twin/ripgrep
1
0
Fork 0
mirror of https://github.com/BurntSushi/ripgrep.git synced 2025-07-26 17:51:56 -07:00
Code Issues Packages Projects Releases Wiki Activity

Compare commits

base: twin:ignore-0.1.1
Branches Tags
twin:master twin:ag/fix-ci twin:ag/index twin:ag/fix-multiline-replace-duplicate twin:ag/fix-cve-2021-3013 twin:ag/moarci twin:ag/ghactions twin:ag/snafu twin:ag/disable-bom-sniff twin:ag/partial-bstr-migration twin:ag/bstr-migration twin:ag/prepare-0.10.0 twin:ag/libripgrep-freeze-2 twin:ag/libripgrep-freeze-1 twin:ag/changelog
twin:globset-0.4.16 twin:14.1.1 twin:14.1.0 twin:ignore-0.4.22 twin:14.0.3 twin:14.0.2 twin:grep-0.3.1 twin:grep-printer-0.2.1 twin:grep-searcher-0.1.13 twin:14.0.1 twin:grep-0.3.0 twin:grep-printer-0.2.0 twin:14.0.0 twin:grep-0.2.13 twin:grep-searcher-0.1.12 twin:grep-pcre2-0.1.7 twin:grep-regex-0.1.12 twin:grep-matcher-0.1.7 twin:grep-cli-0.1.10 twin:ignore-0.4.21 twin:globset-0.4.14 twin:globset-0.4.13 twin:globset-0.4.12 twin:grep-cli-0.1.9 twin:globset-0.4.11 twin:grep-cli-0.1.8 twin:grep-0.2.12 twin:ignore-0.4.20 twin:grep-0.2.11 twin:grep-printer-0.1.7 twin:grep-searcher-0.1.11 twin:grep-pcre2-0.1.6 twin:grep-regex-0.1.11 twin:grep-matcher-0.1.6 twin:grep-cli-0.1.7 twin:ignore-0.4.19 twin:globset-0.4.10 twin:grep-0.2.10 twin:grep-searcher-0.1.10 twin:grep-0.2.9 twin:grep-searcher-0.1.9 twin:grep-regex-0.1.10 twin:globset-0.4.9 twin:globset-0.4.8 twin:13.0.0 twin:grep-0.2.8 twin:grep-printer-0.1.6 twin:grep-searcher-0.1.8 twin:grep-pcre2-0.1.5 twin:grep-regex-0.1.9 twin:grep-matcher-0.1.5 twin:grep-cli-0.1.6 twin:ignore-0.4.18 twin:globset-0.4.7 twin:ignore-0.4.17 twin:globset-0.4.6 twin:12.1.1 twin:grep-0.2.7 twin:grep-cli-0.1.5 twin:ignore-0.4.16 twin:12.1.0 twin:grep-0.2.6 twin:grep-printer-0.1.5 twin:grep-regex-0.1.8 twin:ignore-0.4.15 twin:12.0.1 twin:ignore-0.4.14 twin:ignore-0.4.13 twin:grep-regex-0.1.7 twin:12.0.0 twin:grep-searcher-0.1.7 twin:grep-regex-0.1.6 twin:grep-printer-0.1.4 twin:grep-pcre2-0.1.4 twin:grep-matcher-0.1.4 twin:ignore-0.4.12 twin:grep-0.2.5 twin:globset-0.4.5 twin:grep-cli-0.1.4 twin:ignore-0.4.11 twin:grep-regex-0.1.5 twin:grep-searcher-0.1.6 twin:ignore-0.4.10 twin:11.0.2 twin:grep-matcher-0.1.3 twin:grep-regex-0.1.4 twin:ignore-0.4.9 twin:ignore-0.4.8 twin:grep-searcher-0.1.5 twin:grep-printer-0.1.3 twin:grep-cli-0.1.3 twin:globset-0.4.4 twin:11.0.1 twin:11.0.0 twin:grep-0.2.4 twin:grep-cli-0.1.2 twin:grep-printer-0.1.2 twin:grep-searcher-0.1.4 twin:grep-regex-0.1.3 twin:grep-pcre2-0.1.3 twin:grep-matcher-0.1.2 twin:ignore-0.4.7 twin:globset-0.4.3 twin:grep-regex-0.1.2 twin:grep-searcher-0.1.3 twin:grep-searcher-0.1.2 twin:ignore-0.4.6 twin:ignore-0.4.5 twin:grep-0.2.3 twin:grep-pcre2-0.1.2 twin:0.10.0 twin:grep-searcher-0.1.1 twin:globset-0.4.2 twin:grep-pcre2-0.1.1 twin:grep-0.2.2 twin:grep-printer-0.1.1 twin:ignore-0.4.4 twin:grep-regex-0.1.1 twin:grep-matcher-0.1.1 twin:grep-cli-0.1.1 twin:grep-0.2.1 twin:grep-cli-0.1.0 twin:grep-0.2.0 twin:grep-search-0.1.0 twin:grep-matcher-0.1.0 twin:grep-printer-0.1.0 twin:grep-pcre2-0.1.0 twin:grep-regex-0.1.0 twin:0.9.0 twin:grep-0.1.9 twin:ignore-0.4.3 twin:globset-0.4.1 twin:ignore-0.4.2 twin:globset-0.4.0 twin:termcolor-0.3.6 twin:0.8.1 twin:termcolor-0.3.5 twin:ignore-0.4.1 twin:0.8.0 twin:ignore-0.4.0 twin:globset-0.3.0 twin:termcolor-0.3.4 twin:wincolor-0.1.6 twin:grep-0.1.8 twin:wincolor-0.1.5 twin:0.7.1 twin:ignore-0.3.1 twin:ignore-0.3.0 twin:grep-0.1.7 twin:globset-0.2.1 twin:0.7.0 twin:termcolor-0.3.3 twin:0.6.0 twin:ignore-0.2.2 twin:ignore-0.2.1 twin:wincolor-0.1.4 twin:0.5.2 twin:ignore-0.1.9 twin:0.5.1 twin:termcolor-0.3.2 twin:0.5.0 twin:grep-0.1.6 twin:ignore-0.1.8 twin:globset-0.1.4 twin:termcolor-0.3.1 twin:wincolor-0.1.3 twin:termcolor-0.3.0 twin:wincolor-0.1.2 twin:0.4.0 twin:0.3.2 twin:ignore-0.1.6 twin:0.3.1 twin:termcolor-0.1.1 twin:0.3.0 twin:termcolor-0.1.0 twin:wincolor-0.1.0 twin:0.2.9 twin:ignore-0.1.5 twin:0.2.8 twin:0.2.7 twin:ignore-0.1.4 twin:grep-0.1.4 twin:globset-0.1.2 twin:0.2.6 twin:ignore-0.1.3 twin:ignore-0.1.2 twin:0.2.5 twin:0.2.4 twin:ignore-0.1.1 twin:ignore-0.1.0 twin:globset-0.1.1 twin:0.2.3 twin:0.2.2 twin:globset-0.1.0 twin:0.2.1 twin:0.2.0 twin:0.1.18 twin:0.1.17 twin:0.1.16 twin:0.1.15 twin:0.1.14 twin:0.1.13 twin:0.1.12 twin:0.1.11 twin:0.1.10 twin:0.1.9 twin:0.1.8 twin:0.1.7 twin:0.1.6 twin:0.1.5 twin:0.1.4 twin:0.1.3 twin:0.1.2 twin:0.1.1 twin:0.1.0 twin:0.0.19 twin:0.0.18 twin:0.0.17 twin:0.0.16 twin:0.0.15 twin:0.0.14 twin:0.0.13 twin:0.0.12 twin:0.0.11 twin:0.0.10 twin:0.0.9 twin:0.0.8 twin:0.0.7 twin:0.0.6 twin:0.0.5 twin:0.0.4 twin:0.0.3 twin:0.0.2 twin:0.0.1
...
compare: twin:termcolor-0.1.0
Branches Tags
twin:master twin:ag/fix-ci twin:ag/index twin:ag/fix-multiline-replace-duplicate twin:ag/fix-cve-2021-3013 twin:ag/moarci twin:ag/ghactions twin:ag/snafu twin:ag/disable-bom-sniff twin:ag/partial-bstr-migration twin:ag/bstr-migration twin:ag/prepare-0.10.0 twin:ag/libripgrep-freeze-2 twin:ag/libripgrep-freeze-1 twin:ag/changelog
twin:globset-0.4.16 twin:14.1.1 twin:14.1.0 twin:ignore-0.4.22 twin:14.0.3 twin:14.0.2 twin:grep-0.3.1 twin:grep-printer-0.2.1 twin:grep-searcher-0.1.13 twin:14.0.1 twin:grep-0.3.0 twin:grep-printer-0.2.0 twin:14.0.0 twin:grep-0.2.13 twin:grep-searcher-0.1.12 twin:grep-pcre2-0.1.7 twin:grep-regex-0.1.12 twin:grep-matcher-0.1.7 twin:grep-cli-0.1.10 twin:ignore-0.4.21 twin:globset-0.4.14 twin:globset-0.4.13 twin:globset-0.4.12 twin:grep-cli-0.1.9 twin:globset-0.4.11 twin:grep-cli-0.1.8 twin:grep-0.2.12 twin:ignore-0.4.20 twin:grep-0.2.11 twin:grep-printer-0.1.7 twin:grep-searcher-0.1.11 twin:grep-pcre2-0.1.6 twin:grep-regex-0.1.11 twin:grep-matcher-0.1.6 twin:grep-cli-0.1.7 twin:ignore-0.4.19 twin:globset-0.4.10 twin:grep-0.2.10 twin:grep-searcher-0.1.10 twin:grep-0.2.9 twin:grep-searcher-0.1.9 twin:grep-regex-0.1.10 twin:globset-0.4.9 twin:globset-0.4.8 twin:13.0.0 twin:grep-0.2.8 twin:grep-printer-0.1.6 twin:grep-searcher-0.1.8 twin:grep-pcre2-0.1.5 twin:grep-regex-0.1.9 twin:grep-matcher-0.1.5 twin:grep-cli-0.1.6 twin:ignore-0.4.18 twin:globset-0.4.7 twin:ignore-0.4.17 twin:globset-0.4.6 twin:12.1.1 twin:grep-0.2.7 twin:grep-cli-0.1.5 twin:ignore-0.4.16 twin:12.1.0 twin:grep-0.2.6 twin:grep-printer-0.1.5 twin:grep-regex-0.1.8 twin:ignore-0.4.15 twin:12.0.1 twin:ignore-0.4.14 twin:ignore-0.4.13 twin:grep-regex-0.1.7 twin:12.0.0 twin:grep-searcher-0.1.7 twin:grep-regex-0.1.6 twin:grep-printer-0.1.4 twin:grep-pcre2-0.1.4 twin:grep-matcher-0.1.4 twin:ignore-0.4.12 twin:grep-0.2.5 twin:globset-0.4.5 twin:grep-cli-0.1.4 twin:ignore-0.4.11 twin:grep-regex-0.1.5 twin:grep-searcher-0.1.6 twin:ignore-0.4.10 twin:11.0.2 twin:grep-matcher-0.1.3 twin:grep-regex-0.1.4 twin:ignore-0.4.9 twin:ignore-0.4.8 twin:grep-searcher-0.1.5 twin:grep-printer-0.1.3 twin:grep-cli-0.1.3 twin:globset-0.4.4 twin:11.0.1 twin:11.0.0 twin:grep-0.2.4 twin:grep-cli-0.1.2 twin:grep-printer-0.1.2 twin:grep-searcher-0.1.4 twin:grep-regex-0.1.3 twin:grep-pcre2-0.1.3 twin:grep-matcher-0.1.2 twin:ignore-0.4.7 twin:globset-0.4.3 twin:grep-regex-0.1.2 twin:grep-searcher-0.1.3 twin:grep-searcher-0.1.2 twin:ignore-0.4.6 twin:ignore-0.4.5 twin:grep-0.2.3 twin:grep-pcre2-0.1.2 twin:0.10.0 twin:grep-searcher-0.1.1 twin:globset-0.4.2 twin:grep-pcre2-0.1.1 twin:grep-0.2.2 twin:grep-printer-0.1.1 twin:ignore-0.4.4 twin:grep-regex-0.1.1 twin:grep-matcher-0.1.1 twin:grep-cli-0.1.1 twin:grep-0.2.1 twin:grep-cli-0.1.0 twin:grep-0.2.0 twin:grep-search-0.1.0 twin:grep-matcher-0.1.0 twin:grep-printer-0.1.0 twin:grep-pcre2-0.1.0 twin:grep-regex-0.1.0 twin:0.9.0 twin:grep-0.1.9 twin:ignore-0.4.3 twin:globset-0.4.1 twin:ignore-0.4.2 twin:globset-0.4.0 twin:termcolor-0.3.6 twin:0.8.1 twin:termcolor-0.3.5 twin:ignore-0.4.1 twin:0.8.0 twin:ignore-0.4.0 twin:globset-0.3.0 twin:termcolor-0.3.4 twin:wincolor-0.1.6 twin:grep-0.1.8 twin:wincolor-0.1.5 twin:0.7.1 twin:ignore-0.3.1 twin:ignore-0.3.0 twin:grep-0.1.7 twin:globset-0.2.1 twin:0.7.0 twin:termcolor-0.3.3 twin:0.6.0 twin:ignore-0.2.2 twin:ignore-0.2.1 twin:wincolor-0.1.4 twin:0.5.2 twin:ignore-0.1.9 twin:0.5.1 twin:termcolor-0.3.2 twin:0.5.0 twin:grep-0.1.6 twin:ignore-0.1.8 twin:globset-0.1.4 twin:termcolor-0.3.1 twin:wincolor-0.1.3 twin:termcolor-0.3.0 twin:wincolor-0.1.2 twin:0.4.0 twin:0.3.2 twin:ignore-0.1.6 twin:0.3.1 twin:termcolor-0.1.1 twin:0.3.0 twin:termcolor-0.1.0 twin:wincolor-0.1.0 twin:0.2.9 twin:ignore-0.1.5 twin:0.2.8 twin:0.2.7 twin:ignore-0.1.4 twin:grep-0.1.4 twin:globset-0.1.2 twin:0.2.6 twin:ignore-0.1.3 twin:ignore-0.1.2 twin:0.2.5 twin:0.2.4 twin:ignore-0.1.1 twin:ignore-0.1.0 twin:globset-0.1.1 twin:0.2.3 twin:0.2.2 twin:globset-0.1.0 twin:0.2.1 twin:0.2.0 twin:0.1.18 twin:0.1.17 twin:0.1.16 twin:0.1.15 twin:0.1.14 twin:0.1.13 twin:0.1.12 twin:0.1.11 twin:0.1.10 twin:0.1.9 twin:0.1.8 twin:0.1.7 twin:0.1.6 twin:0.1.5 twin:0.1.4 twin:0.1.3 twin:0.1.2 twin:0.1.1 twin:0.1.0 twin:0.0.19 twin:0.0.18 twin:0.0.17 twin:0.0.16 twin:0.0.15 twin:0.0.14 twin:0.0.13 twin:0.0.12 twin:0.0.11 twin:0.0.10 twin:0.0.9 twin:0.0.8 twin:0.0.7 twin:0.0.6 twin:0.0.5 twin:0.0.4 twin:0.0.3 twin:0.0.2 twin:0.0.1

102 Commits
ignore-0.1 ... termcolor-

Author SHA1 Message Date
Andrew Gallant
df72d8d1e0 Make wincolor crate compilable on non-Windows platforms. 2016-11-20 15:44:19 -05:00
Andrew Gallant
d06f84ced3 Get rid of special mmap decision on Windows. 
I spent some quality time on my Windows 10 laptop and it appears to
suffer from a similar trade-off as on Linux: mmaps are bad for large
directory traversals but good for single large files.

Darwin continues to reject memory maps in all cases (unless explicitly
requested), but more testing should be done there.
 2016-11-20 15:32:50 -05:00
Andrew Gallant
9598331fa8 Propagate no_messages option to worker. 
Fixes #241
 2016-11-20 15:01:37 -05:00
Andrew Gallant
883d8fc72f Merge pull request #240 from BurntSushi/color 
Completely re-work colored output and tty handling.
 2016-11-20 15:01:11 -05:00
Andrew Gallant
e8a30cb893 Completely re-work colored output and tty handling. 
This commit completely guts all of the color handling code and replaces
most of it with two new crates: wincolor and termcolor. wincolor
provides a simple API to coloring using the Windows console and
termcolor provides a platform independent coloring API tuned for
multithreaded command line programs. This required a lot more
flexibility than what the `term` crate provided, so it was dropped.
We instead switch to writing ANSI escape sequences directly and ignore
the TERMINFO database.

In addition to fixing several bugs, this commit also permits end users
to customize colors to a certain extent. For example, this command will
set the match color to magenta and the line number background to yellow:

    rg --colors 'match:fg:magenta' --colors 'line:bg:yellow' foo

For tty handling, we've adopted a hack from `git` to do tty detection in
MSYS/mintty terminals. As a result, ripgrep should get both color
detection and piping correct on Windows regardless of which terminal you
use.

Finally, switch to line buffering. Performance doesn't seem to be
impacted and it's an otherwise more user friendly option.

Fixes #37, Fixes #51, Fixes #94, Fixes #117, Fixes #182, Fixes #231
 2016-11-20 11:14:52 -05:00
Andrew Gallant
03f7605322 Rename --files-without-matches to --files-without-match. 
This is to be consistent with grep.
 2016-11-19 20:15:41 -05:00
Andrew Gallant
61663e2307 Merge pull request #239 from mernen/files-without-matches 
Add --files-without-matches flag.
 2016-11-19 20:05:00 -05:00
Daniel Luz
bd3e7eedb1 Add --files-without-matches flag. 
Performs the opposite of --files-with-matches: only shows paths of
files that contain zero matches.

Closes #138
 2016-11-19 21:48:59 -02:00
Andrew Gallant
1e6c2ac8e3 Merge pull request #238 from jacwah/refactor 
`Ignore` refactorings
 2016-11-18 23:11:17 -05:00
Andrew Gallant
0302d58eb8 Fix stdin bug with --file. 
When `rg -f-` is used, the default search path should be `./` and not
`-`.
 2016-11-17 20:48:11 -05:00
Andrew Gallant
e37f783fc0 Fix issue number mixup. 
Thanks @bluss!
 2016-11-17 20:30:18 -05:00
Andrew Gallant
495e13cc61 Merge pull request #233 from BurntSushi/clap 
Switch from Docopt to Clap.
 2016-11-17 20:22:16 -05:00
Andrew Gallant
92dc402f7f Switch from Docopt to Clap. 
There were two important reasons for the switch:

1. Performance. Docopt does poorly when the argv becomes large, which is
   a reasonable common use case for search tools. (e.g., use with xargs)
2. Better failure modes. Clap knows a lot more about how a particular
   argv might be invalid, and can therefore provide much clearer error
   messages.

While both were important, (1) made it urgent.

Note that since Clap requires at least Rust 1.11, this will in turn
increase the minimum Rust version supported by ripgrep from Rust 1.9 to
Rust 1.11. It is therefore a breaking change, so the soonest release of
ripgrep with Clap will have to be 0.3.

There is also at least one subtle breaking change in real usage.
Previous to this commit, this used to work:

    rg -e -foo

Where this would cause ripgrep to search for the string `-foo`. Clap
currently has problems supporting this use case
(see: https://github.com/kbknapp/clap-rs/issues/742),
but it can be worked around by using this instead:

    rg -e [-]foo

or even

    rg [-]foo

and this still works:

    rg -- -foo

This commit also adds Bash, Fish and PowerShell completion files to the
release, fixes a bug that prevented ripgrep from working on file
paths containing invalid UTF-8 and shows short descriptions in the
output of `-h` but longer descriptions in the output of `--help`.

Fixes #136, Fixes #189, Fixes #210, Fixes #230
 2016-11-17 19:53:41 -05:00
Andrew Gallant
a3f5e0c3d5 Use env::home_dir() instead of env::var_os(HOME). 
Thanks @steveklabnik!
 2016-11-17 16:54:39 -05:00
Andrew Gallant
39e1a0d694 Merge pull request #227 from emk/issue-7 
Allow specifying patterns with `-f FILE` and `-f-`
 2016-11-15 18:15:05 -05:00
Eric Kidd
e9cd0a1cc3 Allow specifying patterns with -f FILE and -f- 
This is a somewhat basic implementation of `-f-` (#7), with unit tests.
Changes include:

1. The internals of the `pattern` function have been refactored to avoid
   code duplication, but there's a lot more we could do.  Right now we
   read the entire pattern list into a `Vec`.
2. There's now a `WorkDir::pipe` command that allows sending standard
   input to `rg` when testing.

Not implemented: aho-corasick.
 2016-11-15 13:00:16 -05:00
Andrew Gallant
cc35ae0748 Merge pull request #235 from jinyeow/master 
Added elixir to types
 2016-11-15 06:36:19 -05:00
Justin Puah
5ee175beaf Added elixir to types 2016-11-15 20:57:38 +11:00
Andrew Gallant
4b18f82899 Disable symlink tests on Windows. 
For some reason, these work on AppVeyor but not in other build systems.
Let's just disable them.

