]>
Commit | Line | Data |
---|---|---|
6e411d20 SP |
1 | git-fast-import(1) |
2 | ================== | |
3 | ||
4 | NAME | |
5 | ---- | |
7a33631f | 6 | git-fast-import - Backend for fast Git data importers |
6e411d20 SP |
7 | |
8 | ||
9 | SYNOPSIS | |
10 | -------- | |
7791a1d9 | 11 | [verse] |
de613050 | 12 | frontend | 'git fast-import' [<options>] |
6e411d20 SP |
13 | |
14 | DESCRIPTION | |
15 | ----------- | |
16 | This program is usually not what the end user wants to run directly. | |
17 | Most end users want to use one of the existing frontend programs, | |
18 | which parses a specific type of foreign source and feeds the contents | |
0b444cdb | 19 | stored there to 'git fast-import'. |
6e411d20 | 20 | |
882227f1 | 21 | fast-import reads a mixed command/data stream from standard input and |
6e411d20 SP |
22 | writes one or more packfiles directly into the current repository. |
23 | When EOF is received on standard input, fast import writes out | |
24 | updated branch and tag refs, fully updating the current repository | |
25 | with the newly imported data. | |
26 | ||
882227f1 | 27 | The fast-import backend itself can import into an empty repository (one that |
0b444cdb | 28 | has already been initialized by 'git init') or incrementally |
6e411d20 SP |
29 | update an existing populated repository. Whether or not incremental |
30 | imports are supported from a particular foreign source depends on | |
31 | the frontend program in use. | |
32 | ||
33 | ||
34 | OPTIONS | |
35 | ------- | |
63e0c8b3 | 36 | |
7073e69e SP |
37 | --force:: |
38 | Force updating modified existing branches, even if doing | |
39 | so would cause commits to be lost (as the new commit does | |
40 | not contain the old commit). | |
41 | ||
29b1b21f | 42 | --quiet:: |
f55c979b EN |
43 | Disable the output shown by --stats, making fast-import usually |
44 | be silent when it is successful. However, if the import stream | |
45 | has directives intended to show user output (e.g. `progress` | |
46 | directives), the corresponding messages will still be shown. | |
6e411d20 | 47 | |
29b1b21f JK |
48 | --stats:: |
49 | Display some basic statistics about the objects fast-import has | |
50 | created, the packfiles they were stored into, and the | |
51 | memory used by fast-import during this run. Showing this output | |
1c262bb7 | 52 | is currently the default, but can be disabled with --quiet. |
5eef828b | 53 | |
29b1b21f JK |
54 | Options for Frontends |
55 | ~~~~~~~~~~~~~~~~~~~~~ | |
6e411d20 | 56 | |
29b1b21f | 57 | --cat-blob-fd=<fd>:: |
28c7b1f7 | 58 | Write responses to `get-mark`, `cat-blob`, and `ls` queries to the |
a96e8078 JH |
59 | file descriptor <fd> instead of `stdout`. Allows `progress` |
60 | output intended for the end-user to be separated from other | |
61 | output. | |
29b1b21f JK |
62 | |
63 | --date-format=<fmt>:: | |
64 | Specify the type of dates the frontend will supply to | |
65 | fast-import within `author`, `committer` and `tagger` commands. | |
66 | See ``Date Formats'' below for details about which formats | |
67 | are supported, and their syntax. | |
68 | ||
69 | --done:: | |
70 | Terminate with error if there is no `done` command at the end of | |
71 | the stream. This option might be useful for detecting errors | |
72 | that cause the frontend to terminate before it has started to | |
73 | write a stream. | |
74 | ||
75 | Locations of Marks Files | |
76 | ~~~~~~~~~~~~~~~~~~~~~~~~ | |
6e411d20 SP |
77 | |
78 | --export-marks=<file>:: | |
79 | Dumps the internal marks table to <file> when complete. | |
80 | Marks are written one per line as `:markid SHA-1`. | |
81 | Frontends can use this file to validate imports after they | |
e8438420 SP |
82 | have been completed, or to save the marks table across |
83 | incremental runs. As <file> is only opened and truncated | |
84 | at checkpoint (or completion) the same path can also be | |
1c262bb7 | 85 | safely given to --import-marks. |
e8438420 SP |
86 | |
87 | --import-marks=<file>:: | |
88 | Before processing any input, load the marks specified in | |
89 | <file>. The input file must exist, must be readable, and | |
1c262bb7 | 90 | must use the same format as produced by --export-marks. |
e8438420 SP |
91 | Multiple options may be supplied to import more than one |
92 | set of marks. If a mark is defined to different values, | |
93 | the last file wins. | |
6e411d20 | 94 | |
dded4f12 RR |
95 | --import-marks-if-exists=<file>:: |
96 | Like --import-marks but instead of erroring out, silently | |
97 | skips the file if it does not exist. | |
98 | ||
c8a9f3d3 | 99 | --[no-]relative-marks:: |
9fee24ca | 100 | After specifying --relative-marks the paths specified |
bc3c79ae SR |
101 | with --import-marks= and --export-marks= are relative |
102 | to an internal directory in the current repository. | |
103 | In git-fast-import this means that the paths are relative | |
104 | to the .git/info/fast-import directory. However, other | |
105 | importers may use a different location. | |
c8a9f3d3 JK |
106 | + |
107 | Relative and non-relative marks may be combined by interweaving | |
108 | --(no-)-relative-marks with the --(import|export)-marks= options. | |
bc3c79ae | 109 | |
29b1b21f JK |
110 | Performance and Compression Tuning |
111 | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ | |
bc3c79ae | 112 | |
29b1b21f JK |
113 | --active-branches=<n>:: |
114 | Maximum number of branches to maintain active at once. | |
115 | See ``Memory Utilization'' below for details. Default is 5. | |
85c62395 | 116 | |
29b1b21f JK |
117 | --big-file-threshold=<n>:: |
118 | Maximum size of a blob that fast-import will attempt to | |
119 | create a delta for, expressed in bytes. The default is 512m | |
120 | (512 MiB). Some importers may wish to lower this on systems | |
121 | with constrained memory. | |
122 | ||
123 | --depth=<n>:: | |
124 | Maximum delta depth, for blob and tree deltification. | |
4f2220e6 | 125 | Default is 50. |
be56862f | 126 | |
bdf1c06d SP |
127 | --export-pack-edges=<file>:: |
128 | After creating a packfile, print a line of data to | |
129 | <file> listing the filename of the packfile and the last | |
130 | commit on each branch that was written to that packfile. | |
131 | This information may be useful after importing projects | |
132 | whose total object set exceeds the 4 GiB packfile limit, | |
133 | as these commits can be used as edge points during calls | |
0b444cdb | 134 | to 'git pack-objects'. |
bdf1c06d | 135 | |
29b1b21f JK |
136 | --max-pack-size=<n>:: |
137 | Maximum size of each output packfile. | |
138 | The default is unlimited. | |
c499d768 | 139 | |
d9545c7f EW |
140 | fastimport.unpackLimit:: |
141 | See linkgit:git-config[1] | |
c499d768 | 142 | |
76a8788c | 143 | PERFORMANCE |
6e411d20 | 144 | ----------- |
882227f1 | 145 | The design of fast-import allows it to import large projects in a minimum |
6e411d20 | 146 | amount of memory usage and processing time. Assuming the frontend |
882227f1 | 147 | is able to keep up with fast-import and feed it a constant stream of data, |
6e411d20 SP |
148 | import times for projects holding 10+ years of history and containing |
149 | 100,000+ individual commits are generally completed in just 1-2 | |
150 | hours on quite modest (~$2,000 USD) hardware. | |
151 | ||
152 | Most bottlenecks appear to be in foreign source data access (the | |
882227f1 | 153 | source just cannot extract revisions fast enough) or disk IO (fast-import |
6e411d20 SP |
154 | writes as fast as the disk will take the data). Imports will run |
155 | faster if the source data is stored on a different drive than the | |
156 | destination Git repository (due to less IO contention). | |
157 | ||
158 | ||
76a8788c | 159 | DEVELOPMENT COST |
6e411d20 | 160 | ---------------- |
882227f1 | 161 | A typical frontend for fast-import tends to weigh in at approximately 200 |
6e411d20 SP |
162 | lines of Perl/Python/Ruby code. Most developers have been able to |
163 | create working importers in just a couple of hours, even though it | |
882227f1 | 164 | is their first exposure to fast-import, and sometimes even to Git. This is |
6e411d20 SP |
165 | an ideal situation, given that most conversion tools are throw-away |
166 | (use once, and never look back). | |
167 | ||
168 | ||
76a8788c | 169 | PARALLEL OPERATION |
6e411d20 | 170 | ------------------ |
0b444cdb | 171 | Like 'git push' or 'git fetch', imports handled by fast-import are safe to |
6e411d20 | 172 | run alongside parallel `git repack -a -d` or `git gc` invocations, |
0b444cdb | 173 | or any other Git operation (including 'git prune', as loose objects |
882227f1 | 174 | are never used by fast-import). |
6e411d20 | 175 | |
882227f1 SP |
176 | fast-import does not lock the branch or tag refs it is actively importing. |
177 | After the import, during its ref update phase, fast-import tests each | |
7073e69e SP |
178 | existing branch ref to verify the update will be a fast-forward |
179 | update (the commit stored in the ref is contained in the new | |
180 | history of the commit to be written). If the update is not a | |
882227f1 SP |
181 | fast-forward update, fast-import will skip updating that ref and instead |
182 | prints a warning message. fast-import will always attempt to update all | |
7073e69e SP |
183 | branch refs, and does not stop on the first failure. |
184 | ||
1c262bb7 JK |
185 | Branch updates can be forced with --force, but it's recommended that |
186 | this only be used on an otherwise quiet repository. Using --force | |
