librarie stack instead of my document stack, which was not what I needed. This one took me quite a while to discover. after setting the defaultStack at the beginning of the handler the mouseLoc returned the right coordinates relative to the document stack. (actualy I had discoverd it and forgot about it again till I got bitten again a year later, and again it got me up the walls before I figured out what was going on and then I remembered it had happened to me before) I plan to create a stack Mistakes I made and things I learned It could be very usefull, even for myself. Anyway thanks a lot Claudi On 7-jan-07, at 14:45, Mark Schonewille wrote: Hi Claudi, Revert simply reads the stackfile from disk again. It doesn't only revert the substacks but also the mainstack. I don't think this is what you want. Perhaps you want to delete stacks stacks B, C and D and replace them with a new version. First, not that this won't work if your stacks are part of a standalone. delete stack B go to stack /disk/folder/stackB.rev set the mainstack of stackB to Your Mainstack save stack Your Mainstack Same with stacks C and D. You cannot open the new stack before deleting the old one. However, you could give the new stacks a different name at first. Suppose that StackB Update is a substack of your updating stack: delete stack B go to stack StackB Update set the mainstack of stackB Update to Your Mainstack set the name of stack StackB Update to StackB save stack Your Mainstack I hope this gives you some ideas. Best, Mark -- Economy-x-Talk Consultancy and Software Engineering http://economy-x-talk.com http://www.salery.biz Get your store on-line within minutes with Salery Web Store software. Download at http://www.salery.biz Op 7-jan-2007, om 14:04 heeft Claudi Cornaz het volgende geschreven: Hi all, Currently I am working on a updater for my app. I can download the new versions of the stacks, the app is a splash with a lot of seperate lib stacks. In my updater stack I then have a list of stacks that need to be reverted to use the new version, but how to do that? All stacks that need to be reverted are invissible I just can't get it to work. Each time the udater stack itself gets reverted, which is of course the current stack. So I tried to use send in time a myRevert message to each stack on MyRevert set the defaultStack to me revert end myRevert Alas still no dice. How can I do this, if somone has a suggestion please, it would be very helpfull. Thanks a lot, Claudi ___ 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
Leaving the browser
Hi all, I just came across an interesting link: http://labs.adobe.com/wiki/index.php/Apollo#General_Apollo_Information Web and desktop combined. Well I guess, in a way we have that capability with rev. Interesting trend. Some nice toughts for during the hollidays. Happy coding and best wishes, Claudi ___ 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: Resizable stack with a custom shape
Hi Derek, I am currently working with windows with emty decorations. (I create my own titelbar) Resizing then surely looks ugly when growing a stack. (the stack resizes but the grown, newly exposed parts are first seen as white and then later get the backcolor) By the way this always happen if liveResizing is off. (lock screen doesn't help, it's in the engin I guess) So I am sure interested in the external you mentioned and I dowloaded the externals stack. Unfortunatly I couldn't find the external for resizing in it. If you have the external and a example I would really appreciate it. I have done some tests with resizing with a custom shape for the window. Besides the darned flashing when you grow a window and some sluggishness It worked more or less. (this was some time ago so I don't remember exactly all the problems.) I used a img in a hidden stack which I resized first to the right size and then set the windowshape again for my main window (something like that. I remember ditching it at the time since the whole stack slowed down I mean every script became slower, which I never quite understood.) For what I am developing resizable custom shaped windows are actualy the most preffered way and I will certainly try it again. So anything which will help me on my way is very, very appreciated. Anything that would improve resizing with liveResizing off, surely is already a big help. Best wishes Claudi On 24-okt-06, at 01:28, Derek Bump wrote: Bill, I encountered the same issue as you a while back, but with a different question. How does one use a custom windowshape and still keep the taskBar button visible (cause rev won't let ya!)? The answer to our questions was use an external to draw the windowshape. Now at this point it doesn't support alpha-transparency... but it does support resizeable rounded-rectangle windows. The now defunct Externals Collection has a command that allows you to pick a custom windowshape. Combine that with a resize stack button (like the one available within the Objects Library) and you get a resizeable window with a round rectangle. The download site is located: http://support.runrev.com/resources/unsupported.php If you have any questions on how to accomplish this, let me know. Derek Bump Dreamscape Software ___ Compress photos quickly and easily with JPEGCompress! http://www.dreamscapesoftware.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: Resizable stack with a custom shape
