6 git-fast-import - Backend for fast Git data importers
11 frontend | 'git fast-import' [options]
15 This program is usually not what the end user wants to run directly.
16 Most end users want to use one of the existing frontend programs,
17 which parses a specific type of foreign source and feeds the contents
18 stored there to 'git-fast-import'.
20 fast-import reads a mixed command/data stream from standard input and
21 writes one or more packfiles directly into the current repository.
22 When EOF is received on standard input, fast import writes out
23 updated branch and tag refs, fully updating the current repository
24 with the newly imported data.
26 The fast-import backend itself can import into an empty repository (one that
27 has already been initialized by 'git-init') or incrementally
28 update an existing populated repository. Whether or not incremental
29 imports are supported from a particular foreign source depends on
30 the frontend program in use.
36 Specify the type of dates the frontend will supply to
37 fast-import within `author`, `committer` and `tagger` commands.
38 See ``Date Formats'' below for details about which formats
39 are supported, and their syntax.
42 Force updating modified existing branches, even if doing
43 so would cause commits to be lost (as the new commit does
44 not contain the old commit).
47 Maximum size of each output packfile, expressed in MiB.
48 The default is 4096 (4 GiB) as that is the maximum allowed
49 packfile size (due to file format limitations). Some
50 importers may wish to lower this, such as to ensure the
51 resulting packfiles fit on CDs.
54 Maximum delta depth, for blob and tree deltification.
57 --active-branches=<n>::
58 Maximum number of branches to maintain active at once.
59 See ``Memory Utilization'' below for details. Default is 5.
61 --export-marks=<file>::
62 Dumps the internal marks table to <file> when complete.
63 Marks are written one per line as `:markid SHA-1`.
64 Frontends can use this file to validate imports after they
65 have been completed, or to save the marks table across
66 incremental runs. As <file> is only opened and truncated
67 at checkpoint (or completion) the same path can also be
68 safely given to \--import-marks.
70 --import-marks=<file>::
71 Before processing any input, load the marks specified in
72 <file>. The input file must exist, must be readable, and
73 must use the same format as produced by \--export-marks.
74 Multiple options may be supplied to import more than one
75 set of marks. If a mark is defined to different values,
78 --export-pack-edges=<file>::
79 After creating a packfile, print a line of data to
80 <file> listing the filename of the packfile and the last
81 commit on each branch that was written to that packfile.
82 This information may be useful after importing projects
83 whose total object set exceeds the 4 GiB packfile limit,
84 as these commits can be used as edge points during calls
85 to 'git-pack-objects'.
88 Disable all non-fatal output, making fast-import silent when it
89 is successful. This option disables the output shown by
93 Display some basic statistics about the objects fast-import has
94 created, the packfiles they were stored into, and the
95 memory used by fast-import during this run. Showing this output
96 is currently the default, but can be disabled with \--quiet.
101 The design of fast-import allows it to import large projects in a minimum
102 amount of memory usage and processing time. Assuming the frontend
103 is able to keep up with fast-import and feed it a constant stream of data,
104 import times for projects holding 10+ years of history and containing
105 100,000+ individual commits are generally completed in just 1-2
106 hours on quite modest (~$2,000 USD) hardware.
108 Most bottlenecks appear to be in foreign source data access (the
109 source just cannot extract revisions fast enough) or disk IO (fast-import
110 writes as fast as the disk will take the data). Imports will run
111 faster if the source data is stored on a different drive than the
112 destination Git repository (due to less IO contention).
117 A typical frontend for fast-import tends to weigh in at approximately 200
118 lines of Perl/Python/Ruby code. Most developers have been able to
119 create working importers in just a couple of hours, even though it
120 is their first exposure to fast-import, and sometimes even to Git. This is
121 an ideal situation, given that most conversion tools are throw-away
122 (use once, and never look back).
127 Like 'git-push' or 'git-fetch', imports handled by fast-import are safe to
128 run alongside parallel `git repack -a -d` or `git gc` invocations,
129 or any other Git operation (including 'git-prune', as loose objects
130 are never used by fast-import).
132 fast-import does not lock the branch or tag refs it is actively importing.
133 After the import, during its ref update phase, fast-import tests each
134 existing branch ref to verify the update will be a fast-forward
135 update (the commit stored in the ref is contained in the new
136 history of the commit to be written). If the update is not a
137 fast-forward update, fast-import will skip updating that ref and instead
138 prints a warning message. fast-import will always attempt to update all
139 branch refs, and does not stop on the first failure.
141 Branch updates can be forced with \--force, but its recommended that
142 this only be used on an otherwise quiet repository. Using \--force
143 is not necessary for an initial import into an empty repository.
148 fast-import tracks a set of branches in memory. Any branch can be created
149 or modified at any point during the import process by sending a
150 `commit` command on the input stream. This design allows a frontend
151 program to process an unlimited number of branches simultaneously,
152 generating commits in the order they are available from the source
153 data. It also simplifies the frontend programs considerably.
