Zip

If I get what you ask, what you are looking for already exists, not sure why no one mentioned it. you can use the Keyword '#Region' '#EndRegion' for this.

wow ..Jos ,why do you feel that way? not sure what I have said to upset you, I was referring to the lack of interest from the forum, you're the only one that showed interest, & immediately recognized the issues in my bug reports, I only posted this so you can get some idea of what I meant , following our previous discussion. so trust me when I say this was in no way whatsoever directed at you, but as you wish, will not post again in this thread. @Valik , made the colors stand out, so the screenshots are clear, obviously, it can be defined like anything else, but I know where you stand.

seriously ?

well, despite the lack of interest , decided to get a basic proof of concept done, in an attempt to showcase the usefulness of my advanced comment block concept, made examples displaying: in document Autoit Help, Code fragments, Color highlighting, Comment folding. All the best.

You act & might think, like you hold the crystal ball of people's minds & concepts, but you dont. In fact I find some of the choices you made in your comments& the pesky issues you chose to focus on, quite alarming. But I couldn't care less about that. I do find it unethical you exercising some way too liberal & offending choice of words, just because your'e a mod here, instead of the other way around. If you find it an acceptable form of communication & believe it strengthens your view fine. Just make sure you do not fall into believing people let you get away with it, because your point is a valid one.

Thanks a lot Jos for verifying & commenting on this issue,when I said "for that to happen, to keep as simple as possible, this should be done regardless ..., " , I was afraid there's no way to workaround it but do some kind of additional logic . no, have not done any debugging on it, so far, I only figured out what I did by following the code. Either way, it's in your hands. Here's what I had in mind in more detail: My concept of it was, with the lexer being able to do the above, When inside a #CS/#CE block, using transparency, instead of a single color, Notes & code examples/variations/Usage instruction can be very clear ,& visually distinctive, just like regular code. Thinking further, this can further enhance Autoit's code examples. - if it all appealing, I can probably work out a demo variation of the lexer as an example, showing a demo document, but chances are implementation will still be needed to be done by a pro. @Valik > please read the above, & do not address me further regarding anything but the lexer in this thread.

I see. Heres my take on your view: Maybe you noticed, I did not say bad code. you just assumed / inserted it in my words to give your words merit/make a point. Whatever the intentions you might have had saying what you did, are lost. overall throughout ,I find the bits you pick & choose to focus on quite puzzling.

with all due respect ... nothing I have done or said is stupid .. care to elaborate ?

lol ... old ...right ..... in that case ... Scite>New>Paste : #CS Switch ;~ EndSwitch ; #CE #CS ;~ Switch EndSwitch ; #CEetc etc On why do I want this .. 1. Ill give you the medium-short version ... (you do mean aside from being that kind of person ... cant stand something going funky on me, when It's close to my heart ..on WHY Autoit is close to my heart .. well, I will not bore the whole forum with that. That means .. I was (& still am) new to Autoit. which actually means I was new to programming altogether. With Newbie Scars still fresh, I got to a point that I just cannot continue unless I get a working more friendly Scite, (note that I was blaming Scite for the Au3 lexer bugs as well ....as every devoted newbie should ..) which led me to halt my Autoit projects, & do something about that, so here I am. Bottom Line: Newbies need a fast way of ditching old code/examples & documenting their progress & hiding it, inside the document, (pros like you don't).

Actually now that I think about it, for that to happen, to keep as simple as possible, this should be done regardless ... your thoughts?

@Jos , lol your'e asking me why (straight answer now ..) ? tbh, I (not being a pro) would never have remembered why it is I did it this way, or what was I attempting to fix, this way I can remember I am trying to workaround to force processing for the line, from looking at the diff alone, & not wonder why. Another issue I noticed & now thinking would be nice to be fixed, is having code fragments (read: invalid code) inside #CS#CE blocks, without throwing the lexer into a fit & invalidating both folds+highlights , (with the added benefit some may find in not scaring a newbie writing invalid code inside the block to death )

@Jos, OK, had some time today & worked with some of my Au3 code, did have some instances where highlights are being expanded beyond the closing tag, & last close tag not folding,( & it did start to annoy me), wrote this small patch to get rid of these issues (attached). LexAu3 v5 to V5+ FIX Last lines end tags not folding + fold highlight might show invalid range.diff.txt

oops hehe , will do

@Jos, nah man, thank you for listening & doing the fixes so fast. if possible, an updated lexau3, in the beta dir, will help me keep up to date.

