+2 -2

Documentation/CodingGuidelines

··· 75 75 76 76 - If you want to find out if a command is available on the user's 77 77 $PATH, you should use 'type <command>', instead of 'which <command>'. 78 78 - The output of 'which' is not machine parseable and its exit code 78 78 + The output of 'which' is not machine parsable and its exit code 79 79 is not reliable across platforms. 80 80 81 81 - We use POSIX compliant parameter substitutions and avoid bashisms; ··· 203 203 . since early 2012 with e1327023ea, we have been using an enum 204 204 definition whose last element is followed by a comma. This, like 205 205 an array initializer that ends with a trailing comma, can be used 206 206 - to reduce the patch noise when adding a new identifer at the end. 206 206 + to reduce the patch noise when adding a new identifier at the end. 207 207 208 208 . since mid 2017 with cbc0f81d, we have been using designated 209 209 initializers for struct (e.g. "struct t v = { .val = 'a' };").

+2 -2

Documentation/RelNotes/1.5.0.txt

··· 251 251 the repository when that happens. 252 252 253 253 254 254 - * Crufts removal 254 254 + * Cruft removal 255 255 256 256 - We used to say "old commits are retrievable using reflog and 257 257 'master@{yesterday}' syntax as long as you haven't run ··· 379 379 - The value of i18n.commitencoding in the originating 380 380 repository is recorded in the commit object on the "encoding" 381 381 header, if it is not UTF-8. git-log and friends notice this, 382 382 - and reencodes the message to the log output encoding when 382 382 + and re-encodes the message to the log output encoding when 383 383 displaying, if they are different. The log output encoding 384 384 is determined by "git log --encoding=<encoding>", 385 385 i18n.logoutputencoding configuration, or i18n.commitencoding

+1 -1

Documentation/RelNotes/1.6.5.4.txt

··· 10 10 11 11 * "git prune-packed" gave progress output even when its standard error is 12 12 not connected to a terminal; this caused cron jobs that run it to 13 13 - produce crufts. 13 13 + produce cruft. 14 14 15 15 * "git pack-objects --all-progress" is an option to ask progress output 16 16 from write-object phase _if_ progress output were to be produced, and

+1 -1

Documentation/RelNotes/1.7.0.2.txt

··· 34 34 * "git status" in 1.7.0 lacked the optimization we used to have in 1.6.X series 35 35 to speed up scanning of large working tree. 36 36 37 37 - * "gitweb" did not diagnose parsing errors properly while reading tis configuration 37 37 + * "gitweb" did not diagnose parsing errors properly while reading its configuration 38 38 file. 39 39 40 40 And other minor fixes and documentation updates.

+1 -1

Documentation/RelNotes/1.7.10.4.txt

··· 7 7 * The message file for Swedish translation has been updated a bit. 8 8 9 9 * A name taken from mailmap was copied into an internal buffer 10 10 - incorrectly and could overun the buffer if it is too long. 10 10 + incorrectly and could overrun the buffer if it is too long. 11 11 12 12 * A malformed commit object that has a header line chomped in the 13 13 middle could kill git with a NULL pointer dereference.

+1 -1

Documentation/RelNotes/1.7.12.3.txt

··· 25 25 its Accept-Encoding header. 26 26 27 27 * "git receive-pack" (the counterpart to "git push") did not give 28 28 - progress output while processing objects it received to the puser 28 28 + progress output while processing objects it received to the user 29 29 when run over the smart-http protocol. 30 30 31 31 * "git status" honored the ignore=dirty settings in .gitmodules but

+1 -1

Documentation/RelNotes/1.7.5.3.txt

··· 22 22 * "git log --stdin path" with an input that has additional pathspec 23 23 used to corrupt memory. 24 24 25 25 - * "git send-pack" (hence "git push") over smalt-HTTP protocol could 25 25 + * "git send-pack" (hence "git push") over smart-HTTP protocol could 26 26 deadlock when the client side pack-object died early. 27 27 28 28 * Compressed tarball gitweb generates used to be made with the timestamp

