Commits


got: don't leave tmp "got-logmsg-*" files in the work tree Reported by stsp on irc, and reviewed as part of the following regress commit covering the commit aspect of logmsg refs. Delete temp merge_log files if created during commits involving backed-out/cherrypicked changes. ok stsp@


gotsh.1: show how to set up anonymous public read-only repository access


use VISUAL instead of EDITOR in histedit_mesg_filemode_change VISUAL is preferred and relying on EDITOR may cause test failures in some environments. pointed out by op and jamsek


simplify file content comparison in a regress test; suggested by op@


fix histedit -m on a commit which only changes filemode bits The commit was being miscategorized as a no-op change and dropped. Now the commit is retained and its log message is updated as expected. ok op, jamsek


remove TODO item regarding 'got send -d' support in gotd


got.1: editing merged log messages is no longer optional, tweak related wording


got.1: Mark the commit argument to cherrypick and backout optional patch by Josiah Frentsos


abort commit with merged log messages if the log message is left unmodified ok op@


avoid creation of temporary file if no merged log messages will be used


rename a variable for clarity


clarify documentation of cherrypick/backout log message recording ok jamsek


uppercase cherrypick/backout -X progress output ok jamsek


improve heuristics for prepopulated log messages Small tweak to cover the case where users specify paths to be committed such that none of the paths modified by a backout or cherrypick operation are in the commit. In this case, the logmsg ref will remain in refs/got/worktree to be used when the user does run a commit involving the affected paths. ok stsp@


fix invalid assumption in commit_path_changed_in_worktree() Make sure we have a parent id first to avoid dereferencing a NULL pointer with the got_object_open_as_commit() call. ok op@ and stsp@


extend test_delete_branch - try to delete using git - try to delete multiple branches in one go - verify the ref-list reported by the server before and after the deletion suggested / ok stsp


gotd: implement the delete-refs capability Allow clients to run "got send -d" against gotd. Clients will send a zero-id as new id for a reference and, in the special but more common case of just deleting and not updating, no pack file will be sent. ok and tweaks by stsp@


regress for new backout/cherrypick logmsg references Tests only cover the -lX interface to list and delete logmsg refs created by previous backout and cherrypick operations. Further tests are needed for revert and commit coverage to ensure that the correct log messages of bo/cy commits are loaded into the editor when 'got commit' is run, and that only the refs used to populate the editor for the new commit are deleted. Potential printf issues observed by op remain in the diff; noted here as a reminder to do a sweep at some point in the future as there are other such instances in cherrypick.sh and possibly elsewhere in regress. Reviewed as part of previous commit introducing logmsg ref support for backed-out/cherrypicked commits. ok stsp@


got: load editor with backout/cherrypick commit log messages Create new logmsg references used to populate the editor with the log messages of backed-out and cherrypicked commits when 'got commit' is run. We use versioned path state heuristics during revert and commit commands in an effort to ensure the editor is only populated with the log messages of backed-out and/or cherrypicked commits involving paths with local changes that are being committed. This is an implementation of an initial idea and algorithm designed by stsp. inputs, improvments and ok stsp@


up the checkout regress ulimit to 33, to accomodate for sparc64 failing at 32. ok stsp


add TODO item for a histedit -m bug I have run into


remove \n from log message


delete trailing whitespaces


add another pre-commit check idea to the gotd TODO list


avoid traversing enumerated commits more than once in got-read-pack Keep track of parent commits that will be processed as part of looping over the commit queue provided by the main process, and do not add these commits to the queue again. Fixes pointless traversal of commits on the queue which will simply be skipped. The end result is the same either way. ok tracey