call unveil earlier in 'got commit' We now know that unveil(2) will never traverse exec. No need to wait with unveil until the editor has been run. ok op@

call unveil earlier in 'got import' We now know that unveil(2) will never traverse exec. No need to wait with unveil until the editor has been run. ok op@

rename got_commit_graph_iter_start() to got_commit_graph_bfsort() This function begins a breadth-first traversal. The new name makes it easier to distinguish from got_commit_graph_toposort().

make 'got rebase' find a merge base with topological sorting if needed Fixes a problematic case of spurious conflicts encountered by naddy@ on landry's firefox package git repository. The current implementation of toposort is expensive, so this might make rebase appear to run slowly on large repositories. However, this is better than letting users deal with spurious conflicts. ok op@

add log -t option which enables topological sorting of commits Because the current implementation of toposort is expensive, add a flag which enables it. I would rather not have this option and just use toposort by default, however more work is required to achieve acceptable performance. ok op@

use a define for vi(1) path This is intended to aid -portable, since other systems may have vi installed in a different place, or maybe prefer to ship with a different default editor. ok stsp@

plug got checkout and update worktree leaks spotted by op@

plug got status worktree leak found by Kyle Ackerman; ok op@

simplify usage of the 'mesg' histedit script command The 'mesg' script command now requires a commit ID as its argument, rather than being tied to a preceding 'pick' or 'edit' command. The old model was too confusing for new users, in particular for people used to Git's rebase -i squash semantics. The 'mesg' command is now semantically equivalent to the 'pick' command and additionally opens the log message in an editor. The new syntax is simpler to use but also requires that we drop support for one-line log messages inside the histedit script, with a commit ID taking its place in the argument space. We don't believe this feature was used much, and that a simplified usage model is more beneficial overall. Patch by Lorenz (xha) ok jamsek

copy remote repo info out of work tree data before closing the work tree Fixes a crash regression introduced when fetch/send were made to close the work tree earlier.

make 'got fetch' and 'got send' release the work tree lock earlier This avoids having work tree operations blocked while network transfers are in progress. ok op@

prevent overlapping repo and work tree in 'got checkout' Some people are eager to try to make Got work just like Git by overlaying the repository and work tree. This causes problems with unveil conflicts at run-time. Fail as early as possible during 'got checkout' when users attempt this. cosmetic tweaks + ok op@

fix 'got log -dPp' diffstat duplication bug Only collect changed paths once if both -d and -P are specified; we already checked for -d and -p. Reported by Lucas on IRC. Regress for this and the previous (got log -x keyword) commit still due. ok stsp@ and op@

got: allow 'got log -x' to accept keywords Suggested by Lucas on IRC. ok stsp@ and op@

zap unused got_commit_object var; ok op@

add cancellation support to the mixed-commits checker in worktree.c

plug got_reference leak in cmd_update() ok op@

got: improve reporting accuracy in branch listing output As suggested by stsp, show the out-of-date symbol for the work tree branch if it is not only out of date in relation to the branch tip but also if it contains mixed commits. Update regress and add new test to check for this case. Also, explain the mixed commit condition of * and ~ markers denoting work tree state in got(1) branch -l docs. ok stsp@

use a separate .cvg meta data directory for cvg(1) work trees This prevents mixing the use of got and cvg clients in the same work tree. Avoids confusion going forward while cvg differentiates itself further. tog(1) and gotadmin(1) remain compatible with both work tree formats. These tools only read work trees to locate the corresponding repository.

got: enable more commands to accept commit keywords More work adding commit keyword support to the blame, cat, ref, tag, and tree commands. With this, all Got commands that take a <commit> option argument or operand now support keywords. Regress flub saved by op! fixes + ok op@

expand support for commit keywords to more got commands Add the ability to use keywords in the backout, branch, checkout, cherrypick, and patch commands. Includes some basic regress tests for the new commands, and also some more contrived test cases for 'got log -c[:]keyword[:(+|-)[N]]'. ok stsp

implement support for keywords as got <commit> arguments This begins enabling the use of keywords in got wherever commit ids or references are used, with more work intended to expand support across all such instances (e.g., branch, checkout, etc.), and add more keywords. The keywords ":base" and ":head" can be passed to 'got {diff,log,update} -c' commands as a substitute for the corresponding commit hash id. Keywords and references can also be modified by appending a ':+' or ':-' and an optional integer N to specify by first parent traversal the Nth generation descendant or antecedent, respectively. If N is omitted, a '1' is implicitly appended. tweaks + ok op and stsp

abort histedit if the user quits the editor without saving the script Also document that the commit/import/tag operations are aborted when the user fails to save the log message from the invoked editor. ok jamsek stsp

always report stat() error with path ok stsp

plug got_object_id leak in cmd_log() ok stsp@