+1 -1

Documentation/RelNotes/1.8.0.txt

··· 233 233 together, misdetected branches. 234 234 235 235 * "git receive-pack" (the counterpart to "git push") did not give 236 236 - progress output while processing objects it received to the puser 236 236 + progress output while processing objects it received to the user 237 237 when run over the smart-http protocol. 238 238 239 239 * When you misspell the command name you give to the "exec" action in

+1 -1

Documentation/RelNotes/1.8.4.1.txt

··· 15 15 in 1.8.4-rc1). 16 16 17 17 * "git rebase -i" and other scripted commands were feeding a 18 18 - random, data dependant error message to 'echo' and expecting it 18 18 + random, data dependent error message to 'echo' and expecting it 19 19 to come out literally. 20 20 21 21 * Setting the "submodule.<name>.path" variable to the empty

+1 -1

Documentation/RelNotes/2.1.3.txt

··· 13 13 they are new enough to support the `--output` option. 14 14 15 15 * "git pack-objects" forgot to disable the codepath to generate 16 16 - object recheability bitmap when it needs to split the resulting 16 16 + object reachability bitmap when it needs to split the resulting 17 17 pack. 18 18 19 19 * "gitweb" used deprecated CGI::startfrom, which was removed from

+1 -1

Documentation/RelNotes/2.10.0.txt

··· 478 478 * One part of "git am" had an oddball helper function that called 479 479 stuff from outside "his" as opposed to calling what we have "ours", 480 480 which was not gender-neutral and also inconsistent with the rest of 481 481 - the system where outside stuff is usuall called "theirs" in 481 481 + the system where outside stuff is usually called "theirs" in 482 482 contrast to "ours". 483 483 484 484 * "git blame file" allowed the lineage of lines in the uncommitted,

+1 -1

Documentation/RelNotes/2.10.2.txt

··· 86 86 by refusing to check out a branch that is already checked out in 87 87 another worktree. However, this also prevented checking out a 88 88 branch, which is designated as the primary branch of a bare 89 89 - reopsitory, in a worktree that is connected to the bare 89 89 + repository, in a worktree that is connected to the bare 90 90 repository. The check has been corrected to allow it. 91 91 92 92 * "git rebase" immediately after "git clone" failed to find the fork

+1 -1

Documentation/RelNotes/2.11.1.txt

··· 104 104 "git difftool --dir-diff" from a subdirectory never worked. This 105 105 has been fixed. 106 106 107 107 - * "git p4" that tracks multile p4 paths imported a single changelist 107 107 + * "git p4" that tracks multiple p4 paths imported a single changelist 108 108 that touches files in these multiple paths as one commit, followed 109 109 by many empty commits. This has been fixed. 110 110

+1 -1

Documentation/RelNotes/2.12.0.txt

··· 315 315 "git difftool --dir-diff" from a subdirectory never worked. This 316 316 has been fixed. 317 317 318 318 - * "git p4" that tracks multile p4 paths imported a single changelist 318 318 + * "git p4" that tracks multiple p4 paths imported a single changelist 319 319 that touches files in these multiple paths as one commit, followed 320 320 by many empty commits. This has been fixed. 321 321

+1 -1

Documentation/RelNotes/2.13.0.txt

··· 177 177 been changed to enable "--decorate". 178 178 179 179 * The output from "git status --short" has been extended to show 180 180 - various kinds of dirtyness in submodules differently; instead of to 180 180 + various kinds of dirtiness in submodules differently; instead of to 181 181 "M" for modified, 'm' and '?' can be shown to signal changes only 182 182 to the working tree of the submodule but not the commit that is 183 183 checked out.

