According to information I've read on the web, when a commit is made on a local machine to a remote server, exact copies (i.e. non-RCS format) of the files being committed are created temporarily on the server so that they are available for commitinfo scripts to act on them. I need to write a commitinfo script that will parse committed files to make sure they have appropriate copyright statements. To run that script on the server, I'll need to get access to those non-RCS format files. However, I'm having trouble locating where those files are on the server, and I'm beginning to doubt the veracity of the information I found. I've even run a commitinfo script which runs an slocate command on each filename in the commit, but the only results seem to be the RCS format files - no non-RCS format files seem to exist. I've been assuming that the name of the non-RCS format file copy on the server is equivalent to the file name extractable from the parameters that commitinfo passes to the script, so that (for example) if I commit file module\topdir\subdir\foo.txt, my commitinfo script should be able to find a file called foo.txt somewhere on the server - but this doesn't seem to be the case - all that slocate locates is the associated RCS format file (module\topdir\subdir\foo.txt,v).
Does anyone know if these non-RCS format files do actually exist (albeit temporarily) on the server and if so where? If not, does anyone have any advice on writing a commitinfo script which (i) will run on the server side and (ii) will parse the entire contents of the file being committed? Thanks in advance _______________________________________________ Info-cvs mailing list [email protected] http://lists.gnu.org/mailman/listinfo/info-cvs