7073e69e | 187 | is not necessary for an initial import into an empty repository. |
6e411d20 SP |
188 | |
189 | ||
76a8788c | 190 | TECHNICAL DISCUSSION |
6e411d20 | 191 | -------------------- |
882227f1 | 192 | fast-import tracks a set of branches in memory. Any branch can be created |
6e411d20 SP |
193 | or modified at any point during the import process by sending a |
194 | `commit` command on the input stream. This design allows a frontend | |
195 | program to process an unlimited number of branches simultaneously, | |
196 | generating commits in the order they are available from the source | |
197 | data. It also simplifies the frontend programs considerably. | |
198 | ||
882227f1 | 199 | fast-import does not use or alter the current working directory, or any |
6e411d20 SP |
200 | file within it. (It does however update the current Git repository, |
201 | as referenced by `GIT_DIR`.) Therefore an import frontend may use | |
202 | the working directory for its own purposes, such as extracting file | |
203 | revisions from the foreign source. This ignorance of the working | |
882227f1 | 204 | directory also allows fast-import to run very quickly, as it does not |
6e411d20 SP |
205 | need to perform any costly file update operations when switching |
206 | between branches. | |
207 | ||
76a8788c | 208 | INPUT FORMAT |
6e411d20 SP |
209 | ------------ |
210 | With the exception of raw file data (which Git does not interpret) | |
882227f1 | 211 | the fast-import input format is text (ASCII) based. This text based |
6e411d20 SP |
212 | format simplifies development and debugging of frontend programs, |
213 | especially when a higher level language such as Perl, Python or | |
214 | Ruby is being used. | |
215 | ||
882227f1 | 216 | fast-import is very strict about its input. Where we say SP below we mean |
8dc6a373 DB |
217 | *exactly* one space. Likewise LF means one (and only one) linefeed |
218 | and HT one (and only one) horizontal tab. | |
6e411d20 SP |
219 | Supplying additional whitespace characters will cause unexpected |
220 | results, such as branch names or file names with leading or trailing | |
882227f1 | 221 | spaces in their name, or early termination of fast-import when it encounters |
6e411d20 SP |
222 | unexpected input. |
223 | ||
401d53fa SP |
224 | Stream Comments |
225 | ~~~~~~~~~~~~~~~ | |
226 | To aid in debugging frontends fast-import ignores any line that | |
227 | begins with `#` (ASCII pound/hash) up to and including the line | |
228 | ending `LF`. A comment line may contain any sequence of bytes | |
229 | that does not contain an LF and therefore may be used to include | |
230 | any detailed debugging information that might be specific to the | |
231 | frontend and useful when inspecting a fast-import data stream. | |
232 | ||
63e0c8b3 SP |
233 | Date Formats |
234 | ~~~~~~~~~~~~ | |
235 | The following date formats are supported. A frontend should select | |
236 | the format it will use for this import by passing the format name | |
1c262bb7 | 237 | in the --date-format=<fmt> command-line option. |
63e0c8b3 SP |
238 | |
239 | `raw`:: | |
9b92c82f | 240 | This is the Git native format and is `<time> SP <offutc>`. |
1c262bb7 | 241 | It is also fast-import's default format, if --date-format was |
63e0c8b3 SP |
242 | not specified. |
243 | + | |
244 | The time of the event is specified by `<time>` as the number of | |
245 | seconds since the UNIX epoch (midnight, Jan 1, 1970, UTC) and is | |
246 | written as an ASCII decimal integer. | |
247 | + | |
9b92c82f SP |
248 | The local offset is specified by `<offutc>` as a positive or negative |
249 | offset from UTC. For example EST (which is 5 hours behind UTC) | |
250 | would be expressed in `<tz>` by ``-0500'' while UTC is ``+0000''. | |
251 | The local offset does not affect `<time>`; it is used only as an | |
252 | advisement to help formatting routines display the timestamp. | |
63e0c8b3 | 253 | + |
9b92c82f SP |
254 | If the local offset is not available in the source material, use |
255 | ``+0000'', or the most common local offset. For example many | |
63e0c8b3 | 256 | organizations have a CVS repository which has only ever been accessed |
0ffa154b | 257 | by users who are located in the same location and time zone. In this |
f842fdb0 | 258 | case a reasonable offset from UTC could be assumed. |
63e0c8b3 SP |
259 | + |
260 | Unlike the `rfc2822` format, this format is very strict. Any | |
882227f1 | 261 | variation in formatting will cause fast-import to reject the value. |
63e0c8b3 SP |
262 | |
263 | `rfc2822`:: | |
264 | This is the standard email format as described by RFC 2822. | |
265 | + | |
266 | An example value is ``Tue Feb 6 11:22:18 2007 -0500''. The Git | |
f842fdb0 | 267 | parser is accurate, but a little on the lenient side. It is the |
0b444cdb | 268 | same parser used by 'git am' when applying patches |
63e0c8b3 SP |
269 | received from email. |
270 | + | |
271 | Some malformed strings may be accepted as valid dates. In some of | |
272 | these cases Git will still be able to obtain the correct date from | |
273 | the malformed string. There are also some types of malformed | |
274 | strings which Git will parse wrong, and yet consider valid. | |
275 | Seriously malformed strings will be rejected. | |
276 | + | |
0ffa154b | 277 | Unlike the `raw` format above, the time zone/UTC offset information |
9b92c82f SP |
278 | contained in an RFC 2822 date string is used to adjust the date |
279 | value to UTC prior to storage. Therefore it is important that | |
280 | this information be as accurate as possible. | |
281 | + | |
f842fdb0 | 282 | If the source material uses RFC 2822 style dates, |
882227f1 | 283 | the frontend should let fast-import handle the parsing and conversion |
63e0c8b3 SP |
284 | (rather than attempting to do it itself) as the Git parser has |
285 | been well tested in the wild. | |
286 | + | |
287 | Frontends should prefer the `raw` format if the source material | |
f842fdb0 | 288 | already uses UNIX-epoch format, can be coaxed to give dates in that |
02783075 | 289 | format, or its format is easily convertible to it, as there is no |
f842fdb0 | 290 | ambiguity in parsing. |
63e0c8b3 SP |
291 | |
292 | `now`:: | |
0ffa154b | 293 | Always use the current time and time zone. The literal |
63e0c8b3 SP |
294 | `now` must always be supplied for `<when>`. |
295 | + | |
0ffa154b | 296 | This is a toy format. The current time and time zone of this system |
63e0c8b3 | 297 | is always copied into the identity string at the time it is being |
882227f1 | 298 | created by fast-import. There is no way to specify a different time or |
0ffa154b | 299 | time zone. |
63e0c8b3 | 300 | + |
6a5d0b0a | 301 | This particular format is supplied as it's short to implement and |
63e0c8b3 SP |
302 | may be useful to a process that wants to create a new commit |
303 | right now, without needing to use a working directory or | |
0b444cdb | 304 | 'git update-index'. |
63e0c8b3 SP |
305 | + |
306 | If separate `author` and `committer` commands are used in a `commit` | |
307 | the timestamps may not match, as the system clock will be polled | |
308 | twice (once for each command). The only way to ensure that both | |
309 | author and committer identity information has the same timestamp | |
310 | is to omit `author` (thus copying from `committer`) or to use a | |
311 | date format other than `now`. | |
312 | ||
6e411d20 SP |
313 | Commands |
314 | ~~~~~~~~ | |
882227f1 | 315 | fast-import accepts several commands to update the current repository |
6e411d20 SP |
316 | and control the current import process. More detailed discussion |
317 | (with examples) of each command follows later. | |
318 | ||
319 | `commit`:: | |
320 | Creates a new branch or updates an existing branch by | |
321 | creating a new commit and updating the branch to point at | |
322 | the newly created commit. | |
323 | ||
324 | `tag`:: | |
325 | Creates an annotated tag object from an existing commit or | |
326 | branch. Lightweight tags are not supported by this command, | |
327 | as they are not recommended for recording meaningful points | |
328 | in time. | |
329 | ||
330 | `reset`:: | |
331 | Reset an existing branch (or a new branch) to a specific | |
332 | revision. This command must be used to change a branch to | |
333 | a specific revision without making a commit on it. | |
334 | ||
335 | `blob`:: | |
336 | Convert raw file data into a blob, for future use in a | |
337 | `commit` command. This command is optional and is not | |
338 | needed to perform an import. | |
339 | ||
340 | `checkpoint`:: | |
882227f1 | 341 | Forces fast-import to close the current packfile, generate its |
6e411d20 SP |
342 | unique SHA-1 checksum and index, and start a new packfile. |
343 | This command is optional and is not needed to perform | |
344 | an import. | |
345 | ||
ac053c02 SP |
346 | `progress`:: |
347 | Causes fast-import to echo the entire line to its own | |
348 | standard output. This command is optional and is not needed | |
349 | to perform an import. | |
350 | ||
be56862f SR |
351 | `done`:: |
352 | Marks the end of the stream. This command is optional | |
353 | unless the `done` feature was requested using the | |
06ab60c0 | 354 | `--done` command-line option or `feature done` command. |
be56862f | 355 | |
28c7b1f7 MH |
356 | `get-mark`:: |
357 | Causes fast-import to print the SHA-1 corresponding to a mark | |
358 | to the file descriptor set with `--cat-blob-fd`, or `stdout` if | |
359 | unspecified. | |
360 | ||
85c62395 DB |
361 | `cat-blob`:: |