+2 -2

Documentation/RelNotes/2.13.3.txt

··· 25 25 * The code to pick up and execute command alias definition from the 26 26 configuration used to switch to the top of the working tree and 27 27 then come back when the expanded alias was executed, which was 28 28 - unnecessarilyl complex. Attempt to simplify the logic by using the 28 28 + unnecessarily complex. Attempt to simplify the logic by using the 29 29 early-config mechanism that does not chdir around. 30 30 31 31 * "git add -p" were updated in 2.12 timeframe to cope with custom ··· 35 35 * Fix a recent regression to "git rebase -i" and add tests that would 36 36 have caught it and others. 37 37 38 38 - * An unaligned 32-bit access in pack-bitmap code ahs been corrected. 38 38 + * An unaligned 32-bit access in pack-bitmap code has been corrected. 39 39 40 40 * Tighten error checks for invalid "git apply" input. 41 41

+2 -2

Documentation/RelNotes/2.14.0.txt

··· 141 141 * Some platforms have ulong that is smaller than time_t, and our 142 142 historical use of ulong for timestamp would mean they cannot 143 143 represent some timestamp that the platform allows. Invent a 144 144 - separate and dedicated timestamp_t (so that we can distingiuish 144 144 + separate and dedicated timestamp_t (so that we can distinguish 145 145 timestamps and a vanilla ulongs, which along is already a good 146 146 move), and then declare uintmax_t is the type to be used as the 147 147 timestamp_t. ··· 442 442 * The code to pick up and execute command alias definition from the 443 443 configuration used to switch to the top of the working tree and 444 444 then come back when the expanded alias was executed, which was 445 445 - unnecessarilyl complex. Attempt to simplify the logic by using the 445 445 + unnecessarily complex. Attempt to simplify the logic by using the 446 446 early-config mechanism that does not chdir around. 447 447 448 448 * Fix configuration codepath to pay proper attention to commondir

+1 -1

Documentation/RelNotes/2.16.0.txt

··· 407 407 (merge eef3df5a93 bw/pathspec-match-submodule-boundary later to maint). 408 408 409 409 * Amending commits in git-gui broke the author name that is non-ascii 410 410 - due to incorrect enconding conversion. 410 410 + due to incorrect encoding conversion. 411 411 412 412 * Recent update to the submodule configuration code broke "diff-tree" 413 413 by accidentally stopping to read from the index upfront.

+1 -1

Documentation/RelNotes/2.16.3.txt

··· 24 24 25 25 * The http tracing code, often used to debug connection issues, 26 26 learned to redact potentially sensitive information from its output 27 27 - so that it can be more safely sharable. 27 27 + so that it can be more safely shareable. 28 28 29 29 * Crash fix for a corner case where an error codepath tried to unlock 30 30 what it did not acquire lock on.

+1 -1

Documentation/RelNotes/2.17.0.txt

··· 216 216 217 217 * The http tracing code, often used to debug connection issues, 218 218 learned to redact potentially sensitive information from its output 219 219 - so that it can be more safely sharable. 219 219 + so that it can be more safely shareable. 220 220 (merge 8ba18e6fa4 jt/http-redact-cookies later to maint). 221 221 222 222 * Crash fix for a corner case where an error codepath tried to unlock

+1 -1

Documentation/RelNotes/2.18.0.txt

··· 179 179 (merge 00a3da2a13 nd/remove-ignore-env-field later to maint). 180 180 181 181 * Code to find the length to uniquely abbreviate object names based 182 182 - on packfile content, which is a relatively recent addtion, has been 182 182 + on packfile content, which is a relatively recent addition, has been 183 183 optimized to use the same fan-out table. 184 184 185 185 * The mechanism to use parse-options API to automate the command line

+1 -1

Documentation/RelNotes/2.19.0.txt

··· 106 106 * The conversion to pass "the_repository" and then "a_repository" 107 107 throughout the object access API continues. 108 108 109 109 - * Continuing with the idea to programatically enumerate various 109 109 + * Continuing with the idea to programmatically enumerate various 110 110 pieces of data required for command line completion, teach the 111 111 codebase to report the list of configuration variables 112 112 subcommands care about to help complete them.

