Hi, Explanation:
"intelligent merging after rename". If a file in a child branch is renamed, a subsequent change to the file in the parent branch (old name) will propagate to the file in the child branch (new name) when the parent is merged onto the child. Use case: I have been trying to use Fossil to manage a build of Tcl on Windows. My idea was to have various Fossil branches tracking the CVS repositories for Tcl, Tk, Threads, Itcl, etc.; and to have a tcl-maintenance branch that merges in from all of the CVS-tracking branches. I can then commit bug fixes (in particular Makefile fixes) and perform builds in the tcl-maintenance branch while I wait for the submitted bugs to be fixed in CVS. This allows me to track and version my fixes. Two specific problems arose that caused me to have trouble with this approach: (1) To get 'threads' and 'tk' into the tcl-maintenance branch, I did a 'fossil merge' on each, then 'fossil mv' to get the folders under 'tcl/pkgs'. I was surprised to find a couple of days later that a 'cvs update ; fossil addremove ; fossil commit' in the cvs-tk branch followed by 'fossil merge cvs-tk' in the tcl-maintenance branch didn't merge any updates. That's when I realised that Fossil doesn't have "intelligent merging after rename". I worked around the problem by renaming the folder in the cvs-tk branch, so that the files/paths in both branches are the same. (2) There is a further complication with the 'threads' package. For historical reasons the Windows makefile is in the wrong folder. The contents of the Makefile are being corrected so that 'threads' will integrate with Tcl's package building system, but CVS doesn't support moving the Makefile, so the choice is to delete/add and break the history, or leave it alone and break the future. My plan was to address this in Fossil by moving the Makefile in the tcl-maintenance branch, then updates on the threads-cvs branch would merge across into the (moved) file in the tcl-maintenance branch. Without intelligent merging this doesn't work, and I can't see a way to work around it (suggestions welcome). Regards, Twylite _______________________________________________ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
