MAN page from RedHat Other patch-2.1-4.i386.rpm

PATCH

Section: User Commands (1)
Updated: LOCAL
Index

NAME

patch - apply a diff file to an original

SYNOPSIS

patch[options] [origfile [patchfile]] [+ [options] [origfile]]...

but usually just

patch<patchfile

DESCRIPTION

Patchwill take a patch file containing any of the four forms of differencelisting produced by thediffprogram and apply those differences to an original file, producing a patchedversion.By default, the patched version is put in place of the original, withthe original file backed up to the same name with theextension ``.orig'' (``~'' on systems that do notsupport long file names), or as specified by the-b (--suffix),-B (--prefix),or-V (--version-control)options.The extension used for making backup files may also be specified in theSIMPLE_BACKUP_SUFFIXenvironment variable, which is overridden by the above options.

If the backup file already exists,patchcreates a new backup file name by changing the first lowercase letterin the last component of the file's name into uppercase. If there areno more lowercase letters in the name, it removes the first characterfrom the name. It repeats this process until it comes up with abackup file that does not already exist.

You may also specify where you want the output to go with a-o (--output)option; if that file already exists, it is backed up first.

Ifpatchfileis omitted, or is a hyphen, the patch will be read from standard input.

Upon startup, patch will attempt to determine the type of the diff listing,unless over-ruled by a-c (--context),-e (--ed),-n (--normal),or-u (--unified)option.Context diffs (old-style, new-style, and unified) andnormal diffs are applied by thepatchprogram itself, whileeddiffs are simply fed to theededitor via a pipe.

Patchwill try to skip any leading garbage, apply the diff,and then skip any trailing garbage.Thus you could feed an article or message containing adiff listing topatch,and it should work.If the entire diff is indented by a consistent amount,this will be taken into account.

With context diffs, and to a lesser extent with normal diffs,patchcan detect when the line numbers mentioned in the patch are incorrect,and will attempt to find the correct place to apply each hunk of the patch.As a first guess, it takes the line number mentioned for the hunk, plus orminus any offset used in applying the previous hunk.If that is not the correct place,patchwill scan both forwards and backwards for a set of lines matching the contextgiven in the hunk.Firstpatchlooks for a place where all lines of the context match.If no such place is found, and it's a context diff, and the maximum fuzz factoris set to 1 or more, then another scan takes place ignoring the first and lastline of context.If that fails, and the maximum fuzz factor is set to 2 or more,the first two and last two lines of context are ignored,and another scan is made.(The default maximum fuzz factor is 2.)Ifpatchcannot find a place to install that hunk of the patch, it will put thehunk out to a reject file, which normally is the name of the output fileplus ``.rej'' (``#'' on systems that do not supportlong file names).(Note that the rejected hunk will come out in context diff form whether theinput patch was a context diff or a normal diff.If the input was a normal diff, many of the contexts will simply be null.)The line numbers on the hunks in the reject file may be different thanin the patch file: they reflect the approximate location patch thinks thefailed hunks belong in the new file rather than the old one.

As each hunk is completed, you will be told whether the hunk succeeded orfailed, and which line (in the new file)patchthought the hunk should go on.If this is different from the line number specified in the diff you willbe told the offset.A single large offset MAY be an indication that a hunk was installed in thewrong place.You will also be told if a fuzz factor was used to make the match, in whichcase you should also be slightly suspicious.

If no original file is specified on the command line,patchwill try to figure out from the leading garbage what the name of the fileto edit is.In the header of a context diff, the file name is found from lines beginningwith ``***'' or ``---'', with the shortest name of an existingfile winning.Only context diffs have lines like that, but if there is an ``Index:''line in the leading garbage,patchwill try to use the file name from that line.The context diff header takes precedence over an Index line.If no file name can be intuited from the leading garbage, you will be askedfor the name of the file to patch.

If the original file cannot be found or is read-only, but a suitableSCCS or RCS file is handy,patchwill attempt to get or check out the file.

Additionally, if the leading garbage contains a ``Prereq: '' line,patchwill take the first word from the prerequisites line (normally a versionnumber) and check the input file to see if that word can be found.If not,patchwill ask for confirmation before proceeding.