+1 -1

Documentation/RelNotes/2.20.0.txt

··· 119 119 alias expansion. 120 120 121 121 * The documentation of "git gc" has been updated to mention that it 122 122 - is no longer limited to "pruning away crufts" but also updates 122 122 + is no longer limited to "pruning away cruft" but also updates 123 123 ancillary files like commit-graph as a part of repository 124 124 optimization. 125 125

+1 -1

Documentation/RelNotes/2.3.3.txt

··· 12 12 * Description given by "grep -h" for its --exclude-standard option 13 13 was phrased poorly. 14 14 15 15 - * Documentaton for "git remote add" mentioned "--tags" and 15 15 + * Documentation for "git remote add" mentioned "--tags" and 16 16 "--no-tags" and it was not clear that fetch from the remote in 17 17 the future will use the default behaviour when neither is given 18 18 to override it.

+1 -1

Documentation/RelNotes/2.3.7.txt

··· 4 4 Fixes since v2.3.6 5 5 ------------------ 6 6 7 7 - * An earlier update to the parser that disects a URL broke an 7 7 + * An earlier update to the parser that dissects a URL broke an 8 8 address, followed by a colon, followed by an empty string (instead 9 9 of the port number), e.g. ssh://example.com:/path/to/repo. 10 10

+1 -1

Documentation/RelNotes/2.4.3.txt

··· 66 66 * Some time ago, "git blame" (incorrectly) lost the convert_to_git() 67 67 call when synthesizing a fake "tip" commit that represents the 68 68 state in the working tree, which broke folks who record the history 69 69 - with LF line ending to make their project portabile across 69 69 + with LF line ending to make their project portable across 70 70 platforms while terminating lines in their working tree files with 71 71 CRLF for their platform. 72 72

+1 -1

Documentation/RelNotes/2.7.0.txt

··· 40 40 41 41 * "git interpret-trailers" can now run outside of a Git repository. 42 42 43 43 - * "git p4" learned to reencode the pathname it uses to communicate 43 43 + * "git p4" learned to re-encode the pathname it uses to communicate 44 44 with the p4 depot with a new option. 45 45 46 46 * Give progress meter to "git filter-branch".

+1 -1

Documentation/RelNotes/2.8.0.txt

··· 189 189 * Some calls to strcpy(3) triggers a false warning from static 190 190 analyzers that are less intelligent than humans, and reducing the 191 191 number of these false hits helps us notice real issues. A few 192 192 - calls to strcpy(3) in a couple of protrams that are already safe 192 192 + calls to strcpy(3) in a couple of programs that are already safe 193 193 has been rewritten to avoid false warnings. 194 194 195 195 * The "name_path" API was an attempt to reduce the need to construct

+1 -1

Documentation/RelNotes/2.9.3.txt

··· 36 36 * One part of "git am" had an oddball helper function that called 37 37 stuff from outside "his" as opposed to calling what we have "ours", 38 38 which was not gender-neutral and also inconsistent with the rest of 39 39 - the system where outside stuff is usuall called "theirs" in 39 39 + the system where outside stuff is usually called "theirs" in 40 40 contrast to "ours". 41 41 42 42 * The test framework learned a new helper test_match_signal to

+1 -1

Documentation/config.txt

··· 142 142 143 143 `gitdir/i`:: 144 144 This is the same as `gitdir` except that matching is done 145 145 - case-insensitively (e.g. on case-insensitive file sytems) 145 145 + case-insensitively (e.g. on case-insensitive file systems) 146 146 147 147 `onbranch`:: 148 148 The data that follows the keyword `onbranch:` is taken to be a