362 | Causes fast-import to print a blob in 'cat-file --batch' | |
363 | format to the file descriptor set with `--cat-blob-fd` or | |
364 | `stdout` if unspecified. | |
365 | ||
8dc6a373 DB |
366 | `ls`:: |
367 | Causes fast-import to print a line describing a directory | |
368 | entry in 'ls-tree' format to the file descriptor set with | |
369 | `--cat-blob-fd` or `stdout` if unspecified. | |
370 | ||
f963bd5d | 371 | `feature`:: |
87c9a140 MM |
372 | Enable the specified feature. This requires that fast-import |
373 | supports the specified feature, and aborts if it does not. | |
f963bd5d | 374 | |
9c8398f0 SR |
375 | `option`:: |
376 | Specify any of the options listed under OPTIONS that do not | |
377 | change stream semantic to suit the frontend's needs. This | |
378 | command is optional and is not needed to perform an import. | |
379 | ||
6e411d20 SP |
380 | `commit` |
381 | ~~~~~~~~ | |
382 | Create or update a branch with a new commit, recording one logical | |
383 | change to the project. | |
384 | ||
385 | .... | |
386 | 'commit' SP <ref> LF | |
387 | mark? | |
a965bb31 | 388 | original-oid? |
74fbd118 SP |
389 | ('author' (SP <name>)? SP LT <email> GT SP <when> LF)? |
390 | 'committer' (SP <name>)? SP LT <email> GT SP <when> LF | |
6e411d20 | 391 | data |
a8a5406a RH |
392 | ('from' SP <commit-ish> LF)? |
393 | ('merge' SP <commit-ish> LF)? | |
a8dd2e7d | 394 | (filemodify | filedelete | filecopy | filerename | filedeleteall | notemodify)* |
1fdb649c | 395 | LF? |
6e411d20 SP |
396 | .... |
397 | ||
398 | where `<ref>` is the name of the branch to make the commit on. | |
399 | Typically branch names are prefixed with `refs/heads/` in | |
400 | Git, so importing the CVS branch symbol `RELENG-1_0` would use | |
401 | `refs/heads/RELENG-1_0` for the value of `<ref>`. The value of | |
402 | `<ref>` must be a valid refname in Git. As `LF` is not valid in | |
403 | a Git refname, no quoting or escaping syntax is supported here. | |
404 | ||
882227f1 | 405 | A `mark` command may optionally appear, requesting fast-import to save a |
6e411d20 SP |
406 | reference to the newly created commit for future use by the frontend |
407 | (see below for format). It is very common for frontends to mark | |
408 | every commit they create, thereby allowing future branch creation | |
409 | from any imported commit. | |
410 | ||
411 | The `data` command following `committer` must supply the commit | |
412 | message (see below for `data` command syntax). To import an empty | |
413 | commit message use a 0 length data. Commit messages are free-form | |
414 | and are not interpreted by Git. Currently they must be encoded in | |
882227f1 | 415 | UTF-8, as fast-import does not permit other encodings to be specified. |
6e411d20 | 416 | |
a8dd2e7d JH |
417 | Zero or more `filemodify`, `filedelete`, `filecopy`, `filerename`, |
418 | `filedeleteall` and `notemodify` commands | |
825769a8 SP |
419 | may be included to update the contents of the branch prior to |
420 | creating the commit. These commands may be supplied in any order. | |
02783075 | 421 | However it is recommended that a `filedeleteall` command precede |
a8dd2e7d JH |
422 | all `filemodify`, `filecopy`, `filerename` and `notemodify` commands in |
423 | the same commit, as `filedeleteall` wipes the branch clean (see below). | |
6e411d20 | 424 | |
62edbec7 EN |
425 | The `LF` after the command is optional (it used to be required). Note |
426 | that for reasons of backward compatibility, if the commit ends with a | |
427 | `data` command (i.e. it has has no `from`, `merge`, `filemodify`, | |
428 | `filedelete`, `filecopy`, `filerename`, `filedeleteall` or | |
429 | `notemodify` commands) then two `LF` commands may appear at the end of | |
430 | the command instead of just one. | |
1fdb649c | 431 | |
6e411d20 SP |
432 | `author` |
433 | ^^^^^^^^ | |
434 | An `author` command may optionally appear, if the author information | |
435 | might differ from the committer information. If `author` is omitted | |
882227f1 | 436 | then fast-import will automatically use the committer's information for |
6e411d20 SP |
437 | the author portion of the commit. See below for a description of |
438 | the fields in `author`, as they are identical to `committer`. | |
439 | ||
440 | `committer` | |
441 | ^^^^^^^^^^^ | |
442 | The `committer` command indicates who made this commit, and when | |
443 | they made it. | |
444 | ||
445 | Here `<name>` is the person's display name (for example | |
446 | ``Com M Itter'') and `<email>` is the person's email address | |
f430ed8b | 447 | (``\cm@example.com''). `LT` and `GT` are the literal less-than (\x3c) |
6e411d20 SP |
448 | and greater-than (\x3e) symbols. These are required to delimit |
449 | the email address from the other fields in the line. Note that | |
4b4963c0 DI |
450 | `<name>` and `<email>` are free-form and may contain any sequence |
451 | of bytes, except `LT`, `GT` and `LF`. `<name>` is typically UTF-8 encoded. | |
6e411d20 | 452 | |
63e0c8b3 | 453 | The time of the change is specified by `<when>` using the date format |
1c262bb7 | 454 | that was selected by the --date-format=<fmt> command-line option. |
63e0c8b3 SP |
455 | See ``Date Formats'' above for the set of supported formats, and |
456 | their syntax. | |
6e411d20 SP |
457 | |
458 | `from` | |
459 | ^^^^^^ | |
ea5e370a SP |
460 | The `from` command is used to specify the commit to initialize |
461 | this branch from. This revision will be the first ancestor of the | |
e7052886 ER |
462 | new commit. The state of the tree built at this commit will begin |
463 | with the state at the `from` commit, and be altered by the content | |
464 | modifications in this commit. | |
ea5e370a SP |
465 | |
466 | Omitting the `from` command in the first commit of a new branch | |
467 | will cause fast-import to create that commit with no ancestor. This | |
468 | tends to be desired only for the initial commit of a project. | |
9b33fa08 EB |
469 | If the frontend creates all files from scratch when making a new |
470 | branch, a `merge` command may be used instead of `from` to start | |
471 | the commit with an empty tree. | |
ea5e370a SP |
472 | Omitting the `from` command on existing branches is usually desired, |
473 | as the current commit on that branch is automatically assumed to | |
474 | be the first ancestor of the new commit. | |
6e411d20 SP |
475 | |
476 | As `LF` is not valid in a Git refname or SHA-1 expression, no | |
a8a5406a | 477 | quoting or escaping syntax is supported within `<commit-ish>`. |
6e411d20 | 478 | |
a8a5406a | 479 | Here `<commit-ish>` is any of the following: |
6e411d20 | 480 | |
882227f1 | 481 | * The name of an existing branch already in fast-import's internal branch |
6a5d0b0a | 482 | table. If fast-import doesn't know the name, it's treated as a SHA-1 |
6e411d20 SP |
483 | expression. |
484 | ||
485 | * A mark reference, `:<idnum>`, where `<idnum>` is the mark number. | |
486 | + | |
882227f1 | 487 | The reason fast-import uses `:` to denote a mark reference is this character |
6e411d20 | 488 | is not legal in a Git branch name. The leading `:` makes it easy |
02783075 | 489 | to distinguish between the mark 42 (`:42`) and the branch 42 (`42` |
6e411d20 SP |
490 | or `refs/heads/42`), or an abbreviated SHA-1 which happened to |
491 | consist only of base-10 digits. | |
492 | + | |
493 | Marks must be declared (via `mark`) before they can be used. | |
494 | ||
495 | * A complete 40 byte or abbreviated commit SHA-1 in hex. | |
496 | ||
497 | * Any valid Git SHA-1 expression that resolves to a commit. See | |
9d83e382 | 498 | ``SPECIFYING REVISIONS'' in linkgit:gitrevisions[7] for details. |
6e411d20 | 499 | |
4ee1b225 FC |
500 | * The special null SHA-1 (40 zeros) specifies that the branch is to be |
501 | removed. | |
502 | ||
6e411d20 SP |
503 | The special case of restarting an incremental import from the |
504 | current branch value should be written as: | |
505 | ---- | |
506 | from refs/heads/branch^0 | |
507 | ---- | |
6cf378f0 | 508 | The `^0` suffix is necessary as fast-import does not permit a branch to |
6e411d20 | 509 | start from itself, and the branch is created in memory before the |
6cf378f0 | 510 | `from` command is even read from the input. Adding `^0` will force |
882227f1 | 511 | fast-import to resolve the commit through Git's revision parsing library, |
6e411d20 SP |
512 | rather than its internal branch table, thereby loading in the |
513 | existing value of the branch. | |
514 | ||
515 | `merge` | |
516 | ^^^^^^^ | |
e7052886 ER |
517 | Includes one additional ancestor commit. The additional ancestry |
518 | link does not change the way the tree state is built at this commit. | |
519 | If the `from` command is | |
9b33fa08 EB |
520 | omitted when creating a new branch, the first `merge` commit will be |
521 | the first ancestor of the current commit, and the branch will start | |
522 | out with no files. An unlimited number of `merge` commands per | |
882227f1 | 523 | commit are permitted by fast-import, thereby establishing an n-way merge. |
6e411d20 | 524 | |
a8a5406a | 525 | Here `<commit-ish>` is any of the commit specification expressions |
6e411d20 SP |
526 | also accepted by `from` (see above). |
527 | ||
528 | `filemodify` | |
ef94edb5 | 529 | ^^^^^^^^^^^^ |
6e411d20 SP |
530 | Included in a `commit` command to add a new file or change the |
531 | content of an existing file. This command has two different means | |
532 | of specifying the content of the file. | |
533 | ||
534 | External data format:: | |