155 fast-import does not use or alter the current working directory, or any
156 file within it. (It does however update the current Git repository,
157 as referenced by `GIT_DIR`.) Therefore an import frontend may use
158 the working directory for its own purposes, such as extracting file
159 revisions from the foreign source. This ignorance of the working
160 directory also allows fast-import to run very quickly, as it does not
161 need to perform any costly file update operations when switching
166 With the exception of raw file data (which Git does not interpret)
167 the fast-import input format is text (ASCII) based. This text based
168 format simplifies development and debugging of frontend programs,
169 especially when a higher level language such as Perl, Python or
172 fast-import is very strict about its input. Where we say SP below we mean
173 *exactly* one space. Likewise LF means one (and only one) linefeed.
174 Supplying additional whitespace characters will cause unexpected
175 results, such as branch names or file names with leading or trailing
176 spaces in their name, or early termination of fast-import when it encounters
181 To aid in debugging frontends fast-import ignores any line that
182 begins with `#` (ASCII pound/hash) up to and including the line
183 ending `LF`. A comment line may contain any sequence of bytes
184 that does not contain an LF and therefore may be used to include
185 any detailed debugging information that might be specific to the
186 frontend and useful when inspecting a fast-import data stream.
190 The following date formats are supported. A frontend should select
191 the format it will use for this import by passing the format name
192 in the \--date-format=<fmt> command line option.
195 This is the Git native format and is `<time> SP <offutc>`.
196 It is also fast-import's default format, if \--date-format was
199 The time of the event is specified by `<time>` as the number of
200 seconds since the UNIX epoch (midnight, Jan 1, 1970, UTC) and is
201 written as an ASCII decimal integer.
203 The local offset is specified by `<offutc>` as a positive or negative
204 offset from UTC. For example EST (which is 5 hours behind UTC)
205 would be expressed in `<tz>` by ``-0500'' while UTC is ``+0000''.
206 The local offset does not affect `<time>`; it is used only as an
207 advisement to help formatting routines display the timestamp.
209 If the local offset is not available in the source material, use
210 ``+0000'', or the most common local offset. For example many
211 organizations have a CVS repository which has only ever been accessed
212 by users who are located in the same location and timezone. In this
213 case a reasonable offset from UTC could be assumed.
215 Unlike the `rfc2822` format, this format is very strict. Any
216 variation in formatting will cause fast-import to reject the value.
219 This is the standard email format as described by RFC 2822.
221 An example value is ``Tue Feb 6 11:22:18 2007 -0500''. The Git
222 parser is accurate, but a little on the lenient side. It is the
223 same parser used by 'git-am' when applying patches
226 Some malformed strings may be accepted as valid dates. In some of
227 these cases Git will still be able to obtain the correct date from
228 the malformed string. There are also some types of malformed
229 strings which Git will parse wrong, and yet consider valid.
230 Seriously malformed strings will be rejected.
232 Unlike the `raw` format above, the timezone/UTC offset information
233 contained in an RFC 2822 date string is used to adjust the date
234 value to UTC prior to storage. Therefore it is important that
235 this information be as accurate as possible.
237 If the source material uses RFC 2822 style dates,
238 the frontend should let fast-import handle the parsing and conversion
239 (rather than attempting to do it itself) as the Git parser has
240 been well tested in the wild.
242 Frontends should prefer the `raw` format if the source material
243 already uses UNIX-epoch format, can be coaxed to give dates in that
244 format, or its format is easily convertible to it, as there is no
245 ambiguity in parsing.
248 Always use the current time and timezone. The literal
249 `now` must always be supplied for `<when>`.
251 This is a toy format. The current time and timezone of this system
252 is always copied into the identity string at the time it is being
253 created by fast-import. There is no way to specify a different time or
256 This particular format is supplied as its short to implement and
257 may be useful to a process that wants to create a new commit
258 right now, without needing to use a working directory or
261 If separate `author` and `committer` commands are used in a `commit`
262 the timestamps may not match, as the system clock will be polled
263 twice (once for each command). The only way to ensure that both
264 author and committer identity information has the same timestamp
265 is to omit `author` (thus copying from `committer`) or to use a
266 date format other than `now`.
270 fast-import accepts several commands to update the current repository
271 and control the current import process. More detailed discussion
272 (with examples) of each command follows later.
275 Creates a new branch or updates an existing branch by
276 creating a new commit and updating the branch to point at
277 the newly created commit.
280 Creates an annotated tag object from an existing commit or
281 branch. Lightweight tags are not supported by this command,
282 as they are not recommended for recording meaningful points
286 Reset an existing branch (or a new branch) to a specific
287 revision. This command must be used to change a branch to
288 a specific revision without making a commit on it.
291 Convert raw file data into a blob, for future use in a
292 `commit` command. This command is optional and is not
293 needed to perform an import.