+1 -1

Documentation/config/tag.txt

··· 13 13 Use of this option when running in an automated script can 14 14 result in a large number of tags being signed. It is therefore 15 15 convenient to use an agent to avoid typing your gpg passphrase 16 16 - several times. Note that this option doesn't affects tag signing 16 16 + several times. Note that this option doesn't affect tag signing 17 17 behavior enabled by "-u <keyid>" or "--local-user=<keyid>" options. 18 18 19 19 tar.umask::

+1 -1

Documentation/git-bisect-lk2009.txt

··· 158 158 supposed to be used so that all the tests are checked after each 159 159 commit. This means that they are not very efficient, because many 160 160 tests are run for no interesting result, and they suffer from 161 161 - combinational explosion. 161 161 + combinatorial explosion. 162 162 163 163 In fact the problem is that big software often has many different 164 164 configuration options and that each test case should pass for each

+1 -1

Documentation/git-check-attr.txt

··· 32 32 instead of from the command-line. 33 33 34 34 -z:: 35 35 - The output format is modified to be machine-parseable. 35 35 + The output format is modified to be machine-parsable. 36 36 If `--stdin` is also given, input paths are separated 37 37 with a NUL character instead of a linefeed character. 38 38

+1 -1

Documentation/git-check-ignore.txt

··· 39 39 instead of from the command-line. 40 40 41 41 -z:: 42 42 - The output format is modified to be machine-parseable (see 42 42 + The output format is modified to be machine-parsable (see 43 43 below). If `--stdin` is also given, input paths are separated 44 44 with a NUL character instead of a linefeed character. 45 45

+1 -1

Documentation/git-cvsserver.txt

··· 294 294 Git directory name 295 295 %g:: 296 296 Git directory name, where all characters except for 297 297 - alpha-numeric ones, `.`, and `-` are replaced with 297 297 + alphanumeric ones, `.`, and `-` are replaced with 298 298 `_` (this should make it easier to use the directory 299 299 name in a filename if wanted) 300 300 %m::

+1 -1

Documentation/git-fast-export.txt

··· 142 142 Specify how to handle `encoding` header in commit objects. When 143 143 asking to 'abort' (which is the default), this program will die 144 144 when encountering such a commit object. With 'yes', the commit 145 145 - message will be reencoded into UTF-8. With 'no', the original 145 145 + message will be re-encoded into UTF-8. With 'no', the original 146 146 encoding will be preserved. 147 147 148 148 --refspec::

+1 -1

Documentation/git-filter-branch.txt

··· 534 534 https://github.com/newren/git-filter-repo/blob/master/contrib/filter-repo-demos/filter-lamely[filter-lamely], 535 535 a drop-in git-filter-branch replacement (with a few caveats). While 536 536 filter-lamely suffers from all the same safety issues as 537 537 - git-filter-branch, it at least ameloriates the performance issues a 537 537 + git-filter-branch, it at least ameliorates the performance issues a 538 538 little. 539 539 540 540 [[SAFETY]]

+1 -1

Documentation/git-range-diff.txt

··· 242 242 243 243 The overall time needed to compute this algorithm is the time needed to 244 244 compute n+m commit diffs and then n*m diffs of patches, plus the time 245 245 - needed to compute the least-cost assigment between n and m diffs. Git 245 245 + needed to compute the least-cost assignment between n and m diffs. Git 246 246 uses an implementation of the Jonker-Volgenant algorithm to solve the 247 247 assignment problem, which has cubic runtime complexity. The matching 248 248 found in this case will look like this:

+1 -1

Documentation/git-tag.txt

··· 65 65 --sign:: 66 66 Make a GPG-signed tag, using the default e-mail address's key. 67 67 The default behavior of tag GPG-signing is controlled by `tag.gpgSign` 68 68 - configuration variable if it exists, or disabled oder otherwise. 68 68 + configuration variable if it exists, or disabled otherwise. 69 69 See linkgit:git-config[1]. 70 70 71 71 --no-sign::

