[
Date Prev][
Date Next][
Thread Prev][
Thread Next][
Date Index][
Thread Index]
[
List Home]
|
Re: [cdt-core-dev] Another Parser Document
|
Thomas Fletcher wrote:
Long term we shouldn't have any specific binding, but like everything else
it should be a plugable syntax thing (ie we feed the comment to you, you
parse and tell us something).
Right, though bear in mind that some comment parses could be "chattier"
than you describe - if you feed me a comment containing "@see
someThing()", I'm probably going to want to call back on the CDT indexer
to make an informed guess as to what "someThing()" is, which in turn
could stress the indexer in unexpected ways. (Because the parser context
at the point of the comment is probably not the same as the context of
whatever it is the comment is describing.)
> For a first (second?) cut however a quick and
dirty answer is needed which people can start shipping in products and which
can be used to draw C/C++ developers into Eclipse.
Which brings me back to *my* original question on this thread; if we
want something quickly, wouldn't it be less effort to add
acceptComment() to ISourceElementRequester than implement a whole
separate parsing pass?
My vote stays with:
1) JavaDoc -- since it is supported from JDT it should be easy to lift
2) Doxygen -- used by a lot of open source projects
Agreed - JavaDoc is a sensible first step, especially as Doxygen markup
syntax is largly a superset of JavaDoc. Doxygen syntax later - fine,
though it's sufficiently large (and enough of a moving target) than a
generalized (data-driven rather than hardcoded) comment parser might
turn out to be the easiest way to get there.
Cheers,
Mike