296 Forces fast-import to close the current packfile, generate its
297 unique SHA-1 checksum and index, and start a new packfile.
298 This command is optional and is not needed to perform
302 Causes fast-import to echo the entire line to its own
303 standard output. This command is optional and is not needed
304 to perform an import.
307 Require that fast-import supports the specified feature, or
308 abort if it does not.
312 Create or update a branch with a new commit, recording one logical
313 change to the project.
318 ('author' SP <name> SP LT <email> GT SP <when> LF)?
319 'committer' SP <name> SP LT <email> GT SP <when> LF
321 ('from' SP <committish> LF)?
322 ('merge' SP <committish> LF)?
323 (filemodify | filedelete | filecopy | filerename | filedeleteall | notemodify)*
327 where `<ref>` is the name of the branch to make the commit on.
328 Typically branch names are prefixed with `refs/heads/` in
329 Git, so importing the CVS branch symbol `RELENG-1_0` would use
330 `refs/heads/RELENG-1_0` for the value of `<ref>`. The value of
331 `<ref>` must be a valid refname in Git. As `LF` is not valid in
332 a Git refname, no quoting or escaping syntax is supported here.
334 A `mark` command may optionally appear, requesting fast-import to save a
335 reference to the newly created commit for future use by the frontend
336 (see below for format). It is very common for frontends to mark
337 every commit they create, thereby allowing future branch creation
338 from any imported commit.
340 The `data` command following `committer` must supply the commit
341 message (see below for `data` command syntax). To import an empty
342 commit message use a 0 length data. Commit messages are free-form
343 and are not interpreted by Git. Currently they must be encoded in
344 UTF-8, as fast-import does not permit other encodings to be specified.
346 Zero or more `filemodify`, `filedelete`, `filecopy`, `filerename`,
347 `filedeleteall` and `notemodify` commands
348 may be included to update the contents of the branch prior to
349 creating the commit. These commands may be supplied in any order.
350 However it is recommended that a `filedeleteall` command precede
351 all `filemodify`, `filecopy`, `filerename` and `notemodify` commands in
352 the same commit, as `filedeleteall` wipes the branch clean (see below).
354 The `LF` after the command is optional (it used to be required).
358 An `author` command may optionally appear, if the author information
359 might differ from the committer information. If `author` is omitted
360 then fast-import will automatically use the committer's information for
361 the author portion of the commit. See below for a description of
362 the fields in `author`, as they are identical to `committer`.
366 The `committer` command indicates who made this commit, and when
369 Here `<name>` is the person's display name (for example
370 ``Com M Itter'') and `<email>` is the person's email address
371 (``cm@example.com''). `LT` and `GT` are the literal less-than (\x3c)
372 and greater-than (\x3e) symbols. These are required to delimit
373 the email address from the other fields in the line. Note that
374 `<name>` is free-form and may contain any sequence of bytes, except
375 `LT` and `LF`. It is typically UTF-8 encoded.
377 The time of the change is specified by `<when>` using the date format
378 that was selected by the \--date-format=<fmt> command line option.
379 See ``Date Formats'' above for the set of supported formats, and
384 The `from` command is used to specify the commit to initialize
385 this branch from. This revision will be the first ancestor of the
388 Omitting the `from` command in the first commit of a new branch
389 will cause fast-import to create that commit with no ancestor. This
390 tends to be desired only for the initial commit of a project.
391 If the frontend creates all files from scratch when making a new
392 branch, a `merge` command may be used instead of `from` to start
393 the commit with an empty tree.
394 Omitting the `from` command on existing branches is usually desired,
395 as the current commit on that branch is automatically assumed to
396 be the first ancestor of the new commit.
398 As `LF` is not valid in a Git refname or SHA-1 expression, no
399 quoting or escaping syntax is supported within `<committish>`.
401 Here `<committish>` is any of the following:
403 * The name of an existing branch already in fast-import's internal branch
404 table. If fast-import doesn't know the name, its treated as a SHA-1
407 * A mark reference, `:<idnum>`, where `<idnum>` is the mark number.
409 The reason fast-import uses `:` to denote a mark reference is this character
410 is not legal in a Git branch name. The leading `:` makes it easy
411 to distinguish between the mark 42 (`:42`) and the branch 42 (`42`
412 or `refs/heads/42`), or an abbreviated SHA-1 which happened to
413 consist only of base-10 digits.
415 Marks must be declared (via `mark`) before they can be used.
417 * A complete 40 byte or abbreviated commit SHA-1 in hex.
419 * Any valid Git SHA-1 expression that resolves to a commit. See
420 ``SPECIFYING REVISIONS'' in linkgit:git-rev-parse[1] for details.