See: https://github.com/rust-lang/rust/pull/37149
 2016-11-11 06:44:23 -05:00
Andrew Gallant
5462af4434 Pin rustc-serialize to 0.3.19. 
See: https://github.com/rust-lang-nursery/rustc-serialize/pull/159
 2016-11-09 20:28:58 -05:00
Andrew Gallant
d2e70da040 0.2.9 2016-11-09 19:07:25 -05:00
Andrew Gallant
64dc9b6709 update deps 2016-11-09 18:54:22 -05:00
Andrew Gallant
9ffd4c421f ignore-0.1.5 2016-11-09 18:52:52 -05:00
Andrew Gallant
d862b80afb changelog 0.2.9 2016-11-09 18:52:08 -05:00
Andrew Gallant
5b73dcc8ab Rework parallelism in directory iterator. 
Previously, ignore::WalkParallel would invoke the callback for all
*explicitly* given file paths in a single thread, which effectively
meant that `rg pattern foo bar baz ...` didn't actually search foo, bar
and baz in parallel.

The code was structured that way to avoid spinning up workers if no
directory paths were given. The original intention was probably to have
a separate pool of threads responsible for searching, but ripgrep ended
up just reusing the ignore::WalkParallel workers themselves for searching,
and thereby subjected to its sub-par performance in this case.

The code has been restructured so that file paths are sent to the workers,
which brings back parallelism.

Fixes #226
 2016-11-09 17:19:40 -05:00
Andrew Gallant
2dce0dc0df Fix a bug with handling --ignore-file. 
Namely, passing a directory to --ignore-file caused ripgrep to allocate
memory without bound.

The issue was that I got a bit overzealous with partial error
reporting. Namely, when processing a gitignore file, we should try
to use every pattern even if some patterns are invalid globs (e.g.,
a**b). In the process, I applied the same logic to I/O errors. In this
case, it manifest by attempting to read lines from a directory, which
appears to yield Results forever, where each Result is an error of the
form "you can't read from a directory silly." Since I treated it as a
partial error, ripgrep was just spinning and accruing each error in
memory, which caused the OOM killer to kick in.

Fixes #228
 2016-11-09 16:45:23 -05:00
Andrew Gallant
2e5c3c05e8 reword 2016-11-06 19:48:49 -05:00
Andrew Gallant
6884eea2f5 reword 2016-11-06 19:48:17 -05:00
Andrew Gallant
a3a2f0be6a ucg author says it's not a bug per se 2016-11-06 19:45:18 -05:00
Andrew Gallant
f24873c70b Don't ever search directories. 2016-11-06 19:02:14 -05:00
Andrew Gallant
58126ffe15 touchups 2016-11-06 18:51:00 -05:00
Andrew Gallant
17644a76c0 typo 2016-11-06 18:49:07 -05:00
Andrew Gallant
9fc9f368f5 Always search paths given by user. 
This permits doing `rg -a test /dev/sda1` for example, where as before
/dev/sda1 was skipped because it wasn't a regular file.
 2016-11-06 18:23:50 -05:00
Andrew Gallant
9cab076a72 touchups 2016-11-06 18:04:55 -05:00
Andrew Gallant
7aa9652f3c touchups 2016-11-06 18:02:45 -05:00
Andrew Gallant
7187f61ca8 touchups 2016-11-06 18:01:55 -05:00
Andrew Gallant
f869c58a5a touchups 2016-11-06 17:59:57 -05:00
Andrew Gallant
3538ba3577 Update README with more/updated benchmarks 2016-11-06 17:55:38 -05:00
Andrew Gallant
a454fa75b9 Update brew tap. 2016-11-06 16:51:14 -05:00
Andrew Gallant
18943b9317 0.2.8 2016-11-06 16:16:48 -05:00
Andrew Gallant
68427b5b79 changelog 0.2.8 2016-11-06 16:16:19 -05:00
Andrew Gallant
4ca15a8a51 simd-accel should not invoke avx-accel. 
This was a silly transcription error.
 2016-11-06 16:15:23 -05:00
Andrew Gallant
2daef51fe5 0.2.7 2016-11-06 15:49:25 -05:00
Andrew Gallant
43ed91dc5c changelog 0.2.7 2016-11-06 15:48:52 -05:00
Andrew Gallant
dada75d2a7 Update sub-crate dependency versions. 2016-11-06 15:48:40 -05:00
Andrew Gallant
76b9f01ad2 ignore-0.1.4 2016-11-06 15:35:21 -05:00
Andrew Gallant
8baa0e56b7 grep-0.1.4 2016-11-06 15:35:17 -05:00
Andrew Gallant
301ee6d3f5 globset-0.1.2 2016-11-06 15:35:05 -05:00
Andrew Gallant
77ad7588ae Add --no-messages flag. 
This flag is similar to what's found in grep: it will suppress all error
messages, such as those shown when a particular file couldn't be read.

Closes #149
 2016-11-06 14:36:08 -05:00
Andrew Gallant
58aca2efb2 Add -m/--max-count flag. 
This flag limits the number of matches printed *per file*.

Closes #159
 2016-11-06 13:09:53 -05:00
Andrew Gallant
351eddc17e Add new 'h' file type. 
This is intended to correspond to C/C++ header files.

Fixes #186
 2016-11-06 12:23:42 -05:00
Andrew Gallant
277dda544c Include the name "ripgrep" in more places. 
Fixes #203
 2016-11-06 12:21:36 -05:00
Andrew Gallant
8c869cbd87 update man page 2016-11-06 12:10:55 -05:00
Andrew Gallant
598b162fea Note -e/--regexp's additional usefulness. 
Specifically, it can be used when searching for patterns that start
with a dash.

Fixes #215
 2016-11-06 12:10:27 -05:00
Andrew Gallant
0222e024fe Fixes a bug with --smart-case. 
This was a subtle bug, but the big picture was that the smart case
information wasn't being carried through to the literal extraction in
some cases. When this happened, it was possible to get back an incomplete
set of literals, which would therefore miss some valid matches.

The fix to this is to actually parse the regex and determine whether
smart case applies before doing anything else. It's a little extra work,
but parsing is pretty fast.

Fixes #199
 2016-11-06 12:07:47 -05:00
Andrew Gallant
5bd0edbbe1 Actually use simd/avx optimizations in bytecount crate. 
Also update compile script.
 2016-11-05 22:44:33 -04:00
Andrew Gallant
4368913d8f Merge branch 'fast_linecount' 2016-11-05 22:29:42 -04:00
Andre Bogus
02de97b8ce Use the bytecount crate for fast line counting. 
Fixes #128
 2016-11-05 22:29:26 -04:00
Andrew Gallant
32db773d51 Merge pull request #223 from BurntSushi/ignore-parallel 
Add parallel recursive directory iterator.
 2016-11-05 22:19:47 -04:00
Andrew Gallant
b272be25fa Add parallel recursive directory iterator. 
This adds a new walk type in the `ignore` crate, `WalkParallel`, which
provides a way for recursively iterating over a set of paths in parallel
while respecting various ignore rules.

The API is a bit strange, as a closure producing a closure isn't
something one often sees, but it does seem to work well.

This also allowed us to simplify much of the worker logic in ripgrep
proper, where MultiWorker is now gone.
 2016-11-05 21:45:55 -04:00
Jacob Wahlgren
f63c168563 Rename IgnoreOptions::has_ignores 
The name has_ignores is not descriptive in my opinion. I think
has_any_ignore_options more clearly states this method's purpose. I also
considered simply IgnoreOptions::any though I went with the more verbose
option.
 2016-11-06 01:17:41 +01:00
Jacob Wahlgren
a05671c8d7 Use new Match::or to simplify return 2016-11-06 01:07:11 +01:00
Andrew Gallant
1aeae3e22d update ripgrep 2016-11-04 21:12:08 -04:00
Andrew Gallant
60d537c43d Merge pull request #220 from theamazingfedex/adding-mak-type 
adding .mak extension for makefile filetype.
 2016-11-03 20:46:29 -04:00
Andrew Gallant
ef5c07476b Merge pull request #219 from theamazingfedex/adding-pdf-filetype 
adding .pdf filetype to available types.
 2016-11-03 20:46:07 -04:00
Andrew Gallant
4f6f34307c Merge pull request #218 from theamazingfedex/adding-cs-filetype 
adding cs filetype for ease of use.
 2016-11-03 20:46:02 -04:00
Daniel Wood
7cf560d27c adding .mak extension for makefile filetype. 
Fixes #217
 2016-11-03 15:52:10 -06:00
Daniel Wood
15b263ff55 adding cs filetype for ease of use. 2016-11-03 15:33:00 -06:00
Daniel Wood
53121e0733 adding .pdf filetype to available types. 2016-11-03 15:30:58 -06:00
Andrew Gallant
404785f950 Merge pull request #214 from lyuha/master 
Add textile, org, creole, rdoc, wiki filetype
 2016-11-02 11:53:22 -04:00
Lyuha
103c4c953c Add pod filetype 2016-11-03 00:06:14 +09:00
Lyuha
82abf883c5 Add wiki filetype 2016-11-03 00:06:14 +09:00
Lyuha
a2315d5ee5 Add creole filetype 2016-11-03 00:06:14 +09:00
Lyuha
201d0cb8c1 Add org filetype 2016-11-03 00:06:14 +09:00
Lyuha
6f45478a7d Add rdoc filetype 2016-11-03 00:06:14 +09:00
Lyuha
9c2c569624 Add textile filetype 2016-11-03 00:06:14 +09:00
Andrew Gallant
a1e4e0f85c Merge pull request #213 from tjdgus3537/master 
add asciidoc filetype and update markdown filetype
 2016-11-02 10:44:31 -04:00
tjdgus3537
caf31a769b Add .markdown, .mdown, .mkdn extension to md and markdown 2016-11-02 22:50:59 +09:00
tjdgus3537
920112e640 Add .adoc, .asc, asciidoc extension to asciidoc 2016-11-02 22:49:31 +09:00
Andrew Gallant
a84ffe603b Merge pull request #212 from radhermit/gentoo 
Add Gentoo info to the README
 2016-11-02 07:05:47 -04:00
Tim Harder
e4f83f3161 Add Gentoo info to the README 2016-11-01 22:03:00 -04:00
Andrew Gallant
fbca4a0332 Merge pull request #209 from dueyfinster/patch-1 
Added taskpaper as a file type
 2016-11-01 10:29:15 -04:00
Neil Grogan
65c7df1c25 Added taskpaper as a file type 2016-11-01 13:59:59 +00:00
Alexander Altman
18237da9b2 Add Agda and improve TeX ignore support (#207) 
Add Agda and improve TeX ignore support
 2016-11-01 06:56:53 -04:00
Andrew Gallant
f147f3aa39 0.2.6 2016-10-31 20:01:37 -04:00
Andrew Gallant
599c4fc3f3 changelog 0.2.6 2016-10-31 20:01:31 -04:00
Andrew Gallant
d85a6dd5c8 update ignore dependency 2016-10-31 20:01:31 -04:00
Andrew Gallant
40abade8ee ignore-0.1.3 2016-10-31 19:54:47 -04:00
Andrew Gallant
fca4fdf6ea ignore-0.1.2 2016-10-31 19:54:38 -04:00
Andrew Gallant
16975797fe Fixes a matching bug in the glob override matcher. 
This was probably a transcription error when moving the ignore matcher
code out of ripgrep core. Specifically, the override glob matcher should
not ignore directories if they don't match.

Fixes #206
 2016-10-31 19:54:38 -04:00
Andrew Gallant
6507a48f97 Ignore ignore/Cargo.lock 2016-10-31 19:54:38 -04:00
Andrew Gallant
c8e2fa1869 update Cargo.lock 2016-10-31 19:54:38 -04:00
Andrew Gallant
f728708ce9 Merge pull request #204 from lyuha/master 
Add .fish extension to fish shell
 2016-10-30 17:31:51 -04:00
Lyuha
c302995d05 Add .fish extension to fish shell 2016-10-31 00:52:45 +09:00
Andrew Gallant
4a77cc8100 update brew tap to 0.2.5 2016-10-29 23:23:45 -04:00
Andrew Gallant
dc86666044 changelog 0.2.5 2016-10-29 22:44:07 -04:00
Andrew Gallant
6b038511c7 0.2.5 2016-10-29 22:42:28 -04:00
Andrew Gallant
c767bccade fix appveyor, sigh 2016-10-29 22:42:22 -04:00
Andrew Gallant
a075a462fa 0.2.4 2016-10-29 22:40:02 -04:00
Andrew Gallant
24f753c306 changelog 0.2.4 2016-10-29 22:27:39 -04:00
Andrew Gallant
1aae2759ad update deps 2016-10-29 22:27:29 -04:00
Andrew Gallant
91646f6cca bump ignore to 0.1.1 2016-10-29 22:19:00 -04:00
47 changed files with 5524 additions and 1997 deletions
Expand all files Collapse all files

3
.gitignore vendored
View File

@@ -3,3 +3,6 @@ tags
target
/grep/Cargo.lock
/globset/Cargo.lock
/ignore/Cargo.lock
/termcolor/Cargo.lock
/wincolor/Cargo.lock

7
.travis.yml
View File

@@ -30,13 +30,10 @@ matrix:
env: TARGET=x86_64-apple-darwin
# Minimum Rust supported channel.
- os: linux
rust: 1.9.0
env: TARGET=x86_64-unknown-linux-musl
- os: linux
rust: 1.9.0
rust: 1.11.0
env: TARGET=x86_64-unknown-linux-gnu
- os: osx
rust: 1.9.0
rust: 1.11.0
env: TARGET=x86_64-apple-darwin
before_install:

101
CHANGELOG.md
View File

@@ -1,3 +1,104 @@
0.2.9
=====
Bug fixes:
* [BUG #226](https://github.com/BurntSushi/ripgrep/issues/226):
File paths explicitly given on the command line weren't searched in parallel.
(This was a regression in `0.2.7`.)
* [BUG #228](https://github.com/BurntSushi/ripgrep/issues/228):
If a directory was given to `--ignore-file`, ripgrep's memory usage would
grow without bound.
0.2.8
=====
Bug fixes:
* Fixed a bug with the SIMD/AVX features for using bytecount in commit
`4ca15a`.
0.2.7
=====
Performance improvements:
* [PERF #223](https://github.com/BurntSushi/ripgrep/pull/223):
Added a parallel recursive directory iterator. This results in major
performance improvements on large repositories.
* [PERF #11](https://github.com/BurntSushi/ripgrep/pull/11):
ripgrep now uses the `bytecount` library for counting new lines. In some
cases, ripgrep runs twice as fast. Use
`RUSTFLAGS="-C target-cpu=native" cargo build --release --features 'simd-accel avx-accel'`
to get the fastest possible binary.
Feature enhancements:
* Added or improved file type filtering for Agda, Tex, Taskpaper, Markdown,
asciidoc, textile, rdoc, org, creole, wiki, pod, C#, PDF, C, C++.
* [FEATURE #149](https://github.com/BurntSushi/ripgrep/issues/149):
Add a new `--no-messages` flag that suppresses error messages.
Note that `rg foo 2> /dev/null` also works.
* [FEATURE #159](https://github.com/BurntSushi/ripgrep/issues/159):
Add a new `-m/--max-count` flag that limits the total number of matches
printed for each file searched.
Bug fixes:
* [BUG #199](https://github.com/BurntSushi/ripgrep/issues/199):
Fixed a bug where `-S/--smart-case` wasn't being applied correctly to
literal optimizations.
* [BUG #203](https://github.com/BurntSushi/ripgrep/issues/203):
Mention the full name, ripgrep, in more places. It now appears in
the output of `--help` and `--version`. The repository URL is now also
in the output of `--help` and the man page.
* [BUG #215](https://github.com/BurntSushi/ripgrep/issues/215):
Include small note about how to search for a pattern that starts with a `-`.
0.2.6
=====
Feature enhancements:
* Added or improved file type filtering for Fish.
Bug fixes:
* [BUG #206](https://github.com/BurntSushi/ripgrep/issues/206):
Fixed a regression with `-g/--glob` flag in `0.2.5`.
0.2.5
=====
Feature enhancements:
* Added or improved file type filtering for Groovy, Handlebars, Tcl, zsh and
Python.
* [FEATURE #9](https://github.com/BurntSushi/ripgrep/issues/9):
Support global gitignore config and `.git/info/exclude` files.
* [FEATURE #45](https://github.com/BurntSushi/ripgrep/issues/45):
Add --ignore-file flag for specifying additional ignore files.
* [FEATURE #202](https://github.com/BurntSushi/ripgrep/pull/202):
Introduce a new
[`ignore`](https://github.com/BurntSushi/ripgrep/tree/master/ignore)
crate that encapsulates all of ripgrep's gitignore matching logic.
Bug fixes:
* [BUG #44](https://github.com/BurntSushi/ripgrep/issues/44):
ripgrep runs slowly when given lots of positional arguments that are
directories.
* [BUG #119](https://github.com/BurntSushi/ripgrep/issues/119):
ripgrep didn't reset terminal colors if it was interrupted by `^C`.
Fixed in [PR #187](https://github.com/BurntSushi/ripgrep/pull/187).
* [BUG #184](https://github.com/BurntSushi/ripgrep/issues/184):
Fixed a bug related to interpreting gitignore files in parent directories.
0.2.4
=====
SKIPPED.
0.2.3
=====
Bug fixes:

188
Cargo.lock generated
View File

@@ -1,23 +1,22 @@
[root]
name = "ripgrep"
version = "0.2.3"
version = "0.2.9"
dependencies = [
"bytecount 0.1.4 (registry+https://github.com/rust-lang/crates.io-index)",
"clap 2.18.0 (registry+https://github.com/rust-lang/crates.io-index)",
"ctrlc 2.0.1 (registry+https://github.com/rust-lang/crates.io-index)",
"deque 0.3.1 (registry+https://github.com/rust-lang/crates.io-index)",
"docopt 0.6.86 (registry+https://github.com/rust-lang/crates.io-index)",
"env_logger 0.3.5 (registry+https://github.com/rust-lang/crates.io-index)",
"grep 0.1.3",
"ignore 0.1.0",
"grep 0.1.4",
"ignore 0.1.5",
"kernel32-sys 0.2.2 (registry+https://github.com/rust-lang/crates.io-index)",
"lazy_static 0.2.1 (registry+https://github.com/rust-lang/crates.io-index)",
"lazy_static 0.2.2 (registry+https://github.com/rust-lang/crates.io-index)",
"libc 0.2.17 (registry+https://github.com/rust-lang/crates.io-index)",
"log 0.3.6 (registry+https://github.com/rust-lang/crates.io-index)",
"memchr 0.1.11 (registry+https://github.com/rust-lang/crates.io-index)",
"memmap 0.5.0 (registry+https://github.com/rust-lang/crates.io-index)",
"num_cpus 1.1.0 (registry+https://github.com/rust-lang/crates.io-index)",
"regex 0.1.80 (registry+https://github.com/rust-lang/crates.io-index)",
"rustc-serialize 0.3.19 (registry+https://github.com/rust-lang/crates.io-index)",
"term 0.4.4 (registry+https://github.com/rust-lang/crates.io-index)",
"termcolor 0.1.0",
"winapi 0.2.8 (registry+https://github.com/rust-lang/crates.io-index)",
]
@@ -29,6 +28,44 @@ dependencies = [
"memchr 0.1.11 (registry+https://github.com/rust-lang/crates.io-index)",
]
[[package]]
name = "ansi_term"
version = "0.9.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
[[package]]
name = "bitflags"
version = "0.7.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
[[package]]
name = "bytecount"
version = "0.1.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
dependencies = [
"simd 0.1.1 (registry+https://github.com/rust-lang/crates.io-index)",
]
[[package]]
name = "clap"
version = "2.18.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
dependencies = [
"ansi_term 0.9.0 (registry+https://github.com/rust-lang/crates.io-index)",
"bitflags 0.7.0 (registry+https://github.com/rust-lang/crates.io-index)",
"libc 0.2.17 (registry+https://github.com/rust-lang/crates.io-index)",
"strsim 0.5.1 (registry+https://github.com/rust-lang/crates.io-index)",
"term_size 0.2.1 (registry+https://github.com/rust-lang/crates.io-index)",
"unicode-segmentation 0.1.2 (registry+https://github.com/rust-lang/crates.io-index)",
"unicode-width 0.1.3 (registry+https://github.com/rust-lang/crates.io-index)",
"vec_map 0.6.0 (registry+https://github.com/rust-lang/crates.io-index)",
]
[[package]]
name = "crossbeam"
version = "0.2.10"
source = "registry+https://github.com/rust-lang/crates.io-index"
[[package]]
name = "ctrlc"
version = "2.0.1"
@@ -39,25 +76,6 @@ dependencies = [
"winapi 0.2.8 (registry+https://github.com/rust-lang/crates.io-index)",
]
[[package]]
name = "deque"
version = "0.3.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
dependencies = [
"rand 0.3.14 (registry+https://github.com/rust-lang/crates.io-index)",
]
[[package]]
name = "docopt"
version = "0.6.86"
source = "registry+https://github.com/rust-lang/crates.io-index"
dependencies = [
"lazy_static 0.2.1 (registry+https://github.com/rust-lang/crates.io-index)",
"regex 0.1.80 (registry+https://github.com/rust-lang/crates.io-index)",
"rustc-serialize 0.3.19 (registry+https://github.com/rust-lang/crates.io-index)",
"strsim 0.5.1 (registry+https://github.com/rust-lang/crates.io-index)",
]
[[package]]
name = "env_logger"
version = "0.3.5"
@@ -84,11 +102,11 @@ dependencies = [
[[package]]
name = "globset"
version = "0.1.0"
version = "0.1.2"
dependencies = [
"aho-corasick 0.5.3 (registry+https://github.com/rust-lang/crates.io-index)",
"fnv 1.0.5 (registry+https://github.com/rust-lang/crates.io-index)",
"lazy_static 0.2.1 (registry+https://github.com/rust-lang/crates.io-index)",
"lazy_static 0.2.2 (registry+https://github.com/rust-lang/crates.io-index)",
"log 0.3.6 (registry+https://github.com/rust-lang/crates.io-index)",
"memchr 0.1.11 (registry+https://github.com/rust-lang/crates.io-index)",
"regex 0.1.80 (registry+https://github.com/rust-lang/crates.io-index)",
@@ -96,7 +114,7 @@ dependencies = [
[[package]]
name = "grep"
version = "0.1.3"
version = "0.1.4"
dependencies = [
"log 0.3.6 (registry+https://github.com/rust-lang/crates.io-index)",
"memchr 0.1.11 (registry+https://github.com/rust-lang/crates.io-index)",
@@ -107,14 +125,15 @@ dependencies = [
[[package]]
name = "ignore"
version = "0.1.0"
version = "0.1.5"
dependencies = [
"globset 0.1.0",
"lazy_static 0.2.1 (registry+https://github.com/rust-lang/crates.io-index)",
"crossbeam 0.2.10 (registry+https://github.com/rust-lang/crates.io-index)",
"globset 0.1.2",
"lazy_static 0.2.2 (registry+https://github.com/rust-lang/crates.io-index)",
"log 0.3.6 (registry+https://github.com/rust-lang/crates.io-index)",
"memchr 0.1.11 (registry+https://github.com/rust-lang/crates.io-index)",
"regex 0.1.80 (registry+https://github.com/rust-lang/crates.io-index)",
"thread_local 0.2.7 (registry+https://github.com/rust-lang/crates.io-index)",
"thread_local 0.3.0 (registry+https://github.com/rust-lang/crates.io-index)",
"walkdir 1.0.1 (registry+https://github.com/rust-lang/crates.io-index)",
]
@@ -129,7 +148,7 @@ dependencies = [
[[package]]
name = "lazy_static"
version = "0.2.1"
version = "0.2.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
[[package]]
@@ -169,14 +188,6 @@ dependencies = [
"libc 0.2.17 (registry+https://github.com/rust-lang/crates.io-index)",
]
[[package]]
name = "rand"
version = "0.3.14"
source = "registry+https://github.com/rust-lang/crates.io-index"
dependencies = [
"libc 0.2.17 (registry+https://github.com/rust-lang/crates.io-index)",
]
[[package]]
name = "regex"
version = "0.1.80"
@@ -195,11 +206,6 @@ name = "regex-syntax"
version = "0.3.9"
source = "registry+https://github.com/rust-lang/crates.io-index"
[[package]]
name = "rustc-serialize"
version = "0.3.19"
source = "registry+https://github.com/rust-lang/crates.io-index"
[[package]]
name = "simd"
version = "0.1.1"
@@ -211,14 +217,22 @@ version = "0.5.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
[[package]]
name = "term"
version = "0.4.4"
name = "term_size"
version = "0.2.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
dependencies = [
"kernel32-sys 0.2.2 (registry+https://github.com/rust-lang/crates.io-index)",
"libc 0.2.17 (registry+https://github.com/rust-lang/crates.io-index)",
"winapi 0.2.8 (registry+https://github.com/rust-lang/crates.io-index)",
]
[[package]]
name = "termcolor"
version = "0.1.0"
dependencies = [
"wincolor 0.1.0",
]
[[package]]
name = "thread-id"
version = "2.0.0"
@@ -228,6 +242,15 @@ dependencies = [
"libc 0.2.17 (registry+https://github.com/rust-lang/crates.io-index)",
]
[[package]]
name = "thread-id"
version = "3.0.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
dependencies = [
"kernel32-sys 0.2.2 (registry+https://github.com/rust-lang/crates.io-index)",
"libc 0.2.17 (registry+https://github.com/rust-lang/crates.io-index)",
]
[[package]]
name = "thread_local"
version = "0.2.7"
@@ -236,11 +259,48 @@ dependencies = [
"thread-id 2.0.0 (registry+https://github.com/rust-lang/crates.io-index)",
]
[[package]]
name = "thread_local"
version = "0.3.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
dependencies = [
"thread-id 3.0.0 (registry+https://github.com/rust-lang/crates.io-index)",
"unreachable 0.1.1 (registry+https://github.com/rust-lang/crates.io-index)",
]
[[package]]
name = "unicode-segmentation"
version = "0.1.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
[[package]]
name = "unicode-width"
version = "0.1.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
[[package]]
name = "unreachable"
version = "0.1.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
dependencies = [
"void 1.0.2 (registry+https://github.com/rust-lang/crates.io-index)",
]
[[package]]
name = "utf8-ranges"
version = "0.1.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
[[package]]
name = "vec_map"
version = "0.6.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
[[package]]
name = "void"
version = "1.0.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
[[package]]
name = "walkdir"
version = "1.0.1"
@@ -260,31 +320,47 @@ name = "winapi-build"
version = "0.1.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
[[package]]
name = "wincolor"
version = "0.1.0"
dependencies = [
"kernel32-sys 0.2.2 (registry+https://github.com/rust-lang/crates.io-index)",
"winapi 0.2.8 (registry+https://github.com/rust-lang/crates.io-index)",
]
[metadata]
"checksum aho-corasick 0.5.3 (registry+https://github.com/rust-lang/crates.io-index)" = "ca972c2ea5f742bfce5687b9aef75506a764f61d37f8f649047846a9686ddb66"
"checksum ansi_term 0.9.0 (registry+https://github.com/rust-lang/crates.io-index)" = "23ac7c30002a5accbf7e8987d0632fa6de155b7c3d39d0067317a391e00a2ef6"
"checksum bitflags 0.7.0 (registry+https://github.com/rust-lang/crates.io-index)" = "aad18937a628ec6abcd26d1489012cc0e18c21798210f491af69ded9b881106d"
"checksum bytecount 0.1.4 (registry+https://github.com/rust-lang/crates.io-index)" = "49e3c21915578e2300b08d3c174a8ac887e0c6421dff86fdc4d741dc29e5d413"
"checksum clap 2.18.0 (registry+https://github.com/rust-lang/crates.io-index)" = "40046b8a004bf3ba43b9078bf4b9b6d1708406a234848f925dbd7160a374c8a8"
"checksum crossbeam 0.2.10 (registry+https://github.com/rust-lang/crates.io-index)" = "0c5ea215664ca264da8a9d9c3be80d2eaf30923c259d03e870388eb927508f97"
"checksum ctrlc 2.0.1 (registry+https://github.com/rust-lang/crates.io-index)" = "77f98bb69e3fefadcc5ca80a1368a55251f70295168203e01165bcaecb270891"
"checksum deque 0.3.1 (registry+https://github.com/rust-lang/crates.io-index)" = "1614659040e711785ed8ea24219140654da1729f3ec8a47a9719d041112fe7bf"
"checksum docopt 0.6.86 (registry+https://github.com/rust-lang/crates.io-index)" = "4a7ef30445607f6fc8720f0a0a2c7442284b629cf0d049286860fae23e71c4d9"
"checksum env_logger 0.3.5 (registry+https://github.com/rust-lang/crates.io-index)" = "15abd780e45b3ea4f76b4e9a26ff4843258dd8a3eed2775a0e7368c2e7936c2f"
"checksum fnv 1.0.5 (registry+https://github.com/rust-lang/crates.io-index)" = "6cc484842f1e2884faf56f529f960cc12ad8c71ce96cc7abba0a067c98fee344"
"checksum fs2 0.3.0 (registry+https://github.com/rust-lang/crates.io-index)" = "640001e1bd865c7c32806292822445af576a6866175b5225aa2087ca5e3de551"
"checksum kernel32-sys 0.2.2 (registry+https://github.com/rust-lang/crates.io-index)" = "7507624b29483431c0ba2d82aece8ca6cdba9382bff4ddd0f7490560c056098d"
"checksum lazy_static 0.2.1 (registry+https://github.com/rust-lang/crates.io-index)" = "49247ec2a285bb3dcb23cbd9c35193c025e7251bfce77c1d5da97e6362dffe7f"
"checksum lazy_static 0.2.2 (registry+https://github.com/rust-lang/crates.io-index)" = "6abe0ee2e758cd6bc8a2cd56726359007748fbf4128da998b65d0b70f881e19b"
"checksum libc 0.2.17 (registry+https://github.com/rust-lang/crates.io-index)" = "044d1360593a78f5c8e5e710beccdc24ab71d1f01bc19a29bcacdba22e8475d8"
"checksum log 0.3.6 (registry+https://github.com/rust-lang/crates.io-index)" = "ab83497bf8bf4ed2a74259c1c802351fcd67a65baa86394b6ba73c36f4838054"
"checksum memchr 0.1.11 (registry+https://github.com/rust-lang/crates.io-index)" = "d8b629fb514376c675b98c1421e80b151d3817ac42d7c667717d282761418d20"
"checksum memmap 0.5.0 (registry+https://github.com/rust-lang/crates.io-index)" = "065ce59af31c18ea2c419100bda6247dd4ec3099423202b12f0bd32e529fabd2"
"checksum num_cpus 1.1.0 (registry+https://github.com/rust-lang/crates.io-index)" = "8890e6084723d57d0df8d2720b0d60c6ee67d6c93e7169630e4371e88765dcad"
"checksum rand 0.3.14 (registry+https://github.com/rust-lang/crates.io-index)" = "2791d88c6defac799c3f20d74f094ca33b9332612d9aef9078519c82e4fe04a5"
"checksum regex 0.1.80 (registry+https://github.com/rust-lang/crates.io-index)" = "4fd4ace6a8cf7860714a2c2280d6c1f7e6a413486c13298bbc86fd3da019402f"
"checksum regex-syntax 0.3.9 (registry+https://github.com/rust-lang/crates.io-index)" = "f9ec002c35e86791825ed294b50008eea9ddfc8def4420124fbc6b08db834957"
"checksum rustc-serialize 0.3.19 (registry+https://github.com/rust-lang/crates.io-index)" = "6159e4e6e559c81bd706afe9c8fd68f547d3e851ce12e76b1de7914bab61691b"
"checksum simd 0.1.1 (registry+https://github.com/rust-lang/crates.io-index)" = "63b5847c2d766ca7ce7227672850955802fabd779ba616aeabead4c2c3877023"
"checksum strsim 0.5.1 (registry+https://github.com/rust-lang/crates.io-index)" = "50c069df92e4b01425a8bf3576d5d417943a6a7272fbabaf5bd80b1aaa76442e"
"checksum term 0.4.4 (registry+https://github.com/rust-lang/crates.io-index)" = "3deff8a2b3b6607d6d7cc32ac25c0b33709453ca9cceac006caac51e963cf94a"
"checksum term_size 0.2.1 (registry+https://github.com/rust-lang/crates.io-index)" = "3f7f5f3f71b0040cecc71af239414c23fd3c73570f5ff54cf50e03cef637f2a0"
"checksum thread-id 2.0.0 (registry+https://github.com/rust-lang/crates.io-index)" = "a9539db560102d1cef46b8b78ce737ff0bb64e7e18d35b2a5688f7d097d0ff03"
"checksum thread-id 3.0.0 (registry+https://github.com/rust-lang/crates.io-index)" = "4437c97558c70d129e40629a5b385b3fb1ffac301e63941335e4d354081ec14a"
"checksum thread_local 0.2.7 (registry+https://github.com/rust-lang/crates.io-index)" = "8576dbbfcaef9641452d5cf0df9b0e7eeab7694956dd33bb61515fb8f18cfdd5"
"checksum thread_local 0.3.0 (registry+https://github.com/rust-lang/crates.io-index)" = "50057ca52c629a39aed52d8eb253800cb727875fa6fc7c4b1445f0ac3b50c27c"
"checksum unicode-segmentation 0.1.2 (registry+https://github.com/rust-lang/crates.io-index)" = "b905d0fc2a1f0befd86b0e72e31d1787944efef9d38b9358a9e92a69757f7e3b"
"checksum unicode-width 0.1.3 (registry+https://github.com/rust-lang/crates.io-index)" = "2d6722facc10989f63ee0e20a83cd4e1714a9ae11529403ac7e0afd069abc39e"
"checksum unreachable 0.1.1 (registry+https://github.com/rust-lang/crates.io-index)" = "1f2ae5ddb18e1c92664717616dd9549dde73f539f01bd7b77c2edb2446bdff91"
"checksum utf8-ranges 0.1.3 (registry+https://github.com/rust-lang/crates.io-index)" = "a1ca13c08c41c9c3e04224ed9ff80461d97e121589ff27c753a16cb10830ae0f"
"checksum vec_map 0.6.0 (registry+https://github.com/rust-lang/crates.io-index)" = "cac5efe5cb0fa14ec2f84f83c701c562ee63f6dcc680861b21d65c682adfb05f"
"checksum void 1.0.2 (registry+https://github.com/rust-lang/crates.io-index)" = "6a02e4885ed3bc0f2de90ea6dd45ebcbb66dacffe03547fadbb0eeae2770887d"
"checksum walkdir 1.0.1 (registry+https://github.com/rust-lang/crates.io-index)" = "98da26f00240118fbb7a06fa29579d1b39d34cd6e0505ea5c125b26d5260a967"
"checksum winapi 0.2.8 (registry+https://github.com/rust-lang/crates.io-index)" = "167dc9d6949a9b857f3451275e911c3f44255842c1f7a76f33c55103a909087a"
"checksum winapi-build 0.1.1 (registry+https://github.com/rust-lang/crates.io-index)" = "2d315eee3b34aca4797b2da6b13ed88266e6d612562a0c46390af8299fc699bc"

25
Cargo.toml
View File

@@ -1,6 +1,6 @@
[package]
name = "ripgrep"
version = "0.2.3" #:version
version = "0.2.9" #:version
authors = ["Andrew Gallant <jamslam@gmail.com>"]
description = """
Line oriented search tool using Rust's regex library. Combines the raw
@@ -13,6 +13,7 @@ readme = "README.md"
keywords = ["regex", "grep", "egrep", "search", "pattern"]
license = "Unlicense/MIT"
exclude = ["HomebrewFormula"]
build = "build.rs"
[[bin]]
bench = false
@@ -24,12 +25,12 @@ name = "integration"
path = "tests/tests.rs"
[dependencies]
bytecount = "0.1.4"
clap = "2.18"
ctrlc = "2.0"
deque = "0.3"
docopt = "0.6"
env_logger = "0.3"
grep = { version = "0.1.3", path = "grep" }
ignore = { version = "0.1.0", path = "ignore" }
grep = { version = "0.1.4", path = "grep" }
ignore = { version = "0.1.5", path = "ignore" }
lazy_static = "0.2"
libc = "0.2"
log = "0.3"
@@ -37,15 +38,19 @@ memchr = "0.1"
memmap = "0.5"
num_cpus = "1"
regex = "0.1.77"
rustc-serialize = "0.3"
term = "0.4"
termcolor = { version = "0.1.0", path = "termcolor" }
[target.'cfg(windows)'.dependencies]
kernel32-sys = "0.2"
winapi = "0.2"
kernel32-sys = "0.2.2"
winapi = "0.2.8"
[build-dependencies]
clap = "2.18"
lazy_static = "0.2"
[features]
simd-accel = ["regex/simd-accel"]
avx-accel = ["bytecount/avx-accel"]
simd-accel = ["bytecount/simd-accel", "regex/simd-accel"]
[profile.release]
debug = true

49
README.md
View File

@@ -15,11 +15,12 @@ Dual-licensed under MIT or the [UNLICENSE](http://unlicense.org).
[![A screenshot of a sample search with ripgrep](http://burntsushi.net/stuff/ripgrep1.png)](http://burntsushi.net/stuff/ripgrep1.png)
### Quick example comparing tools
### Quick examples comparing tools
This example searches the entire Linux kernel source tree (after running
`make defconfig && make -j8`) for `[A-Z]+_SUSPEND`, where all matches must be
words. Timings were collected on a system with an Intel i7-6900K 3.2 GHz.
words. Timings were collected on a system with an Intel i7-6900K 3.2 GHz, and
ripgrep was compiled using the `compile` script in this repo.
Please remember that a single benchmark is never enough! See my
[blog post on `ripgrep`](http://blog.burntsushi.net/ripgrep/)
@@ -27,17 +28,41 @@ for a very detailed comparison with more benchmarks and analysis.
| Tool | Command | Line count | Time |
| ---- | ------- | ---------- | ---- |
| ripgrep | `rg -n -w '[A-Z]+_SUSPEND'` | 450 | **0.245s** |
| ripgrep (Unicode) | `rg -n -w '[A-Z]+_SUSPEND'` | 450 | **0.134s** |
| [The Silver Searcher](https://github.com/ggreer/the_silver_searcher) | `ag -w '[A-Z]+_SUSPEND'` | 450 | 0.753s |
| [git grep](https://www.kernel.org/pub/software/scm/git/docs/git-grep.html) | `LC_ALL=C git grep -E -n -w '[A-Z]+_SUSPEND'` | 450 | 0.823s |
| [git grep (Unicode)](https://www.kernel.org/pub/software/scm/git/docs/git-grep.html) | `LC_ALL=en_US.UTF-8 git grep -E -n -w '[A-Z]+_SUSPEND'` | 450 | 2.880s |
| [sift](https://github.com/svent/sift) | `sift --git -n -w '[A-Z]+_SUSPEND'` | 450 | 3.656s |
| [The Platinum Searcher](https://github.com/monochromegane/the_platinum_searcher) | `pt -w -e '[A-Z]+_SUSPEND'` | 450 | 12.369s |
| [ack](http://beyondgrep.com/) | `ack -w '[A-Z]+_SUSPEND'` | 1878 | 16.952s |
| [ack](https://github.com/petdance/ack2) | `ack -w '[A-Z]+_SUSPEND'` | 1878 | 16.952s |
(Yes, `ack` [has](https://github.com/petdance/ack2/issues/445) a
[bug](https://github.com/petdance/ack2/issues/14).)
Here's another benchmark that disregards gitignore files and searches with a
whitelist instead. The corpus is the same as in the previous benchmark, and the
flags passed to each command ensures that they are doing equivalent work:
| Tool | Command | Line count | Time |
| ---- | ------- | ---------- | ---- |
| ripgrep | `rg -L -u -tc -n -w '[A-Z]+_SUSPEND'` | 404 | **0.108s** |
| [ucg](https://github.com/gvansickle/ucg) | `ucg --type=cc -w '[A-Z]+_SUSPEND'` | 392 | 0.219s |
| [GNU grep](https://www.gnu.org/software/grep/) | `egrep -R -n --include='*.c' --include='*.h' -w '[A-Z]+_SUSPEND'` | 404 | 0.733s |
(`ucg` [has slightly different behavior in the presence of symbolic links](https://github.com/gvansickle/ucg/issues/106).)
And finally, a straight up comparison between ripgrep and GNU grep on a single
large file (~9.3GB,
[`OpenSubtitles2016.raw.en.gz`](http://opus.lingfil.uu.se/OpenSubtitles2016/mono/OpenSubtitles2016.raw.en.gz)):
| Tool | Command | Line count | Time |
| ---- | ------- | ---------- | ---- |
| ripgrep | `rg -w 'Sherlock [A-Z]\w+'` | 5268 | **2.520s** |
| [GNU grep](https://www.gnu.org/software/grep/) | `LC_ALL=C egrep -w 'Sherlock [A-Z]\w+'` | 5268 | 7.143s |
In the above benchmark, passing the `-n` flag (for showing line numbers)
increases the times to `3.081s` for ripgrep and `11.403s` for GNU grep.
### Why should I use `ripgrep`?
* It can replace both The Silver Searcher and GNU grep because it is faster
@@ -82,8 +107,9 @@ Summarizing, `ripgrep` is fast because:
[`RegexSet`](https://doc.rust-lang.org/regex/regex/struct.RegexSet.html).
That means a single file path can be matched against multiple glob patterns
simultaneously.
* Uses a Chase-Lev work-stealing queue for quickly distributing work to
multiple threads.
* It uses a lock-free parallel recursive directory iterator, courtesy of
[`crossbeam`](https://docs.rs/crossbeam) and
[`ignore`](https://docs.rs/ignore).
### Installation
@@ -118,6 +144,12 @@ If you're an **Arch Linux** user, then you can install `ripgrep` from the offici
$ pacman -S ripgrep
```
If you're a **Gentoo** user, you can install `ripgrep` from the [official repo](https://packages.gentoo.org/packages/sys-apps/ripgrep):
```
$ emerge ripgrep
```
If you're a **Fedora 24+** user, you can install `ripgrep` from [copr](https://copr.fedorainfracloud.org/coprs/carlgeorge/ripgrep/):
```
@@ -276,9 +308,12 @@ If you have a Rust nightly compiler, then you can enable optional SIMD
acceleration like so:
```
RUSTFLAGS="-C target-cpu=native" cargo build --release --features simd-accel
RUSTFLAGS="-C target-cpu=native" cargo build --release --features 'simd-accel avx-accel'
```
If your machine doesn't support AVX instructions, then simply remove
`avx-accel` from the features list. Similarly for SIMD.
### Running tests
`ripgrep` is relatively well tested, including both unit tests and integration