Hi Derek, I am currently working with windows with emty decorations. (I create my own titelbar) Resizing then surely looks ugly when growing a stack. (the stack resizes but the grown, newly exposed parts are first seen as white and then later get the backcolor) By the way this always happen if liveResizing is off. (lock screen doesn't help, it's in the engin I guess) So I am sure interested in the external you mentioned and I dowloaded the externals stack. Unfortunatly I couldn't find the external for resizing in it. If you have the external and a example I would really appreciate it. I have done some tests with resizing with a custom shape for the window. Besides the darned flashing when you grow a window and some sluggishness It worked more or less. (this was some time ago so I don't remember exactly all the problems.) I used a img in a hidden stack which I resized first to the right size and then set the windowshape again for my main window (something like that. I remember ditching it at the time since the whole stack slowed down I mean every script became slower, which I never quite understood.) For what I am developing resizable custom shaped windows are actualy the most preffered way and I will certainly try it again. So anything which will help me on my way is very, very appreciated. Anything that would improve resizing with liveResizing off, surely is already a big help. Best wishes Claudi On 24-okt-06, at 01:28, Derek Bump wrote: Bill, I encountered the same issue as you a while back, but with a different question. How does one use a custom windowshape and still keep the taskBar button visible (cause rev won't let ya!)? The answer to our questions was use an external to draw the windowshape. Now at this point it doesn't support alpha-transparency... but it does support resizeable rounded-rectangle windows. The now defunct Externals Collection has a command that allows you to pick a custom windowshape. Combine that with a resize stack button (like the one available within the Objects Library) and you get a resizeable window with a round rectangle. The download site is located: http://support.runrev.com/resources/unsupported.php If you have any questions on how to accomplish this, let me know. Derek Bump Dreamscape Software ___ Compress photos quickly and easily with JPEGCompress! http://www.dreamscapesoftware.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
blendlevel magic
Hi, I just discovered a way to cool trick with blendlevels and (list)fields. You can have multicolored hilitedLines for a field. No problem. No special coding neccesary, you just set the hilitedLines of 2 flds a top of each other. That's all. Create 2 flds same size on top of each other. Set the opaque of the top one to false Put as many empty lines in the bottom one as you have lines of text in the top one. Now set the hiliteColor of the 2 flds to something different, let's say one is bright blue and the other is quite red. Next set the blendlevel of the top fld to somewhere between 30 and 70% also play with the inks. (a nice one is for instance bothl blendlevels 0 and ink for the bottom layer to bitwise scrCopy and the ink for the top one to arithmic blend) Anyway now set the hilitedLines of the base fld, the one with the empty lines, to lets say 1, 2 and of the top fld to 2,3. Now we have 3 differently colored lines. Sync the 2 flds up and you can have magic, Heck you can do it with 3 flds aswell. Alternating colored lines, it's easy. Except for syncing the scroll of the 2 fld's, no extra code necessary. Neat, ... ... very neat. Happy coding you guys and gals Claudi ___ 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: Why do group ID's change when cloning a stack
Hi all, Well, I do name all my controls, because ofcourse it's a lot clearer to read the code. The thing I am working on now is a frame-border kind of object. I design it in a 3D program slice into the right bits and import the images into runrev where they get assembled. There is provision for different states, like focused, unfocused, dimmed, dragEnter etc. There may also be many different styles and substyles, which ofcourse have theire different states. So actualy the component is a grp of grps which contain grps which contain img's. All these style grp's have basicaly the same setup, with the same names for theire subgrps and subparts. Now I can treat each grp the same way with the same code because the names of their subgroups/parts are all the same. So I only need the id of the enclosing grp to be very specific because id's are unique, whereas in this setup the names are NOT different on PURPOSE. That's the nice thing about unique id's combined with names which needen't be unique. With storing the unique id of the parent grp I can adress very specificly all parts in that grp although there might be tens or hundred of similar grps nested in other grps, which are nested in other grps. One id makes it very precise. Because of the unique id's I only need to store one ID istead of the whole chain of names till I reach a unique part which might be several levels up (img left of grp focused of grp border of grp greenstyle of grp defaultStyles of grp styles of grp fieldFrame. Instead of img left of grp id 1003) Well sure the id's stay unique, but they are not safe because they get changed. This is a real nuissance. What I am doing now is only about 3 levels deep, so it is manageable altough now I need to insert code to check for duplicate names otherwise it might get screwed up completly. (Names might be changed on the fly) I know I will get into situations where the number of levels get way beyond 3 and then it becomes very awkward. Also I believe this should be documented somewhere 'cause it might waste people a lot of time. I am also interested WHY this should happen. If you have stack A with 2 grp's and you clone the stack then the id's of the 2 grp's will stay the same but every control in the grp will get a new ID. If I clone the same stack A again I now have 2 cloned stacks which are identical that is to say all the ids of all controls in both cloned stacks are exactly the same and both differ from the original. So Why? Best wishes Claudi ___ 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
Why do group ID's change when cloning a stack