422 The special case of restarting an incremental import from the
423 current branch value should be written as:
425 from refs/heads/branch^0
427 The `{caret}0` suffix is necessary as fast-import does not permit a branch to
428 start from itself, and the branch is created in memory before the
429 `from` command is even read from the input. Adding `{caret}0` will force
430 fast-import to resolve the commit through Git's revision parsing library,
431 rather than its internal branch table, thereby loading in the
432 existing value of the branch.
436 Includes one additional ancestor commit. If the `from` command is
437 omitted when creating a new branch, the first `merge` commit will be
438 the first ancestor of the current commit, and the branch will start
439 out with no files. An unlimited number of `merge` commands per
440 commit are permitted by fast-import, thereby establishing an n-way merge.
441 However Git's other tools never create commits with more than 15
442 additional ancestors (forming a 16-way merge). For this reason
443 it is suggested that frontends do not use more than 15 `merge`
444 commands per commit; 16, if starting a new, empty branch.
446 Here `<committish>` is any of the commit specification expressions
447 also accepted by `from` (see above).
451 Included in a `commit` command to add a new file or change the
452 content of an existing file. This command has two different means
453 of specifying the content of the file.
455 External data format::
456 The data content for the file was already supplied by a prior
457 `blob` command. The frontend just needs to connect it.
460 'M' SP <mode> SP <dataref> SP <path> LF
463 Here `<dataref>` can be either a mark reference (`:<idnum>`)
464 set by a prior `blob` command, or a full 40-byte SHA-1 of an
465 existing Git blob object.
468 The data content for the file has not been supplied yet.
469 The frontend wants to supply it as part of this modify
473 'M' SP <mode> SP 'inline' SP <path> LF
477 See below for a detailed description of the `data` command.
479 In both formats `<mode>` is the type of file entry, specified
480 in octal. Git only supports the following modes:
482 * `100644` or `644`: A normal (not-executable) file. The majority
483 of files in most projects use this mode. If in doubt, this is
485 * `100755` or `755`: A normal, but executable, file.
486 * `120000`: A symlink, the content of the file will be the link target.
487 * `160000`: A gitlink, SHA-1 of the object refers to a commit in
488 another repository. Git links can only be specified by SHA or through
489 a commit mark. They are used to implement submodules.
491 In both formats `<path>` is the complete path of the file to be added
492 (if not already existing) or modified (if already existing).
494 A `<path>` string must use UNIX-style directory separators (forward
495 slash `/`), may contain any byte other than `LF`, and must not
496 start with double quote (`"`).
498 If an `LF` or double quote must be encoded into `<path>` shell-style
499 quoting should be used, e.g. `"path/with\n and \" in it"`.
501 The value of `<path>` must be in canonical form. That is it must not:
503 * contain an empty directory component (e.g. `foo//bar` is invalid),
504 * end with a directory separator (e.g. `foo/` is invalid),
505 * start with a directory separator (e.g. `/foo` is invalid),
506 * contain the special component `.` or `..` (e.g. `foo/./bar` and
507 `foo/../bar` are invalid).
509 It is recommended that `<path>` always be encoded using UTF-8.
513 Included in a `commit` command to remove a file or recursively
514 delete an entire directory from the branch. If the file or directory
515 removal makes its parent directory empty, the parent directory will
516 be automatically removed too. This cascades up the tree until the
517 first non-empty directory or the root is reached.
523 here `<path>` is the complete path of the file or subdirectory to
524 be removed from the branch.
525 See `filemodify` above for a detailed description of `<path>`.
529 Recursively copies an existing file or subdirectory to a different
530 location within the branch. The existing file or directory must
531 exist. If the destination exists it will be completely replaced
532 by the content copied from the source.
535 'C' SP <path> SP <path> LF
538 here the first `<path>` is the source location and the second
539 `<path>` is the destination. See `filemodify` above for a detailed
540 description of what `<path>` may look like. To use a source path
541 that contains SP the path must be quoted.
543 A `filecopy` command takes effect immediately. Once the source
544 location has been copied to the destination any future commands
545 applied to the source location will not impact the destination of
550 Renames an existing file or subdirectory to a different location
551 within the branch. The existing file or directory must exist. If
552 the destination exists it will be replaced by the source directory.
555 'R' SP <path> SP <path> LF
558 here the first `<path>` is the source location and the second
559 `<path>` is the destination. See `filemodify` above for a detailed
560 description of what `<path>` may look like. To use a source path
561 that contains SP the path must be quoted.
563 A `filerename` command takes effect immediately. Once the source
564 location has been renamed to the destination any future commands
565 applied to the source location will create new files there and not
566 impact the destination of the rename.
568 Note that a `filerename` is the same as a `filecopy` followed by a
569 `filedelete` of the source location. There is a slight performance
570 advantage to using `filerename`, but the advantage is so small
571 that it is never worth trying to convert a delete/add pair in
572 source material into a rename for fast-import. This `filerename`
573 command is provided just to simplify frontends that already have
574 rename information and don't want bother with decomposing it into a
575 `filecopy` followed by a `filedelete`.