535 | The data content for the file was already supplied by a prior | |
536 | `blob` command. The frontend just needs to connect it. | |
537 | + | |
538 | .... | |
539 | 'M' SP <mode> SP <dataref> SP <path> LF | |
540 | .... | |
541 | + | |
334fba65 | 542 | Here usually `<dataref>` must be either a mark reference (`:<idnum>`) |
6e411d20 | 543 | set by a prior `blob` command, or a full 40-byte SHA-1 of an |
334fba65 JN |
544 | existing Git blob object. If `<mode>` is `040000`` then |
545 | `<dataref>` must be the full 40-byte SHA-1 of an existing | |
546 | Git tree object or a mark reference set with `--import-marks`. | |
6e411d20 SP |
547 | |
548 | Inline data format:: | |
549 | The data content for the file has not been supplied yet. | |
550 | The frontend wants to supply it as part of this modify | |
551 | command. | |
552 | + | |
553 | .... | |
554 | 'M' SP <mode> SP 'inline' SP <path> LF | |
555 | data | |
556 | .... | |
557 | + | |
558 | See below for a detailed description of the `data` command. | |
559 | ||
560 | In both formats `<mode>` is the type of file entry, specified | |
561 | in octal. Git only supports the following modes: | |
562 | ||
563 | * `100644` or `644`: A normal (not-executable) file. The majority | |
564 | of files in most projects use this mode. If in doubt, this is | |
565 | what you want. | |
566 | * `100755` or `755`: A normal, but executable, file. | |
9981b6d9 | 567 | * `120000`: A symlink, the content of the file will be the link target. |
03db4525 AG |
568 | * `160000`: A gitlink, SHA-1 of the object refers to a commit in |
569 | another repository. Git links can only be specified by SHA or through | |
570 | a commit mark. They are used to implement submodules. | |
334fba65 JN |
571 | * `040000`: A subdirectory. Subdirectories can only be specified by |
572 | SHA or through a tree mark set with `--import-marks`. | |
6e411d20 SP |
573 | |
574 | In both formats `<path>` is the complete path of the file to be added | |
575 | (if not already existing) or modified (if already existing). | |
576 | ||
c4431d38 | 577 | A `<path>` string must use UNIX-style directory separators (forward |
6e411d20 SP |
578 | slash `/`), may contain any byte other than `LF`, and must not |
579 | start with double quote (`"`). | |
580 | ||
7c65b2eb MM |
581 | A path can use C-style string quoting; this is accepted in all cases |
582 | and mandatory if the filename starts with double quote or contains | |
583 | `LF`. In C-style quoting, the complete name should be surrounded with | |
584 | double quotes, and any `LF`, backslash, or double quote characters | |
585 | must be escaped by preceding them with a backslash (e.g., | |
586 | `"path/with\n, \\ and \" in it"`). | |
6e411d20 | 587 | |
02783075 | 588 | The value of `<path>` must be in canonical form. That is it must not: |
6e411d20 SP |
589 | |
590 | * contain an empty directory component (e.g. `foo//bar` is invalid), | |
c4431d38 JK |
591 | * end with a directory separator (e.g. `foo/` is invalid), |
592 | * start with a directory separator (e.g. `/foo` is invalid), | |
6e411d20 SP |
593 | * contain the special component `.` or `..` (e.g. `foo/./bar` and |
594 | `foo/../bar` are invalid). | |
595 | ||
e5959106 JN |
596 | The root of the tree can be represented by an empty string as `<path>`. |
597 | ||
6e411d20 SP |
598 | It is recommended that `<path>` always be encoded using UTF-8. |
599 | ||
6e411d20 | 600 | `filedelete` |
ef94edb5 | 601 | ^^^^^^^^^^^^ |
512e44b2 SP |
602 | Included in a `commit` command to remove a file or recursively |
603 | delete an entire directory from the branch. If the file or directory | |
604 | removal makes its parent directory empty, the parent directory will | |
6e411d20 SP |
605 | be automatically removed too. This cascades up the tree until the |
606 | first non-empty directory or the root is reached. | |
607 | ||
608 | .... | |
609 | 'D' SP <path> LF | |
610 | .... | |
611 | ||
512e44b2 SP |
612 | here `<path>` is the complete path of the file or subdirectory to |
613 | be removed from the branch. | |
6e411d20 SP |
614 | See `filemodify` above for a detailed description of `<path>`. |
615 | ||
b6f3481b | 616 | `filecopy` |
a367b869 | 617 | ^^^^^^^^^^ |
b6f3481b SP |
618 | Recursively copies an existing file or subdirectory to a different |
619 | location within the branch. The existing file or directory must | |
620 | exist. If the destination exists it will be completely replaced | |
621 | by the content copied from the source. | |
622 | ||
623 | .... | |
624 | 'C' SP <path> SP <path> LF | |
625 | .... | |
626 | ||
627 | here the first `<path>` is the source location and the second | |
628 | `<path>` is the destination. See `filemodify` above for a detailed | |
629 | description of what `<path>` may look like. To use a source path | |
630 | that contains SP the path must be quoted. | |
631 | ||
632 | A `filecopy` command takes effect immediately. Once the source | |
633 | location has been copied to the destination any future commands | |
634 | applied to the source location will not impact the destination of | |
635 | the copy. | |
636 | ||
f39a946a SP |
637 | `filerename` |
638 | ^^^^^^^^^^^^ | |
639 | Renames an existing file or subdirectory to a different location | |
640 | within the branch. The existing file or directory must exist. If | |
641 | the destination exists it will be replaced by the source directory. | |
642 | ||
643 | .... | |
644 | 'R' SP <path> SP <path> LF | |
645 | .... | |
646 | ||
647 | here the first `<path>` is the source location and the second | |
648 | `<path>` is the destination. See `filemodify` above for a detailed | |
649 | description of what `<path>` may look like. To use a source path | |
650 | that contains SP the path must be quoted. | |
651 | ||
652 | A `filerename` command takes effect immediately. Once the source | |
653 | location has been renamed to the destination any future commands | |
654 | applied to the source location will create new files there and not | |
655 | impact the destination of the rename. | |
656 | ||
b6f3481b SP |
657 | Note that a `filerename` is the same as a `filecopy` followed by a |
658 | `filedelete` of the source location. There is a slight performance | |
659 | advantage to using `filerename`, but the advantage is so small | |
660 | that it is never worth trying to convert a delete/add pair in | |
661 | source material into a rename for fast-import. This `filerename` | |
662 | command is provided just to simplify frontends that already have | |
663 | rename information and don't want bother with decomposing it into a | |
664 | `filecopy` followed by a `filedelete`. | |
665 | ||
825769a8 SP |
666 | `filedeleteall` |
667 | ^^^^^^^^^^^^^^^ | |
668 | Included in a `commit` command to remove all files (and also all | |
669 | directories) from the branch. This command resets the internal | |
670 | branch structure to have no files in it, allowing the frontend | |
671 | to subsequently add all interesting files from scratch. | |
672 | ||
673 | .... | |
674 | 'deleteall' LF | |
675 | .... | |
676 | ||
677 | This command is extremely useful if the frontend does not know | |
678 | (or does not care to know) what files are currently on the branch, | |
679 | and therefore cannot generate the proper `filedelete` commands to | |
680 | update the content. | |
681 | ||
682 | Issuing a `filedeleteall` followed by the needed `filemodify` | |
683 | commands to set the correct content will produce the same results | |
684 | as sending only the needed `filemodify` and `filedelete` commands. | |
882227f1 | 685 | The `filedeleteall` approach may however require fast-import to use slightly |
825769a8 SP |
686 | more memory per active branch (less than 1 MiB for even most large |
687 | projects); so frontends that can easily obtain only the affected | |
688 | paths for a commit are encouraged to do so. | |
689 | ||
a8dd2e7d JH |
690 | `notemodify` |
691 | ^^^^^^^^^^^^ | |
b421812b | 692 | Included in a `commit` `<notes_ref>` command to add a new note |
a8a5406a RH |
693 | annotating a `<commit-ish>` or change this annotation contents. |
694 | Internally it is similar to filemodify 100644 on `<commit-ish>` | |
b421812b DI |
695 | path (maybe split into subdirectories). It's not advised to |
696 | use any other commands to write to the `<notes_ref>` tree except | |
697 | `filedeleteall` to delete all existing notes in this tree. | |
698 | This command has two different means of specifying the content | |
699 | of the note. | |
a8dd2e7d JH |
700 | |
701 | External data format:: | |
702 | The data content for the note was already supplied by a prior | |
703 | `blob` command. The frontend just needs to connect it to the | |
704 | commit that is to be annotated. | |
705 | + | |
706 | .... | |
a8a5406a | 707 | 'N' SP <dataref> SP <commit-ish> LF |
a8dd2e7d JH |
708 | .... |
709 | + | |
710 | Here `<dataref>` can be either a mark reference (`:<idnum>`) | |
711 | set by a prior `blob` command, or a full 40-byte SHA-1 of an | |
712 | existing Git blob object. | |
713 | ||
714 | Inline data format:: | |
715 | The data content for the note has not been supplied yet. | |
716 | The frontend wants to supply it as part of this modify | |
717 | command. | |
718 | + | |
719 | .... | |
a8a5406a | 720 | 'N' SP 'inline' SP <commit-ish> LF |
a8dd2e7d JH |
721 | data |
722 | .... | |
723 | + | |
724 | See below for a detailed description of the `data` command. | |
725 | ||
a8a5406a | 726 | In both formats `<commit-ish>` is any of the commit specification |