3
appveyor.yml
View File

@@ -31,6 +31,8 @@ test_script:
- cargo test --verbose --manifest-path grep/Cargo.toml
- cargo test --verbose --manifest-path globset/Cargo.toml
- cargo test --verbose --manifest-path ignore/Cargo.toml
- cargo test --verbose --manifest-path wincolor/Cargo.toml
- cargo test --verbose --manifest-path termcolor/Cargo.toml
before_deploy:
# Generate artifacts for release
@@ -60,6 +62,7 @@ deploy:
branches:
only:
- /\d+\.\d+\.\d+/
- master
# - appveyor
# - /\d+\.\d+\.\d+/

23
build.rs Normal file
View File

@@ -0,0 +1,23 @@
#[macro_use]
extern crate clap;
#[macro_use]
extern crate lazy_static;
use std::fs;
use clap::Shell;
#[allow(dead_code)]
#[path = "src/app.rs"]
mod app;
fn main() {
fs::create_dir_all(env!("OUT_DIR")).unwrap();
let mut app = app::app_short();
app.gen_completions("rg", Shell::Bash, env!("OUT_DIR"));
app.gen_completions("rg", Shell::Fish, env!("OUT_DIR"));
// Zsh seems to fail with a panic.
// app.gen_completions("rg", Shell::Zsh, env!("OUT_DIR"));
app.gen_completions("rg", Shell::PowerShell, env!("OUT_DIR"));
}

1
ci/before_deploy.sh
View File

@@ -19,6 +19,7 @@ mk_tarball() {
cp target/$TARGET/release/rg "$td/$name/"
cp {doc/rg.1,README.md,UNLICENSE,COPYING,LICENSE-MIT} "$td/$name/"
cp target/$TARGET/release/build/ripgrep-*/out/{_rg.,rg.}* "$td/$name/"
pushd $td
tar czf "$out_dir/$name.tar.gz" *

2
ci/script.sh
View File

@@ -25,6 +25,8 @@ run_test_suite() {
cargo test --target $TARGET --verbose --manifest-path globset/Cargo.toml
cargo build --target $TARGET --verbose --manifest-path ignore/Cargo.toml
cargo test --target $TARGET --verbose --manifest-path ignore/Cargo.toml
cargo build --target $TARGET --verbose --manifest-path termcolor/Cargo.toml
cargo test --target $TARGET --verbose --manifest-path termcolor/Cargo.toml
# sanity check the file type
file target/$TARGET/debug/rg

6
compile
View File

@@ -1,5 +1,5 @@
#!/bin/sh
export RUSTFLAGS="-C target-feature=+ssse3"
# export RUSTFLAGS="-C target-cpu=native"
cargo build --release --features simd-accel
# export RUSTFLAGS="-C target-feature=+ssse3"
export RUSTFLAGS="-C target-cpu=native"
cargo build --release --features 'simd-accel avx-accel'

69
doc/rg.1
View File

@@ -1,4 +1,4 @@
.\" Automatically generated by Pandoc 1.17.2
.\" Automatically generated by Pandoc 1.18
.\"
.TH "rg" "1"
.hy
@@ -7,11 +7,11 @@
rg \- recursively search current directory for lines matching a pattern
.SH SYNOPSIS
.PP
rg [\f[I]options\f[]] \-e PATTERN ...
[\f[I]<\f[]path\f[I]> ...\f[]]
.PP
rg [\f[I]options\f[]] <\f[I]pattern\f[]> [\f[I]<\f[]path\f[I]> ...\f[]]
.PP
rg [\f[I]options\f[]] (\-e PATTERN | \-f FILE) ...
[\f[I]<\f[]path\f[I]> ...\f[]]
.PP
rg [\f[I]options\f[]] \-\-files [\f[I]<\f[]path\f[I]> ...\f[]]
.PP
rg [\f[I]options\f[]] \-\-type\-list
@@ -21,8 +21,10 @@ rg [\f[I]options\f[]] \-\-help
rg [\f[I]options\f[]] \-\-version
.SH DESCRIPTION
.PP
rg (ripgrep) combines the usability of The Silver Searcher (an ack
ripgrep (rg) combines the usability of The Silver Searcher (an ack
clone) with the raw speed of grep.
.PP
Project home page: https://github.com/BurntSushi/ripgrep
.SH COMMON OPTIONS
.TP
.B \-a, \-\-text
@@ -46,6 +48,7 @@ Valid values are never, always or auto.
Use PATTERN to search.
This option can be provided multiple times, where all patterns given are
searched.
This is also useful when searching for patterns that start with a dash.
.RS
.RE
.TP
@@ -140,6 +143,28 @@ Show NUM lines before and after each match.
.RS
.RE
.TP
.B \-\-colors \f[I]SPEC\f[] ...
This flag specifies color settings for use in the output.
This flag may be provided multiple times.
Settings are applied iteratively.
Colors are limited to one of eight choices: red, blue, green, cyan,
magenta, yellow, white and black.
Styles are limited to either nobold or bold.
.RS
.PP
The format of the flag is {type}:{attribute}:{value}.
{type} should be one of path, line or match.
{attribute} can be fg, bg or style.
Value is either a color (for fg and bg) or a text style.
A special format, {type}:none, will clear all color settings for {type}.
.PP
For example, the following command will change the match color to
magenta and the background color for line numbers to yellow:
.PP
rg \-\-colors \[aq]match:fg:magenta\[aq] \-\-colors
\[aq]line:bg:yellow\[aq] foo.
.RE
.TP
.B \-\-column
Show column numbers (1 based) in output.
This only shows the column numbers for the first match on each line.
@@ -160,6 +185,15 @@ Show debug messages.
.RS
.RE
.TP
.B \-f, \-\-file FILE ...
Search for patterns from the given file, with one pattern per line.
When this flag is used or multiple times or in combination with the
\-e/\-\-regexp flag, then all patterns provided are searched.
Empty pattern lines will match all input lines, and the newline is not
counted as part of the pattern.
.RS
.RE
.TP
.B \-\-files
Print each file that would be searched (but don\[aq]t search).
.RS
@@ -170,6 +204,11 @@ Only show path of each file with matches.
.RS
.RE
.TP
.B \-\-files\-without\-match
Only show path of each file with no matches.
.RS
.RE
.TP
.B \-H, \-\-with\-filename
Prefix each match with the file name that contains it.
This is the default when more than one file is searched.
@@ -199,11 +238,26 @@ Search hidden directories and files.
.RS
.RE
.TP
.B \-\-ignore\-file FILE ...
Specify additional ignore files for filtering file paths.
Ignore files should be in the gitignore format and are matched relative
to the current working directory.
These ignore files have lower precedence than all other ignore files.
When specifying multiple ignore files, earlier files have lower
precedence than later files.
.RS
.RE
.TP
.B \-L, \-\-follow
Follow symlinks.
.RS
.RE
.TP
.B \-m, \-\-max\-count NUM
Limit the number of matching lines per file searched to NUM.
.RS
.RE
.TP
.B \-\-maxdepth \f[I]NUM\f[]
Descend at most NUM directories below the command line arguments.
A value of zero searches only the starting\-points themselves.
@@ -218,6 +272,11 @@ context related options.)
.RS
.RE
.TP
.B \-\-no\-messages
Suppress all error messages.
.RS
.RE
.TP
.B \-\-no\-mmap
Never use memory maps, even when they might be faster.
.RS

50
doc/rg.1.md
View File

@@ -4,10 +4,10 @@ rg - recursively search current directory for lines matching a pattern
# SYNOPSIS
rg [*options*] -e PATTERN ... [*<*path*> ...*]
rg [*options*] <*pattern*> [*<*path*> ...*]
rg [*options*] (-e PATTERN | -f FILE) ... [*<*path*> ...*]
rg [*options*] --files [*<*path*> ...*]
rg [*options*] --type-list
@@ -18,9 +18,11 @@ rg [*options*] --version
# DESCRIPTION
rg (ripgrep) combines the usability of The Silver Searcher (an ack clone) with
ripgrep (rg) combines the usability of The Silver Searcher (an ack clone) with
the raw speed of grep.
Project home page: https://github.com/BurntSushi/ripgrep
# COMMON OPTIONS
-a, --text
@@ -35,7 +37,8 @@ the raw speed of grep.
-e, --regexp *PATTERN* ...
: Use PATTERN to search. This option can be provided multiple times, where all
patterns given are searched.
patterns given are searched. This is also useful when searching for patterns
that start with a dash.
-F, --fixed-strings
: Treat the pattern as a literal string instead of a regular expression.
@@ -92,6 +95,22 @@ the raw speed of grep.
-C, --context *NUM*
: Show NUM lines before and after each match.
--colors *SPEC* ...
: This flag specifies color settings for use in the output. This flag may be
provided multiple times. Settings are applied iteratively. Colors are limited
to one of eight choices: red, blue, green, cyan, magenta, yellow, white and
black. Styles are limited to either nobold or bold.
The format of the flag is {type}:{attribute}:{value}. {type} should be one
of path, line or match. {attribute} can be fg, bg or style. Value is either
a color (for fg and bg) or a text style. A special format, {type}:none,
will clear all color settings for {type}.
For example, the following command will change the match color to magenta
and the background color for line numbers to yellow:
rg --colors 'match:fg:magenta' --colors 'line:bg:yellow' foo.
--column
: Show column numbers (1 based) in output. This only shows the column
numbers for the first match on each line. Note that this doesn't try
@@ -104,12 +123,21 @@ the raw speed of grep.
--debug
: Show debug messages.
-f, --file FILE ...
: Search for patterns from the given file, with one pattern per line. When this
flag is used or multiple times or in combination with the -e/--regexp flag,
then all patterns provided are searched. Empty pattern lines will match all
input lines, and the newline is not counted as part of the pattern.
--files
: Print each file that would be searched (but don't search).
-l, --files-with-matches
: Only show path of each file with matches.
--files-without-match
: Only show path of each file with no matches.
-H, --with-filename
: Prefix each match with the file name that contains it. This is the
default when more than one file is searched.
@@ -129,9 +157,20 @@ the raw speed of grep.
: Search hidden directories and files. (Hidden directories and files are
skipped by default.)
--ignore-file FILE ...
: Specify additional ignore files for filtering file paths.
Ignore files should be in the gitignore format and are matched
relative to the current working directory. These ignore files
have lower precedence than all other ignore files. When
specifying multiple ignore files, earlier files have lower
precedence than later files.
-L, --follow
: Follow symlinks.
-m, --max-count NUM
: Limit the number of matching lines per file searched to NUM.
--maxdepth *NUM*
: Descend at most NUM directories below the command line arguments.
A value of zero searches only the starting-points themselves.
@@ -141,6 +180,9 @@ the raw speed of grep.
when ripgrep thinks it will be faster. (Note that mmap searching
doesn't currently support the various context related options.)
--no-messages
: Suppress all error messages.
--no-mmap
: Never use memory maps, even when they might be faster.

2
globset/Cargo.toml
View File

@@ -1,6 +1,6 @@
[package]
name = "globset"
version = "0.1.1" #:version
version = "0.1.2" #:version
authors = ["Andrew Gallant <jamslam@gmail.com>"]
description = """
Cross platform single glob and glob set matching. Glob set matching is the

7
globset/src/lib.rs
View File

@@ -130,13 +130,6 @@ pub use glob::{Glob, GlobBuilder, GlobMatcher};
mod glob;
mod pathutil;
macro_rules! eprintln {
($($tt:tt)*) => {{
use std::io::Write;
let _ = writeln!(&mut ::std::io::stderr(), $($tt)*);
}}
}
/// Represents an error that can occur when parsing a glob pattern.
#[derive(Clone, Debug, Eq, PartialEq)]
pub enum Error {

2
grep/Cargo.toml
View File

@@ -1,6 +1,6 @@
[package]
name = "grep"
version = "0.1.3" #:version
version = "0.1.4" #:version
authors = ["Andrew Gallant <jamslam@gmail.com>"]
description = """
Fast line oriented regex searching as a library.

10
grep/src/literals.rs
View File

@@ -9,7 +9,7 @@ principled.
*/
use std::cmp;
use regex::bytes::Regex;
use regex::bytes::RegexBuilder;
use syntax::{
Expr, Literals, Lit,
ByteClass, ByteRange, CharClass, ClassRange, Repeater,
@@ -33,7 +33,7 @@ impl LiteralSets {
}
}
pub fn to_regex(&self) -> Option<Regex> {
pub fn to_regex_builder(&self) -> Option<RegexBuilder> {
if self.prefixes.all_complete() && !self.prefixes.is_empty() {
debug!("literal prefixes detected: {:?}", self.prefixes);
// When this is true, the regex engine will do a literal scan.
@@ -79,14 +79,12 @@ impl LiteralSets {
debug!("required literals found: {:?}", req_lits);
let alts: Vec<String> =
req_lits.into_iter().map(|x| bytes_to_regex(x)).collect();
// Literals always compile.
Some(Regex::new(&alts.join("|")).unwrap())
Some(RegexBuilder::new(&alts.join("|")))
} else if lit.is_empty() {
None
} else {
// Literals always compile.
debug!("required literal found: {:?}", show(lit));
Some(Regex::new(&bytes_to_regex(lit)).unwrap())
Some(RegexBuilder::new(&bytes_to_regex(lit)))
}
}
}

52
grep/src/search.rs
View File

@@ -144,14 +144,19 @@ impl GrepBuilder {
let expr = try!(self.parse());
let literals = LiteralSets::create(&expr);
let re = try!(self.regex(&expr));
let required = literals.to_regex().or_else(|| {
let expr = match strip_unicode_word_boundaries(&expr) {
None => return None,
Some(expr) => expr,
};
debug!("Stripped Unicode word boundaries. New AST:\n{:?}", expr);
self.regex(&expr).ok()
});
let required = match literals.to_regex_builder() {
Some(builder) => Some(try!(self.regex_build(builder))),
None => {
match strip_unicode_word_boundaries(&expr) {
None => None,
Some(expr) => {
debug!("Stripped Unicode word boundaries. \
New AST:\n{:?}", expr);
self.regex(&expr).ok()
}
}
}
};
Ok(Grep {
re: re,
required: required,
@@ -162,11 +167,12 @@ impl GrepBuilder {
/// Creates a new regex from the given expression with the current
/// configuration.
fn regex(&self, expr: &Expr) -> Result<Regex> {
let casei =
self.opts.case_insensitive
|| (self.opts.case_smart && !has_uppercase_literal(expr));
RegexBuilder::new(&expr.to_string())
.case_insensitive(casei)
self.regex_build(RegexBuilder::new(&expr.to_string()))
}
/// Builds a new regex from the given builder using the caller's settings.
fn regex_build(&self, builder: RegexBuilder) -> Result<Regex> {
builder
.multi_line(true)
.unicode(true)
.size_limit(self.opts.size_limit)
@@ -182,12 +188,30 @@ impl GrepBuilder {
try!(syntax::ExprBuilder::new()
.allow_bytes(true)
.unicode(true)
.case_insensitive(self.opts.case_insensitive)
.case_insensitive(try!(self.is_case_insensitive()))
.parse(&self.pattern));
let expr = try!(nonl::remove(expr, self.opts.line_terminator));
debug!("regex ast:\n{:#?}", expr);
Ok(expr)
}
/// Determines whether the case insensitive flag should be enabled or not.
///
/// An error is returned if the regex could not be parsed.
fn is_case_insensitive(&self) -> Result<bool> {
if self.opts.case_insensitive {
return Ok(true);
}
if !self.opts.case_smart {
return Ok(false);
}
let expr =
try!(syntax::ExprBuilder::new()
.allow_bytes(true)
.unicode(true)
.parse(&self.pattern));
Ok(!has_uppercase_literal(&expr))
}
}
impl Grep {

5
ignore/Cargo.toml
View File

@@ -1,6 +1,6 @@
[package]
name = "ignore"
version = "0.1.1" #:version
version = "0.1.5" #:version
authors = ["Andrew Gallant <jamslam@gmail.com>"]
description = """
A fast library for efficiently matching ignore files such as `.gitignore`
@@ -18,7 +18,8 @@ name = "ignore"
bench = false
[dependencies]
globset = { version = "0.1.1", path = "../globset" }
crossbeam = "0.2"
globset = { version = "0.1.2", path = "../globset" }
lazy_static = "0.2"
log = "0.3"
memchr = "0.1"

98
ignore/examples/walk.rs
View File

@@ -1,28 +1,92 @@
/*
#![allow(dead_code, unused_imports, unused_mut, unused_variables)]
extern crate crossbeam;
extern crate ignore;
extern crate walkdir;
use std::env;
use std::io::{self, Write};
use std::os::unix::ffi::OsStrExt;
use std::path::Path;
use std::sync::Arc;
use std::sync::atomic::{AtomicUsize, Ordering};
use std::thread;
use ignore::ignore::IgnoreBuilder;
use crossbeam::sync::MsQueue;
use ignore::WalkBuilder;
use walkdir::WalkDir;
fn main() {
let path = env::args().nth(1).unwrap();
let ig = IgnoreBuilder::new().build();
let wd = WalkDir::new(path);
let walker = ignore::walk::Iter::new(ig, wd);
let mut stdout = io::BufWriter::new(io::stdout());
// let mut count = 0;
for dirent in walker {
// count += 1;
stdout.write(dirent.path().as_os_str().as_bytes()).unwrap();
stdout.write(b"\n").unwrap();
let mut path = env::args().nth(1).unwrap();
let mut parallel = false;
let mut simple = false;
let queue: Arc<MsQueue<Option<DirEntry>>> = Arc::new(MsQueue::new());
if path == "parallel" {
path = env::args().nth(2).unwrap();
parallel = true;
} else if path == "walkdir" {
path = env::args().nth(2).unwrap();
simple = true;
}
// println!("{}", count);
let stdout_queue = queue.clone();
let stdout_thread = thread::spawn(move || {
let mut stdout = io::BufWriter::new(io::stdout());
while let Some(dent) = stdout_queue.pop() {
write_path(&mut stdout, dent.path());
}
});
if parallel {
let walker = WalkBuilder::new(path).threads(6).build_parallel();
walker.run(|| {
let queue = queue.clone();
Box::new(move |result| {
use ignore::WalkState::*;
queue.push(Some(DirEntry::Y(result.unwrap())));
Continue
})
});
} else if simple {
let mut stdout = io::BufWriter::new(io::stdout());
let walker = WalkDir::new(path);
for result in walker {
queue.push(Some(DirEntry::X(result.unwrap())));
}
} else {
let mut stdout = io::BufWriter::new(io::stdout());
let walker = WalkBuilder::new(path).build();
for result in walker {
queue.push(Some(DirEntry::Y(result.unwrap())));
}
}
queue.push(None);
stdout_thread.join().unwrap();
}
enum DirEntry {
X(walkdir::DirEntry),
Y(ignore::DirEntry),
}
impl DirEntry {
fn path(&self) -> &Path {
match *self {
DirEntry::X(ref x) => x.path(),
DirEntry::Y(ref y) => y.path(),
}
}
}
#[cfg(unix)]
fn write_path<W: Write>(mut wtr: W, path: &Path) {
use std::os::unix::ffi::OsStrExt;
wtr.write(path.as_os_str().as_bytes()).unwrap();
wtr.write(b"\n").unwrap();
}
#[cfg(not(unix))]
fn write_path<W: Write>(mut wtr: W, path: &Path) {
wtr.write(path.to_string_lossy().as_bytes()).unwrap();
wtr.write(b"\n").unwrap();
}
*/
fn main() {}

31
ignore/src/dir.rs
View File

@@ -75,7 +75,7 @@ struct IgnoreOptions {
impl IgnoreOptions {
/// Returns true if at least one type of ignore rules should be matched.
fn should_ignores(&self) -> bool {
fn has_any_ignore_options(&self) -> bool {
self.ignore || self.git_global || self.git_ignore || self.git_exclude
}
}
@@ -137,6 +137,11 @@ impl Ignore {
self.0.parent.is_none()
}
/// Returns true if this matcher was added via the `add_parents` method.
pub fn is_absolute_parent(&self) -> bool {
self.0.is_absolute_parent
}
/// Return this matcher's parent, if one exists.
pub fn parent(&self) -> Option<Ignore> {
self.0.parent.clone()
@@ -280,7 +285,7 @@ impl Ignore {
}
}
let mut whitelisted = Match::None;
if self.0.opts.should_ignores() {
if self.0.opts.has_any_ignore_options() {
let mat = self.matched_ignore(path, is_dir);
if mat.is_ignore() {
return mat;
@@ -360,23 +365,12 @@ impl Ignore {
}
let m_global = self.0.git_global_matcher.matched(&path, is_dir)
.map(IgnoreMatch::gitignore);
if !m_ignore.is_none() {
m_ignore
} else if !m_gi.is_none() {
m_gi
} else if !m_gi_exclude.is_none() {
m_gi_exclude
} else if !m_global.is_none() {
m_global
} else if !m_explicit.is_none() {
m_explicit
} else {
Match::None
}
m_ignore.or(m_gi).or(m_gi_exclude).or(m_global).or(m_explicit)
}
/// Returns an iterator over parent ignore matchers, including this one.
fn parents(&self) -> Parents {
pub fn parents(&self) -> Parents {
Parents(Some(self))
}
@@ -387,7 +381,10 @@ impl Ignore {
}
}
struct Parents<'a>(Option<&'a Ignore>);
/// An iterator over all parents of an ignore matcher, including itself.
///
/// The lifetime `'a` refers to the lifetime of the initial `Ignore` matcher.
pub struct Parents<'a>(Option<&'a Ignore>);
impl<'a> Iterator for Parents<'a> {
type Item = &'a Ignore;