579 Included in a `commit` command to remove all files (and also all
580 directories) from the branch. This command resets the internal
581 branch structure to have no files in it, allowing the frontend
582 to subsequently add all interesting files from scratch.
588 This command is extremely useful if the frontend does not know
589 (or does not care to know) what files are currently on the branch,
590 and therefore cannot generate the proper `filedelete` commands to
593 Issuing a `filedeleteall` followed by the needed `filemodify`
594 commands to set the correct content will produce the same results
595 as sending only the needed `filemodify` and `filedelete` commands.
596 The `filedeleteall` approach may however require fast-import to use slightly
597 more memory per active branch (less than 1 MiB for even most large
598 projects); so frontends that can easily obtain only the affected
599 paths for a commit are encouraged to do so.
603 Included in a `commit` command to add a new note (annotating a given
604 commit) or change the content of an existing note. This command has
605 two different means of specifying the content of the note.
607 External data format::
608 The data content for the note was already supplied by a prior
609 `blob` command. The frontend just needs to connect it to the
610 commit that is to be annotated.
613 'N' SP <dataref> SP <committish> LF
616 Here `<dataref>` can be either a mark reference (`:<idnum>`)
617 set by a prior `blob` command, or a full 40-byte SHA-1 of an
618 existing Git blob object.
621 The data content for the note has not been supplied yet.
622 The frontend wants to supply it as part of this modify
626 'N' SP 'inline' SP <committish> LF
630 See below for a detailed description of the `data` command.
632 In both formats `<committish>` is any of the commit specification
633 expressions also accepted by `from` (see above).
637 Arranges for fast-import to save a reference to the current object, allowing
638 the frontend to recall this object at a future point in time, without
639 knowing its SHA-1. Here the current object is the object creation
640 command the `mark` command appears within. This can be `commit`,
641 `tag`, and `blob`, but `commit` is the most common usage.
644 'mark' SP ':' <idnum> LF
647 where `<idnum>` is the number assigned by the frontend to this mark.
648 The value of `<idnum>` is expressed as an ASCII decimal integer.
649 The value 0 is reserved and cannot be used as
650 a mark. Only values greater than or equal to 1 may be used as marks.
652 New marks are created automatically. Existing marks can be moved
653 to another object simply by reusing the same `<idnum>` in another
658 Creates an annotated tag referring to a specific commit. To create
659 lightweight (non-annotated) tags see the `reset` command below.
663 'from' SP <committish> LF
664 'tagger' SP <name> SP LT <email> GT SP <when> LF
668 where `<name>` is the name of the tag to create.
670 Tag names are automatically prefixed with `refs/tags/` when stored
671 in Git, so importing the CVS branch symbol `RELENG-1_0-FINAL` would
672 use just `RELENG-1_0-FINAL` for `<name>`, and fast-import will write the
673 corresponding ref as `refs/tags/RELENG-1_0-FINAL`.
675 The value of `<name>` must be a valid refname in Git and therefore
676 may contain forward slashes. As `LF` is not valid in a Git refname,
677 no quoting or escaping syntax is supported here.
679 The `from` command is the same as in the `commit` command; see
682 The `tagger` command uses the same format as `committer` within
683 `commit`; again see above for details.
685 The `data` command following `tagger` must supply the annotated tag
686 message (see below for `data` command syntax). To import an empty
687 tag message use a 0 length data. Tag messages are free-form and are
688 not interpreted by Git. Currently they must be encoded in UTF-8,
689 as fast-import does not permit other encodings to be specified.
691 Signing annotated tags during import from within fast-import is not
692 supported. Trying to include your own PGP/GPG signature is not
693 recommended, as the frontend does not (easily) have access to the
694 complete set of bytes which normally goes into such a signature.
695 If signing is required, create lightweight tags from within fast-import with
696 `reset`, then create the annotated versions of those tags offline
697 with the standard 'git-tag' process.
701 Creates (or recreates) the named branch, optionally starting from
702 a specific revision. The reset command allows a frontend to issue
703 a new `from` command for an existing branch, or to create a new
704 branch from an existing commit without creating a new commit.
708 ('from' SP <committish> LF)?
712 For a detailed description of `<ref>` and `<committish>` see above
713 under `commit` and `from`.
715 The `LF` after the command is optional (it used to be required).
717 The `reset` command can also be used to create lightweight
718 (non-annotated) tags. For example:
725 would create the lightweight tag `refs/tags/938` referring to
726 whatever commit mark `:938` references.
730 Requests writing one file revision to the packfile. The revision
731 is not connected to any commit; this connection must be formed in
732 a subsequent `commit` command by referencing the blob through an
741 The mark command is optional here as some frontends have chosen
742 to generate the Git SHA-1 for the blob on their own, and feed that
743 directly to `commit`. This is typically more work than its worth
744 however, as marks are inexpensive to store and easy to use.