a8dd2e7d JH |
727 | expressions also accepted by `from` (see above). |
728 | ||
6e411d20 SP |
729 | `mark` |
730 | ~~~~~~ | |
882227f1 | 731 | Arranges for fast-import to save a reference to the current object, allowing |
6e411d20 SP |
732 | the frontend to recall this object at a future point in time, without |
733 | knowing its SHA-1. Here the current object is the object creation | |
734 | command the `mark` command appears within. This can be `commit`, | |
735 | `tag`, and `blob`, but `commit` is the most common usage. | |
736 | ||
737 | .... | |
738 | 'mark' SP ':' <idnum> LF | |
739 | .... | |
740 | ||
741 | where `<idnum>` is the number assigned by the frontend to this mark. | |
ef94edb5 SP |
742 | The value of `<idnum>` is expressed as an ASCII decimal integer. |
743 | The value 0 is reserved and cannot be used as | |
6e411d20 SP |
744 | a mark. Only values greater than or equal to 1 may be used as marks. |
745 | ||
746 | New marks are created automatically. Existing marks can be moved | |
747 | to another object simply by reusing the same `<idnum>` in another | |
748 | `mark` command. | |
749 | ||
a965bb31 EN |
750 | `original-oid` |
751 | ~~~~~~~~~~~~~~ | |
752 | Provides the name of the object in the original source control system. | |
753 | fast-import will simply ignore this directive, but filter processes | |
754 | which operate on and modify the stream before feeding to fast-import | |
755 | may have uses for this information | |
756 | ||
757 | .... | |
758 | 'original-oid' SP <object-identifier> LF | |
759 | .... | |
760 | ||
761 | where `<object-identifer>` is any string not containing LF. | |
762 | ||
6e411d20 SP |
763 | `tag` |
764 | ~~~~~ | |
765 | Creates an annotated tag referring to a specific commit. To create | |
766 | lightweight (non-annotated) tags see the `reset` command below. | |
767 | ||
768 | .... | |
769 | 'tag' SP <name> LF | |
a8a5406a | 770 | 'from' SP <commit-ish> LF |
a965bb31 | 771 | original-oid? |
74fbd118 | 772 | 'tagger' (SP <name>)? SP LT <email> GT SP <when> LF |
6e411d20 | 773 | data |
6e411d20 SP |
774 | .... |
775 | ||
776 | where `<name>` is the name of the tag to create. | |
777 | ||
778 | Tag names are automatically prefixed with `refs/tags/` when stored | |
779 | in Git, so importing the CVS branch symbol `RELENG-1_0-FINAL` would | |
882227f1 | 780 | use just `RELENG-1_0-FINAL` for `<name>`, and fast-import will write the |
6e411d20 SP |
781 | corresponding ref as `refs/tags/RELENG-1_0-FINAL`. |
782 | ||
783 | The value of `<name>` must be a valid refname in Git and therefore | |
784 | may contain forward slashes. As `LF` is not valid in a Git refname, | |
785 | no quoting or escaping syntax is supported here. | |
786 | ||
787 | The `from` command is the same as in the `commit` command; see | |
788 | above for details. | |
789 | ||
790 | The `tagger` command uses the same format as `committer` within | |
791 | `commit`; again see above for details. | |
792 | ||
793 | The `data` command following `tagger` must supply the annotated tag | |
794 | message (see below for `data` command syntax). To import an empty | |
795 | tag message use a 0 length data. Tag messages are free-form and are | |
796 | not interpreted by Git. Currently they must be encoded in UTF-8, | |
882227f1 | 797 | as fast-import does not permit other encodings to be specified. |
6e411d20 | 798 | |
882227f1 | 799 | Signing annotated tags during import from within fast-import is not |
6e411d20 SP |
800 | supported. Trying to include your own PGP/GPG signature is not |
801 | recommended, as the frontend does not (easily) have access to the | |
802 | complete set of bytes which normally goes into such a signature. | |
882227f1 | 803 | If signing is required, create lightweight tags from within fast-import with |
6e411d20 | 804 | `reset`, then create the annotated versions of those tags offline |
0b444cdb | 805 | with the standard 'git tag' process. |
6e411d20 SP |
806 | |
807 | `reset` | |
808 | ~~~~~~~ | |
809 | Creates (or recreates) the named branch, optionally starting from | |
810 | a specific revision. The reset command allows a frontend to issue | |
811 | a new `from` command for an existing branch, or to create a new | |
812 | branch from an existing commit without creating a new commit. | |
813 | ||
814 | .... | |
815 | 'reset' SP <ref> LF | |
a8a5406a | 816 | ('from' SP <commit-ish> LF)? |
1fdb649c | 817 | LF? |
6e411d20 SP |
818 | .... |
819 | ||
a8a5406a | 820 | For a detailed description of `<ref>` and `<commit-ish>` see above |
6e411d20 SP |
821 | under `commit` and `from`. |
822 | ||
1fdb649c SP |
823 | The `LF` after the command is optional (it used to be required). |
824 | ||
6e411d20 SP |
825 | The `reset` command can also be used to create lightweight |
826 | (non-annotated) tags. For example: | |
827 | ||
828 | ==== | |
829 | reset refs/tags/938 | |
830 | from :938 | |
831 | ==== | |
832 | ||
833 | would create the lightweight tag `refs/tags/938` referring to | |
834 | whatever commit mark `:938` references. | |
835 | ||
836 | `blob` | |
837 | ~~~~~~ | |
838 | Requests writing one file revision to the packfile. The revision | |
839 | is not connected to any commit; this connection must be formed in | |
840 | a subsequent `commit` command by referencing the blob through an | |
841 | assigned mark. | |
842 | ||
843 | .... | |
844 | 'blob' LF | |
845 | mark? | |
a965bb31 | 846 | original-oid? |
6e411d20 SP |
847 | data |
848 | .... | |
849 | ||
850 | The mark command is optional here as some frontends have chosen | |
851 | to generate the Git SHA-1 for the blob on their own, and feed that | |
6a5d0b0a | 852 | directly to `commit`. This is typically more work than it's worth |
6e411d20 SP |
853 | however, as marks are inexpensive to store and easy to use. |
854 | ||
855 | `data` | |
856 | ~~~~~~ | |
857 | Supplies raw data (for use as blob/file content, commit messages, or | |
882227f1 | 858 | annotated tag messages) to fast-import. Data can be supplied using an exact |
6e411d20 SP |
859 | byte count or delimited with a terminating line. Real frontends |
860 | intended for production-quality conversions should always use the | |
861 | exact byte count format, as it is more robust and performs better. | |
882227f1 | 862 | The delimited format is intended primarily for testing fast-import. |
6e411d20 | 863 | |
401d53fa SP |
864 | Comment lines appearing within the `<raw>` part of `data` commands |
865 | are always taken to be part of the body of the data and are therefore | |
866 | never ignored by fast-import. This makes it safe to import any | |
867 | file/message content whose lines might start with `#`. | |
868 | ||
ef94edb5 SP |
869 | Exact byte count format:: |
870 | The frontend must specify the number of bytes of data. | |
871 | + | |
6e411d20 SP |
872 | .... |
873 | 'data' SP <count> LF | |
2c570cde | 874 | <raw> LF? |
6e411d20 | 875 | .... |
ef94edb5 | 876 | + |
6e411d20 | 877 | where `<count>` is the exact number of bytes appearing within |
ef94edb5 SP |
878 | `<raw>`. The value of `<count>` is expressed as an ASCII decimal |
879 | integer. The `LF` on either side of `<raw>` is not | |
6e411d20 | 880 | included in `<count>` and will not be included in the imported data. |
2c570cde SP |
881 | + |
882 | The `LF` after `<raw>` is optional (it used to be required) but | |
883 | recommended. Always including it makes debugging a fast-import | |
884 | stream easier as the next command always starts in column 0 | |
885 | of the next line, even if `<raw>` did not end with an `LF`. | |
6e411d20 | 886 | |
ef94edb5 SP |
887 | Delimited format:: |
888 | A delimiter string is used to mark the end of the data. | |
882227f1 | 889 | fast-import will compute the length by searching for the delimiter. |
02783075 | 890 | This format is primarily useful for testing and is not |
ef94edb5 SP |
891 | recommended for real data. |
892 | + | |
6e411d20 SP |
893 | .... |
894 | 'data' SP '<<' <delim> LF | |
895 | <raw> LF | |
896 | <delim> LF | |
2c570cde | 897 | LF? |
6e411d20 | 898 | .... |
ef94edb5 | 899 | + |
6e411d20 SP |
900 | where `<delim>` is the chosen delimiter string. The string `<delim>` |
901 | must not appear on a line by itself within `<raw>`, as otherwise | |
882227f1 | 902 | fast-import will think the data ends earlier than it really does. The `LF` |
6e411d20 SP |
903 | immediately trailing `<raw>` is part of `<raw>`. This is one of |
904 | the limitations of the delimited format, it is impossible to supply | |
905 | a data chunk which does not have an LF as its last byte. | |
2c570cde SP |
906 | + |
907 | The `LF` after `<delim> LF` is optional (it used to be required). | |
6e411d20 SP |
908 | |
909 | `checkpoint` | |
910 | ~~~~~~~~~~~~ | |
882227f1 | 911 | Forces fast-import to close the current packfile, start a new one, and to |
820b9310 | 912 | save out all current branch refs, tags and marks. |
6e411d20 SP |
913 | |
914 | .... | |
915 | 'checkpoint' LF | |
1fdb649c | 916 | LF? |
6e411d20 SP |
917 | .... |
918 | ||
882227f1 | 919 | Note that fast-import automatically switches packfiles when the current |
1c262bb7 | 920 | packfile reaches --max-pack-size, or 4 GiB, whichever limit is |
882227f1 | 921 | smaller. During an automatic packfile switch fast-import does not update |
820b9310 SP |
922 | the branch refs, tags or marks. |
923 | ||
924 | As a `checkpoint` can require a significant amount of CPU time and | |
925 | disk IO (to compute the overall pack SHA-1 checksum, generate the | |
926 | corresponding index file, and update the refs) it can easily take | |