+1 -1

Documentation/git-update-index.txt

··· 426 426 linkgit:git-config[1]). 427 427 428 428 To avoid deleting a shared index file that is still used, its 429 429 - modification time is updated to the current time everytime a new split 429 429 + modification time is updated to the current time every time a new split 430 430 index based on the shared index file is either created or read from. 431 431 432 432 UNTRACKED CACHE

+2 -2

Documentation/git.txt

··· 271 271 the working tree. 272 272 273 273 274 274 - Synching repositories 275 275 - ~~~~~~~~~~~~~~~~~~~~~ 274 274 + Syncing repositories 275 275 + ~~~~~~~~~~~~~~~~~~~~ 276 276 277 277 include::cmds-synchingrepositories.txt[] 278 278

+3 -3

Documentation/gitattributes.txt

··· 293 293 294 294 In these cases you can tell Git the encoding of a file in the working 295 295 directory with the `working-tree-encoding` attribute. If a file with this 296 296 - attribute is added to Git, then Git reencodes the content from the 296 296 + attribute is added to Git, then Git re-encodes the content from the 297 297 specified encoding to UTF-8. Finally, Git stores the UTF-8 encoded 298 298 content in its internal data structure (called "the index"). On checkout 299 299 - the content is reencoded back to the specified encoding. 299 299 + the content is re-encoded back to the specified encoding. 300 300 301 301 Please note that using the `working-tree-encoding` attribute may have a 302 302 number of pitfalls: ··· 498 498 When Git encounters the first file that needs to be cleaned or smudged, 499 499 it starts the filter and performs the handshake. In the handshake, the 500 500 welcome message sent by Git is "git-filter-client", only version 2 is 501 501 - suppported, and the supported capabilities are "clean", "smudge", and 501 501 + supported, and the supported capabilities are "clean", "smudge", and 502 502 "delay". 503 503 504 504 Afterwards Git sends a list of "key=value" pairs terminated with

+1 -1

Documentation/gitmodules.txt

··· 81 81 Committed differences and modifications to tracked files will show 82 82 up. 83 83 84 84 - none;; No modifiations to submodules are ignored, all of committed 84 84 + none;; No modifications to submodules are ignored, all of committed 85 85 differences, and modifications to tracked and untracked files are 86 86 shown. This is the default option. 87 87

+1 -1

Documentation/howto/separating-topic-branches.txt

··· 81 81 o---o---o---o---o---o 82 82 83 83 The last diff better not to show anything other than cleanups 84 84 - for crufts. Then I can finally clean things up: 84 84 + for cruft. Then I can finally clean things up: 85 85 86 86 $ git branch -D topic 87 87 $ git reset --hard HEAD^ ;# nuke pretend merge

+1 -1

Documentation/merge-options.txt

··· 34 34 35 35 --cleanup=<mode>:: 36 36 This option determines how the merge message will be cleaned up before 37 37 - commiting. See linkgit:git-commit[1] for more details. In addition, if 37 37 + committing. See linkgit:git-commit[1] for more details. In addition, if 38 38 the '<mode>' is given a value of `scissors`, scissors will be appended 39 39 to `MERGE_MSG` before being passed on to the commit machinery in the 40 40 case of a merge conflict.

+1 -1

Documentation/technical/api-submodule-config.txt

··· 58 58 59 59 Whenever a submodule configuration is parsed in `parse_submodule_config_option` 60 60 via e.g. `gitmodules_config()`, it will overwrite the null_sha1 entry. 61 61 - So in the normal case, when HEAD:.gitmodules is parsed first and then overlayed 61 61 + So in the normal case, when HEAD:.gitmodules is parsed first and then overlaid 62 62 with the repository configuration, the null_sha1 entry contains the local 63 63 configuration of a submodule (e.g. consolidated values from local git 64 64 configuration and the .gitmodules file in the worktree).