748 Supplies raw data (for use as blob/file content, commit messages, or
749 annotated tag messages) to fast-import. Data can be supplied using an exact
750 byte count or delimited with a terminating line. Real frontends
751 intended for production-quality conversions should always use the
752 exact byte count format, as it is more robust and performs better.
753 The delimited format is intended primarily for testing fast-import.
755 Comment lines appearing within the `<raw>` part of `data` commands
756 are always taken to be part of the body of the data and are therefore
757 never ignored by fast-import. This makes it safe to import any
758 file/message content whose lines might start with `#`.
760 Exact byte count format::
761 The frontend must specify the number of bytes of data.
768 where `<count>` is the exact number of bytes appearing within
769 `<raw>`. The value of `<count>` is expressed as an ASCII decimal
770 integer. The `LF` on either side of `<raw>` is not
771 included in `<count>` and will not be included in the imported data.
773 The `LF` after `<raw>` is optional (it used to be required) but
774 recommended. Always including it makes debugging a fast-import
775 stream easier as the next command always starts in column 0
776 of the next line, even if `<raw>` did not end with an `LF`.
779 A delimiter string is used to mark the end of the data.
780 fast-import will compute the length by searching for the delimiter.
781 This format is primarily useful for testing and is not
782 recommended for real data.
785 'data' SP '<<' <delim> LF
791 where `<delim>` is the chosen delimiter string. The string `<delim>`
792 must not appear on a line by itself within `<raw>`, as otherwise
793 fast-import will think the data ends earlier than it really does. The `LF`
794 immediately trailing `<raw>` is part of `<raw>`. This is one of
795 the limitations of the delimited format, it is impossible to supply
796 a data chunk which does not have an LF as its last byte.
798 The `LF` after `<delim> LF` is optional (it used to be required).
802 Forces fast-import to close the current packfile, start a new one, and to
803 save out all current branch refs, tags and marks.
810 Note that fast-import automatically switches packfiles when the current
811 packfile reaches \--max-pack-size, or 4 GiB, whichever limit is
812 smaller. During an automatic packfile switch fast-import does not update
813 the branch refs, tags or marks.
815 As a `checkpoint` can require a significant amount of CPU time and
816 disk IO (to compute the overall pack SHA-1 checksum, generate the
817 corresponding index file, and update the refs) it can easily take
818 several minutes for a single `checkpoint` command to complete.
820 Frontends may choose to issue checkpoints during extremely large
821 and long running imports, or when they need to allow another Git
822 process access to a branch. However given that a 30 GiB Subversion
823 repository can be loaded into Git through fast-import in about 3 hours,
824 explicit checkpointing may not be necessary.
826 The `LF` after the command is optional (it used to be required).
830 Causes fast-import to print the entire `progress` line unmodified to
831 its standard output channel (file descriptor 1) when the command is
832 processed from the input stream. The command otherwise has no impact
833 on the current import, or on any of fast-import's internal state.
836 'progress' SP <any> LF
840 The `<any>` part of the command may contain any sequence of bytes
841 that does not contain `LF`. The `LF` after the command is optional.
842 Callers may wish to process the output through a tool such as sed to
843 remove the leading part of the line, for example:
846 frontend | git fast-import | sed 's/^progress //'
849 Placing a `progress` command immediately after a `checkpoint` will
850 inform the reader when the `checkpoint` has been completed and it
851 can safely access the refs that fast-import updated.
855 Require that fast-import supports the specified feature, or abort if
859 'feature' SP <feature> LF
862 The <feature> part of the command may be any string matching
863 ^[a-zA-Z][a-zA-Z-]*$ and should be understood by fast-import.
865 Feature work identical as their option counterparts.
867 The following features are currently supported:
876 If fast-import is supplied invalid input it will terminate with a
877 non-zero exit status and create a crash report in the top level of
878 the Git repository it was importing into. Crash reports contain
879 a snapshot of the internal fast-import state as well as the most
880 recent commands that lead up to the crash.
882 All recent commands (including stream comments, file changes and
883 progress commands) are shown in the command history within the crash
884 report, but raw file data and commit messages are excluded from the
885 crash report. This exclusion saves space within the report file
886 and reduces the amount of buffering that fast-import must perform
889 After writing a crash report fast-import will close the current
890 packfile and export the marks table. This allows the frontend
891 developer to inspect the repository state and resume the import from
892 the point where it crashed. The modified branches and tags are not
893 updated during a crash, as the import did not complete successfully.
894 Branch and tag information can be found in the crash report and
895 must be applied manually if the update is needed.
900 $ cat >in <<END_OF_INPUT
901 # my very first test commit
902 commit refs/heads/master
903 committer Shawn O. Pearce <spearce> 19283 -0400
904 # who is that guy anyway?
