gotwebd: abort tree and blob as well

gotwebd: abort blame if the client disconnect midway thru Without setting err when the callback fails we continue the blame machinery.

make new log message modication check more robust; fixes regress fallout ok jamsek

garbage-collect redundant read_logmsg() parameter As identified and suggested by stsp, with the previous commit, 'strip_comments' is now always set thus no longer needed. ok stsp@

got: use timestamp and emptiness to validate log message As suggested by naddy: consider commit log messages valid provided the temp file time stamp has changed and the file is not empty. This heuristic provides the desired behaviour (i.e., reusing cherrypicked/backed-out log messages) that's currently provided but is simpler to grok. Improved by and ok stsp@

convert two files to utf8 ok stsp

add a TODO item for gotadmin pack command

open_blob: copy whole id not just the sha1

remove recv_disconnect() from repo_write.c; missed in previous commit

do not expect to see a DISCONNECT message in repo processes The parent no longer sends this message. Perform related cleanup in the shutdown path instead. ok op@

add functions to extract object ids from fileindex entries ok stsp

fix gotd segfault in libevent while disconnecting clients Sending a DISCONNECT message to the repo process and then killing it and freeing its structure won't work. The message ends up on internal libevent queues and will soon crash because we have freed the memory for the event structre. Sending that message is a leftover from the times when the repo procs were persistent.

rm unused boolean grammar. ok stsp@

make edits made to comments count as a log message modification This supports use of cherrypick/backout without requiring the user to modify the log message of the original commit. ok jamsek, op

fix gotd and gotsh usage() function declaration "function declaration isn't a prototype" should fix the error seen by tracey@ on sparc64

make gotd parent dispatch handlers more robust during teardown We have observed gotd exiting after "cannot find client for fd N" errors. If this occurs then we are likely in the process of disconnecting a client session while processing an event from a child process. Treat the above error as non-fatal and stop processing more events from the child process.

tog: reset 1 second refresh rate after starting search This fixes an issue reported by Mikhail where CPU usage continuously increases if search is started after the refresh rate has dropped from 100ms to 1s by resetting the 1s refresh rate if search is started after fast initialisation has finished. This is not needed in limit_log_view() because the 1s refresh rate is reset in draw_commits() once the screen is redrawn (i.e., commits_needed == 0). Tested by and based on initial patch from Mikhail. ok stsp@

drop double process name from some gotd logs i.e. "listen: listen: shutting down" -> "listen: shutting down" the procname is already prepended by vlog() ok jamsek

only fetch the work tree's branch by default if it is inside "refs/heads/" ok jamsek

got: use intermediate pointers to plug leak on realloc And save worktree_branch_len for reuse. ok stsp@ and op@

nix unnecessary colon in gotd log output ok op@

fix interaction of 'got fetch -b', got.conf, and work tree Without branches in got.conf for a remote, and without -b/-R options, the fallback to HEAD would only work when not invoked in a work tree. With this fix 'got fetch' should behave as described in the man page. The -b option now overrides both got.conf and the fallback to the work tree's branch. And fallback to HEAD works as expected when invoked in a repository. Also, do not strictly require remote repositories to provide a branch from the refs/heads/ namespace. In such cases users should be able to use -R to select something to fetch. ok jamsek

hide unexpected diff output from xfail test

add xfail test for 'got status' ignore patterns with a trailing slash Patch by Lucas

make 'got import' -I option match directory names with a trailing slash reported by Lucas on IRC, who sent a patch which this commit was based on ok jamsek