8
ignore/src/gitignore.rs
View File

@@ -311,7 +311,7 @@ impl GitignoreBuilder {
Ok(line) => line,
Err(err) => {
errs.push(Error::Io(err).tagged(path, lineno));
continue;
break;
}
};
if let Err(err) = self.add_line(Some(path.to_path_buf()), &line) {
@@ -453,9 +453,9 @@ fn gitconfig_contents() -> Option<Vec<u8>> {
/// Specifically, this respects XDG_CONFIG_HOME.
fn excludes_file_default() -> Option<PathBuf> {
env::var_os("XDG_CONFIG_HOME")
.and_then(|x| if x.is_empty() { None } else { Some(x) })
.or_else(|| env::var_os("HOME"))
.map(|x| PathBuf::from(x).join("git/ignore"))
.and_then(|x| if x.is_empty() { None } else { Some(PathBuf::from(x)) })
.or_else(|| env::home_dir())
.map(|x| x.join("git/ignore"))
}
/// Extract git's `core.excludesfile` config setting from the raw file contents

71
ignore/src/lib.rs
View File

@@ -44,6 +44,7 @@ for result in WalkBuilder::new("./").hidden(false).build() {
See the documentation for `WalkBuilder` for many other options.
*/
extern crate crossbeam;
extern crate globset;
#[macro_use]
extern crate lazy_static;
@@ -61,7 +62,7 @@ use std::fmt;
use std::io;
use std::path::{Path, PathBuf};
pub use walk::{DirEntry, Walk, WalkBuilder};
pub use walk::{DirEntry, Walk, WalkBuilder, WalkParallel, WalkState};
mod dir;
pub mod gitignore;
@@ -80,6 +81,12 @@ pub enum Error {
WithLineNumber { line: u64, err: Box<Error> },
/// An error associated with a particular file path.
WithPath { path: PathBuf, err: Box<Error> },
/// An error associated with a particular directory depth when recursively
/// walking a directory.
WithDepth { depth: usize, err: Box<Error> },
/// An error that occurs when a file loop is detected when traversing
/// symbolic links.
Loop { ancestor: PathBuf, child: PathBuf },
/// An error that occurs when doing I/O, such as reading an ignore file.
Io(io::Error),
/// An error that occurs when trying to parse a glob.
@@ -101,6 +108,7 @@ impl Error {
Error::Partial(_) => true,
Error::WithLineNumber { ref err, .. } => err.is_partial(),
Error::WithPath { ref err, .. } => err.is_partial(),
Error::WithDepth { ref err, .. } => err.is_partial(),
_ => false,
}
}
@@ -111,6 +119,8 @@ impl Error {
Error::Partial(ref errs) => errs.len() == 1 && errs[0].is_io(),
Error::WithLineNumber { ref err, .. } => err.is_io(),
Error::WithPath { ref err, .. } => err.is_io(),
Error::WithDepth { ref err, .. } => err.is_io(),
Error::Loop { .. } => false,
Error::Io(_) => true,
Error::Glob(_) => false,
Error::UnrecognizedFileType(_) => false,
@@ -118,6 +128,16 @@ impl Error {
}
}
/// Returns a depth associated with recursively walking a directory (if
/// this error was generated from a recursive directory iterator).
pub fn depth(&self) -> Option<usize> {
match *self {
Error::WithPath { ref err, .. } => err.depth(),
Error::WithDepth { depth, .. } => Some(depth),
_ => None,
}
}
/// Turn an error into a tagged error with the given file path.
fn with_path<P: AsRef<Path>>(self, path: P) -> Error {
Error::WithPath {
@@ -126,6 +146,14 @@ impl Error {
}
}
/// Turn an error into a tagged error with the given depth.
fn with_depth(self, depth: usize) -> Error {
Error::WithDepth {
depth: depth,
err: Box::new(self),
}
}
/// Turn an error into a tagged error with the given file path and line
/// number. If path is empty, then it is omitted from the error.
fn tagged<P: AsRef<Path>>(self, path: P, lineno: u64) -> Error {
@@ -146,6 +174,8 @@ impl error::Error for Error {
Error::Partial(_) => "partial error",
Error::WithLineNumber { ref err, .. } => err.description(),
Error::WithPath { ref err, .. } => err.description(),
Error::WithDepth { ref err, .. } => err.description(),
Error::Loop { .. } => "file system loop found",
Error::Io(ref err) => err.description(),
Error::Glob(ref msg) => msg,
Error::UnrecognizedFileType(_) => "unrecognized file type",
@@ -168,6 +198,12 @@ impl fmt::Display for Error {
Error::WithPath { ref path, ref err } => {
write!(f, "{}: {}", path.display(), err)
}
Error::WithDepth { ref err, .. } => err.fmt(f),
Error::Loop { ref ancestor, ref child } => {
write!(f, "File system loop found: \
{} points to an ancestor {}",
child.display(), ancestor.display())
}
Error::Io(ref err) => err.fmt(f),
Error::Glob(ref msg) => write!(f, "{}", msg),
Error::UnrecognizedFileType(ref ty) => {
@@ -187,6 +223,30 @@ impl From<io::Error> for Error {
}
}
impl From<walkdir::Error> for Error {
fn from(err: walkdir::Error) -> Error {
let depth = err.depth();
if let (Some(anc), Some(child)) = (err.loop_ancestor(), err.path()) {
return Error::WithDepth {
depth: depth,
err: Box::new(Error::Loop {
ancestor: anc.to_path_buf(),
child: child.to_path_buf(),
}),
};
}
let path = err.path().map(|p| p.to_path_buf());
let mut ig_err = Error::Io(io::Error::from(err));
if let Some(path) = path {
ig_err = Error::WithPath {
path: path,
err: Box::new(ig_err),
};
}
ig_err
}
}
#[derive(Debug, Default)]
struct PartialErrorBuilder(Vec<Error>);
@@ -297,4 +357,13 @@ impl<T> Match<T> {
Match::Whitelist(t) => Match::Whitelist(f(t)),
}
}
/// Return the match if it is not none. Otherwise, return other.
pub fn or(self, other: Self) -> Self {
if self.is_none() {
other
} else {
self
}
}
}

23
ignore/src/overrides.rs
View File

@@ -79,8 +79,9 @@ impl Override {
///
/// If there are no overrides, then this always returns `Match::None`.
///
/// If there is at least one whitelist override, then this never returns
/// `Match::None`, since non-matches are interpreted as ignored.
/// If there is at least one whitelist override and `is_dir` is false, then
/// this never returns `Match::None`, since non-matches are interpreted as
/// ignored.
///
/// The given path is matched to the globs relative to the path given
/// when building the override matcher. Specifically, before matching
@@ -97,7 +98,7 @@ impl Override {
return Match::None;
}
let mat = self.0.matched(path, is_dir).invert();
if mat.is_none() && self.num_whitelists() > 0 {
if mat.is_none() && self.num_whitelists() > 0 && !is_dir {
return Match::Ignore(Glob::unmatched());
}
mat.map(move |giglob| Glob(GlobInner::Matched(giglob)))
@@ -166,7 +167,7 @@ mod tests {
assert!(ov.matched("a.foo", false).is_whitelist());
assert!(ov.matched("a.foo", true).is_whitelist());
assert!(ov.matched("a.rs", false).is_ignore());
assert!(ov.matched("a.rs", true).is_ignore());
assert!(ov.matched("a.rs", true).is_none());
assert!(ov.matched("a.bar", false).is_ignore());
assert!(ov.matched("a.bar", true).is_ignore());
}
@@ -199,4 +200,18 @@ mod tests {
assert!(ov.matched("baz/a", false).is_whitelist());
assert!(ov.matched("baz/a/b", false).is_whitelist());
}
#[test]
fn allow_directories() {
// This tests that directories are NOT ignored when they are unmatched.
let ov = ov(&["*.rs"]);
assert!(ov.matched("foo.rs", false).is_whitelist());
assert!(ov.matched("foo.c", false).is_ignore());
assert!(ov.matched("foo", false).is_ignore());
assert!(ov.matched("foo", true).is_none());
assert!(ov.matched("src/foo.rs", false).is_whitelist());
assert!(ov.matched("src/foo.c", false).is_ignore());
assert!(ov.matched("src/foo", false).is_ignore());
assert!(ov.matched("src/foo", true).is_none());
}
}

22
ignore/src/types.rs
View File

@@ -80,6 +80,8 @@ use pathutil::file_name;
use {Error, Match};
const DEFAULT_TYPES: &'static [(&'static str, &'static [&'static str])] = &[
("agda", &["*.agda", "*.lagda"]),
("asciidoc", &["*.adoc", "*.asc", "*.asciidoc"]),
("asm", &["*.asm", "*.s", "*.S"]),
("awk", &["*.awk"]),
("c", &["*.c", "*.h", "*.H"]),
@@ -87,18 +89,22 @@ const DEFAULT_TYPES: &'static [(&'static str, &'static [&'static str])] = &[
("clojure", &["*.clj", "*.cljc", "*.cljs", "*.cljx"]),
("cmake", &["*.cmake", "CMakeLists.txt"]),
("coffeescript", &["*.coffee"]),
("creole", &["*.creole"]),
("config", &["*.config"]),
("cpp", &[
"*.C", "*.cc", "*.cpp", "*.cxx",
"*.h", "*.H", "*.hh", "*.hpp",
]),
("cs", &["*.cs"]),
("csharp", &["*.cs"]),
("css", &["*.css"]),
("cython", &["*.pyx"]),
("dart", &["*.dart"]),
("d", &["*.d"]),
("elisp", &["*.el"]),
("elixir", &["*.ex", "*.exs"]),
("erlang", &["*.erl", "*.hrl"]),
("fish", &["*.fish"]),
("fortran", &[
"*.f", "*.F", "*.f77", "*.F77", "*.pfo",
"*.f90", "*.F90", "*.f95", "*.F95",
@@ -106,6 +112,7 @@ const DEFAULT_TYPES: &'static [(&'static str, &'static [&'static str])] = &[
("fsharp", &["*.fs", "*.fsx", "*.fsi"]),
("go", &["*.go"]),
("groovy", &["*.groovy", "*.gradle"]),
("h", &["*.h", "*.hpp"]),
("hbs", &["*.hbs"]),
("haskell", &["*.hs", "*.lhs"]),
("html", &["*.htm", "*.html"]),
@@ -119,9 +126,9 @@ const DEFAULT_TYPES: &'static [(&'static str, &'static [&'static str])] = &[
("lisp", &["*.el", "*.jl", "*.lisp", "*.lsp", "*.sc", "*.scm"]),
("lua", &["*.lua"]),
("m4", &["*.ac", "*.m4"]),
("make", &["gnumakefile", "Gnumakefile", "makefile", "Makefile", "*.mk"]),
("markdown", &["*.md"]),
("md", &["*.md"]),
("make", &["gnumakefile", "Gnumakefile", "makefile", "Makefile", "*.mk", "*.mak"]),
("markdown", &["*.markdown", "*.md", "*.mdown", "*.mkdn"]),
("md", &["*.markdown", "*.md", "*.mdown", "*.mkdn"]),
("matlab", &["*.m"]),
("mk", &["mkfile"]),
("ml", &["*.ml"]),
@@ -129,11 +136,15 @@ const DEFAULT_TYPES: &'static [(&'static str, &'static [&'static str])] = &[
("objc", &["*.h", "*.m"]),
("objcpp", &["*.h", "*.mm"]),
("ocaml", &["*.ml", "*.mli", "*.mll", "*.mly"]),
("org", &["*.org"]),
("perl", &["*.perl", "*.pl", "*.PL", "*.plh", "*.plx", "*.pm"]),
("pdf", &["*.pdf"]),
("php", &["*.php", "*.php3", "*.php4", "*.php5", "*.phtml"]),
("pod", &["*.pod"]),
("py", &["*.py"]),
("readme", &["README*", "*README"]),
("r", &["*.R", "*.r", "*.Rmd", "*.Rnw"]),
("rdoc", &["*.rdoc"]),
("rst", &["*.rst"]),
("ruby", &["*.rb"]),
("rust", &["*.rs"]),
@@ -143,14 +154,17 @@ const DEFAULT_TYPES: &'static [(&'static str, &'static [&'static str])] = &[
("sql", &["*.sql"]),
("sv", &["*.v", "*.vg", "*.sv", "*.svh", "*.h"]),
("swift", &["*.swift"]),
("taskpaper", &["*.taskpaper"]),
("tcl", &["*.tcl"]),
("tex", &["*.tex", "*.cls", "*.sty"]),
("tex", &["*.tex", "*.ltx", "*.cls", "*.sty", "*.bib"]),
("textile", &["*.textile"]),
("ts", &["*.ts", "*.tsx"]),
("txt", &["*.txt"]),
("toml", &["*.toml", "Cargo.lock"]),
("vala", &["*.vala"]),
("vb", &["*.vb"]),
("vimscript", &["*.vim"]),
("wiki", &["*.mediawiki", "*.wiki"]),
("xml", &["*.xml"]),
("yacc", &["*.y"]),
("yaml", &["*.yaml", "*.yml"]),

1026
ignore/src/walk.rs
View File

File diff suppressed because it is too large Load Diff

4
pkg/brew/ripgrep-bin.rb
View File

@@ -1,9 +1,9 @@
class RipgrepBin < Formula
version '0.2.3'
version '0.2.8'
desc "Search tool like grep and The Silver Searcher."
homepage "https://github.com/BurntSushi/ripgrep"
url "https://github.com/BurntSushi/ripgrep/releases/download/#{version}/ripgrep-#{version}-x86_64-apple-darwin.tar.gz"
sha256 "7407555dfe040a2631a7efdd1eea62cf1d1c50e5a6ecf8ee82e0bef9d5f37298"
sha256 "349aba7561028e869932bae8fd27cd5ce45a68f47f05d426d6701a50a8474aa0"
conflicts_with "ripgrep"