Hi all, By accident I just discovered that group ID's change when you clone a stack. This can be very annoying when you work with tmplt stacks which you clone to use as, lets say a document or a viewer. I just did some testing, but button id's don't seem to change. Neither the id's of fields. But sure enough group id's do change, altough always in the same way. They start with the highest group id in the stack you clone incremented by one and so further up. Why? Why only groups. (well to be honest I didn't try with all controls, so there might be other controls who's id will also get changed when cloning a stack. This is a real nuissance for me since I rely heavilly on group id's in the stuff I am making now, which will get used as a tmplt stack and thus gets cloned for actual use. Lots of nested grps and I designed it around the id's, because that's safer there might be duplicate names. But it seems now I can't use the id's because they pull the rug underneath my feet when cloning the stack for use. This is in rev 2.7.2 (OsX) Am I missing something. Is it supposed to behave like that, is this a bug? I have no idea except that this is gonna cost me quite some work and planning of a new strategie to get this working as it should. Damn (lots of rewrites in several library's) Well you do learn each day, but it is tough going like this. If somebody can shed some light on this dark matter, at least it is for me, please do. Best wishes Claudi ___ 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: Why do group ID's change when cloning a stack
I did some further testing and it turns out like this: All objects (at least the ones I tested and I did most) don't change id when the stack is cloned. That is all controls that are not part of a grp! The id of all outermost grps don't change but the id's of ALL controls in a grp will change, including the id's of nested grp's. So each and every control in a grp get's it's id changed when the stack is cloned, except the id's of the outermost grp's. What would be the reason behind this? Why doesn't the id of the outermost grp's change but each and every control in each grp get's a new ID. What will happen when I clone a stack with hundred of grp's. How much time wil this take if each control in all these grp's need to be reIDed. I can live with the fact, well I will have to, there obviously is no choice, but I sure like to know! Best wishes, Claudi ___ 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: I am glad I can donate something
Hi all, Fixed the cursor, I hope it's good this time (I checked it with rev 2.7.1 In the update I send before I had accidently updated the cursor in a document instead of the tmplt - stupid me). Martin I am almost sure there will be some cross-platform thingy's. That's still something I'll have to start learning. (like many other things) Best wishes Claudi 2Gather is a outliner (treeview) You can download it from: www.cc-imaginering.nl/runrev/2Gather/2Gather.rev.zip ___ 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: I am glad I can donate something
Hi, I am glad you like it. Questions about tree views have been on the list quite often, so I hope this can help some people to get started. Much more could be said about treeviews / hierarchical lists and list fields in general. I am working on a librarie for this kind of thing right now. It's quite complex to get it all right. It has to handle, folders, treeviews, and other compound datastructures, divider lines, checkboxes, and other kinds of icons like an eye or a checkmark etc in front of a line (or someplace inbetween). All this with full drag and drop etc. It's a lot of work but slowly it's getting along. Anyway I have changed the cursor id, I believe it was a standart cursor in rev 2.5. It's updated now and I also removed the breakpoints. You can download it from: www.cc-imaginering.nl/runrev/2Gather/2Gather.rev.zip For those who like it you, can now also download the standalone (unfortunatly only for mac, I didn't try it on windows, if anyone get's it working on windows I would appreciate a copy so I can learn some of the windows part) You can download the standalone from: www.cc-imaginering.nl/runrev/2Gather/2Gather.app.zip Best wishes to you all, Claudi ___ 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
I am glad I can donate something
Hi all, Perhaps some of you have occasionaly seen my name on this list, asking some, somtimes dumb, question, but I read the list with much interest. So you might say I am a lurker. (God I hate that word) Well I have learned an awfull lot of you all and I feel it's time to give something back. I give you all ... ... drum roll ... ...2Gather. I have uploaded a screenshot and the link to download it to my userspace claudi under the categorie general. You can also just download it from: www.cc-imaginering.nl/runrev/2Gather/2Gather.rev.zip (This way I can keep track if there is interrest) 2Gather is an outliner. It's quite standard except that each entry can have a text / an image / and audio data all at the same time. I know it works quite well on my mac, since I am using it as a standalone to document all my programming and quite some other stuff as well. It started out as a tmplt/exercise for apps that use documents to save data. Things surely can take strange turn. Anyway, it's not finished or polished. It's more like a sketchbook. Me learning rev (long ago I used to dabble in hypercard) and trying to get to grips with this strange and wonderfull world of revolution. So there are a lot of ideas embedded in it just to get the hang of it. (how to handel documents with the same name, how to work with menus, how to have a good text menu, a window menu, preferences, etc). Unfortunatly it's not much documented, and probably it's part englisch and part dutch, I have not looked at the code for a while. It's in the same state as I compiled it last (quite some month ago). If there is enough interest I might be able to fix that, although that