diff options
| author | Alyssa Rosenzweig <alyssa@rosenzweig.io> | 2017-04-04 09:44:10 -0700 |
|---|---|---|
| committer | Leah Rowe <info@minifree.org> | 2017-04-05 00:21:00 +0100 |
| commit | 15ee46bb2995319f2866089b7cf50f467ab223b7 (patch) | |
| tree | a1e6290c5b6e7f09bd43d91718f532b5529d4533 /docs/misc/patch.md | |
| parent | 4b0281fb2e5ba648058c9075f4d8b45bb76dbc66 (diff) | |
| download | librebootfr-15ee46bb2995319f2866089b7cf50f467ab223b7.tar.gz librebootfr-15ee46bb2995319f2866089b7cf50f467ab223b7.zip | |
Migrate a file to TOC
Diffstat (limited to 'docs/misc/patch.md')
| -rw-r--r-- | docs/misc/patch.md | 99 |
1 files changed, 0 insertions, 99 deletions
diff --git a/docs/misc/patch.md b/docs/misc/patch.md deleted file mode 100644 index 296a5853..00000000 --- a/docs/misc/patch.md +++ /dev/null @@ -1,99 +0,0 @@ -% Diff and patch - -This is just a quick guide for reference, use 'man' to know more. - -[Back to index](./) - -Apply a patch -============= - -To apply a patch to a single file, do that in it's directory: - - $ patch < foo.patch - -Assuming that the patch is distributed in unified format identifying the -file the patch should be applied to, the above will work. Otherwise: - - $ patch foo.txt < bar.patch - -You can apply a patch to an entire directory, but note the "p level". -What this means is that inside patch files will be the files that you -intend to patch, identified by path names that might be different when -the files ane located on your own computer instead of on the computer -where the patch was created. 'p' level instructs the 'patch' utility -to ignore parts of the path name to identify the files correctly. -Usually a p level of 1 will work, so you would use: - - $ patch -p1 < baz.patch - -Change to the top level directory before running this. If a patch level -of 1 cannot identify the files to patch, then inspect the patch file for -file names. For example:\ -**/home/user/do/not/panic/yet.c** - -and you are working in a directory that contains panic/yet.c, use: - - $ patch -p5 < baz.patch - -You usually count one up for each path separator (forward slash) removed -from the beginning of the path, until you are left with a path that -exists in the current working directory. The count is the p level. - -Removing a patch using the -R flag - - $ patch -p5 -R < baz.patch - -Create a patch with diff -======================== - -Diff can create a patch for a single file: - - $ diff -u original.c new.c > original.patch - -For diff'ing a source tree: - - $ cp -R original new - -Do whatever you want in new/ and then diff it: - - $ diff -rupN original/ new/ > original.patch - -git diff -======== - -git is something special. - -Note: this won't show new files created. - -Just make whatever changes you want to a git clone and then: - - $ git diff > patch.git - -Note the git revision that you did this with: - - $ git log - -Alternatively (better yet), commit your changes and then use: - $ git format-patch -N -Replace N with the number of commits that you want to show. - -git apply -========= - -it really is. - -Now to apply that patch in the future, just git clone it again and do -with the git revision you found from above: - - $ git reset --hard REVISIONNUMBER - -Now put patch.git in the git clone directory and do: - - $ git apply patch.git - -If you use a patch from git format-patch, then use **git am patch.git** -instead of **git apply patch.git**. git-am will re-create the commits -aswell, instead of just applying the patch. - -Copyright © 2014, 2015 Leah Rowe <info@minifree.org>\ -This page is available under the [CC BY SA 4.0](../cc-by-sa-4.0.txt) |