455
src/app.rs Normal file
View File

@@ -0,0 +1,455 @@
use std::collections::HashMap;
use clap::{App, AppSettings, Arg};
const ABOUT: &'static str = "
ripgrep (rg) recursively searches your current directory for a regex pattern.
Project home page: https://github.com/BurntSushi/ripgrep
Use -h for short descriptions and --help for more details.";
const USAGE: &'static str = "
rg [OPTIONS] <pattern> [<path> ...]
rg [OPTIONS] [-e PATTERN | -f FILE ]... [<path> ...]
rg [OPTIONS] --files [<path> ...]
rg [OPTIONS] --type-list";
const TEMPLATE: &'static str = "\
{bin} {version}
{author}
{about}
USAGE:{usage}
ARGS:
{positionals}
OPTIONS:
{unified}";
/// Build a clap application with short help strings.
pub fn app_short() -> App<'static, 'static> {
app(false, |k| USAGES[k].short)
}
/// Build a clap application with long help strings.
pub fn app_long() -> App<'static, 'static> {
app(true, |k| USAGES[k].long)
}
/// Build a clap application parameterized by usage strings.
///
/// The function given should take a clap argument name and return a help
/// string. `app` will panic if a usage string is not defined.
///
/// This is an intentionally stand-alone module so that it can be used easily
/// in a `build.rs` script to build shell completion files.
fn app<F>(next_line_help: bool, doc: F) -> App<'static, 'static>
where F: Fn(&'static str) -> &'static str {
let arg = |name| {
Arg::with_name(name).help(doc(name)).next_line_help(next_line_help)
};
let flag = |name| arg(name).long(name);
App::new("ripgrep")
.author(crate_authors!())
.version(crate_version!())
.about(ABOUT)
.max_term_width(100)
.setting(AppSettings::UnifiedHelpMessage)
.usage(USAGE)
.template(TEMPLATE)
// Handle help/version manually to make their output formatting
// consistent with short/long views.
.arg(arg("help-short").short("h"))
.arg(flag("help"))
.arg(flag("version").short("V"))
// First, set up primary positional/flag arguments.
.arg(arg("pattern")
.required_unless_one(&[
"file", "files", "help-short", "help", "regexp", "type-list",
"version",
]))
.arg(arg("path").multiple(true))
.arg(flag("regexp").short("e")
.takes_value(true).multiple(true).number_of_values(1)
.value_name("pattern"))
.arg(flag("files")
// This should also conflict with `pattern`, but the first file
// path will actually be in `pattern`.
.conflicts_with_all(&["file", "regexp", "type-list"]))
.arg(flag("type-list")
.conflicts_with_all(&["file", "files", "pattern", "regexp"]))
// Second, set up common flags.
.arg(flag("text").short("a"))
.arg(flag("count").short("c"))
.arg(flag("color")
.value_name("WHEN")
.takes_value(true)
.hide_possible_values(true)
.possible_values(&["never", "auto", "always", "ansi"]))
.arg(flag("colors").value_name("SPEC")
.takes_value(true).multiple(true).number_of_values(1))
.arg(flag("fixed-strings").short("F"))
.arg(flag("glob").short("g")
.takes_value(true).multiple(true).number_of_values(1)
.value_name("GLOB"))
.arg(flag("ignore-case").short("i"))
.arg(flag("line-number").short("n"))
.arg(flag("no-line-number").short("N"))
.arg(flag("quiet").short("q"))
.arg(flag("type").short("t")
.takes_value(true).multiple(true).number_of_values(1)
.value_name("TYPE"))
.arg(flag("type-not").short("T")
.takes_value(true).multiple(true).number_of_values(1)
.value_name("TYPE"))
.arg(flag("unrestricted").short("u")
.multiple(true))
.arg(flag("invert-match").short("v"))
.arg(flag("word-regexp").short("w"))
// Third, set up less common flags.
.arg(flag("after-context").short("A")
.value_name("NUM").takes_value(true)
.validator(validate_number))
.arg(flag("before-context").short("B")
.value_name("NUM").takes_value(true)
.validator(validate_number))
.arg(flag("context").short("C")
.value_name("NUM").takes_value(true)
.validator(validate_number))
.arg(flag("column"))
.arg(flag("context-separator").value_name("ARG").takes_value(true))
.arg(flag("debug"))
.arg(flag("file").short("f")
.value_name("FILE").takes_value(true)
.multiple(true).number_of_values(1))
.arg(flag("files-with-matches").short("l"))
.arg(flag("files-without-match"))
.arg(flag("with-filename").short("H"))
.arg(flag("no-filename"))
.arg(flag("heading"))
.arg(flag("no-heading"))
.arg(flag("hidden"))
.arg(flag("ignore-file")
.value_name("FILE").takes_value(true)
.multiple(true).number_of_values(1))
.arg(flag("follow").short("L"))
.arg(flag("max-count")
.short("m").value_name("NUM").takes_value(true)
.validator(validate_number))
.arg(flag("maxdepth")
.value_name("NUM").takes_value(true)
.validator(validate_number))
.arg(flag("mmap"))
.arg(flag("no-messages"))
.arg(flag("no-mmap"))
.arg(flag("no-ignore"))
.arg(flag("no-ignore-parent"))
.arg(flag("no-ignore-vcs"))
.arg(flag("null"))
.arg(flag("pretty").short("p"))
.arg(flag("replace").short("r").value_name("ARG").takes_value(true))
.arg(flag("case-sensitive").short("s"))
.arg(flag("smart-case").short("S"))
.arg(flag("threads")
.short("j").value_name("ARG").takes_value(true)
.validator(validate_number))
.arg(flag("vimgrep"))
.arg(flag("type-add")
.value_name("TYPE").takes_value(true)
.multiple(true).number_of_values(1))
.arg(flag("type-clear")
.value_name("TYPE").takes_value(true)
.multiple(true).number_of_values(1))
}
struct Usage {
short: &'static str,
long: &'static str,
}
macro_rules! doc {
($map:expr, $name:expr, $short:expr) => {
doc!($map, $name, $short, $short)
};
($map:expr, $name:expr, $short:expr, $long:expr) => {
$map.insert($name, Usage {
short: $short,
long: concat!($long, "\n "),
});
};
}
lazy_static! {
static ref USAGES: HashMap<&'static str, Usage> = {
let mut h = HashMap::new();
doc!(h, "help-short",
"Show short help output.",
"Show short help output. Use --help to show more details.");
doc!(h, "help",
"Show verbose help output.",
"When given, more details about flags are provided.");
doc!(h, "version",
"Prints version information.");
doc!(h, "pattern",
"A regular expression used for searching.",
"A regular expression used for searching. Multiple patterns \
may be given. To match a pattern beginning with a -, use [-].");
doc!(h, "regexp",
"A regular expression used for searching.",
"A regular expression used for searching. Multiple patterns \
may be given. To match a pattern beginning with a -, use [-].");
doc!(h, "path",
"A file or directory to search.",
"A file or directory to search. Directories are searched \
recursively.");
doc!(h, "files",
"Print each file that would be searched.",
"Print each file that would be searched without actually \
performing the search. This is useful to determine whether a \
particular file is being searched or not.");
doc!(h, "type-list",
"Show all supported file types.",
"Show all supported file types and their corresponding globs.");
doc!(h, "text",
"Search binary files as if they were text.");
doc!(h, "count",
"Only show count of matches for each file.");
doc!(h, "color",
"When to use color. [default: auto]",
"When to use color in the output. The possible values are \
never, auto, always or ansi. The default is auto. When always \
is used, coloring is attempted based on your environment. When \
ansi used, coloring is forcefully done using ANSI escape color \
codes.");
doc!(h, "colors",
"Configure color settings and styles.",
"This flag specifies color settings for use in the output. \
This flag may be provided multiple times. Settings are applied \
iteratively. Colors are limited to one of eight choices: \
red, blue, green, cyan, magenta, yellow, white and black. \
Styles are limited to either nobold or bold.\n\nThe format \
of the flag is {type}:{attribute}:{value}. {type} should be \
one of path, line or match. {attribute} can be fg, bg or style. \
{value} is either a color (for fg and bg) or a text style. \
A special format, {type}:none, will clear all color settings \
for {type}.\n\nFor example, the following command will change \
the match color to magenta and the background color for line \
numbers to yellow:\n\n\
rg --colors 'match:fg:magenta' --colors 'line:bg:yellow' foo.");
doc!(h, "fixed-strings",
"Treat the pattern as a literal string.",
"Treat the pattern as a literal string instead of a regular \
expression. When this flag is used, special regular expression \
meta characters such as (){}*+. do not need to be escaped.");
doc!(h, "glob",
"Include or exclude files/directories.",
"Include or exclude files/directories for searching that \
match the given glob. This always overrides any other \
ignore logic. Multiple glob flags may be used. Globbing \
rules match .gitignore globs. Precede a glob with a ! \
to exclude it.");
doc!(h, "ignore-case",
"Case insensitive search.",
"Case insensitive search. This is overridden by \
--case-sensitive.");
doc!(h, "line-number",
"Show line numbers.",
"Show line numbers (1-based). This is enabled by default when \
searching in a tty.");
doc!(h, "no-line-number",
"Suppress line numbers.",
"Suppress line numbers. This is enabled by default when NOT \
searching in a tty.");
doc!(h, "quiet",
"Do not print anything to stdout.",
"Do not print anything to stdout. If a match is found in a file, \
stop searching. This is useful when ripgrep is used only for \
its exit code.");
doc!(h, "type",
"Only search files matching TYPE.",
"Only search files matching TYPE. Multiple type flags may be \
provided. Use the --type-list flag to list all available \
types.");
doc!(h, "type-not",
"Do not search files matching TYPE.",
"Do not search files matching TYPE. Multiple type-not flags may \
be provided. Use the --type-list flag to list all available \
types.");
doc!(h, "unrestricted",
"Reduce the level of \"smart\" searching.",
"Reduce the level of \"smart\" searching. A single -u \
won't respect .gitignore (etc.) files. Two -u flags will \
additionally search hidden files and directories. Three \
-u flags will additionally search binary files. -uu is \
roughly equivalent to grep -r and -uuu is roughly \
equivalent to grep -a -r.");
doc!(h, "invert-match",
"Invert matching.",
"Invert matching. Show lines that don't match given patterns.");
doc!(h, "word-regexp",
"Only show matches surrounded by word boundaries.",
"Only show matches surrounded by word boundaries. This is \
equivalent to putting \\b before and after all of the search \
patterns.");
doc!(h, "after-context",
"Show NUM lines after each match.");
doc!(h, "before-context",
"Show NUM lines before each match.");
doc!(h, "context",
"Show NUM lines before and after each match.");
doc!(h, "column",
"Show column numbers",
"Show column numbers (1-based). This only shows the column \
numbers for the first match on each line. This does not try \
to account for Unicode. One byte is equal to one column.");
doc!(h, "context-separator",
"Set the context separator string. [default: --]",
"The string used to separate non-contiguous context lines in the \
output. Escape sequences like \\x7F or \\t may be used. The \
default value is --.");
doc!(h, "debug",
"Show debug messages.",
"Show debug messages. Please use this when filing a bug report.");
doc!(h, "file",
"Search for patterns from the given file.",
"Search for patterns from the given file, with one pattern per \
line. When this flag is used or multiple times or in \
combination with the -e/--regexp flag, then all patterns \
provided are searched. Empty pattern lines will match all input \
lines, and the newline is not counted as part of the pattern.");
doc!(h, "files-with-matches",
"Only show the path of each file with at least one match.");
doc!(h, "files-without-match",
"Only show the path of each file that contains zero matches.");
doc!(h, "with-filename",
"Show file name for each match.",
"Prefix each match with the file name that contains it. This is \
the default when more than one file is searched.");
doc!(h, "no-filename",
"Never show the file name for a match.",
"Never show the file name for a match. This is the default when \
one file is searched.");
doc!(h, "heading",
"Show matches grouped by each file.",
"This shows the file name above clusters of matches from each \
file. This is the default mode at a tty.");
doc!(h, "no-heading",
"Don't group matches by each file.",
"Don't group matches by each file. This is the default mode \
when not at a tty.");
doc!(h, "hidden",
"Search hidden files and directories.",
"Search hidden files and directories. By default, hidden files \
and directories are skipped.");
doc!(h, "ignore-file",
"Specify additional ignore files.",
"Specify additional ignore files for filtering file paths. \
Ignore files should be in the gitignore format and are matched \
relative to the current working directory. These ignore files \
have lower precedence than all other ignore files. When \
specifying multiple ignore files, earlier files have lower \
precedence than later files.");
doc!(h, "follow",
"Follow symbolic links.");
doc!(h, "max-count",
"Limit the number of matches.",
"Limit the number of matching lines per file searched to NUM.");
doc!(h, "maxdepth",
"Descend at most NUM directories.",
"Limit the depth of directory traversal to NUM levels beyond \
the paths given. A value of zero only searches the \
starting-points themselves.\n\nFor example, \
'rg --maxdepth 0 dir/' is a no-op because dir/ will not be \
descended into. 'rg --maxdepth 1 dir/' will search only the \
direct children of dir/.");
doc!(h, "mmap",
"Searching using memory maps when possible.",
"Search using memory maps when possible. This is enabled by \
default when ripgrep thinks it will be faster. Note that memory \
map searching doesn't currently support all options, so if an \
incompatible option (e.g., --context) is given with --mmap, \
then memory maps will not be used.");
doc!(h, "no-messages",
"Suppress all error messages.",
"Suppress all error messages. This is equivalent to redirecting \
stderr to /dev/null.");
doc!(h, "no-mmap",
"Never use memory maps.",
"Never use memory maps, even when they might be faster.");
doc!(h, "no-ignore",
"Don't respect ignore files.",
"Don't respect ignore files (.gitignore, .ignore, etc.). This \
implies --no-ignore-parent and --no-ignore-vcs.");
doc!(h, "no-ignore-parent",
"Don't respect ignore files in parent directories.",
"Don't respect ignore files (.gitignore, .ignore, etc.) in \
parent directories.");
doc!(h, "no-ignore-vcs",
"Don't respect VCS ignore files",
"Don't respect version control ignore files (.gitignore, etc.). \
This implies --no-ignore-parent. Note that .ignore files will \
continue to be respected.");
doc!(h, "null",
"Print NUL byte after file names",
"Whenever a file name is printed, follow it with a NUL byte. \
This includes printing file names before matches, and when \
printing a list of matching files such as with --count, \
--files-with-matches and --files. This option is useful for use \
with xargs.");
doc!(h, "pretty",
"Alias for --color always --heading -n.");
doc!(h, "replace",
"Replace matches with string given.",
"Replace every match with the string given when printing \
results. Neither this flag nor any other flag will modify your \
files.\n\nCapture group indices (e.g., $5) and names \
(e.g., $foo) are supported in the replacement string.");
doc!(h, "case-sensitive",
"Search case sensitively.",
"Search case sensitively. This overrides -i/--ignore-case and \
-S/--smart-case.");
doc!(h, "smart-case",
"Smart case search.",
"Searches case insensitively if the pattern is all lowercase. \
Search case sensitively otherwise. This is overridden by \
either -s/--case-sensitive or -i/--ignore-case.");
doc!(h, "threads",
"The approximate number of threads to use.",
"The approximate number of threads to use. A value of 0 (which \
is the default) causes ripgrep to choose the thread count \
using heuristics.");
doc!(h, "vimgrep",
"Show results in vim compatible format.",
"Show results with every match on its own line, including \
line numbers and column numbers. With this option, a line with \
more than one match will be printed more than once.");
doc!(h, "type-add",
"Add a new glob for a file type.",
"Add a new glob for a particular file type. Only one glob can be \
added at a time. Multiple --type-add flags can be provided. \
Unless --type-clear is used, globs are added to any existing \
globs defined inside of ripgrep.\n\nNote that this MUST be \
passed to every invocation of ripgrep. Type settings are NOT \
persisted.\n\nExample: \
rg --type-add 'foo:*.foo' -tfoo PATTERN.");
doc!(h, "type-clear",
"Clear globs for given file type.",
"Clear the file type globs previously defined for TYPE. This \
only clears the default tpye definitions that are found inside \
of ripgrep.\n\nNote that this MUST be passed to every \
invocation of ripgrep. Type settings are NOT persisted.");
h
};
}
fn validate_number(s: String) -> Result<(), String> {
s.parse::<usize>().map(|_|()).map_err(|err| err.to_string())
}

1261
src/args.rs
View File

File diff suppressed because it is too large Load Diff

111
src/atty.rs
View File

@@ -4,6 +4,11 @@ from (or to) a terminal. Windows and Unix do this differently, so implement
both here.
*/
#[cfg(windows)]
use winapi::minwindef::DWORD;
#[cfg(windows)]
use winapi::winnt::HANDLE;
#[cfg(unix)]
pub fn stdin_is_readable() -> bool {
use std::fs::File;
@@ -44,26 +49,104 @@ pub fn on_stdout() -> bool {
/// Returns true if there is a tty on stdin.
#[cfg(windows)]
pub fn on_stdin() -> bool {
// BUG: https://github.com/BurntSushi/ripgrep/issues/19
// It's not clear to me how to determine whether there is a tty on stdin.
// Checking GetConsoleMode(GetStdHandle(stdin)) != 0 appears to report
// that stdin is a pipe, even if it's not in a cygwin terminal, for
// example.
//
// To fix this, we just assume there is always a tty on stdin. If Windows
// users need to search stdin, they'll have to pass -. Ug.
true
use kernel32::GetStdHandle;
use winapi::winbase::{
STD_INPUT_HANDLE, STD_OUTPUT_HANDLE, STD_ERROR_HANDLE,
};
unsafe {
let stdin = GetStdHandle(STD_INPUT_HANDLE);
if console_on_handle(stdin) {
// False positives aren't possible. If we got a console then
// we definitely have a tty on stdin.
return true;
}
// Otherwise, it's possible to get a false negative. If we know that
// there's a console on stdout or stderr however, then this is a true
// negative.
if console_on_fd(STD_OUTPUT_HANDLE)
|| console_on_fd(STD_ERROR_HANDLE) {
return false;
}
// Otherwise, we can't really tell, so we do a weird hack.
msys_tty_on_handle(stdin)
}
}
/// Returns true if there is a tty on stdout.
#[cfg(windows)]
pub fn on_stdout() -> bool {
use kernel32;
use winapi;
use kernel32::GetStdHandle;
use winapi::winbase::{
STD_INPUT_HANDLE, STD_OUTPUT_HANDLE, STD_ERROR_HANDLE,
};
unsafe {
let fd = winapi::winbase::STD_OUTPUT_HANDLE;
let mut out = 0;
kernel32::GetConsoleMode(kernel32::GetStdHandle(fd), &mut out) != 0
let stdout = GetStdHandle(STD_OUTPUT_HANDLE);
if console_on_handle(stdout) {
// False positives aren't possible. If we got a console then
// we definitely have a tty on stdout.
return true;
}
// Otherwise, it's possible to get a false negative. If we know that
// there's a console on stdin or stderr however, then this is a true
// negative.
if console_on_fd(STD_INPUT_HANDLE) || console_on_fd(STD_ERROR_HANDLE) {
return false;
}
// Otherwise, we can't really tell, so we do a weird hack.
msys_tty_on_handle(stdout)
}
}
/// Returns true if there is an MSYS tty on the given handle.
#[cfg(windows)]
fn msys_tty_on_handle(handle: HANDLE) -> bool {
use std::ffi::OsString;
use std::mem;
use std::os::raw::c_void;
use std::os::windows::ffi::OsStringExt;
use std::slice;
use kernel32::{GetFileInformationByHandleEx};
use winapi::fileapi::FILE_NAME_INFO;
use winapi::minwinbase::FileNameInfo;
use winapi::minwindef::MAX_PATH;
unsafe {
let size = mem::size_of::<FILE_NAME_INFO>();
let mut name_info_bytes = vec![0u8; size + MAX_PATH];
let res = GetFileInformationByHandleEx(
handle,
FileNameInfo,
&mut *name_info_bytes as *mut _ as *mut c_void,
name_info_bytes.len() as u32);
if res == 0 {
return true;
}
let name_info: FILE_NAME_INFO =
*(name_info_bytes[0..size].as_ptr() as *const FILE_NAME_INFO);
let name_bytes =
&name_info_bytes[size..size + name_info.FileNameLength as usize];
let name_u16 = slice::from_raw_parts(
name_bytes.as_ptr() as *const u16, name_bytes.len() / 2);
let name = OsString::from_wide(name_u16)
.as_os_str().to_string_lossy().into_owned();
name.contains("msys-") || name.contains("-pty")
}
}
/// Returns true if there is a console on the given file descriptor.
#[cfg(windows)]
unsafe fn console_on_fd(fd: DWORD) -> bool {
use kernel32::GetStdHandle;
console_on_handle(GetStdHandle(fd))
}
/// Returns true if there is a console on the given handle.
#[cfg(windows)]
fn console_on_handle(handle: HANDLE) -> bool {
use kernel32::GetConsoleMode;
let mut out = 0;
unsafe { GetConsoleMode(handle, &mut out) != 0 }
}