I hear you, well have to discuss it later, I did not do a good job of describing my full intent, maybe I will later on. Back to the matters at hand, Reporting: I think I notice an End Select/Switch not shown issue when unfolded(had the same problem here at 1st, so noticed it immediately) Edit: Im using Array.au3 so we can use the same source file to test, looking at _ArrayDelete, "Switch UBound($avarray, 0)", with all folds folded & opening this Switch only.

@ Jos Just got in & noticed your post, that's great, downloading as I type. at this time it's probably a good idea to version the LexAu3, to make sure were on the same page when discussing it. Here, I have versioned them, as "LexAU3.cxx.V0" with the original I had (2007) "LexAU3.cxx.V1 Before CS Fix" :the Scite4Autoit updated one you gave me when this thread's discussion started, "LexAU3.cxx.V2" #CS fix 1 (not fully working) "LexAU3.cxx.v3" #CS fix 2 (working). So Ill refer to this one you just posted as V4. on V3, as I told you, for my own use, I did end up quickly modifying it slightly to support the fold.highlight, & bring it in line with other languages display wise. (in my case , I did end up just making it use the old fold behavior if fold highlight is disabled, the V3 to V3 + fold highlight quick patch I currently use is attached) well in that case, now that you enables support, ill just use the new one! Thanks ! Actually, both related & unrelated to this, here's another thing I thought would be quite useful, if you mean all these comments detected with ';' maybe using a specific definition for the currently unused comment.box can bring in another viable comment type, which will also serve as a different type comment, & has its own color definition as often used. In Autoit's case can be quite useful to tell between user comments, & udf/function comments etc. grabbing now, reporting in half an hour max. Scintilla_2_to_6_Au3 v3 to v3 with my Foldhighlight Support.diff.txt

( Addictive lil thing aint it , funny thing is, I was not even aware of its existence, until looking at the source code when I started my attempt quest of adding some features I needed from scite for my own use). -great will follow this thread.

Well in that case glad I took the screen cap after all, Yeah its one of these options not always used, but there should have been a warning sticker on it ,as you do get spoiled by it once it's on . sure I am aware of that (reported just to make sure it's known, & within your design goals, as an attempt to give back relevant input). To be honest if this option is off and your'e not coding at all on other languages & used to how this option interacts with the way the fold structures work on them (C++ etc), you never see a problem, it's once you do that it throws you off . In any case, now, thanks to you & your last patches, I can work around that to follow a similar behavior. (for me, I think I'm just gonna throw a bool for fold.highlight props var, & IF the relevant level-- calls, it should be within my coding skills....) so truly, many thanks.

On the 2nd item, I think I better post a screen capture of what I mean, just to make sure my point was clear, All the best.

@Jos On Item 1, yes, (ie I did mean all fold points should remain folded whether or not theyr'e nested another level inside a #CS#CE.) Did assume your'e aware of it, but mentioned it just in case this is NOT by design, for completeness. On Item 2, as I said it also dictates the Scite's fold highlights behavior, so I assumed it is worth mentioning. These 2 are the only 2 points I noticed (so these 2reported), your last patch fixed the bugs, thanks for everything.

@Jos, OK, so far so good, was working with it for a while, reporting back. 1. if using the Stream Comment now, all internal folds are kept intact, except Comment Blocks, (so all these quite large comment blocks get unfolded, ie lexer ignores them. 2. Another thought: I noticed (visually & in the lexer code) how 'levelCurrent' is set for some. While this might looks nice visually so you can see the EndIf's etc when in a folded state, it can look weird or plain confusing if anyone is using Fold.Highlight, (as it follows it). All the best.

@Jos, working good for me, arming the commit button , thanks again.

wow, thanks yet again, will patch, run it through here & report.

@Jos, Thanks again for the fix, back with a report: I can confirm the fix works, but have managed to find scenarios where it doesnt, (no idea why) A quick test can be done by: Define fold.comment=2 in your SciTEUser.properties opening "Array.Au3" select the full text from top to bottom, hit CTRL+SHiFT+Q click the top #CS fold point Result here : Text folded only until line 881. -Update: Tried doing the same on GuiEdit.au3: Result here : Text folded only until line 487. All the best

Thanks a lot ! will test & report.