+7 -7

Documentation/technical/api-trace2.txt

··· 178 178 179 179 == Public API 180 180 181 181 - All Trace2 API functions send a messsage to all of the active 181 181 + All Trace2 API functions send a message to all of the active 182 182 Trace2 Targets. This section describes the set of available 183 183 messages. 184 184 ··· 377 377 and the thread elapsed time. 378 378 + 379 379 This function must be called by the thread-proc before it returns 380 380 - (so that the coorect TLS data is used and cleaned up. It should 380 380 + (so that the correct TLS data is used and cleaned up). It should 381 381 not be called by the caller of `pthread_join()`. 382 382 383 383 === Region and Data Messages ··· 406 406 being started, such as "read_recursive" or "do_read_index". 407 407 + 408 408 The `repo` field, if set, will be used to get the "repo-id", so that 409 409 - recursive oerations can be attributed to the correct repository. 409 409 + recursive operations can be attributed to the correct repository. 410 410 411 411 `void trace2_region_leave(const char *category, const char *label, const struct repository *repo)`:: 412 412 ··· 421 421 and reports the elapsed time of the stack frame. 422 422 + 423 423 The `category`, `label`, and `repo` fields are the same as above. 424 424 - The `category` and `label` do not need to match the correpsonding 424 424 + The `category` and `label` do not need to match the corresponding 425 425 "region_enter" message, but it makes the data stream easier to 426 426 understand. 427 427 ··· 816 816 Note that the session-id of the child process is not available to 817 817 the current/spawning process, so the child's PID is reported here as 818 818 a hint for post-processing. (But it is only a hint because the child 819 819 - proces may be a shell script which doesn't have a session-id.) 819 819 + process may be a shell script which doesn't have a session-id.) 820 820 + 821 821 Note that the `t_rel` field contains the observed run time in seconds 822 822 for the child process (starting before the fork/exec/spawn and ··· 1176 1176 + 1177 1177 Regions may be nested. This causes messages to be indented in the 1178 1178 PERF target, for example. 1179 1179 - Elapsed times are relative to the start of the correpsonding nesting 1179 1179 + Elapsed times are relative to the start of the corresponding nesting 1180 1180 level as expected. For example, if we add region message to: 1181 1181 + 1182 1182 ---------------- ··· 1371 1371 In this example, the preload region took 0.009122 seconds. The 7 threads 1372 1372 took between 0.006069 and 0.008947 seconds to work on their portion of 1373 1373 the index. Thread "th01" worked on 508 items at offset 0. Thread "th02" 1374 1374 - worked on 508 items at offset 2032. Thread "th04" worked on 508 itemts 1374 1374 + worked on 508 items at offset 2032. Thread "th04" worked on 508 items 1375 1375 at offset 508. 1376 1376 + 1377 1377 This example also shows that thread names are assigned in a racy manner

+6 -6

Documentation/technical/commit-graph.txt

··· 22 22 directory of an alternate. 23 23 24 24 The commit-graph file stores the commit graph structure along with some 25 25 - extra metadata to speed up graph walks. By listing commit OIDs in lexi- 26 26 - cographic order, we can identify an integer position for each commit and 27 27 - refer to the parents of a commit using those integer positions. We use 28 28 - binary search to find initial commits and then use the integer positions 29 29 - for fast lookups during the walk. 25 25 + extra metadata to speed up graph walks. By listing commit OIDs in 26 26 + lexicographic order, we can identify an integer position for each commit 27 27 + and refer to the parents of a commit using those integer positions. We 28 28 + use binary search to find initial commits and then use the integer 29 29 + positions for fast lookups during the walk. 30 30 31 31 A consumer may load the following info for a commit from the graph: 32 32 ··· 85 85 Since the commit-graph file is closed under reachability, we can guarantee 86 86 the following weaker condition on all commits: 87 87 88 88 - If A and B are commits with generation numbers N amd M, respectively, 88 88 + If A and B are commits with generation numbers N and M, respectively, 89 89 and N < M, then A cannot reach B. 90 90 91 91 Note how the strict inequality differs from the inequality when we have

