Igor Pechtchanski <[EMAIL PROTECTED]> wrote: >> I'm having a problem building cvs-1.11.0-1. GDBM_EXPORT is not >> picked up as a macro (which I assume it is, as it's used to declare >> functions), and this results in parse errors and gcc being generally >> unhappy. I'm attaching the exact error message. >> This is probably because I have gdbm-1.8.0-4, and cvs lists >> gdbm-1.8.0-3 in its dependences, although Chuck Wilson apparently >> received no complaints during the test period ( >> http://www.cygwin.com/ml/cygwin/2002-02/msg01333.html ).
I ran into this problem, and solved it by backing out Chuck's gdbm changes, instead building cvs in its flat-file mode. I needed to do this so I had a cvs which could act on repositories created using this method. I don't know what you want to use it for, but unless you want to specifically want to use gdbm, this may be acceptable. As the cvs docs say, the performance gain is minimal unless you have a *huge* modules file. Attatched is Chuck's patch as included in the Cygwin source package, spilt into 2: cygupdate is general fixups, and gdbmify is the gdbm additions, which I backed out to build a cvs_flatdbm.exe which I keep in ~/bin. Max.
cvs-1.11.0-1-cygupdate.patch
Description: Binary data
cvs-1.11.0-1-gdbmify.patch
Description: Binary data
-- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
