Mike Roberts:

What would the interest level be in an expanded regular expression
parser/matcher?

  I actually think that including RE support in Scintilla was a
mistake. There are several large, high quality regular expression
libraries available and it would be better for applications to use one
of these. Often an application will be focussed on a particular
language or platform and there will be a regular expression library
and syntax associated with those that should be used. Its also better
if the same library is used for all parts of the application. It is
quite difficult to provide a runtime mechanism for Scintilla to choose
a library and no one has been sufficiently interested to produce a
compile time mechanism. One issue is that libraries often want direct
access through a char* rather than indirect access through a function
call or direct access to a complex structure like Scintilla's split
buffer.

   Since SciTE uses Scintilla's RE other projects often copy this so
if SciTE was changed to use another library then it would be a better
example. It would also ensure that  the support code (such as
WindowAccessor) worked well. I'd be more interested in one of the more
complete libraries like PCRE or Boost.Regex but that's a large amount
of work and may involve additional dependencies - Boost.Regex prefers
using ICU for Unicode... . Too much work for me to look at in the near
future.

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

Reply via email to