might take a while and I spend all my time on something new. Anyway see it as a strange and somewhat coherent set of thoughts/routines. I am not a programmer but I do programming and now with my acquired knowledge am gonna try to create something much more complex than a basic outliner. Ideas, data, creativity is much more three dimensional than a 2 dimensional outliner and human behaviour and thought is also much more fluid and flexible than outliners or minmaps can offer. (at least all the outliners I have seen and I have seen a lot of them and they just don't come close) Well I hope some of you might find 2Gather or some of it's code usefull in some way, you are all free to do with it as you please / like / deem neccesary or just for fun. On the other hand, of course, I can't take any responsability for any damage this code may cause in any way, Use it at youre own risk. The name, the creator codeand the extension I keep reserved for my own use however. So, I hope someone finds some good use for this and perhaps next time you see my name on the list I will be a bit less of a stranger. Best wishes and happy coding you all, Claudi ___ 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
userspace
Hi all, I have a stack, some kind of outliner, which I would like to donate to this community. Of course my userspace would seem logical, but I would like to know how many people download it (if there is a lot of interest it might stimulate me to document the code more and to prune the code. It's like it is now and I am working on something much more than a outliner now and I spend all my time on that). Outliners are nice but they are actualy kindergarten stuff and with the amount of available data nowadays we need something much more flexible and fluid. So my actual question is, is there some way to track the download statistics of my userspace? Otherwise I will probably make a little demo and upload that to my userspace and add a btn to this demo to download the real stack from my site. That's if there is any interest in this at all. I use it on my mac as a standalone on a daily basis. It saves double clickable documents. (well at least it did, although the last compiled version suddenly doesn't anymore. It could have to do with a missing applescript resource in the compiled app I read somewhere on this list) Best wishes Claudi ___ 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: Paragraph Level Text Formatting
I am very interested in this aswell since I will be needing this in the near future. So for me it will certainly be worthy. Best wishes Claudi Op 23-mrt-06 om 20:46 heeft Jonathan Lynch het volgende geschreven: My faerie tale generator stack (still incomplete) handles inputting text that allows formatted text to flow between fields in a series. The fields can be shaped however you wish. I sent it to Ken to check out. If he thinks it worthy, I will post it online whereever he suggests. For my uses, it works quite well. ___ 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: relayering groups within groups?
Hi There is a trick I use to get sub-groups within a master group to the top when they get selected. I have 2 invisible grafics (just empty small rects) at the top of the master group. Let's call them lowBoundry and highBoundry Now when the user selects a sub group and I want to bring it to the front, I first move the sub group inbetween the 2 invisible graphics. In this way the sub group does not become part of the highest sub group. Next I move the the lowBoundry graphic to the top. So now from top to bottom: lowBoundry - highBoundy - the sub group that was moved. So all we need to do now is move the highBoundry again to the top to restore there right order. There is one important little snag though. Say the sub group to move to the top is on layer 10 and consists of 2 controls. The total of layers in the master group (including the 2 invisible graphics) is 16. This means the lowBoundry wil be on layer 15. The moment you change the layer of a sub group it will first be taken out of the stack. In this case the sub group consists of 2 controls, so the total number of layers this sub group occupies is 3. This means the lowBoundry graphic will first fall down to layer 12 (15 - 3). In essence all the controls above the sub group we move, will fall down by the number of layers the moving group occupies. This trick works very fast and it doesn't matter how many sub groups there are. Of course it only works to bring a sub group to the top, but probably when selecting objects, something like this might just be what you need. I use it in a slightly more complex way in an application where I have many sub groups but the sub groups are of different specific types. Let's say some have a triangular shape, others a circular shape and the third kind are of square shape. Now when the user selects a triangle it should move to the front of all triangles but always stay behind the circles and the squares. The same applies of course to the circles and the squares. Each should move to the top of it's kind but not cross the boundry to other types. Here I use 3 invisible graphics between the kinds. So above all triangles there are 3 invis grc's then come the circles with also 3 invis grc's and next come the squares and finaly again 3 invis grc's (actualy for the top, 2 graphics is enough) In between you need 3 graphics of which the topmost never moves, but prevents the other 2 graphics to become part of a sub group when they are switched around. I hope you can use it or that some one else finds this usefull Best wishes Claudi On 1-feb-06, at 21:16, Jonathan Lynch wrote: Hi Terry, I use this technique to adjust the layer of a subgroup that contains an image, which is inside a group that works as a page. A page can contain fields and images. All the images are inside groups in order to allow them to be reversably