908 M 644 inline .gitignore
915 $ git fast-import <in
916 fatal: Corrupt mode: M 777 inline bob
917 fast-import: dumping crash report to .git/fast_import_crash_8434
919 $ cat .git/fast_import_crash_8434
920 fast-import crash report:
921 fast-import process: 8434
922 parent process : 1391
923 at Sat Sep 1 00:58:12 2007
925 fatal: Corrupt mode: M 777 inline bob
927 Most Recent Commands Before Crash
928 ---------------------------------
929 # my very first test commit
930 commit refs/heads/master
931 committer Shawn O. Pearce <spearce> 19283 -0400
932 # who is that guy anyway?
934 M 644 inline .gitignore
940 active_branches = 1 cur, 5 max
943 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
944 1) 0 refs/heads/master
949 status : active loaded dirty
950 tip commit : 0000000000000000000000000000000000000000
951 old tree : 0000000000000000000000000000000000000000
952 cur tree : 0000000000000000000000000000000000000000
963 The following tips and tricks have been collected from various
964 users of fast-import, and are offered here as suggestions.
966 Use One Mark Per Commit
967 ~~~~~~~~~~~~~~~~~~~~~~~
968 When doing a repository conversion, use a unique mark per commit
969 (`mark :<n>`) and supply the \--export-marks option on the command
970 line. fast-import will dump a file which lists every mark and the Git
971 object SHA-1 that corresponds to it. If the frontend can tie
972 the marks back to the source repository, it is easy to verify the
973 accuracy and completeness of the import by comparing each Git
974 commit to the corresponding source revision.
976 Coming from a system such as Perforce or Subversion this should be
977 quite simple, as the fast-import mark can also be the Perforce changeset
978 number or the Subversion revision number.
980 Freely Skip Around Branches
981 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
982 Don't bother trying to optimize the frontend to stick to one branch
983 at a time during an import. Although doing so might be slightly
984 faster for fast-import, it tends to increase the complexity of the frontend
987 The branch LRU builtin to fast-import tends to behave very well, and the
988 cost of activating an inactive branch is so low that bouncing around
989 between branches has virtually no impact on import performance.
993 When importing a renamed file or directory, simply delete the old
994 name(s) and modify the new name(s) during the corresponding commit.
995 Git performs rename detection after-the-fact, rather than explicitly
998 Use Tag Fixup Branches
999 ~~~~~~~~~~~~~~~~~~~~~~
1000 Some other SCM systems let the user create a tag from multiple
1001 files which are not from the same commit/changeset. Or to create
1002 tags which are a subset of the files available in the repository.
1004 Importing these tags as-is in Git is impossible without making at
1005 least one commit which ``fixes up'' the files to match the content
1006 of the tag. Use fast-import's `reset` command to reset a dummy branch
1007 outside of your normal branch space to the base commit for the tag,
1008 then commit one or more file fixup commits, and finally tag the
1011 For example since all normal branches are stored under `refs/heads/`
1012 name the tag fixup branch `TAG_FIXUP`. This way it is impossible for
1013 the fixup branch used by the importer to have namespace conflicts
1014 with real branches imported from the source (the name `TAG_FIXUP`
1015 is not `refs/heads/TAG_FIXUP`).
1017 When committing fixups, consider using `merge` to connect the
1018 commit(s) which are supplying file revisions to the fixup branch.
1019 Doing so will allow tools such as 'git-blame' to track
1020 through the real commit history and properly annotate the source
1023 After fast-import terminates the frontend will need to do `rm .git/TAG_FIXUP`
1024 to remove the dummy branch.
1026 Import Now, Repack Later
1027 ~~~~~~~~~~~~~~~~~~~~~~~~
1028 As soon as fast-import completes the Git repository is completely valid
1029 and ready for use. Typically this takes only a very short time,
1030 even for considerably large projects (100,000+ commits).
1032 However repacking the repository is necessary to improve data
1033 locality and access performance. It can also take hours on extremely
1034 large projects (especially if -f and a large \--window parameter is
1035 used). Since repacking is safe to run alongside readers and writers,
1036 run the repack in the background and let it finish when it finishes.
1037 There is no reason to wait to explore your new Git project!
1039 If you choose to wait for the repack, don't try to run benchmarks
1040 or performance tests until repacking is completed. fast-import outputs
1041 suboptimal packfiles that are simply never seen in real use
1044 Repacking Historical Data
1045 ~~~~~~~~~~~~~~~~~~~~~~~~~
1046 If you are repacking very old imported data (e.g. older than the
1047 last year), consider expending some extra CPU time and supplying
1048 \--window=50 (or higher) when you run 'git-repack'.
1049 This will take longer, but will also produce a smaller packfile.
1050 You only need to expend the effort once, and everyone using your
1051 project will benefit from the smaller repository.
1053 Include Some Progress Messages
1054 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1055 Every once in a while have your frontend emit a `progress` message
1056 to fast-import. The contents of the messages are entirely free-form,
1057 so one suggestion would be to output the current month and year
1058 each time the current commit date moves into the next month.