The upshot of all this is that you should be able to say, while in a newsinterface, the following:

| patch -d /usr/src/local/blurfl

and patch a file in the blurfl directory directly from the article containingthe patch.

If the patch file contains more than one patch,patchwill try to apply each of them as if they came from separate patch files.This means, among other things, that it is assumed that the name of the fileto patch must be determined for each diff listing,and that the garbage before each diff listing willbe examined for interesting things such as file names and revision level, asmentioned previously.You can give options (and another original file name) for the second andsubsequent patches by separating the corresponding argument listsby a `+'.(The argument list for a second or subsequent patch may not specify a newpatch file, however.)

Patchrecognizes the following options:

-b suff, --suffix=suff

causessuffto be interpreted as the backup extension, to beused in place of ``.orig'' or ``~''.

-B pref, --prefix=pref

causesprefto be interpreted as a prefix to the backup filename. If this argument is specified, any argument from-bwill be ignored.

-c, --context

forcespatchto interpret the patch file as a context diff.

-d dir, --directory=dir

causespatchto interpretdiras a directory, and cd to it before doinganything else.

-D sym, --ifdef=sym

causespatchto use the "#ifdef...#endif" construct to mark changes.symwill be used as the differentiating symbol.

-e, --ed

forcespatchto interpret the patch file as anedscript.

-E, --remove-empty-files

causespatchto remove output files that are empty after the patches have been applied.

-f, --force

forcespatchto assume that the user knows exactly what he or she is doing, and to notask any questions. It assumes the following: skip patches for which afile to patch can't be found; patch files even though they have thewrong version for the ``Prereq:'' line in the patch; and assume thatpatches are not reversed even if they look like they are.This option does not suppress commentary; use-sfor that.

-t, --batch

similar to-f,in that it suppresses questions, but makes some different assumptions:skip patches for which a file to patch can't be found (the same as -f);skip patches for which the file has the wrong version for the ``Prereq:'' linein the patch; and assume that patches are reversed if they look likethey are.

-F number, --fuzz=number

sets the maximum fuzz factor.This option only applies to context diffs, and causespatchto ignore up to that many lines in looking for places to install a hunk.Note that a larger fuzz factor increases the odds of a faulty patch.The default fuzz factor is 2, and it may not be set to more thanthe number of lines of context in the context diff, ordinarily 3.

-l, --ignore-whitespace

causes the pattern matching to be done loosely, in case the tabs andspaces have been munged in your input file.Any sequence of whitespace in the pattern line will match any sequencein the input file.Normal characters must still match exactly.Each line of the context must still match a line in the input file.

-n, --normal

forcespatchto interpret the patch file as a normal diff.

-N, --forward

causespatchto ignore patches that it thinks are reversed or already applied.See also-R .

-o file, --output=file

causesfileto be interpreted as the output file name.

-p[number], --strip[=number]

sets the pathname strip count,which controls how pathnames found in the patch file are treated, in casethe you keep your files in a different directory than the person who sentout the patch.The strip count specifies how many slashes are to be stripped fromthe front of the pathname.(Any intervening directory names also go away.)For example, supposing the file name in the patch file was

/u/howard/src/blurfl/blurfl.c

setting-por-p0gives the entire pathname unmodified,-p1gives

u/howard/src/blurfl/blurfl.c

without the leading slash,-p4gives

blurfl/blurfl.c

and not specifying-pat all just gives you "blurfl.c", unless all of the directories in theleading path (u/howard/src/blurfl) exist and that path is relative,in which case you get the entire pathname unmodified.Whatever you end up with is looked for either in the current directory,or the directory specified by the-doption.

-r file, --reject-file=file

causesfileto be interpreted as the reject file name.

-R, --reverse

tellspatchthat this patch was created with the old and new files swapped.(Yes, I'm afraid that does happen occasionally, human nature being what itis.)Patchwill attempt to swap each hunk around before applying it.Rejects will come out in the swapped format.The-Roption will not work witheddiff scripts because there is too littleinformation to reconstruct the reverse operation.

