stop installing git-{receive,upload}-pack symlinks to gotsh in ~/bin Having those links in the user's PATH can make our send/fetch regression tests fail. We do want to talk to git-daemon during those tests, and these symlinks can get in the way of that.

convert got_pack' filesize to off_t; ok stsp@

plug a memory leak on error in got_gotconfig_read() ok op@

add missing check for errors from got_gotconfig_read() in open_worktree()

make got.conf(5) warn about remotes configured in locally-shared repositories

sort getopt() option lists and switch statements; patch by Josiah Frentsos

fix off_t type mismatch in gotd see e082ed67; ok stsp@

check size before calling mmap(2) It's only a preparatory step, as checking whether a size_t is less than SIZE_MAX is moot. In a follow-up commit, however, the `filesize' field of the struct got_pack will become off_t and these checks will kick in. This also makes consistent how we guard mmap(2) against empty files. ok and improvements stsp@

add bounds check when reading a delta offset from a packed object ok stsp@

add missing `goto done;' on error path; ok stsp@

fix spelling of "FastCGI"; patch by Josiah Frentsos; ok op@

fix off_t type mismatch off_t is a signed type and depending on the platform, it can be "long" or "long long", so cast to long long for printf().

rebase.sh: remove accidentally included absolute path to "got"

fix gotd(8) usage() string; patch by Josiah Frentsos

fix regression test failures with Git 3.30.5 / 2.38.1 or later installed The fix for CVE-2022-39253 in Git made our regression tests fail since creating submodules from local clones is no longer allowed by default. Add an override to our invocations of "git submodule add", the same which was added to Git's regression test suite as part of their fix. This CVE doesn't otherwise affect us. We do not implement submodules. Our use of them in regression tests exists only to ensure that Got does not freak out when it sees one. Precisely because automated support for nested repositories and/or working copies (as in Git Submodules, or Subversion Externals, etc.) have been an endless source of security problems in other systems. sudden test failures pointed out by naddy@

change got_pack_parse_offset_delta tslen argument to size_t The tslen argument is always in the (1..9) range so `int' is technically fine (even int8_t would be!) but all the callers are passing a size_t, so change the type accordingly. ok stsp@

gotsh.1: Use Sx for referencing EXAMPLES; patch by Josiah Frentsos

bump version number

CHANGES for 0.77

do not require gotd(8) to be installed in /usr/local; dev builds go to ~/bin/

sync distfile list

avoid printing harmless errors that can occur when tog exits due to Ctrl-C ok jamsek

fix typo in gotsh.1 ok stsp@

fix detection of SIGTERM in tog; this signal was accidentally being ignored ok jamsek

close parent's end of imsg pipe before waiting for a child process to exit Prevents a dead-lock in 'tog log' where tog wants to exit (e.g. because the user pressed Ctrl-C) while a got-read-pack child process wants to send more commits. Closing the parent's pipe descriptor makes writes to the pipe fail in the child process. The child then unwinds via an ERR_EOF error and exits, instead of forever polling its end of the pipe in order to write more data. ok jamsek