927 | several minutes for a single `checkpoint` command to complete. | |
928 | ||
929 | Frontends may choose to issue checkpoints during extremely large | |
930 | and long running imports, or when they need to allow another Git | |
931 | process access to a branch. However given that a 30 GiB Subversion | |
882227f1 | 932 | repository can be loaded into Git through fast-import in about 3 hours, |
820b9310 SP |
933 | explicit checkpointing may not be necessary. |
934 | ||
1fdb649c | 935 | The `LF` after the command is optional (it used to be required). |
820b9310 | 936 | |
ac053c02 SP |
937 | `progress` |
938 | ~~~~~~~~~~ | |
939 | Causes fast-import to print the entire `progress` line unmodified to | |
940 | its standard output channel (file descriptor 1) when the command is | |
941 | processed from the input stream. The command otherwise has no impact | |
942 | on the current import, or on any of fast-import's internal state. | |
943 | ||
944 | .... | |
945 | 'progress' SP <any> LF | |
946 | LF? | |
947 | .... | |
948 | ||
949 | The `<any>` part of the command may contain any sequence of bytes | |
950 | that does not contain `LF`. The `LF` after the command is optional. | |
951 | Callers may wish to process the output through a tool such as sed to | |
952 | remove the leading part of the line, for example: | |
953 | ||
954 | ==== | |
b1889c36 | 955 | frontend | git fast-import | sed 's/^progress //' |
ac053c02 SP |
956 | ==== |
957 | ||
958 | Placing a `progress` command immediately after a `checkpoint` will | |
959 | inform the reader when the `checkpoint` has been completed and it | |
960 | can safely access the refs that fast-import updated. | |
961 | ||
28c7b1f7 MH |
962 | `get-mark` |
963 | ~~~~~~~~~~ | |
964 | Causes fast-import to print the SHA-1 corresponding to a mark to | |
965 | stdout or to the file descriptor previously arranged with the | |
966 | `--cat-blob-fd` argument. The command otherwise has no impact on the | |
967 | current import; its purpose is to retrieve SHA-1s that later commits | |
968 | might want to refer to in their commit messages. | |
969 | ||
970 | .... | |
971 | 'get-mark' SP ':' <idnum> LF | |
972 | .... | |
973 | ||
28c7b1f7 MH |
974 | See ``Responses To Commands'' below for details about how to read |
975 | this output safely. | |
976 | ||
85c62395 DB |
977 | `cat-blob` |
978 | ~~~~~~~~~~ | |
979 | Causes fast-import to print a blob to a file descriptor previously | |
980 | arranged with the `--cat-blob-fd` argument. The command otherwise | |
981 | has no impact on the current import; its main purpose is to | |
982 | retrieve blobs that may be in fast-import's memory but not | |
983 | accessible from the target repository. | |
984 | ||
985 | .... | |
986 | 'cat-blob' SP <dataref> LF | |
987 | .... | |
988 | ||
989 | The `<dataref>` can be either a mark reference (`:<idnum>`) | |
990 | set previously or a full 40-byte SHA-1 of a Git blob, preexisting or | |
991 | ready to be written. | |
992 | ||
898243b8 | 993 | Output uses the same format as `git cat-file --batch`: |
85c62395 DB |
994 | |
995 | ==== | |
996 | <sha1> SP 'blob' SP <size> LF | |
997 | <contents> LF | |
998 | ==== | |
999 | ||
7ffde293 EN |
1000 | This command can be used where a `filemodify` directive can appear, |
1001 | allowing it to be used in the middle of a commit. For a `filemodify` | |
1002 | using an inline directive, it can also appear right before the `data` | |
1003 | directive. | |
777f80d7 | 1004 | |
d57e490a JN |
1005 | See ``Responses To Commands'' below for details about how to read |
1006 | this output safely. | |
1007 | ||
8dc6a373 DB |
1008 | `ls` |
1009 | ~~~~ | |
1010 | Prints information about the object at a path to a file descriptor | |
1011 | previously arranged with the `--cat-blob-fd` argument. This allows | |
1012 | printing a blob from the active commit (with `cat-blob`) or copying a | |
1013 | blob or tree from a previous commit for use in the current one (with | |
1014 | `filemodify`). | |
1015 | ||
a63c54a0 EN |
1016 | The `ls` command can also be used where a `filemodify` directive can |
1017 | appear, allowing it to be used in the middle of a commit. | |
8dc6a373 DB |
1018 | |
1019 | Reading from the active commit:: | |
1020 | This form can only be used in the middle of a `commit`. | |
1021 | The path names a directory entry within fast-import's | |
1022 | active commit. The path must be quoted in this case. | |
1023 | + | |
1024 | .... | |
1025 | 'ls' SP <path> LF | |
1026 | .... | |
1027 | ||
1028 | Reading from a named tree:: | |
1029 | The `<dataref>` can be a mark reference (`:<idnum>`) or the | |
1030 | full 40-byte SHA-1 of a Git tag, commit, or tree object, | |
1031 | preexisting or waiting to be written. | |
1032 | The path is relative to the top level of the tree | |
1033 | named by `<dataref>`. | |
1034 | + | |
1035 | .... | |
1036 | 'ls' SP <dataref> SP <path> LF | |
1037 | .... | |
1038 | ||
1039 | See `filemodify` above for a detailed description of `<path>`. | |
1040 | ||
6cf378f0 | 1041 | Output uses the same format as `git ls-tree <tree> -- <path>`: |
8dc6a373 DB |
1042 | |
1043 | ==== | |
1044 | <mode> SP ('blob' | 'tree' | 'commit') SP <dataref> HT <path> LF | |
1045 | ==== | |
1046 | ||
1047 | The <dataref> represents the blob, tree, or commit object at <path> | |
28c7b1f7 MH |
1048 | and can be used in later 'get-mark', 'cat-blob', 'filemodify', or |
1049 | 'ls' commands. | |
8dc6a373 DB |
1050 | |
1051 | If there is no file or subtree at that path, 'git fast-import' will | |
1052 | instead report | |
1053 | ||
1054 | ==== | |
1055 | missing SP <path> LF | |
1056 | ==== | |
1057 | ||
d57e490a JN |
1058 | See ``Responses To Commands'' below for details about how to read |
1059 | this output safely. | |
1060 | ||
f963bd5d SR |
1061 | `feature` |
1062 | ~~~~~~~~~ | |
1063 | Require that fast-import supports the specified feature, or abort if | |
1064 | it does not. | |
1065 | ||
1066 | .... | |
4980fffb | 1067 | 'feature' SP <feature> ('=' <argument>)? LF |
f963bd5d SR |
1068 | .... |
1069 | ||
4980fffb | 1070 | The <feature> part of the command may be any one of the following: |
f963bd5d | 1071 | |
4980fffb JN |
1072 | date-format:: |
1073 | export-marks:: | |
1074 | relative-marks:: | |
1075 | no-relative-marks:: | |
1076 | force:: | |
1077 | Act as though the corresponding command-line option with | |
04b125de | 1078 | a leading `--` was passed on the command line |
4980fffb | 1079 | (see OPTIONS, above). |
f963bd5d | 1080 | |
4980fffb | 1081 | import-marks:: |
3beb4fc4 | 1082 | import-marks-if-exists:: |
4980fffb | 1083 | Like --import-marks except in two respects: first, only one |
3beb4fc4 DI |
1084 | "feature import-marks" or "feature import-marks-if-exists" |
1085 | command is allowed per stream; second, an --import-marks= | |
1086 | or --import-marks-if-exists command-line option overrides | |
1087 | any of these "feature" commands in the stream; third, | |
1088 | "feature import-marks-if-exists" like a corresponding | |
1089 | command-line option silently skips a nonexistent file. | |
f963bd5d | 1090 | |
28c7b1f7 | 1091 | get-mark:: |
85c62395 | 1092 | cat-blob:: |
8dc6a373 | 1093 | ls:: |
28c7b1f7 MH |
1094 | Require that the backend support the 'get-mark', 'cat-blob', |
1095 | or 'ls' command respectively. | |
8dc6a373 DB |
1096 | Versions of fast-import not supporting the specified command |
1097 | will exit with a message indicating so. | |
85c62395 DB |
1098 | This lets the import error out early with a clear message, |
1099 | rather than wasting time on the early part of an import | |
1100 | before the unsupported command is detected. | |
081751c8 | 1101 | |
547e8b92 JN |
1102 | notes:: |
1103 | Require that the backend support the 'notemodify' (N) | |
1104 | subcommand to the 'commit' command. | |
1105 | Versions of fast-import not supporting notes will exit | |
1106 | with a message indicating so. | |
1107 | ||
be56862f SR |
1108 | done:: |
1109 | Error out if the stream ends without a 'done' command. | |
1110 | Without this feature, errors causing the frontend to end | |
1111 | abruptly at a convenient point in the stream can go | |
3266de10 ER |
1112 | undetected. This may occur, for example, if an import |
1113 | front end dies in mid-operation without emitting SIGTERM | |
1114 | or SIGKILL at its subordinate git fast-import instance. | |
a8e4a594 | 1115 | |
9c8398f0 SR |
1116 | `option` |
1117 | ~~~~~~~~ | |
1118 | Processes the specified option so that git fast-import behaves in a | |
1119 | way that suits the frontend's needs. | |
1120 | Note that options specified by the frontend are overridden by any | |
1121 | options the user may specify to git fast-import itself. | |
1122 | ||
1123 | .... | |
1124 | 'option' SP <option> LF | |
1125 | .... | |
1126 | ||
1127 | The `<option>` part of the command may contain any of the options | |
1128 | listed in the OPTIONS section that do not change import semantics, | |
04b125de | 1129 | without the leading `--` and is treated in the same way. |
9c8398f0 SR |
1130 | |
1131 | Option commands must be the first commands on the input (not counting | |
1132 | feature commands), to give an option command after any non-option | |
1133 | command is an error. | |
1134 | ||
06ab60c0 | 1135 | The following command-line options change import semantics and may therefore |
9c8398f0 SR |
1136 | not be passed as option: |
1137 | ||
1138 | * date-format | |
1139 | * import-marks | |
1140 | * export-marks | |
85c62395 | 1141 | * cat-blob-fd |
9c8398f0 SR |
1142 | * force |