cropped. So, I might have 5 to 10 objects on a page. Doing this, relayering an image group takes maybe 1/5 of a second - something like that. The lag is definitely noticeable. With 100 objects to relayer, I bet you will notice the lag time. If you manage to figure out a better way, please let me know. Good luck, Jonathan On 2/1/06, Terry Judd [EMAIL PROTECTED] wrote: Hi Jonathon - I arrived at a similar solution after much stuffing about last night. My only concern is that I may have to do this with lots of objects (100+?) every time I want to drag one about. Hopefully it's not too slow. Thanks for the info. Terry... I have a solution to this problem. It was a real pain to figure out. The basic problem is in the way Revolution orders items inside of a group. To keep a subgroup intact, when you relayer it inside of a group, you have to set it's layer to the bottom of the group. So, the question arises, how do you move a subgroup up and down, if the only safe method is setting it to the very bottom. The answer is this: When you are moving a subgroup down one layer or up one layer, what you really do is determine when layer the subgroup needs to be at, then move all the other subgroups and objects in the group to the bottom, in the correct order, so that everything else stays in the order it was in, and the subgroup winds up at the layer you want it to be at. It's a giant pain, but it works. I can send you an example of a stack that does this, if you wish. Take care, Jonathan ___ 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 ___ use-revolution
probable bug of fields
Hi all, I found some strange behaviour in fields which I think is a bug. Create a new field with the dontWrap and the hScrollbar both set to true. Now type in some text till the hScroll will have started to scroll. Select most of the text and do menu cut. Wow the hScroll is still active. Now you can even select the rest of the text and hit backspace. The field will be empty by now, but you can still scroll horizontaly. Ugly, but mainly cosmetic. Now check on the formattedWidth of the field. It will be more than the width of the field, but there is no text in the field, it's empty. This one is nasty. It means the formattedWidth is not always correct and can't be trusted. From the moment some text has been cut, the formattedWidth will be wrong till newly typed in text will exceed the old formattedWidth. From then on it's correct again. Can someone confirm this? I use the latest build of 2.6 and OS 10.2.8 Claudi ___ 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
can't get keyboard shortcuts to work
Hi all, Probably I am missing something quite obvious, but I can't get my menu shortcuts to work. In the menubuilder I made a menubar with different menu shortcuts. In the menu I have the usual menuPick handler with a switch statement like on menuPick pItem switch pItem ... ... However when the menu gets invoked with a keyboard shortcut pItem is always empty, so ofcourse none of the cases in the switch ever gets triggerd. (the menuPick handler gets called correctly but no parameter is passed) Should keyboard shortcuts be handled differently or is there some propertie I should set that I am not aware of? I am using runrev 2.6 and OS10.2.8 Thanks a lot in advance Claudi ___ 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
Detecting height of dock and taskbar
Hi all Is there a way to detect the height of the dock in OSX and of the taskbar in windows. I try to create a complete set of routines to: create open and save documents (windows and mac). While opening a document I want it to be completly on the screen (also the bottomRight so it's possible to resize the doc) and have an offset for the topLeft from other open docs. For this I need to know how much room to leave at the bottom for the dock or the taskbar. I guess I need a shell command to do this, but unfortunatly I have no clue I hope someone on this great and very usefull list has a solution. Best wishes Claudi ___ 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: scripting challenge: Large/Smaller text size
woow, I realy feel humble. I have been working on a text menu with bigger smaller as one of the features. I have been working on it quite a while. I have got it working but now as I see these bautifull and elegant solutions, I can start all over again. (putting a couple of checks in of course so not to make text smaller then 7 pts which gives strange results and a pref for how to proceed when the smallest size in the selected text reaches 7pts, stopping or leaving that part at 7 pts and making the rest smaller) There are so many ways to do things in runrev and it's very enlightning to see how the pros tackle a problem. One of the best ways to learn for me (and others as well I guess) Couldn't we have chalenges like this, let's say once a week or once a month? My guess is that a lot of people could realy benefit from these chalenges/examples. In my own solution I used matchChunk to get at all the font tags with a size attribute, which also works but uses a lot more code and altough reaseably fast not in a long shot like the code of Mark. One advantage of all my toiling though, I learned a bit about regex, I am not completly confused by it anymore. Very humbly and awe strikken Claudi ___ 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
How to put a dash in front of a menu item