1059 Your users will feel better knowing how much of the data stream
1063 Packfile Optimization
1064 ---------------------
1065 When packing a blob fast-import always attempts to deltify against the last
1066 blob written. Unless specifically arranged for by the frontend,
1067 this will probably not be a prior version of the same file, so the
1068 generated delta will not be the smallest possible. The resulting
1069 packfile will be compressed, but will not be optimal.
1071 Frontends which have efficient access to all revisions of a
1072 single file (for example reading an RCS/CVS ,v file) can choose
1073 to supply all revisions of that file as a sequence of consecutive
1074 `blob` commands. This allows fast-import to deltify the different file
1075 revisions against each other, saving space in the final packfile.
1076 Marks can be used to later identify individual file revisions during
1077 a sequence of `commit` commands.
1079 The packfile(s) created by fast-import do not encourage good disk access
1080 patterns. This is caused by fast-import writing the data in the order
1081 it is received on standard input, while Git typically organizes
1082 data within packfiles to make the most recent (current tip) data
1083 appear before historical data. Git also clusters commits together,
1084 speeding up revision traversal through better cache locality.
1086 For this reason it is strongly recommended that users repack the
1087 repository with `git repack -a -d` after fast-import completes, allowing
1088 Git to reorganize the packfiles for faster data access. If blob
1089 deltas are suboptimal (see above) then also adding the `-f` option
1090 to force recomputation of all deltas can significantly reduce the
1091 final packfile size (30-50% smaller can be quite typical).
1096 There are a number of factors which affect how much memory fast-import
1097 requires to perform an import. Like critical sections of core
1098 Git, fast-import uses its own memory allocators to amortize any overheads
1099 associated with malloc. In practice fast-import tends to amortize any
1100 malloc overheads to 0, due to its use of large block allocations.
1104 fast-import maintains an in-memory structure for every object written in
1105 this execution. On a 32 bit system the structure is 32 bytes,
1106 on a 64 bit system the structure is 40 bytes (due to the larger
1107 pointer sizes). Objects in the table are not deallocated until
1108 fast-import terminates. Importing 2 million objects on a 32 bit system
1109 will require approximately 64 MiB of memory.
1111 The object table is actually a hashtable keyed on the object name
1112 (the unique SHA-1). This storage configuration allows fast-import to reuse
1113 an existing or already written object and avoid writing duplicates
1114 to the output packfile. Duplicate blobs are surprisingly common
1115 in an import, typically due to branch merges in the source.
1119 Marks are stored in a sparse array, using 1 pointer (4 bytes or 8
1120 bytes, depending on pointer size) per mark. Although the array
1121 is sparse, frontends are still strongly encouraged to use marks
1122 between 1 and n, where n is the total number of marks required for
1127 Branches are classified as active and inactive. The memory usage
1128 of the two classes is significantly different.
1130 Inactive branches are stored in a structure which uses 96 or 120
1131 bytes (32 bit or 64 bit systems, respectively), plus the length of
1132 the branch name (typically under 200 bytes), per branch. fast-import will
1133 easily handle as many as 10,000 inactive branches in under 2 MiB
1136 Active branches have the same overhead as inactive branches, but
1137 also contain copies of every tree that has been recently modified on
1138 that branch. If subtree `include` has not been modified since the
1139 branch became active, its contents will not be loaded into memory,
1140 but if subtree `src` has been modified by a commit since the branch
1141 became active, then its contents will be loaded in memory.
1143 As active branches store metadata about the files contained on that
1144 branch, their in-memory storage size can grow to a considerable size
1147 fast-import automatically moves active branches to inactive status based on
1148 a simple least-recently-used algorithm. The LRU chain is updated on
1149 each `commit` command. The maximum number of active branches can be
1150 increased or decreased on the command line with \--active-branches=.
1154 Trees (aka directories) use just 12 bytes of memory on top of the
1155 memory required for their entries (see ``per active file'' below).
1156 The cost of a tree is virtually 0, as its overhead amortizes out
1157 over the individual file entries.
1159 per active file entry
1160 ~~~~~~~~~~~~~~~~~~~~~
1161 Files (and pointers to subtrees) within active trees require 52 or 64
1162 bytes (32/64 bit platforms) per entry. To conserve space, file and
1163 tree names are pooled in a common string table, allowing the filename
1164 ``Makefile'' to use just 16 bytes (after including the string header
1165 overhead) no matter how many times it occurs within the project.
1167 The active branch LRU, when coupled with the filename string pool
1168 and lazy loading of subtrees, allows fast-import to efficiently import
1169 projects with 2,000+ branches and 45,114+ files in a very limited
1170 memory footprint (less than 2.7 MiB per active branch).
1175 Written by Shawn O. Pearce <spearce@spearce.org>.
1179 Documentation by Shawn O. Pearce <spearce@spearce.org>.
1183 Part of the linkgit:git[1] suite
projects / git / blob