1143 | ||
be56862f SR |
1144 | `done` |
1145 | ~~~~~~ | |
1146 | If the `done` feature is not in use, treated as if EOF was read. | |
1147 | This can be used to tell fast-import to finish early. | |
1148 | ||
06ab60c0 | 1149 | If the `--done` command-line option or `feature done` command is |
be56862f SR |
1150 | in use, the `done` command is mandatory and marks the end of the |
1151 | stream. | |
1152 | ||
76a8788c | 1153 | RESPONSES TO COMMANDS |
d57e490a JN |
1154 | --------------------- |
1155 | New objects written by fast-import are not available immediately. | |
1156 | Most fast-import commands have no visible effect until the next | |
1157 | checkpoint (or completion). The frontend can send commands to | |
1158 | fill fast-import's input pipe without worrying about how quickly | |
1159 | they will take effect, which improves performance by simplifying | |
1160 | scheduling. | |
1161 | ||
1162 | For some frontends, though, it is useful to be able to read back | |
1163 | data from the current repository as it is being updated (for | |
1164 | example when the source material describes objects in terms of | |
1165 | patches to be applied to previously imported objects). This can | |
1166 | be accomplished by connecting the frontend and fast-import via | |
1167 | bidirectional pipes: | |
1168 | ||
1169 | ==== | |
1170 | mkfifo fast-import-output | |
1171 | frontend <fast-import-output | | |
1172 | git fast-import >fast-import-output | |
1173 | ==== | |
1174 | ||
28c7b1f7 MH |
1175 | A frontend set up this way can use `progress`, `get-mark`, `ls`, and |
1176 | `cat-blob` commands to read information from the import in progress. | |
d57e490a JN |
1177 | |
1178 | To avoid deadlock, such frontends must completely consume any | |
28c7b1f7 | 1179 | pending output from `progress`, `ls`, `get-mark`, and `cat-blob` before |
d57e490a JN |
1180 | performing writes to fast-import that might block. |
1181 | ||
76a8788c | 1182 | CRASH REPORTS |
e7e5170f SP |
1183 | ------------- |
1184 | If fast-import is supplied invalid input it will terminate with a | |
1185 | non-zero exit status and create a crash report in the top level of | |
1186 | the Git repository it was importing into. Crash reports contain | |
1187 | a snapshot of the internal fast-import state as well as the most | |
1188 | recent commands that lead up to the crash. | |
1189 | ||
1190 | All recent commands (including stream comments, file changes and | |
1191 | progress commands) are shown in the command history within the crash | |
1192 | report, but raw file data and commit messages are excluded from the | |
1193 | crash report. This exclusion saves space within the report file | |
1194 | and reduces the amount of buffering that fast-import must perform | |
1195 | during execution. | |
1196 | ||
1197 | After writing a crash report fast-import will close the current | |
1198 | packfile and export the marks table. This allows the frontend | |
1199 | developer to inspect the repository state and resume the import from | |
1200 | the point where it crashed. The modified branches and tags are not | |
1201 | updated during a crash, as the import did not complete successfully. | |
1202 | Branch and tag information can be found in the crash report and | |
1203 | must be applied manually if the update is needed. | |
1204 | ||
1205 | An example crash: | |
1206 | ||
1207 | ==== | |
1208 | $ cat >in <<END_OF_INPUT | |
1209 | # my very first test commit | |
1210 | commit refs/heads/master | |
1211 | committer Shawn O. Pearce <spearce> 19283 -0400 | |
1212 | # who is that guy anyway? | |
1213 | data <<EOF | |
1214 | this is my commit | |
1215 | EOF | |
1216 | M 644 inline .gitignore | |
1217 | data <<EOF | |
1218 | .gitignore | |
1219 | EOF | |
1220 | M 777 inline bob | |
1221 | END_OF_INPUT | |
1222 | ||
b1889c36 | 1223 | $ git fast-import <in |
e7e5170f SP |
1224 | fatal: Corrupt mode: M 777 inline bob |
1225 | fast-import: dumping crash report to .git/fast_import_crash_8434 | |
1226 | ||
1227 | $ cat .git/fast_import_crash_8434 | |
1228 | fast-import crash report: | |
1229 | fast-import process: 8434 | |
1230 | parent process : 1391 | |
1231 | at Sat Sep 1 00:58:12 2007 | |
1232 | ||
1233 | fatal: Corrupt mode: M 777 inline bob | |
1234 | ||
1235 | Most Recent Commands Before Crash | |
1236 | --------------------------------- | |
1237 | # my very first test commit | |
1238 | commit refs/heads/master | |
1239 | committer Shawn O. Pearce <spearce> 19283 -0400 | |
1240 | # who is that guy anyway? | |
1241 | data <<EOF | |
1242 | M 644 inline .gitignore | |
1243 | data <<EOF | |
1244 | * M 777 inline bob | |
1245 | ||
1246 | Active Branch LRU | |
1247 | ----------------- | |
1248 | active_branches = 1 cur, 5 max | |
1249 | ||
1250 | pos clock name | |
1251 | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ | |
1252 | 1) 0 refs/heads/master | |
1253 | ||
1254 | Inactive Branches | |
1255 | ----------------- | |
1256 | refs/heads/master: | |
1257 | status : active loaded dirty | |
1258 | tip commit : 0000000000000000000000000000000000000000 | |
1259 | old tree : 0000000000000000000000000000000000000000 | |
1260 | cur tree : 0000000000000000000000000000000000000000 | |
1261 | commit clock: 0 | |
1262 | last pack : | |
1263 | ||
1264 | ||
1265 | ------------------- | |
1266 | END OF CRASH REPORT | |
1267 | ==== | |
1268 | ||
76a8788c | 1269 | TIPS AND TRICKS |
bdd9f424 SP |
1270 | --------------- |
1271 | The following tips and tricks have been collected from various | |
882227f1 | 1272 | users of fast-import, and are offered here as suggestions. |
bdd9f424 SP |
1273 | |
1274 | Use One Mark Per Commit | |
1275 | ~~~~~~~~~~~~~~~~~~~~~~~ | |
1276 | When doing a repository conversion, use a unique mark per commit | |
1c262bb7 | 1277 | (`mark :<n>`) and supply the --export-marks option on the command |
882227f1 | 1278 | line. fast-import will dump a file which lists every mark and the Git |
bdd9f424 SP |
1279 | object SHA-1 that corresponds to it. If the frontend can tie |
1280 | the marks back to the source repository, it is easy to verify the | |
1281 | accuracy and completeness of the import by comparing each Git | |
1282 | commit to the corresponding source revision. | |
1283 | ||
1284 | Coming from a system such as Perforce or Subversion this should be | |
882227f1 | 1285 | quite simple, as the fast-import mark can also be the Perforce changeset |
bdd9f424 SP |
1286 | number or the Subversion revision number. |
1287 | ||
1288 | Freely Skip Around Branches | |
1289 | ~~~~~~~~~~~~~~~~~~~~~~~~~~~ | |
1290 | Don't bother trying to optimize the frontend to stick to one branch | |
1291 | at a time during an import. Although doing so might be slightly | |
882227f1 | 1292 | faster for fast-import, it tends to increase the complexity of the frontend |
bdd9f424 SP |
1293 | code considerably. |
1294 | ||
882227f1 | 1295 | The branch LRU builtin to fast-import tends to behave very well, and the |
bdd9f424 SP |
1296 | cost of activating an inactive branch is so low that bouncing around |
1297 | between branches has virtually no impact on import performance. | |
1298 | ||
c7346156 SP |
1299 | Handling Renames |
1300 | ~~~~~~~~~~~~~~~~ | |
1301 | When importing a renamed file or directory, simply delete the old | |
1302 | name(s) and modify the new name(s) during the corresponding commit. | |
1303 | Git performs rename detection after-the-fact, rather than explicitly | |
1304 | during a commit. | |
1305 | ||
bdd9f424 SP |
1306 | Use Tag Fixup Branches |
1307 | ~~~~~~~~~~~~~~~~~~~~~~ | |
1308 | Some other SCM systems let the user create a tag from multiple | |
1309 | files which are not from the same commit/changeset. Or to create | |
1310 | tags which are a subset of the files available in the repository. | |
1311 | ||
1312 | Importing these tags as-is in Git is impossible without making at | |
1313 | least one commit which ``fixes up'' the files to match the content | |
882227f1 | 1314 | of the tag. Use fast-import's `reset` command to reset a dummy branch |
bdd9f424 SP |
1315 | outside of your normal branch space to the base commit for the tag, |
1316 | then commit one or more file fixup commits, and finally tag the | |
1317 | dummy branch. | |
1318 | ||
1319 | For example since all normal branches are stored under `refs/heads/` | |
1320 | name the tag fixup branch `TAG_FIXUP`. This way it is impossible for | |
1321 | the fixup branch used by the importer to have namespace conflicts | |
1322 | with real branches imported from the source (the name `TAG_FIXUP` | |
1323 | is not `refs/heads/TAG_FIXUP`). | |
1324 | ||
1325 | When committing fixups, consider using `merge` to connect the | |
1326 | commit(s) which are supplying file revisions to the fixup branch. | |
0b444cdb | 1327 | Doing so will allow tools such as 'git blame' to track |
bdd9f424 SP |
1328 | through the real commit history and properly annotate the source |
1329 | files. | |
1330 | ||
882227f1 | 1331 | After fast-import terminates the frontend will need to do `rm .git/TAG_FIXUP` |
bdd9f424 SP |
1332 | to remove the dummy branch. |
1333 | ||
1334 | Import Now, Repack Later | |
1335 | ~~~~~~~~~~~~~~~~~~~~~~~~ | |
882227f1 | 1336 | As soon as fast-import completes the Git repository is completely valid |
02783075 | 1337 | and ready for use. Typically this takes only a very short time, |
bdd9f424 SP |
1338 | even for considerably large projects (100,000+ commits). |
1339 | ||
1340 | However repacking the repository is necessary to improve data | |
1341 | locality and access performance. It can also take hours on extremely | |