Hi all Still working/improving my demo of a textMenu. It's almost ready. I can now also increase/decrease textsizes. Lighten/darken is just a snap away (it's the same as increase/decrease with a slight twist) and the menu displayscorrectly all attributes. There is just one thing I'd like to add to it: putting a dash in front of a menu(item). Is there a way I can achieve this? This way the user could see if a selection has mixed attributes or if it's all the same. The way I found in the documentation, and after some tests (like putting a \ in front of the - which does'n't give quite the desired effect. ) it doesn't work. As far as I discovered we can put a checkMark or a diamond in front of a menu but not a dash. I hope I am wrong, am I? In case we indeed, can't put a dash in front of a menu, wouldn't this be a usefull enhanchement? Should I, in a yet to figure out way, post this as a request? All best wishes Claudi ___ 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: matchChunk to find fontNames
Thanks Björnke, Eric and Marielle, You all helped me. In the mean time I got the matchChunkpart also worked out, and am putting it all together now. I will try to make a complete textMenu example, with all the features you ever wanted, plus a few more I hope (well, we will see about that one) Regex can be really fun and it's not quite that difficult after all and tremendously usefull. It took me a while but I learned quite a lot, and all though I don't know every thing I am certainly not afraid of regex anymore. I am putting the last bits and pieces together and smooth the whole thing a bit. It's still quite experimental at the moment, but as soon as it's descent I will post it In the mean time all the best. Claudi On Thursday, June 23, 2005, at 02:18 AM, Björnke von Gierke wrote: On Jun 23 2005, at 00:29, Claudi Cornaz wrote: Hi all, I am trying to implement a text menu and need to find all the fontNames and text attributes in the selectedChunk first to update my menu and afterwards to handle the choice. As long as all the text has all the same attributes (bold / italic etc) it's no problem but I need also to find all the attributes when it's mixed. I have been trying to get the instances of font face= xxx - xxx [Size=yy] [color=#ZZ] to extract all the parts like, the fontname, the size and the color if applicable with matchChunk. you could get the foregroundColor (or more aptly named the textColor) of the char. The textStyle property reports the current styles delimited by comma, if you query only one style at once. The textSize again returns mixed for a whole selection, but you can get it char per char, and it will give you exact data that way. so for example: --preparation put empty into theColor put empty into theSize put empty into theStyle put word two of the selectedchunk into theStart put word four of the selectedchunk into theEnd put word -1 of the selectedchunk into theField --getting the actual data from the chars repeat with x = theStart to theEnd put (the effective textColor of char x of field theField) comma after theColor put (the effective textSize of char x of field theField) comma after theSize put (the effective textStyle of char x of field theField) comma after theStyle end repeat --now we can do whatever we want with the assembled list --for example to know what the dominant textSize is: put empty into theSizes --adding them up by using an array repeat for each item theItem in theSize add 1 to theSizes[theItem] end repeat --finall cleanup of the data we got combine theSizes using return and comma sort theSizes numeric by item 2 of each put theSizes --and finally you got your most used size! put line -1 of theSizes hope it does what you want? Björnke who wrote this mail Wouter and Mark who gave some vital hints (arrays, counting, sorting) -- http://contest.wecode.org Now running: the first ChatRev coding contest! sponsors: Altuit Andre Garzia Fourth World Karl Becker Runtime Revolution TidBITS in cooperation with eHUG ___ 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
matchChunk to find fontNames
Hi all, I am trying to implement a text menu and need to find all the fontNames and text attributes in the selectedChunk first to update my menu and afterwards to handle the choice. As long as all the text has all the same attributes (bold / italic etc) it's no problem but I need also to find all the attributes when it's mixed. I have been trying to get the instances of font face= xxx - xxx [Size=yy] [color=#ZZ] to extract all the parts like, the fontname, the size and the color if applicable with matchChunk. I start to understand regex a little bit, but I still did not got it to work just right. Here is what I got sofar to get the whole font face tag: matchChunk(tText, (font face=\[A-z\- ]+\), tStart, tEnd) Some fontNames have spaces or - in theire name but if I add them to the regex like in the above example the end goes on to far like: Stone Sans OS ITC TT-Bold color=#FFzeker/font . . . . . . etc I tried to change the regex part with a $ at the end so it should stop at the char. matchChunk(tText, (font face=\[A-z\- ]+\$), tStart, tEnd) Well, no dice, now it finds nothing no more so obviously I am doing something wrong. How should this be done? Sugestions for a better way to find all the style runs are of course very welcome as well, since I just thought of this way to do it, in fact after quite some thought this came as best solution to my mind, but wether this is the best way to do it, well probably not. Anyway much thanks in advance and all help appreciated best wishes Claudi ___ 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: Make icons on a Mac?
Hi Kat, I use Iconographer and with this app I can create icons for mac (Os9 and OSX) aswell as for windows (aslo different formats, so also for XP) It's easy to use and suites my needs very well. (www.mscape.com) Best wishes Claudi On Saturday, June 18, 2005, at 05:27 AM, Kat wrote: I've been following the discussion about icons with great interest. I'd like to create some for my Rev products, but would rather NOT have to use Windows if I can avoid it. Is there a Mac application that makes icons that work in Revolution? Cheers, Kat Katherine Cochrane http://rugusa.linguistix.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 ___ 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: is there a better way to do this?