475
src/main.rs
View File

@@ -1,6 +1,9 @@
#![allow(dead_code, unused_imports)]
extern crate bytecount;
#[macro_use]
extern crate clap;
extern crate ctrlc;
extern crate deque;
extern crate docopt;
extern crate env_logger;
extern crate grep;
extern crate ignore;
@@ -15,36 +18,24 @@ extern crate memchr;
extern crate memmap;
extern crate num_cpus;
extern crate regex;
extern crate rustc_serialize;
extern crate term;
extern crate termcolor;
#[cfg(windows)]
extern crate winapi;
use std::error::Error;
use std::fs::File;
use std::io;
use std::io::Write;
use std::path::Path;
use std::process;
use std::result;
use std::sync::{Arc, Mutex};
use std::sync::atomic::{AtomicBool, Ordering};
use std::sync::atomic::{AtomicBool, AtomicUsize, Ordering};
use std::sync::mpsc;
use std::thread;
use std::cmp;
use deque::{Stealer, Stolen};
use grep::Grep;
use memmap::{Mmap, Protection};
use term::Terminal;
use ignore::DirEntry;
use termcolor::WriteColor;
use args::Args;
use out::{ColoredTerminal, Out};
use pathutil::strip_prefix;
use printer::Printer;
use search_stream::InputBuffer;
#[cfg(windows)]
use terminal_win::WindowsBuffer;
use worker::Work;
macro_rules! errored {
($($tt:tt)*) => {
@@ -59,20 +50,20 @@ macro_rules! eprintln {
}}
}
mod app;
mod args;
mod atty;
mod out;
mod pathutil;
mod printer;
mod search_buffer;
mod search_stream;
#[cfg(windows)]
mod terminal_win;
mod unescape;
mod worker;
pub type Result<T> = result::Result<T, Box<Error + Send + Sync>>;
fn main() {
match Args::parse().and_then(run) {
match Args::parse().map(Arc::new).and_then(run) {
Ok(count) if count == 0 => process::exit(1),
Ok(_) => process::exit(0),
Err(err) => {
@@ -82,95 +73,103 @@ fn main() {
}
}
fn run(args: Args) -> Result<u64> {
let args = Arc::new(args);
let handler_args = args.clone();
ctrlc::set_handler(move || {
let stdout = io::stdout();
let mut stdout = stdout.lock();
let _ = handler_args.stdout().reset();
let _ = stdout.flush();
process::exit(1);
});
let paths = args.paths();
let threads = cmp::max(1, args.threads() - 1);
let isone =
paths.len() == 1 && (paths[0] == Path::new("-") || paths[0].is_file());
fn run(args: Arc<Args>) -> Result<u64> {
if args.never_match() {
return Ok(0);
}
{
let args = args.clone();
ctrlc::set_handler(move || {
let mut writer = args.stdout();
let _ = writer.reset();
let _ = writer.flush();
process::exit(1);
});
}
let threads = args.threads();
if args.files() {
return run_files(args.clone());
}
if args.type_list() {
return run_types(args.clone());
}
if threads == 1 || isone {
return run_one_thread(args.clone());
}
let out = Arc::new(Mutex::new(args.out()));
let quiet_matched = QuietMatched::new(args.quiet());
let mut workers = vec![];
let workq = {
let (workq, stealer) = deque::new();
for _ in 0..threads {
let worker = MultiWorker {
chan_work: stealer.clone(),
quiet_matched: quiet_matched.clone(),
out: out.clone(),
outbuf: Some(args.outbuf()),
worker: Worker {
args: args.clone(),
inpbuf: args.input_buffer(),
grep: args.grep(),
match_count: 0,
},
};
workers.push(thread::spawn(move || worker.run()));
}
workq
};
let mut paths_searched: u64 = 0;
for dent in args.walker() {
if quiet_matched.has_match() {
break;
}
paths_searched += 1;
if dent.is_stdin() {
workq.push(Work::Stdin);
if threads == 1 || args.is_one_path() {
run_files_one_thread(args)
} else {
workq.push(Work::File(dent));
run_files_parallel(args)
}
} else if args.type_list() {
run_types(args)
} else if threads == 1 || args.is_one_path() {
run_one_thread(args)
} else {
run_parallel(args)
}
}
fn run_parallel(args: Arc<Args>) -> Result<u64> {
let bufwtr = Arc::new(args.buffer_writer());
let quiet_matched = QuietMatched::new(args.quiet());
let paths_searched = Arc::new(AtomicUsize::new(0));
let match_count = Arc::new(AtomicUsize::new(0));
args.walker_parallel().run(|| {
let args = args.clone();
let quiet_matched = quiet_matched.clone();
let paths_searched = paths_searched.clone();
let match_count = match_count.clone();
let bufwtr = bufwtr.clone();
let mut buf = bufwtr.buffer();
let mut worker = args.worker();
Box::new(move |result| {
use ignore::WalkState::*;
if quiet_matched.has_match() {
return Quit;
}
let dent = match get_or_log_dir_entry(result, args.no_messages()) {
None => return Continue,
Some(dent) => dent,
};
paths_searched.fetch_add(1, Ordering::SeqCst);
buf.clear();
{
// This block actually executes the search and prints the
// results into outbuf.
let mut printer = args.printer(&mut buf);
let count =
if dent.is_stdin() {
worker.run(&mut printer, Work::Stdin)
} else {
worker.run(&mut printer, Work::DirEntry(dent))
};
match_count.fetch_add(count as usize, Ordering::SeqCst);
if quiet_matched.set_match(count > 0) {
return Quit;
}
}
// BUG(burntsushi): We should handle this error instead of ignoring
// it. See: https://github.com/BurntSushi/ripgrep/issues/200
let _ = bufwtr.print(&buf);
Continue
})
});
if !args.paths().is_empty() && paths_searched.load(Ordering::SeqCst) == 0 {
if !args.no_messages() {
eprint_nothing_searched();
}
}
if !paths.is_empty() && paths_searched == 0 {
eprintln!("No files were searched, which means ripgrep probably \
applied a filter you didn't expect. \
Try running again with --debug.");
}
for _ in 0..workers.len() {
workq.push(Work::Quit);
}
let mut match_count = 0;
for worker in workers {
match_count += worker.join().unwrap();
}
Ok(match_count)
Ok(match_count.load(Ordering::SeqCst) as u64)
}
fn run_one_thread(args: Arc<Args>) -> Result<u64> {
let mut worker = Worker {
args: args.clone(),
inpbuf: args.input_buffer(),
grep: args.grep(),
match_count: 0,
};
let mut term = args.stdout();
let stdout = args.stdout();
let mut stdout = stdout.lock();
let mut worker = args.worker();
let mut paths_searched: u64 = 0;
for dent in args.walker() {
let mut printer = args.printer(&mut term);
if worker.match_count > 0 {
let mut match_count = 0;
for result in args.walker() {
let dent = match get_or_log_dir_entry(result, args.no_messages()) {
None => continue,
Some(dent) => dent,
};
let mut printer = args.printer(&mut stdout);
if match_count > 0 {
if args.quiet() {
break;
}
@@ -179,32 +178,56 @@ fn run_one_thread(args: Arc<Args>) -> Result<u64> {
}
}
paths_searched += 1;
if dent.is_stdin() {
worker.do_work(&mut printer, WorkReady::Stdin);
} else {
let file = match File::open(dent.path()) {
Ok(file) => file,
Err(err) => {
eprintln!("{}: {}", dent.path().display(), err);
continue;
}
match_count +=
if dent.is_stdin() {
worker.run(&mut printer, Work::Stdin)
} else {
worker.run(&mut printer, Work::DirEntry(dent))
};
worker.do_work(&mut printer, WorkReady::DirFile(dent, file));
}
}
if !args.paths().is_empty() && paths_searched == 0 {
eprintln!("No files were searched, which means ripgrep probably \
applied a filter you didn't expect. \
Try running again with --debug.");
if !args.no_messages() {
eprint_nothing_searched();
}
}
Ok(worker.match_count)
Ok(match_count)
}
fn run_files(args: Arc<Args>) -> Result<u64> {
let term = args.stdout();
let mut printer = args.printer(term);
fn run_files_parallel(args: Arc<Args>) -> Result<u64> {
let print_args = args.clone();
let (tx, rx) = mpsc::channel::<ignore::DirEntry>();
let print_thread = thread::spawn(move || {
let stdout = print_args.stdout();
let mut printer = print_args.printer(stdout.lock());
let mut file_count = 0;
for dent in rx.iter() {
printer.path(dent.path());
file_count += 1;
}
file_count
});
let no_messages = args.no_messages();
args.walker_parallel().run(move || {
let tx = tx.clone();
Box::new(move |result| {
if let Some(dent) = get_or_log_dir_entry(result, no_messages) {
tx.send(dent).unwrap();
}
ignore::WalkState::Continue
})
});
Ok(print_thread.join().unwrap())
}
fn run_files_one_thread(args: Arc<Args>) -> Result<u64> {
let stdout = args.stdout();
let mut printer = args.printer(stdout.lock());
let mut file_count = 0;
for dent in args.walker() {
for result in args.walker() {
let dent = match get_or_log_dir_entry(result, args.no_messages()) {
None => continue,
Some(dent) => dent,
};
printer.path(dent.path());
file_count += 1;
}
@@ -212,8 +235,8 @@ fn run_files(args: Arc<Args>) -> Result<u64> {
}
fn run_types(args: Arc<Args>) -> Result<u64> {
let term = args.stdout();
let mut printer = args.printer(term);
let stdout = args.stdout();
let mut printer = args.printer(stdout.lock());
let mut ty_count = 0;
for def in args.type_defs() {
printer.type_def(def);
@@ -222,163 +245,91 @@ fn run_types(args: Arc<Args>) -> Result<u64> {
Ok(ty_count)
}
enum Work {
Stdin,
File(DirEntry),
Quit,
}
enum WorkReady {
Stdin,
DirFile(DirEntry, File),
}
struct MultiWorker {
chan_work: Stealer<Work>,
quiet_matched: QuietMatched,
out: Arc<Mutex<Out>>,
#[cfg(not(windows))]
outbuf: Option<ColoredTerminal<term::TerminfoTerminal<Vec<u8>>>>,
#[cfg(windows)]
outbuf: Option<ColoredTerminal<WindowsBuffer>>,
worker: Worker,
}
struct Worker {
args: Arc<Args>,
inpbuf: InputBuffer,
grep: Grep,
match_count: u64,
}
impl MultiWorker {
fn run(mut self) -> u64 {
loop {
if self.quiet_matched.has_match() {
break;
}
let work = match self.chan_work.steal() {
Stolen::Empty | Stolen::Abort => continue,
Stolen::Data(Work::Quit) => break,
Stolen::Data(Work::Stdin) => WorkReady::Stdin,
Stolen::Data(Work::File(ent)) => {
match File::open(ent.path()) {
Ok(file) => WorkReady::DirFile(ent, file),
Err(err) => {
eprintln!("{}: {}", ent.path().display(), err);
continue;
}
}
}
};
let mut outbuf = self.outbuf.take().unwrap();
outbuf.clear();
let mut printer = self.worker.args.printer(outbuf);
self.worker.do_work(&mut printer, work);
if self.quiet_matched.set_match(self.worker.match_count > 0) {
break;
}
let outbuf = printer.into_inner();
if !outbuf.get_ref().is_empty() {
let mut out = self.out.lock().unwrap();
out.write(&outbuf);
}
self.outbuf = Some(outbuf);
}
self.worker.match_count
}
}
impl Worker {
fn do_work<W: Terminal + Send>(
&mut self,
printer: &mut Printer<W>,
work: WorkReady,
) {
let result = match work {
WorkReady::Stdin => {
let stdin = io::stdin();
let stdin = stdin.lock();
self.search(printer, &Path::new("<stdin>"), stdin)
}
WorkReady::DirFile(ent, file) => {
let mut path = ent.path();
if let Some(p) = strip_prefix("./", path) {
path = p;
}
if self.args.mmap() {
self.search_mmap(printer, path, &file)
} else {
self.search(printer, path, file)
}
}
};
match result {
Ok(count) => {
self.match_count += count;
}
Err(err) => {
fn get_or_log_dir_entry(
result: result::Result<ignore::DirEntry, ignore::Error>,
no_messages: bool,
) -> Option<ignore::DirEntry> {
match result {
Err(err) => {
if !no_messages {
eprintln!("{}", err);
}
None
}
}
fn search<R: io::Read, W: Terminal + Send>(
&mut self,
printer: &mut Printer<W>,
path: &Path,
rdr: R,
) -> Result<u64> {
self.args.searcher(
&mut self.inpbuf,
printer,
&self.grep,
path,
rdr,
).run().map_err(From::from)
}
fn search_mmap<W: Terminal + Send>(
&mut self,
printer: &mut Printer<W>,
path: &Path,
file: &File,
) -> Result<u64> {
if try!(file.metadata()).len() == 0 {
// Opening a memory map with an empty file results in an error.
// However, this may not actually be an empty file! For example,
// /proc/cpuinfo reports itself as an empty file, but it can
// produce data when it's read from. Therefore, we fall back to
// regular read calls.
return self.search(printer, path, file);
Ok(dent) => {
if let Some(err) = dent.error() {
if !no_messages {
eprintln!("{}", err);
}
}
let ft = match dent.file_type() {
None => return Some(dent), // entry is stdin
Some(ft) => ft,
};
// A depth of 0 means the user gave the path explicitly, so we
// should always try to search it.
if dent.depth() == 0 && !ft.is_dir() {
Some(dent)
} else if ft.is_file() {
Some(dent)
} else {
None
}
}
let mmap = try!(Mmap::open(file, Protection::Read));
Ok(self.args.searcher_buffer(
printer,
&self.grep,
path,
unsafe { mmap.as_slice() },
).run())
}
}
fn version() -> String {
let (maj, min, pat) = (
option_env!("CARGO_PKG_VERSION_MAJOR"),
option_env!("CARGO_PKG_VERSION_MINOR"),
option_env!("CARGO_PKG_VERSION_PATCH"),
);
match (maj, min, pat) {
(Some(maj), Some(min), Some(pat)) => {
format!("ripgrep {}.{}.{}", maj, min, pat)
}
_ => "".to_owned(),
}
}
fn eprint_nothing_searched() {
eprintln!("No files were searched, which means ripgrep probably \
applied a filter you didn't expect. \
Try running again with --debug.");
}
/// A simple thread safe abstraction for determining whether a search should
/// stop if the user has requested quiet mode.
#[derive(Clone, Debug)]
struct QuietMatched(Arc<Option<AtomicBool>>);
pub struct QuietMatched(Arc<Option<AtomicBool>>);
impl QuietMatched {
fn new(quiet: bool) -> QuietMatched {
/// Create a new QuietMatched value.
///
/// If quiet is true, then set_match and has_match will reflect whether
/// a search should quit or not because it found a match.
///
/// If quiet is false, then set_match is always a no-op and has_match
/// always returns false.
pub fn new(quiet: bool) -> QuietMatched {
let atomic = if quiet { Some(AtomicBool::new(false)) } else { None };
QuietMatched(Arc::new(atomic))
}
fn has_match(&self) -> bool {
/// Returns true if and only if quiet mode is enabled and a match has
/// occurred.
pub fn has_match(&self) -> bool {
match *self.0 {
None => false,
Some(ref matched) => matched.load(Ordering::SeqCst),
}
}
fn set_match(&self, yes: bool) -> bool {
/// Sets whether a match has occurred or not.
///
/// If quiet mode is disabled, then this is a no-op.
pub fn set_match(&self, yes: bool) -> bool {
match *self.0 {
None => false,
Some(_) if !yes => false,

374
src/out.rs
View File

@@ -1,374 +0,0 @@
use std::io::{self, Write};
use term::{self, Terminal};
#[cfg(not(windows))]
use term::terminfo::TermInfo;
#[cfg(windows)]
use term::WinConsole;
#[cfg(windows)]
use terminal_win::WindowsBuffer;
/// Out controls the actual output of all search results for a particular file
/// to the end user.
///
/// (The difference between Out and Printer is that a Printer works with
/// individual search results where as Out works with search results for each
/// file as a whole. For example, it knows when to print a file separator.)
pub struct Out {
#[cfg(not(windows))]
term: ColoredTerminal<term::TerminfoTerminal<io::BufWriter<io::Stdout>>>,
#[cfg(windows)]
term: ColoredTerminal<WinConsole<io::Stdout>>,
printed: bool,
file_separator: Option<Vec<u8>>,
}
impl Out {
/// Create a new Out that writes to the wtr given.
#[cfg(not(windows))]
pub fn new(color: bool) -> Out {
let wtr = io::BufWriter::new(io::stdout());
Out {
term: ColoredTerminal::new(wtr, color),
printed: false,
file_separator: None,
}
}
/// Create a new Out that writes to the wtr given.
#[cfg(windows)]
pub fn new(color: bool) -> Out {
Out {
term: ColoredTerminal::new_stdout(color),
printed: false,
file_separator: None,
}
}
/// If set, the separator is printed between matches from different files.
/// By default, no separator is printed.
pub fn file_separator(mut self, sep: Vec<u8>) -> Out {
self.file_separator = Some(sep);
self
}
/// Write the search results of a single file to the underlying wtr and
/// flush wtr.
#[cfg(not(windows))]
pub fn write(
&mut self,
buf: &ColoredTerminal<term::TerminfoTerminal<Vec<u8>>>,
) {
self.write_sep();
match *buf {
ColoredTerminal::Colored(ref tt) => {
let _ = self.term.write_all(tt.get_ref());
}
ColoredTerminal::NoColor(ref buf) => {
let _ = self.term.write_all(buf);
}
}
self.write_done();
}
/// Write the search results of a single file to the underlying wtr and
/// flush wtr.
#[cfg(windows)]
pub fn write(
&mut self,
buf: &ColoredTerminal<WindowsBuffer>,
) {
self.write_sep();
match *buf {
ColoredTerminal::Colored(ref tt) => {
tt.print_stdout(&mut self.term);
}
ColoredTerminal::NoColor(ref buf) => {
let _ = self.term.write_all(buf);
}
}
self.write_done();
}
fn write_sep(&mut self) {
if let Some(ref sep) = self.file_separator {
if self.printed {
let _ = self.term.write_all(sep);
let _ = self.term.write_all(b"\n");
}
}
}
fn write_done(&mut self) {
let _ = self.term.flush();
self.printed = true;
}
}
/// ColoredTerminal provides optional colored output through the term::Terminal
/// trait. In particular, it will dynamically configure itself to use coloring
/// if it's available in the environment.
#[derive(Clone, Debug)]
pub enum ColoredTerminal<T: Terminal + Send> {
Colored(T),
NoColor(T::Output),
}
#[cfg(not(windows))]
impl<W: io::Write + Send> ColoredTerminal<term::TerminfoTerminal<W>> {
/// Create a new output buffer.
///
/// When color is true, the buffer will attempt to support coloring.
pub fn new(wtr: W, color: bool) -> Self {
lazy_static! {
// Only pay for parsing the terminfo once.
static ref TERMINFO: Option<TermInfo> = {
match TermInfo::from_env() {
Ok(info) => Some(info),
Err(err) => {
debug!("error loading terminfo for coloring: {}", err);
None
}
}
};
}
// If we want color, build a term::TerminfoTerminal and see if the
// current environment supports coloring. If not, bail with NoColor. To
// avoid losing our writer (ownership), do this the long way.
if !color {
return ColoredTerminal::NoColor(wtr);
}
let terminfo = match *TERMINFO {
None => return ColoredTerminal::NoColor(wtr),
Some(ref ti) => {
// Ug, this should go away with the next release of `term`.
TermInfo {
names: ti.names.clone(),
bools: ti.bools.clone(),
numbers: ti.numbers.clone(),
strings: ti.strings.clone(),
}
}
};
let tt = term::TerminfoTerminal::new_with_terminfo(wtr, terminfo);
if !tt.supports_color() {
debug!("environment doesn't support coloring");
return ColoredTerminal::NoColor(tt.into_inner());
}
ColoredTerminal::Colored(tt)
}
}
#[cfg(not(windows))]
impl ColoredTerminal<term::TerminfoTerminal<Vec<u8>>> {
/// Clear the give buffer of all search results such that it is reusable
/// in another search.
pub fn clear(&mut self) {
match *self {
ColoredTerminal::Colored(ref mut tt) => {
tt.get_mut().clear();
}
ColoredTerminal::NoColor(ref mut buf) => {
buf.clear();
}
}
}
}
#[cfg(windows)]
impl ColoredTerminal<WindowsBuffer> {
/// Create a new output buffer.
///
/// When color is true, the buffer will attempt to support coloring.
pub fn new_buffer(color: bool) -> Self {
if !color {
ColoredTerminal::NoColor(vec![])
} else {
ColoredTerminal::Colored(WindowsBuffer::new())
}
}
/// Clear the give buffer of all search results such that it is reusable
/// in another search.
pub fn clear(&mut self) {
match *self {
ColoredTerminal::Colored(ref mut win) => win.clear(),
ColoredTerminal::NoColor(ref mut buf) => buf.clear(),
}
}
}
#[cfg(windows)]
impl ColoredTerminal<WinConsole<io::Stdout>> {
/// Create a new output buffer.
///
/// When color is true, the buffer will attempt to support coloring.
pub fn new_stdout(color: bool) -> Self {
if !color {
return ColoredTerminal::NoColor(io::stdout());
}
match WinConsole::new(io::stdout()) {
Ok(win) => ColoredTerminal::Colored(win),
Err(_) => ColoredTerminal::NoColor(io::stdout()),
}
}
}
impl<T: Terminal + Send> ColoredTerminal<T> {
fn map_result<F>(
&mut self,
mut f: F,
) -> term::Result<()>
where F: FnMut(&mut T) -> term::Result<()> {
match *self {
ColoredTerminal::Colored(ref mut w) => f(w),
ColoredTerminal::NoColor(_) => Err(term::Error::NotSupported),
}
}
fn map_bool<F>(
&self,
mut f: F,
) -> bool
where F: FnMut(&T) -> bool {
match *self {
ColoredTerminal::Colored(ref w) => f(w),
ColoredTerminal::NoColor(_) => false,
}
}
}
impl<T: Terminal + Send> io::Write for ColoredTerminal<T> {
fn write(&mut self, buf: &[u8]) -> io::Result<usize> {
match *self {
ColoredTerminal::Colored(ref mut w) => w.write(buf),
ColoredTerminal::NoColor(ref mut w) => w.write(buf),
}
}
fn flush(&mut self) -> io::Result<()> {
Ok(())
}
}
impl<T: Terminal + Send> term::Terminal for ColoredTerminal<T> {
type Output = T::Output;
fn fg(&mut self, fg: term::color::Color) -> term::Result<()> {
self.map_result(|w| w.fg(fg))
}
fn bg(&mut self, bg: term::color::Color) -> term::Result<()> {
self.map_result(|w| w.bg(bg))
}
fn attr(&mut self, attr: term::Attr) -> term::Result<()> {
self.map_result(|w| w.attr(attr))
}
fn supports_attr(&self, attr: term::Attr) -> bool {
self.map_bool(|w| w.supports_attr(attr))
}
fn reset(&mut self) -> term::Result<()> {
self.map_result(|w| w.reset())
}
fn supports_reset(&self) -> bool {
self.map_bool(|w| w.supports_reset())
}
fn supports_color(&self) -> bool {
self.map_bool(|w| w.supports_color())
}
fn cursor_up(&mut self) -> term::Result<()> {
self.map_result(|w| w.cursor_up())
}
fn delete_line(&mut self) -> term::Result<()> {
self.map_result(|w| w.delete_line())
}
fn carriage_return(&mut self) -> term::Result<()> {
self.map_result(|w| w.carriage_return())
}
fn get_ref(&self) -> &Self::Output {
match *self {
ColoredTerminal::Colored(ref w) => w.get_ref(),
ColoredTerminal::NoColor(ref w) => w,
}
}
fn get_mut(&mut self) -> &mut Self::Output {
match *self {
ColoredTerminal::Colored(ref mut w) => w.get_mut(),
ColoredTerminal::NoColor(ref mut w) => w,
}
}
fn into_inner(self) -> Self::Output {
match self {
ColoredTerminal::Colored(w) => w.into_inner(),
ColoredTerminal::NoColor(w) => w,
}
}
}
impl<'a, T: Terminal + Send> term::Terminal for &'a mut ColoredTerminal<T> {
type Output = T::Output;
fn fg(&mut self, fg: term::color::Color) -> term::Result<()> {
(**self).fg(fg)
}
fn bg(&mut self, bg: term::color::Color) -> term::Result<()> {
(**self).bg(bg)
}
fn attr(&mut self, attr: term::Attr) -> term::Result<()> {
(**self).attr(attr)
}
fn supports_attr(&self, attr: term::Attr) -> bool {
(**self).supports_attr(attr)
}
fn reset(&mut self) -> term::Result<()> {
(**self).reset()
}
fn supports_reset(&self) -> bool {
(**self).supports_reset()
}
fn supports_color(&self) -> bool {
(**self).supports_color()
}
fn cursor_up(&mut self) -> term::Result<()> {
(**self).cursor_up()
}
fn delete_line(&mut self) -> term::Result<()> {
(**self).delete_line()
}
fn carriage_return(&mut self) -> term::Result<()> {
(**self).carriage_return()
}
fn get_ref(&self) -> &Self::Output {
(**self).get_ref()
}
fn get_mut(&mut self) -> &mut Self::Output {
(**self).get_mut()
}
fn into_inner(self) -> Self::Output {
// Good golly miss molly...
unimplemented!()
}
}