1c262bb7 | 1342 | large projects (especially if -f and a large --window parameter is |
bdd9f424 SP |
1343 | used). Since repacking is safe to run alongside readers and writers, |
1344 | run the repack in the background and let it finish when it finishes. | |
1345 | There is no reason to wait to explore your new Git project! | |
1346 | ||
1347 | If you choose to wait for the repack, don't try to run benchmarks | |
882227f1 | 1348 | or performance tests until repacking is completed. fast-import outputs |
bdd9f424 SP |
1349 | suboptimal packfiles that are simply never seen in real use |
1350 | situations. | |
1351 | ||
1352 | Repacking Historical Data | |
1353 | ~~~~~~~~~~~~~~~~~~~~~~~~~ | |
1354 | If you are repacking very old imported data (e.g. older than the | |
1355 | last year), consider expending some extra CPU time and supplying | |
1c262bb7 | 1356 | --window=50 (or higher) when you run 'git repack'. |
bdd9f424 SP |
1357 | This will take longer, but will also produce a smaller packfile. |
1358 | You only need to expend the effort once, and everyone using your | |
1359 | project will benefit from the smaller repository. | |
1360 | ||
ac053c02 SP |
1361 | Include Some Progress Messages |
1362 | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ | |
1363 | Every once in a while have your frontend emit a `progress` message | |
1364 | to fast-import. The contents of the messages are entirely free-form, | |
1365 | so one suggestion would be to output the current month and year | |
1366 | each time the current commit date moves into the next month. | |
1367 | Your users will feel better knowing how much of the data stream | |
1368 | has been processed. | |
1369 | ||
bdd9f424 | 1370 | |
76a8788c | 1371 | PACKFILE OPTIMIZATION |
6e411d20 | 1372 | --------------------- |
882227f1 | 1373 | When packing a blob fast-import always attempts to deltify against the last |
6e411d20 SP |
1374 | blob written. Unless specifically arranged for by the frontend, |
1375 | this will probably not be a prior version of the same file, so the | |
1376 | generated delta will not be the smallest possible. The resulting | |
1377 | packfile will be compressed, but will not be optimal. | |
1378 | ||
1379 | Frontends which have efficient access to all revisions of a | |
1380 | single file (for example reading an RCS/CVS ,v file) can choose | |
1381 | to supply all revisions of that file as a sequence of consecutive | |
882227f1 | 1382 | `blob` commands. This allows fast-import to deltify the different file |
6e411d20 SP |
1383 | revisions against each other, saving space in the final packfile. |
1384 | Marks can be used to later identify individual file revisions during | |
1385 | a sequence of `commit` commands. | |
1386 | ||
882227f1 SP |
1387 | The packfile(s) created by fast-import do not encourage good disk access |
1388 | patterns. This is caused by fast-import writing the data in the order | |
6e411d20 SP |
1389 | it is received on standard input, while Git typically organizes |
1390 | data within packfiles to make the most recent (current tip) data | |
1391 | appear before historical data. Git also clusters commits together, | |
1392 | speeding up revision traversal through better cache locality. | |
1393 | ||
1394 | For this reason it is strongly recommended that users repack the | |
882227f1 | 1395 | repository with `git repack -a -d` after fast-import completes, allowing |
6e411d20 SP |
1396 | Git to reorganize the packfiles for faster data access. If blob |
1397 | deltas are suboptimal (see above) then also adding the `-f` option | |
1398 | to force recomputation of all deltas can significantly reduce the | |
1399 | final packfile size (30-50% smaller can be quite typical). | |
1400 | ||
73845048 ÆAB |
1401 | Instead of running `git repack` you can also run `git gc |
1402 | --aggressive`, which will also optimize other things after an import | |
1403 | (e.g. pack loose refs). As noted in the "AGGRESSIVE" section in | |
1404 | linkgit:git-gc[1] the `--aggressive` option will find new deltas with | |
1405 | the `-f` option to linkgit:git-repack[1]. For the reasons elaborated | |
1406 | on above using `--aggressive` after a fast-import is one of the few | |
1407 | cases where it's known to be worthwhile. | |
bdd9f424 | 1408 | |
76a8788c | 1409 | MEMORY UTILIZATION |
6e411d20 | 1410 | ------------------ |
882227f1 | 1411 | There are a number of factors which affect how much memory fast-import |
6e411d20 | 1412 | requires to perform an import. Like critical sections of core |
02783075 BH |
1413 | Git, fast-import uses its own memory allocators to amortize any overheads |
1414 | associated with malloc. In practice fast-import tends to amortize any | |
6e411d20 SP |
1415 | malloc overheads to 0, due to its use of large block allocations. |
1416 | ||
1417 | per object | |
1418 | ~~~~~~~~~~ | |
882227f1 | 1419 | fast-import maintains an in-memory structure for every object written in |
6e411d20 SP |
1420 | this execution. On a 32 bit system the structure is 32 bytes, |
1421 | on a 64 bit system the structure is 40 bytes (due to the larger | |
1422 | pointer sizes). Objects in the table are not deallocated until | |
882227f1 | 1423 | fast-import terminates. Importing 2 million objects on a 32 bit system |
6e411d20 SP |
1424 | will require approximately 64 MiB of memory. |
1425 | ||
1426 | The object table is actually a hashtable keyed on the object name | |
882227f1 | 1427 | (the unique SHA-1). This storage configuration allows fast-import to reuse |
6e411d20 SP |
1428 | an existing or already written object and avoid writing duplicates |
1429 | to the output packfile. Duplicate blobs are surprisingly common | |
1430 | in an import, typically due to branch merges in the source. | |
1431 | ||
1432 | per mark | |
1433 | ~~~~~~~~ | |
1434 | Marks are stored in a sparse array, using 1 pointer (4 bytes or 8 | |
1435 | bytes, depending on pointer size) per mark. Although the array | |
1436 | is sparse, frontends are still strongly encouraged to use marks | |
1437 | between 1 and n, where n is the total number of marks required for | |
1438 | this import. | |
1439 | ||
1440 | per branch | |
1441 | ~~~~~~~~~~ | |
1442 | Branches are classified as active and inactive. The memory usage | |
1443 | of the two classes is significantly different. | |
1444 | ||
1445 | Inactive branches are stored in a structure which uses 96 or 120 | |
1446 | bytes (32 bit or 64 bit systems, respectively), plus the length of | |
882227f1 | 1447 | the branch name (typically under 200 bytes), per branch. fast-import will |
6e411d20 SP |
1448 | easily handle as many as 10,000 inactive branches in under 2 MiB |
1449 | of memory. | |
1450 | ||
1451 | Active branches have the same overhead as inactive branches, but | |
1452 | also contain copies of every tree that has been recently modified on | |
1453 | that branch. If subtree `include` has not been modified since the | |
1454 | branch became active, its contents will not be loaded into memory, | |
1455 | but if subtree `src` has been modified by a commit since the branch | |
1456 | became active, then its contents will be loaded in memory. | |
1457 | ||
1458 | As active branches store metadata about the files contained on that | |
1459 | branch, their in-memory storage size can grow to a considerable size | |
1460 | (see below). | |
1461 | ||
882227f1 | 1462 | fast-import automatically moves active branches to inactive status based on |
6e411d20 SP |
1463 | a simple least-recently-used algorithm. The LRU chain is updated on |
1464 | each `commit` command. The maximum number of active branches can be | |
1c262bb7 | 1465 | increased or decreased on the command line with --active-branches=. |
6e411d20 SP |
1466 | |
1467 | per active tree | |
1468 | ~~~~~~~~~~~~~~~ | |
1469 | Trees (aka directories) use just 12 bytes of memory on top of the | |
1470 | memory required for their entries (see ``per active file'' below). | |
02783075 | 1471 | The cost of a tree is virtually 0, as its overhead amortizes out |
6e411d20 SP |
1472 | over the individual file entries. |
1473 | ||
1474 | per active file entry | |
1475 | ~~~~~~~~~~~~~~~~~~~~~ | |
1476 | Files (and pointers to subtrees) within active trees require 52 or 64 | |
1477 | bytes (32/64 bit platforms) per entry. To conserve space, file and | |
1478 | tree names are pooled in a common string table, allowing the filename | |
1479 | ``Makefile'' to use just 16 bytes (after including the string header | |
1480 | overhead) no matter how many times it occurs within the project. | |
1481 | ||
1482 | The active branch LRU, when coupled with the filename string pool | |
882227f1 | 1483 | and lazy loading of subtrees, allows fast-import to efficiently import |
6e411d20 SP |
1484 | projects with 2,000+ branches and 45,114+ files in a very limited |
1485 | memory footprint (less than 2.7 MiB per active branch). | |
1486 | ||
76a8788c | 1487 | SIGNALS |
dc01f59d JN |
1488 | ------- |
1489 | Sending *SIGUSR1* to the 'git fast-import' process ends the current | |
1490 | packfile early, simulating a `checkpoint` command. The impatient | |
1491 | operator can use this facility to peek at the objects and refs from an | |
1492 | import in progress, at the cost of some added running time and worse | |
1493 | compression. | |
6e411d20 | 1494 | |
26726718 MH |
1495 | SEE ALSO |
1496 | -------- | |
1497 | linkgit:git-fast-export[1] | |
1498 | ||
6e411d20 SP |
1499 | GIT |
1500 | --- | |
9e1f0a85 | 1501 | Part of the linkgit:git[1] suite |