Hi Björnke, I can't get your sugestion to work. That's why I do it with a do command. I tried the following script. (uTestSet[uValue] is Yes) on mouseUp put uTestSet into tVar put the tVar[uValue] of this stack into fld debug put cr - cr after fld debug put put the tVar [uValue] of this stack after fld quote debug quote into tCommandStr do tCommandStr end mouseUp The result is: - Yes As you can see the first line is empty and when done with the do construct it works. (the line after --) What am I doing wrong? Best wishes Claudi On Tuesday, June 14, 2005, at 12:52 AM, Björnke von Gierke wrote: On Jun 14 2005, at 00:03, Claudi Cornaz wrote: .. (with uText possibly also in a var?) .. oh, didn't see this part of your mail, you wanted something like this, right? on mouseUp put new into tSetname put uText into tProp put tSetname [ tProp ] into cprop put the cprop of this stack end mouseUp -- http://contest.wecode.org Now running: the first ChatRev coding contest! sponsors: Altuit Andre Garzia Karl Becker Runtime Revolution TidBITS in cooperation with eHUG ___ use-revolution mailing list use-revolution@lists.runrev.com http://lists.runrev.com/mailman/listinfo/use-revolution ___ use-revolution mailing list use-revolution@lists.runrev.com http://lists.runrev.com/mailman/listinfo/use-revolution
Re: is there a better way to do this?
Thanks Björnke I had to read it twice before I got it, but I did. (get it the second time.) The whole propertiySet including the customKeypart should be in one variable, you can't assemble it in one statement. Put it all in a var before you use it and it's ok. Thanks a lot Warmest regards Claudi On Tuesday, June 14, 2005, at 02:30 PM, Björnke von Gierke wrote: Hi Claudi You need to assemble your String that you want to put before you do anything else with it, like this: on mouseUp put uTestSet into tVar put tVar [uValue] into cprop --alternatively if uValue is a variable: --put tvar [ uValue ] into cprop put the cprop of this stack into fld debug put cr - cr after fld debug put put the tVar [uValue] of this stack after fld quote debug quote into tCommandStr do tCommandStr end mouseUp On Jun 14 2005, at 12:03, Claudi Cornaz wrote: Hi Björnke, I can't get your sugestion to work. That's why I do it with a do command. I tried the following script. (uTestSet[uValue] is Yes) on mouseUp put uTestSet into tVar put the tVar[uValue] of this stack into fld debug put cr - cr after fld debug put put the tVar [uValue] of this stack after fld quote debug quote into tCommandStr do tCommandStr end mouseUp The result is: - Yes As you can see the first line is empty and when done with the do construct it works. (the line after --) What am I doing wrong? Best wishes Claudi On Tuesday, June 14, 2005, at 12:52 AM, Björnke von Gierke wrote: On Jun 14 2005, at 00:03, Claudi Cornaz wrote: .. (with uText possibly also in a var?) .. oh, didn't see this part of your mail, you wanted something like this, right? on mouseUp put new into tSetname put uText into tProp put tSetname [ tProp ] into cprop put the cprop of this stack end mouseUp -- http://contest.wecode.org Now running: the first ChatRev coding contest! sponsors: Altuit Andre Garzia Karl Becker Runtime Revolution TidBITS in cooperation with eHUG ___ use-revolution mailing list use-revolution@lists.runrev.com http://lists.runrev.com/mailman/listinfo/use-revolution ___ use-revolution mailing list use-revolution@lists.runrev.com http://lists.runrev.com/mailman/listinfo/use-revolution ___ use-revolution mailing list use-revolution@lists.runrev.com http://lists.runrev.com/mailman/listinfo/use-revolution ___ use-revolution mailing list use-revolution@lists.runrev.com http://lists.runrev.com/mailman/listinfo/use-revolution
is there a better way to do this?
Hi all, I am trying to find a better way to get the value of a customkey of a custompropertySet when the name of the set is in a var. The only way I get it to work is with a do statement like ( tSetName contains the name of a customPropertySet) put put the tSetName [uText] of this stack into tText into tCommandStr do tCommandStr Is there another, easyer and faster way? (with uText possibly also in a var?) Thanks in advance Claudi ___ use-revolution mailing list use-revolution@lists.runrev.com http://lists.runrev.com/mailman/listinfo/use-revolution
Re: imageSource bug?
Hi Sarah and Brian, Thanks for the response. After getting my code more stable, that is now there are only sporadic errors while running the scripts, the problem seems to have disapeared. I suspect sometimes my errors get the engine in quite a confused state so it does sometimes unexpected things. (like turning all chars in all lines of a field with an imageSource in to spaces for instance) Well making mistakes is an important way of learning and boy can I make them. Best wishes Claudi ___ use-revolution mailing list use-revolution@lists.runrev.com http://lists.runrev.com/mailman/listinfo/use-revolution
imageSource bug?
