Very well put, lots of times I have been referred to look at Scite's
source, only to have my eye begin to water and I realize that find/replace
or whatever is interwoven to the point that it would be difficult to
extract. Here's what I'm faced with. Here are my choices:

1) Rewrite the Scintilla stuff I need in C# in my project
2) Work with Garrett, the author of ScintillaNET, to push the logic into
his .NET wrapper which is C++ exposed to .NET, but he very busy and
doesn't have time
3) Work with you to try and extract and encapsulate Scite logic to be
reused in Scintilla
4) Go ahead and spend the money and purchase another Control, with
Numerical Studio's ProgramEdit being the control of choice, after all $49
dollar is less that an hours work at my billiing, this option makes the
most choice, however Sctinlla is well tested.  Look at the feature list of
the ProgramEdit contol.

Syntax highlighting
Configuration files
Undo/Redo support
Search/Replace dialog
Formatting settings
Indicator margin
Line number margin
Collapse/Expand display
Intelli-mouse support
Advanced tags
Non-monospace font support
Splitting view
User margin
Customizable line color
Color waveline support
HTML/RTF format export
Built-in settings for common languages
Unicode support
Intellisense functions
Dynamic settings
Visible white spaces
Print/PrintPreview dialogs
Grid line support

So how much work would it be for us to get the Find, Search/Replace,
HTML/RTF export, smart indent, Print and Print Preview, split views, ect
into scintilla?  Months I would imagine, and then I'm faced with the
orignal reason I posted this thread, lexers are written for specific
languages, you cannot roll your own, ProgramEdit also has that, we can
have our <% %>  use VBScript and JScript and everything else outside be say
TSQL or C++ or C#?  The reason I haven't jumped to ProgramEdit is that
they are new, and about to release a new version in two weeks that should
address the performance problems.

I think it would be possible for Scintilla to do all that ProgramEdit does
but it would take an initiative, a grass roots effort to create "Son of
Scintilla" but it would basically put these guys all out of business which
in way would be interesting.

- Mike


>  Mike:
> > > If someone wants to develop an extended control on top of Scintilla
> > then
> > > that would be great.
> > >
> > > I don't have the time and would much rather work on the major
> > > deficiencies in Scintilla.
> >
> > The only problem with somebody extending the control is that doesn't
> > help
> > us who are already using the control, ScintillaNET and other
> > implementations use the actual control itself? The point of the control
> > (for me anyway) is for reuse otherwise just have Scite, and if it's very
> > difficult for us to reuse the control because we're all rewriting the
> > same
> > logic, over and over again, that is a major deficiency. Logic that one
> > expects to find in Scintilla like smart indent, find/replace and more
> > not
> > being present seem to be major issues, these are tons of common,
> > everyday
> > features in Scite that really should pushed into the control. Why add
> > more
> > features when these core ones are yet missing?
> 
>  One problem is that Scintilla development is closely watched and guided by
>  Neil, which want a consistent component. That's good :-)
> 
>  SciTE developement is a bit more "anarchic"... Still watched and guided by
>  Neil, which is the sole (AFAIK) person allowed to include code in the
>  official
>  release. IMHO, it is still good :-))
> 
>  But for SciTE, Neil is more willing to include features he doesn't use or
>  even he doesn't see usefulness (as long as somebody says it is useful for
>  him/her, and implement him/herself).
> 
>  So we got a lot of useful things into SciTE, but somehow spread all over
>  SciTE's code, so hard to find and isolate...
>  Not only stuff useful for Scintilla, but also for other projects. I wish
>  translation stuff to be usable for my projects, also the execute command
>  stuff,
>  etc.
> 
>  It would be good to take all this and put each feature in a separate file
>  and/or class, so anybody can easily pick the useful features.
>  Note: perhaps we could also put in comments some keywords with special
>  marks
>  (indent, find, etc.), so a grep can point out all stuff needed for a given
>  task. Needs also an index of these keywords.
> 
>  Now, this isn't a small task. I am willing to do some of it, but I have a
>  ton of features to do or to finish on this project, and very little free
>  time... So are most of the contributors.
>  So you are right on your request, but you won't get want you want just by
>  asking ;-)
> 
>  Note: ask Neil pointed out, some features won't go into Scintilla. But as
>  I
>  said, if they are clearly isolated and marked, they would be much easier
>  to
>  reuse, perhaps that's a way to do it. Work on this way will never be lost.
>  I
>  made in the past some breaking out of large source files into smaller
>  parts,
>  by purposes (IO, Bar, Dlg etc.). It made easier to navigate in SciTE. And
>  Neil
>  accepted the changes. So there is hope ;-)
> 
>  Regards.
> 
>  --
>  --=#=--=#=--=#=--=#=--=#=--=#=--=#=--=#=--=#=--
>  Philippe Lhoste (Paris -- France)
>  Professional programmer and amateur artist
>  http://jove.prohosting.com/~philho/
>  --=#=--=#=--=#=--=#=--=#=--=#=--=#=--=#=--=#=--
> 
>  GMX ProMail (250 MB Mailbox, 50 FreeSMS, Virenschutz, 2,99 EUR/Monat...)
>  jetzt 3 Monate GRATIS + 3x DER SPIEGEL +++ http://www.gmx.net/derspiegel
>  +++
> 
>  _______________________________________________
>  Scintilla-interest mailing list
>  Scintilla-interest@[...].org
>  http://mailman.lyra.org/mailman/listinfo/scintilla-interest
> 

_______________________________________________
Scintilla-interest mailing list
Scintilla-interest@[...].org
http://mailman.lyra.org/mailman/listinfo/scintilla-interest