If the first hunk of a patch fails,patchwill reverse the hunk to see if it can be applied that way.If it can, you will be asked if you want to have the-Roption set.If it can't, the patch will continue to be applied normally.(Note: this method cannot detect a reversed patch if it is a normal diffand if the first command is an append (i.e. it should have been a delete)since appends always succeed, due to the fact that a null context will matchanywhere.Luckily, most patches add or change lines rather than delete them, so mostreversed normal diffs will begin with a delete, which will fail, triggeringthe heuristic.)

-s, --silent, --quiet

makespatchdo its work silently, unless an error occurs.

-S, --skip

causespatchto ignore this patch from the patch file, but continue on lookingfor the next patch in the file.Thus

patch -S + -S + <patchfile

will ignore the first and second of three patches.

-u, --unified

forcespatchto interpret the patch file as a unified context diff (a unidiff).

-v, --version

causespatchto print out its revision header and patch level.

-V method, --version--control=method

causesmethodto be interpreted as a method for creatingbackup file names. The type of backups made can also be given in theVERSION_CONTROLenvironment variable, which is overridden by this option.The-Boption overrides this option, causing the prefix to always be used formaking backup file names.The value of theVERSION_CONTROLenvironment variable and the argument to the-Voption are like the GNUEmacs `version-control' variable; they also recognize synonyms thatare more descriptive. The valid values are (unique abbreviations areaccepted):

`t' or `numbered': Always make numbered backups.
`nil' or `existing': Make numbered backups of files that alreadyhave them, simple backups of the others.This is the default.
`never' or `simple': Always make simple backups.

-x number, --debug=number

sets internal debugging flags, and is of interest only topatchpatchers.

AUTHOR

Larry Wall <lwallAATTnetlabs.com>
with many other contributors.

ENVIRONMENT

TMPDIR: Directory to put temporary files in; default is /tmp.
SIMPLE_BACKUP_SUFFIX: Extension to use for backup file names instead of ``.orig'' or``~''.
VERSION_CONTROL: Selects when numbered backup files are made.

FILES

$TMPDIR/patch*

NOTES FOR PATCH SENDERS

There are several things you should bear in mind if you are going tobe sending out patches.First, you can save people a lot of grief by keeping a patchlevel.h filewhich is patched to increment the patch level as the first diff in thepatch file you send out.If you put a Prereq: line in with the patch, it won't let them applypatches out of order without some warning.Second, make sure you've specified the file names right, either in acontext diff header, or with an Index: line.If you are patching something in a subdirectory, be sure to tell the patchuser to specify a -poption as needed.Third, you can create a file by sending out a diff that compares anull file to the file you want to create.This will only work if the file you want to create doesn't exist already inthe target directory.Fourth, take care not to send out reversed patches, since it makes people wonderwhether they already applied the patch.Fifth, while you may be able to get away with putting 582 diff listings intoone file, it is probably wiser to group related patches into separate files incase something goes haywire.

DIAGNOSTICS

Too many to list here, but generally indicative thatpatchcouldn't parse your patch file.

The message ``Hmm...'' indicates that there is unprocessed text inthe patch file and thatpatchis attempting to intuit whether there is a patch in that text and, if so,what kind of patch it is.

Patchwill exit with a non-zero status if any reject files were created.When applying a set of patches in a loop it behooves you to check thisexit status so you don't apply a later patch to a partially patched file.

CAVEATS

Patchcannot tell if the line numbers are off in anedscript, and can only detectbad line numbers in a normal diff when it finds a ``change'' ora ``delete'' command.A context diff using fuzz factor 3 may have the same problem.Until a suitable interactive interface is added, you should probably doa context diff in these cases to see if the changes made sense.Of course, compiling without errors is a pretty good indication that the patchworked, but not always.

Patchusually produces the correct results, even when it has to do a lot ofguessing.However, the results are guaranteed to be correct only when the patch isapplied to exactly the same version of the file that the patch wasgenerated from.

BUGS

Could be smarter about partial matches, excessively deviant offsets andswapped code, but that would take an extra pass.

If code has been duplicated (for instance with #ifdef OLDCODE ... #else ...#endif),patchis incapable of patching both versions, and, if it works at all, will likelypatch the wrong one, and tell you that it succeeded to boot.

If you apply a patch you've already applied,patchwill think it is a reversed patch, and offer to un-apply the patch.This could be construed as a feature.

This document was created byman2html,using the manual pages.