On Wed, 6 Mar 2002 19:40:58 -0800, Brian Ingerson wrote: >On 06/03/02 22:31 -0500, Matthew O. Persico wrote: >>On Wed, 6 Mar 2002 21:41:06 -0500 (EST), Sam Tregar wrote: >>>On Wed, 6 Mar 2002, Matthew O. Persico wrote: >>> >>>>I want to recompile cvs as a library and write just enough >>>>Inline::C >>>>code to access that library's functions. >>> >>>This sounds like a really bad idea to me. >>What is so 'bad'? That is an awfully vague expression. >> >>>Wouldn't it be easier to just run cvs commands using system()? >>> >>Yes. That's what VCS::CVS does. I, however, consider system() to be >>a >>kludge when not using perl in Shell/Glue mode. It will also make >>some >>operations on remote repositories easier to code. I can also extend >> >>CVS functionality w/o having to open up cvs to do it. >>This is also a learning exercise. > >I support this. (But then again I love "bad" ideas) I say go for it >and >let us know your experience! It would make a good story for the >website. > >Cheers, Ingy
Bad news. See this: On Wed, 06 Mar 2002 22:19:58 -0500, Michael Roberts wrote: >Unfortunately, a CVS library has yet to be built -- apparently the >structure of the code is *really*, *really* strange. I haven't >looked >at it, but I do follow some of their discussion (cvshome.org for all >that jazz) and a wrappable CVS library is kind of one of the things >on >the table at the moment, or was about four months ago. > >Michael Oh well, it seemed like a good idea. -- Matthew O. Persico
