|
| 1 | +Git v1.7.8.2 Release Notes |
| 2 | +========================== |
| 3 | + |
| 4 | +Fixes since v1.7.8.1 |
| 5 | +-------------------- |
| 6 | + |
| 7 | + * Porcelain commands like "git reset" did not distinguish deletions |
| 8 | + and type-changes from ordinary modification, and reported them with |
| 9 | + the same 'M' moniker. They now use 'D' (for deletion) and 'T' (for |
| 10 | + type-change) to match "git status -s" and "git diff --name-status". |
| 11 | + |
| 12 | + * The configuration file parser used for sizes (e.g. bigFileThreshold) |
| 13 | + did not correctly interpret 'g' suffix. |
| 14 | + |
| 15 | + * The replacement implemention for snprintf used on platforms with |
| 16 | + native snprintf that is broken did not use va_copy correctly. |
| 17 | + |
| 18 | + * LF-to-CRLF streaming filter replaced all LF with CRLF, which might |
| 19 | + be techinically correct but not friendly to people who are trying |
| 20 | + to recover from earlier mistakes of using CRLF in the repository |
| 21 | + data in the first place. It now refrains from doing so for LF that |
| 22 | + follows a CR. |
| 23 | + |
| 24 | + * git native connection going over TCP (not over SSH) did not set |
| 25 | + SO_KEEPALIVE option which failed to receive link layer errors. |
| 26 | + |
| 27 | + * "git branch -m <current branch> HEAD" is an obvious no-op but was not |
| 28 | + allowed. |
| 29 | + |
| 30 | + * "git checkout -m" did not recreate the conflicted state in a "both |
| 31 | + sides added, without any common ancestor version" conflict |
| 32 | + situation. |
| 33 | + |
| 34 | + * "git cherry-pick $commit" (not a range) created an unnecessary |
| 35 | + sequencer state and interfered with valid workflow to use the |
| 36 | + command during a session to cherry-pick multiple commits. |
| 37 | + |
| 38 | + * You could make "git commit" segfault by giving the "--no-message" |
| 39 | + option. |
| 40 | + |
| 41 | + * "fast-import" did not correctly update an existing notes tree, |
| 42 | + possibly corrupting the fan-out. |
| 43 | + |
| 44 | + * "git fetch-pack" accepted unqualified refs that do not begin with |
| 45 | + refs/ by mistake and compensated it by matching the refspec with |
| 46 | + tail-match, which was doubly wrong. This broke fetching from a |
| 47 | + repository with a funny named ref "refs/foo/refs/heads/master" and a |
| 48 | + 'master' branch with "git fetch-pack refs/heads/master", as the |
| 49 | + command incorrectly considered the former a "match". |
| 50 | + |
| 51 | + * "git log --follow" did not honor the rename threshold score given |
| 52 | + with the -M option (e.g. "-M50%"). |
| 53 | + |
| 54 | + * "git mv" gave suboptimal error/warning messages when it overwrites |
| 55 | + target files. It also did not pay attention to "-v" option. |
| 56 | + |
| 57 | + * Authenticated "git push" over dumb HTTP were broken with a recent |
| 58 | + change and failed without asking for password when username is |
| 59 | + given. |
| 60 | + |
| 61 | + * "git push" to an empty repository over HTTP were broken with a |
| 62 | + recent change to the ref handling. |
| 63 | + |
| 64 | + * "git push -v" forgot how to be verbose by mistake. It now properly |
| 65 | + becomes verbose when asked to. |
| 66 | + |
| 67 | + * When a "reword" action in "git rebase -i" failed to run "commit --amend", |
| 68 | + we did not give the control back to the user to resolve the situation, and |
| 69 | + instead kept the original commit log message. |
| 70 | + |
| 71 | +Also contains minor fixes and documentation updates. |
0 commit comments