Fargo is a command-line find and replace processor for Git repositories. It scans all the tracked files in a git repo and performs substitutions based on input.
Note
If you're looking for a tool to do searches without substitutions,
then you should try the git grep command which already comes with Git.
You can install fargo directly from GitHub:
pip install git+https://github.com/te-je/fargo
That should install all the dependencies for you. If you want to install directly from source, clone the git repository and run the standard python setup.py install command.
- Python 3.5+
fargo [-iP] <text> [<substitution>] [<repository>]
The program scans the tracked files inside a local git repository found at
<repository> and replaces all occurrences of <text> with
<substitution>.
| Option/Argument | Description | Default |
|---|---|---|
<text> |
This is the text that is scanned inside the Git repository. | |
<substitution> |
Occurrences of <text> will be replaced
by this. |
If this is omitted, then occurrences of
<text> are removed. |
<repository> |
A local Git repository that is scanned for
occurrences of <text>. |
If not given, the current directory is used (must be a Git repository). |
-P |
Enable regex patterns. If set,
<substitution> can include capture
groups from the regular expression given in
<text>. |
|
-i |
Run in interactive mode. | |
--help |
Show program help and exit |
Interactive mode is turned on using the -i switch. In interactive mode,
fargo prompts on stdin before substitutions on each line where the search
text was found:
> fargo -i foo bar ./example.rst:L14:You'll find several examples in the |0|foo|bar|ter on Accept replacements? [yes]: ‸
Breaking this down, the first part of the prompt identifies the file
(./example.rst) and line number (14) where the text was found. It's followed
by the full line of text for context. Within the line, there's one
candidate substition consisting of three blocks: |0|foo|bar|. The first
block is the sequence number of the substitution on the line (more about this
below). The second block is the text that was matched, and the third block is
the text that it is about to be replaced with. Finally, there's the prompt
about what to do with the line (Accept Replacements? [yes]: ‸). Entering
either yes or no will accept or reject the line changes.
Sometimes, there might be more than one substitution possible on the same line:
> fargo -i me ME ./chatlog.txt:L2: is not the sa|0|me|ME| as |1|me|ME|eting |2|me|ME| in person. It's Accept replacements? [yes]: ‸
Instead of accepting or rejecting all of the substitutions wholesale, specific
substitutions can be accepted by entering a space separated list of their
sequence numbers. For example, entering 2 above will result in the line:
is not the same as meeting ME in person. It's.
To turn on regular expressions, use the -P switch. Regular expressions
allow the use of capture groups inside subsitutions:
> fargo -P "Name: (\w+) (\w+)" "Name: \2, \1"
Or, you can use named capture groups like so:
> fargo -P "Name: (?P<first>\w+) (?P<last>\w+)" "Name: \g<last>, \g<first>"
Fargo supports unicode text substitutions. Sort of. Fargo tries to guess the encoding of every file that it reads (using chardet). If it can't make a confident guess, then the file is ignored.
In cases where Fargo makes substitutions that can't be re-encoded with the original encoding, then the substitutions are ignored and the file is left untouched.