+2 -2

Documentation/technical/hash-function-transition.txt

··· 531 531 on public-facing Git servers is strongly discouraged. Once Git 532 532 protocol gains SHA-256 support, SHA-256 based servers are likely not 533 533 to support SHA-1 compatibility, to avoid what may be a very expensive 534 534 - hash reencode during clone and to encourage peers to modernize. 534 534 + hash re-encode during clone and to encourage peers to modernize. 535 535 536 536 The design described here allows fetches by SHA-1 clients of a 537 537 personal SHA-256 repository because it's not much more difficult than ··· 602 602 603 603 Choice of Hash 604 604 -------------- 605 605 - In early 2005, around the time that Git was written, Xiaoyun Wang, 605 605 + In early 2005, around the time that Git was written, Xiaoyun Wang, 606 606 Yiqun Lisa Yin, and Hongbo Yu announced an attack finding SHA-1 607 607 collisions in 2^69 operations. In August they published details. 608 608 Luckily, no practical demonstrations of a collision in full SHA-1 were

+2 -2

Documentation/technical/index-format.txt

··· 318 318 == End of Index Entry 319 319 320 320 The End of Index Entry (EOIE) is used to locate the end of the variable 321 321 - length index entries and the begining of the extensions. Code can take 321 321 + length index entries and the beginning of the extensions. Code can take 322 322 advantage of this to quickly locate the index extensions without having 323 323 to parse through all of the index entries. 324 324 ··· 351 351 352 352 - A number of index offset entries each consisting of: 353 353 354 354 - - 32-bit offset from the begining of the file to the first cache entry 354 354 + - 32-bit offset from the beginning of the file to the first cache entry 355 355 in this block of entries. 356 356 357 357 - 32-bit count of cache entries in this block

+1 -1

Documentation/technical/pack-protocol.txt

··· 644 644 command-ok = PKT-LINE("ok" SP refname) 645 645 command-fail = PKT-LINE("ng" SP refname SP error-msg) 646 646 647 647 - error-msg = 1*(OCTECT) ; where not "ok" 647 647 + error-msg = 1*(OCTET) ; where not "ok" 648 648 ---- 649 649 650 650 Updates can be unsuccessful for a number of reasons. The reference can have

+1 -1

Documentation/technical/partial-clone.txt

··· 32 32 33 33 A remote that can later provide the missing objects is called a 34 34 promisor remote, as it promises to send the objects when 35 35 - requested. Initialy Git supported only one promisor remote, the origin 35 35 + requested. Initially Git supported only one promisor remote, the origin 36 36 remote from which the user cloned and that was configured in the 37 37 "extensions.partialClone" config option. Later support for more than 38 38 one promisor remote has been implemented.

+1 -1

Documentation/technical/protocol-v2.txt

··· 252 252 ofs-delta 253 253 Indicate that the client understands PACKv2 with delta referring 254 254 to its base by position in pack rather than by an oid. That is, 255 255 - they can read OBJ_OFS_DELTA (ake type 6) in a packfile. 255 255 + they can read OBJ_OFS_DELTA (aka type 6) in a packfile. 256 256 257 257 If the 'shallow' feature is advertised the following arguments can be 258 258 included in the clients request as well as the potential addition of the

+1 -1

Documentation/technical/rerere.txt

··· 117 117 4 combinations of ("B or C", "C or B") x ("X or Y", "Y or X"). 118 118 119 119 By sorting, the conflict is given its canonical name, namely, "an 120 120 - early part became B or C, a late part becames X or Y", and whenever 120 120 + early part became B or C, a late part became X or Y", and whenever 121 121 any of these four patterns appear, and we can get to the same conflict 122 122 and resolution that we saw earlier. 123 123