Hi all, I am working on some sort of outliner. On each line the first word is like a header. It can have zero or more # chars indicating the level of indent followed by one char which can be ^, * or . This is followed by the name of the topic. I set the imageSource of all the chars in the first word. # is an transparent gif and the other 3 chars are images of a triangle. Ok all works well for a while and suddenly my scripts break. The intendetion and the triangles are still ok but the first word has disapeared. I finaly discovered what happens. All the chars with an imageSource get replaced by ascii 32, a space char, so of course my first word, or what should be my first word is gone since it has become just some spaces. Has anyone expirienced this before and/or does someone know a work around for this unwanted behaviour. I checked the archives but found nothing directly. There was some mentioning of html text to speed things up but that's not quite related. All help is much apreciated since this has cost me already quite a lot of time and frustration. Sincerly Claudi Ps This is with OS X.2 and both with runrev 2.5.1 and 2.6 ___ use-revolution mailing list use-revolution@lists.runrev.com http://lists.runrev.com/mailman/listinfo/use-revolution
custom propertieset woes
Hi List, I hope someone can help me, because I just don't get it. I have a field with some custom propertie sets. The sets are named ux where x is a unique number. Each set has two properties. One is a list of a subset of the set names and the other is just some text, like the example below. setName: u123456 [uSubSet] u234567 u345678 u 456789 [uText] some text setName: u 234567 [uSubSet] u826545 u876543 [uText] some text setName: u u826545 [uSubSet] u858743 u158741 [uText] some text etc. Now I take a initial propertyset and extract the subSet and iterate through this subSet (with of course the iteration through that subSet's subSet) to get all the subSets. The weird thing is the first iteration works perfect. The second and all subsequent ones failed, returning a empty subList (which isn't empty at all) After putting a line in the script to put the subList into a field for checking, the second iteration suddenly started to work but the third and all subsequent ones still returned empty. I use thefollowing construct to get the subSet (where tPropName is the name of a property set) put put the tPropName [uSubList] of fld quote topicList quote into tSubList into tCommandStr do tCommandStr So in the debugger I copied tCommandStr and tried it in the msg box (without the part into tSubList offcourse). While the script is halted, 'cause of a breakpoint the command in the msg box will return empty. After aborting the script at that point and hitting return in the msg box again I get the expected result. Why will it work in the msg box and not in the script? Next I added a line in the script to output all the tCommandStr's to a field. When I copy them to the msg box they all work as expected but still the script returns empty after the second iteration. How is this possible? It's driving me nuts. Why should a line in a script work 2 iterations but not a third time? I hope someone can shed some light on this so I can understand what's going on. I will be using propertiesets extensibly in almost all the programming I do so it's quite important for me to understand what is happening. Best regards, Claudi ___ use-revolution mailing list use-revolution@lists.runrev.com http://lists.runrev.com/mailman/listinfo/use-revolution
Re: custom propertieset woes
Hi Xavier, Here is the script -- *** function GatherAllSubs pLine put word 2 to -1 of pLine into tGatheredList put GetPropName(tGatheredList) into tPropName put put the tPropName [uSubList] of fld quote topicList quote into tSubList into tCommandStr do tCommandStr -- tSubList now contains the list exactly as it would appear in the topicList fld. Now proces this list repeat for each line tTopic in tSubList -- each topic might have a. no subs b. closed subsc. open subs if last char of word 1 of tTopic is kClosedSubs then put cr GatherAllSubs(tTopic) after tGatheredList else put cr word 2 to -1 of tTopic after tGatheredList end if end repeat return tGatheredList end GatherAllSubs pLine will be in the form: ##^ This is topic 1 word 1 is like a header with some info I use also in other places. kClosedSubs is a constant for the char the function GetPropName() will return the propertie sets name associated with This is topic 1 (or whatever will be in pLine) (so it will return like u126543) After the line do tCommandStr I put a line to output tCommandStr to a field to see if there something wrong with it. sample output is: put the uCC835628448213[uSubList] of fld topicList into tSubList Copying this line to the msg box (without the into tSubList part) it works perfectly as expected returning the right values, but executed in the script it puts strangly enough empty into tSubList. Why would a line of transcript not work as expected in a script, but the same line works perfectly well in the msg box? unless offcourse there is something wrong with the into tSubList part. I also restarted runrev and now it's back to doing only the first iteration. Ps. In my first mail I put u24536 etc in the subList of the set for simplicity. In reality I get the sets name via the GetPropName function 'cause in the subList will be text like: ###* And this is another topic ^ and this one is one level deeper Hope this helps to spot the problem. It has me completly baffled. Best wishes Claudi___ use-revolution mailing list use-revolution@lists.runrev.com http://lists.runrev.com/mailman/listinfo/use-revolution