438
src/printer.rs
View File

@@ -1,8 +1,10 @@
use std::error;
use std::fmt;
use std::path::Path;
use std::str::FromStr;
use regex::bytes::Regex;
use term::{Attr, Terminal};
use term::color;
use termcolor::{Color, ColorSpec, ParseColorError, WriteColor};
use pathutil::strip_prefix;
use ignore::types::FileTypeDef;
@@ -40,38 +42,12 @@ pub struct Printer<W> {
replace: Option<Vec<u8>>,
/// Whether to prefix each match with the corresponding file name.
with_filename: bool,
/// The choice of colors.
color_choice: ColorChoice
/// The color specifications.
colors: ColorSpecs,
}
struct ColorChoice {
matched_line: color::Color,
heading: color::Color,
line_number: color::Color
}
impl ColorChoice {
#[cfg(unix)]
pub fn new() -> ColorChoice {
ColorChoice {
matched_line: color::RED,
heading: color::GREEN,
line_number: color::BLUE
}
}
#[cfg(not(unix))]
pub fn new() -> ColorChoice {
ColorChoice {
matched_line: color::BRIGHT_RED,
heading: color::BRIGHT_GREEN,
line_number: color::BRIGHT_BLUE
}
}
}
impl<W: Terminal + Send> Printer<W> {
/// Create a new printer that writes to wtr.
impl<W: WriteColor> Printer<W> {
/// Create a new printer that writes to wtr with the given color settings.
pub fn new(wtr: W) -> Printer<W> {
Printer {
wtr: wtr,
@@ -85,10 +61,16 @@ impl<W: Terminal + Send> Printer<W> {
null: false,
replace: None,
with_filename: false,
color_choice: ColorChoice::new()
colors: ColorSpecs::default(),
}
}
/// Set the color specifications.
pub fn colors(mut self, colors: ColorSpecs) -> Printer<W> {
self.colors = colors;
self
}
/// When set, column numbers will be printed for the first match on each
/// line.
pub fn column(mut self, yes: bool) -> Printer<W> {
@@ -158,6 +140,7 @@ impl<W: Terminal + Send> Printer<W> {
}
/// Flushes the underlying writer and returns it.
#[allow(dead_code)]
pub fn into_inner(mut self) -> W {
let _ = self.wtr.flush();
self.wtr
@@ -284,8 +267,7 @@ impl<W: Terminal + Send> Printer<W> {
let mut last_written = 0;
for (s, e) in re.find_iter(buf) {
self.write(&buf[last_written..s]);
let _ = self.wtr.fg(self.color_choice.matched_line);
let _ = self.wtr.attr(Attr::Bold);
let _ = self.wtr.set_color(self.colors.matched());
self.write(&buf[s..e]);
let _ = self.wtr.reset();
last_written = e;
@@ -322,30 +304,20 @@ impl<W: Terminal + Send> Printer<W> {
}
fn write_heading<P: AsRef<Path>>(&mut self, path: P) {
if self.wtr.supports_color() {
let _ = self.wtr.fg(self.color_choice.heading);
let _ = self.wtr.attr(Attr::Bold);
}
let _ = self.wtr.set_color(self.colors.path());
self.write_path(path.as_ref());
let _ = self.wtr.reset();
if self.null {
self.write(b"\x00");
} else {
self.write_eol();
}
if self.wtr.supports_color() {
let _ = self.wtr.reset();
}
}
fn write_non_heading_path<P: AsRef<Path>>(&mut self, path: P) {
if self.wtr.supports_color() {
let _ = self.wtr.fg(self.color_choice.heading);
let _ = self.wtr.attr(Attr::Bold);
}
let _ = self.wtr.set_color(self.colors.path());
self.write_path(path.as_ref());
if self.wtr.supports_color() {
let _ = self.wtr.reset();
}
let _ = self.wtr.reset();
if self.null {
self.write(b"\x00");
} else {
@@ -354,14 +326,9 @@ impl<W: Terminal + Send> Printer<W> {
}
fn line_number(&mut self, n: u64, sep: u8) {
if self.wtr.supports_color() {
let _ = self.wtr.fg(self.color_choice.line_number);
let _ = self.wtr.attr(Attr::Bold);
}
let _ = self.wtr.set_color(self.colors.line());
self.write(n.to_string().as_bytes());
if self.wtr.supports_color() {
let _ = self.wtr.reset();
}
let _ = self.wtr.reset();
self.write(&[sep]);
}
@@ -396,3 +363,362 @@ impl<W: Terminal + Send> Printer<W> {
}
}
}
/// An error that can occur when parsing color specifications.
#[derive(Clone, Debug, Eq, PartialEq)]
pub enum Error {
/// This occurs when an unrecognized output type is used.
UnrecognizedOutType(String),
/// This occurs when an unrecognized spec type is used.
UnrecognizedSpecType(String),
/// This occurs when an unrecognized color name is used.
UnrecognizedColor(String, String),
/// This occurs when an unrecognized style attribute is used.
UnrecognizedStyle(String),
/// This occurs when the format of a color specification is invalid.
InvalidFormat(String),
}
impl error::Error for Error {
fn description(&self) -> &str {
match *self {
Error::UnrecognizedOutType(_) => "unrecognized output type",
Error::UnrecognizedSpecType(_) => "unrecognized spec type",
Error::UnrecognizedColor(_, _) => "unrecognized color name",
Error::UnrecognizedStyle(_) => "unrecognized style attribute",
Error::InvalidFormat(_) => "invalid color spec",
}
}
fn cause(&self) -> Option<&error::Error> {
None
}
}
impl fmt::Display for Error {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
match *self {
Error::UnrecognizedOutType(ref name) => {
write!(f, "Unrecognized output type '{}'. Choose from: \
path, line, match.", name)
}
Error::UnrecognizedSpecType(ref name) => {
write!(f, "Unrecognized spec type '{}'. Choose from: \
fg, bg, style, none.", name)
}
Error::UnrecognizedColor(_, ref msg) => {
write!(f, "{}", msg)
}
Error::UnrecognizedStyle(ref name) => {
write!(f, "Unrecognized style attribute '{}'. Choose from: \
nobold, bold.", name)
}
Error::InvalidFormat(ref original) => {
write!(f, "Invalid color speci format: '{}'. Valid format \
is '(path|line|match):(fg|bg|style):(value)'.",
original)
}
}
}
}
impl From<ParseColorError> for Error {
fn from(err: ParseColorError) -> Error {
Error::UnrecognizedColor(err.invalid().to_string(), err.to_string())
}
}
/// A merged set of color specifications.
#[derive(Clone, Debug, Default, Eq, PartialEq)]
pub struct ColorSpecs {
path: ColorSpec,
line: ColorSpec,
matched: ColorSpec,
}
/// A single color specification provided by the user.
///
/// A `ColorSpecs` can be built by merging a sequence of `Spec`s.
///
/// ## Example
///
/// The only way to build a `Spec` is to parse it from a string. Once multiple
/// `Spec`s have been constructed, then can be merged into a single
/// `ColorSpecs` value.
///
/// ```rust
/// use termcolor::{Color, ColorSpecs, Spec};
///
/// let spec1: Spec = "path:fg:blue".parse().unwrap();
/// let spec2: Spec = "match:bg:green".parse().unwrap();
/// let specs = ColorSpecs::new(&[spec1, spec2]);
///
/// assert_eq!(specs.path().fg(), Some(Color::Blue));
/// assert_eq!(specs.matched().bg(), Some(Color::Green));
/// ```
///
/// ## Format
///
/// The format of a `Spec` is a triple: `{type}:{attribute}:{value}`. Each
/// component is defined as follows:
///
/// * `{type}` can be one of `path`, `line` or `match`.
/// * `{attribute}` can be one of `fg`, `bg` or `style`. `{attribute}` may also
/// be the special value `none`, in which case, `{value}` can be omitted.
/// * `{value}` is either a color name (for `fg`/`bg`) or a style instruction.
///
/// `{type}` controls which part of the output should be styled and is
/// application dependent.
///
/// When `{attribute}` is `none`, then this should cause any existing color
/// settings to be cleared.
///
/// `{value}` should be a color when `{attribute}` is `fg` or `bg`, or it
/// should be a style instruction when `{attribute}` is `style`. When
/// `{attribute}` is `none`, `{value}` must be omitted.
///
/// Valid colors are `black`, `blue`, `green`, `red`, `cyan`, `magenta`,
/// `yellow`, `white`.
///
/// Valid style instructions are `nobold` and `bold`.
#[derive(Clone, Debug, Eq, PartialEq)]
pub struct Spec {
ty: OutType,
value: SpecValue,
}
/// The actual value given by the specification.
#[derive(Clone, Debug, Eq, PartialEq)]
enum SpecValue {
None,
Fg(Color),
Bg(Color),
Style(Style),
}
/// The set of configurable portions of ripgrep's output.
#[derive(Clone, Debug, Eq, PartialEq)]
enum OutType {
Path,
Line,
Match,
}
/// The specification type.
#[derive(Clone, Debug, Eq, PartialEq)]
enum SpecType {
Fg,
Bg,
Style,
None,
}
/// The set of available styles for use in the terminal.
#[derive(Clone, Debug, Eq, PartialEq)]
enum Style {
Bold,
NoBold,
}
impl ColorSpecs {
/// Create color specifications from a list of user supplied
/// specifications.
pub fn new(user_specs: &[Spec]) -> ColorSpecs {
let mut specs = ColorSpecs::default();
for user_spec in user_specs {
match user_spec.ty {
OutType::Path => user_spec.merge_into(&mut specs.path),
OutType::Line => user_spec.merge_into(&mut specs.line),
OutType::Match => user_spec.merge_into(&mut specs.matched),
}
}
specs
}
/// Return the color specification for coloring file paths.
fn path(&self) -> &ColorSpec {
&self.path
}
/// Return the color specification for coloring line numbers.
fn line(&self) -> &ColorSpec {
&self.line
}
/// Return the color specification for coloring matched text.
fn matched(&self) -> &ColorSpec {
&self.matched
}
}
impl Spec {
/// Merge this spec into the given color specification.
fn merge_into(&self, cspec: &mut ColorSpec) {
self.value.merge_into(cspec);
}
}
impl SpecValue {
/// Merge this spec value into the given color specification.
fn merge_into(&self, cspec: &mut ColorSpec) {
match *self {
SpecValue::None => cspec.clear(),
SpecValue::Fg(ref color) => { cspec.set_fg(Some(color.clone())); }
SpecValue::Bg(ref color) => { cspec.set_bg(Some(color.clone())); }
SpecValue::Style(ref style) => {
match *style {
Style::Bold => { cspec.set_bold(true); }
Style::NoBold => { cspec.set_bold(false); }
}
}
}
}
}
impl FromStr for Spec {
type Err = Error;
fn from_str(s: &str) -> Result<Spec, Error> {
let pieces: Vec<&str> = s.split(":").collect();
if pieces.len() <= 1 || pieces.len() > 3 {
return Err(Error::InvalidFormat(s.to_string()));
}
let otype: OutType = try!(pieces[0].parse());
match try!(pieces[1].parse()) {
SpecType::None => Ok(Spec { ty: otype, value: SpecValue::None }),
SpecType::Style => {
if pieces.len() < 3 {
return Err(Error::InvalidFormat(s.to_string()));
}
let style: Style = try!(pieces[2].parse());
Ok(Spec { ty: otype, value: SpecValue::Style(style) })
}
SpecType::Fg => {
if pieces.len() < 3 {
return Err(Error::InvalidFormat(s.to_string()));
}
let color: Color = try!(pieces[2].parse());
Ok(Spec { ty: otype, value: SpecValue::Fg(color) })
}
SpecType::Bg => {
if pieces.len() < 3 {
return Err(Error::InvalidFormat(s.to_string()));
}
let color: Color = try!(pieces[2].parse());
Ok(Spec { ty: otype, value: SpecValue::Bg(color) })
}
}
}
}
impl FromStr for OutType {
type Err = Error;
fn from_str(s: &str) -> Result<OutType, Error> {
match &*s.to_lowercase() {
"path" => Ok(OutType::Path),
"line" => Ok(OutType::Line),
"match" => Ok(OutType::Match),
_ => Err(Error::UnrecognizedOutType(s.to_string())),
}
}
}
impl FromStr for SpecType {
type Err = Error;
fn from_str(s: &str) -> Result<SpecType, Error> {
match &*s.to_lowercase() {
"fg" => Ok(SpecType::Fg),
"bg" => Ok(SpecType::Bg),
"style" => Ok(SpecType::Style),
"none" => Ok(SpecType::None),
_ => Err(Error::UnrecognizedSpecType(s.to_string())),
}
}
}
impl FromStr for Style {
type Err = Error;
fn from_str(s: &str) -> Result<Style, Error> {
match &*s.to_lowercase() {
"bold" => Ok(Style::Bold),
"nobold" => Ok(Style::NoBold),
_ => Err(Error::UnrecognizedStyle(s.to_string())),
}
}
}
#[cfg(test)]
mod tests {
use termcolor::{Color, ColorSpec};
use super::{ColorSpecs, Error, OutType, Spec, SpecValue, Style};
#[test]
fn merge() {
let user_specs: &[Spec] = &[
"match:fg:blue".parse().unwrap(),
"match:none".parse().unwrap(),
"match:style:bold".parse().unwrap(),
];
let mut expect_matched = ColorSpec::new();
expect_matched.set_bold(true);
assert_eq!(ColorSpecs::new(user_specs), ColorSpecs {
path: ColorSpec::default(),
line: ColorSpec::default(),
matched: expect_matched,
});
}
#[test]
fn specs() {
let spec: Spec = "path:fg:blue".parse().unwrap();
assert_eq!(spec, Spec {
ty: OutType::Path,
value: SpecValue::Fg(Color::Blue),
});
let spec: Spec = "path:bg:red".parse().unwrap();
assert_eq!(spec, Spec {
ty: OutType::Path,
value: SpecValue::Bg(Color::Red),
});
let spec: Spec = "match:style:bold".parse().unwrap();
assert_eq!(spec, Spec {
ty: OutType::Match,
value: SpecValue::Style(Style::Bold),
});
let spec: Spec = "line:none".parse().unwrap();
assert_eq!(spec, Spec {
ty: OutType::Line,
value: SpecValue::None,
});
}
#[test]
fn spec_errors() {
let err = "line:nonee".parse::<Spec>().unwrap_err();
assert_eq!(err, Error::UnrecognizedSpecType("nonee".to_string()));
let err = "".parse::<Spec>().unwrap_err();
assert_eq!(err, Error::InvalidFormat("".to_string()));
let err = "foo".parse::<Spec>().unwrap_err();
assert_eq!(err, Error::InvalidFormat("foo".to_string()));
let err = "line:style:italic".parse::<Spec>().unwrap_err();
assert_eq!(err, Error::UnrecognizedStyle("italic".to_string()));
let err = "line:fg:brown".parse::<Spec>().unwrap_err();
match err {
Error::UnrecognizedColor(name, _) => assert_eq!(name, "brown"),
err => assert!(false, "unexpected error: {:?}", err),
}
let err = "foo:fg:brown".parse::<Spec>().unwrap_err();
assert_eq!(err, Error::UnrecognizedOutType("foo".to_string()));
}
}

67
src/search_buffer.rs
View File

@@ -10,7 +10,7 @@ use std::cmp;
use std::path::Path;
use grep::Grep;
use term::Terminal;
use termcolor::WriteColor;
use printer::Printer;
use search_stream::{IterLines, Options, count_lines, is_binary};
@@ -26,7 +26,7 @@ pub struct BufferSearcher<'a, W: 'a> {
last_line: usize,
}
impl<'a, W: Send + Terminal> BufferSearcher<'a, W> {
impl<'a, W: WriteColor> BufferSearcher<'a, W> {
pub fn new(
printer: &'a mut Printer<W>,
grep: &'a Grep,
@@ -61,6 +61,15 @@ impl<'a, W: Send + Terminal> BufferSearcher<'a, W> {
self
}
/// If enabled, searching will print the path of files that *don't* match
/// the given pattern.
///
/// Disabled by default.
pub fn files_without_matches(mut self, yes: bool) -> Self {
self.opts.files_without_matches = yes;
self
}
/// Set the end-of-line byte used by this searcher.
pub fn eol(mut self, eol: u8) -> Self {
self.opts.eol = eol;
@@ -81,6 +90,14 @@ impl<'a, W: Send + Terminal> BufferSearcher<'a, W> {
self
}
/// Limit the number of matches to the given count.
///
/// The default is None, which corresponds to no limit.
pub fn max_count(mut self, count: Option<u64>) -> Self {
self.opts.max_count = count;
self
}
/// If enabled, don't show any output and quit searching after the first
/// match is found.
pub fn quiet(mut self, yes: bool) -> Self {
@@ -111,11 +128,11 @@ impl<'a, W: Send + Terminal> BufferSearcher<'a, W> {
self.print_match(m.start(), m.end());
}
last_end = m.end();
if self.opts.stop_after_first_match() {
if self.opts.terminate(self.match_count) {
break;
}
}
if self.opts.invert_match {
if self.opts.invert_match && !self.opts.terminate(self.match_count) {
let upto = self.buf.len();
self.print_inverted_matches(last_end, upto);
}
@@ -125,6 +142,9 @@ impl<'a, W: Send + Terminal> BufferSearcher<'a, W> {
if self.opts.files_with_matches && self.match_count > 0 {
self.printer.path(self.path);
}
if self.opts.files_without_matches && self.match_count == 0 {
self.printer.path(self.path);
}
self.match_count
}
@@ -146,6 +166,9 @@ impl<'a, W: Send + Terminal> BufferSearcher<'a, W> {
debug_assert!(self.opts.invert_match);
let mut it = IterLines::new(self.opts.eol, start);
while let Some((s, e)) = it.next(&self.buf[..end]) {
if self.opts.terminate(self.match_count) {
return;
}
self.print_match(s, e);
}
}
@@ -173,10 +196,9 @@ mod tests {
use std::path::Path;
use grep::GrepBuilder;
use term::{Terminal, TerminfoTerminal};
use out::ColoredTerminal;
use printer::Printer;
use termcolor;
use super::BufferSearcher;
@@ -193,15 +215,14 @@ and exhibited clearly, with a label attached.\
&Path::new("/baz.rs")
}
type TestSearcher<'a> =
BufferSearcher<'a, ColoredTerminal<TerminfoTerminal<Vec<u8>>>>;
type TestSearcher<'a> = BufferSearcher<'a, termcolor::NoColor<Vec<u8>>>;
fn search<F: FnMut(TestSearcher) -> TestSearcher>(
pat: &str,
haystack: &str,
mut map: F,
) -> (u64, String) {
let outbuf = ColoredTerminal::NoColor(vec![]);
let outbuf = termcolor::NoColor::new(vec![]);
let mut pp = Printer::new(outbuf).with_filename(true);
let grep = GrepBuilder::new(pat).build().unwrap();
let count = {
@@ -266,6 +287,34 @@ and exhibited clearly, with a label attached.\
assert_eq!(out, "/baz.rs\n");
}
#[test]
fn files_without_matches() {
let (count, out) = search(
"zzzz", SHERLOCK, |s| s.files_without_matches(true));
assert_eq!(0, count);
assert_eq!(out, "/baz.rs\n");
}
#[test]
fn max_count() {
let (count, out) = search(
"Sherlock", SHERLOCK, |s| s.max_count(Some(1)));
assert_eq!(1, count);
assert_eq!(out, "\
/baz.rs:For the Doctor Watsons of this world, as opposed to the Sherlock
");
}
#[test]
fn invert_match_max_count() {
let (count, out) = search(
"zzzz", SHERLOCK, |s| s.invert_match(true).max_count(Some(1)));
assert_eq!(1, count);
assert_eq!(out, "\
/baz.rs:For the Doctor Watsons of this world, as opposed to the Sherlock
");
}
#[test]
fn invert_match() {
let (count, out) = search(

183
src/search_stream.rs
View File

@@ -10,9 +10,10 @@ use std::fmt;
use std::io;
use std::path::{Path, PathBuf};
use bytecount;
use grep::{Grep, Match};
use memchr::{memchr, memrchr};
use term::Terminal;
use termcolor::WriteColor;
use printer::Printer;
@@ -81,9 +82,11 @@ pub struct Options {
pub before_context: usize,
pub count: bool,
pub files_with_matches: bool,
pub files_without_matches: bool,
pub eol: u8,
pub invert_match: bool,
pub line_number: bool,
pub max_count: Option<u64>,
pub quiet: bool,
pub text: bool,
}
@@ -95,9 +98,11 @@ impl Default for Options {
before_context: 0,
count: false,
files_with_matches: false,
files_without_matches: false,
eol: b'\n',
invert_match: false,
line_number: false,
max_count: None,
quiet: false,
text: false,
}
@@ -106,20 +111,32 @@ impl Default for Options {
}
impl Options {
/// Several options (--quiet, --count, --files-with-matches) imply that
/// we shouldn't ever display matches.
/// Several options (--quiet, --count, --files-with-matches,
/// --files-without-match) imply that we shouldn't ever display matches.
pub fn skip_matches(&self) -> bool {
self.count || self.files_with_matches || self.quiet
self.count || self.files_with_matches || self.files_without_matches
|| self.quiet
}
/// Some options (--quiet, --files-with-matches) imply that we can stop
/// searching after the first match.
/// Some options (--quiet, --files-with-matches, --files-without-match)
/// imply that we can stop searching after the first match.
pub fn stop_after_first_match(&self) -> bool {
self.files_with_matches || self.quiet
self.files_with_matches || self.files_without_matches || self.quiet
}
/// Returns true if the search should terminate based on the match count.
pub fn terminate(&self, match_count: u64) -> bool {
if match_count > 0 && self.stop_after_first_match() {
return true;
}
if self.max_count.map_or(false, |max| match_count >= max) {
return true;
}
false
}
}
impl<'a, R: io::Read, W: Terminal + Send> Searcher<'a, R, W> {
impl<'a, R: io::Read, W: WriteColor> Searcher<'a, R, W> {
/// Create a new searcher.
///
/// `inp` is a reusable input buffer that is used as scratch space by this
@@ -185,6 +202,14 @@ impl<'a, R: io::Read, W: Terminal + Send> Searcher<'a, R, W> {
self
}
/// If enabled, searching will print the path of files without any matches.
///
/// Disabled by default.
pub fn files_without_matches(mut self, yes: bool) -> Self {
self.opts.files_without_matches = yes;
self
}
/// Set the end-of-line byte used by this searcher.
pub fn eol(mut self, eol: u8) -> Self {
self.opts.eol = eol;
@@ -205,6 +230,14 @@ impl<'a, R: io::Read, W: Terminal + Send> Searcher<'a, R, W> {
self
}
/// Limit the number of matches to the given count.
///
/// The default is None, which corresponds to no limit.
pub fn max_count(mut self, count: Option<u64>) -> Self {
self.opts.max_count = count;
self
}
/// If enabled, don't show any output and quit searching after the first
/// match is found.
pub fn quiet(mut self, yes: bool) -> Self {
@@ -274,13 +307,15 @@ impl<'a, R: io::Read, W: Terminal + Send> Searcher<'a, R, W> {
} else if self.opts.files_with_matches {
self.printer.path(self.path);
}
} else if self.match_count == 0 && self.opts.files_without_matches {
self.printer.path(self.path);
}
Ok(self.match_count)
}
#[inline(always)]
fn terminate(&self) -> bool {
self.match_count > 0 && self.opts.stop_after_first_match()
self.opts.terminate(self.match_count)
}
#[inline(always)]
@@ -317,6 +352,9 @@ impl<'a, R: io::Read, W: Terminal + Send> Searcher<'a, R, W> {
debug_assert!(self.opts.invert_match);
let mut it = IterLines::new(self.opts.eol, self.inp.pos);
while let Some((start, end)) = it.next(&self.inp.buf[..upto]) {
if self.terminate() {
return;
}
self.print_match(start, end);
self.inp.pos = end;
}
@@ -524,9 +562,10 @@ impl InputBuffer {
keep_from: usize,
) -> Result<bool, io::Error> {
// Rollover bytes from buf[keep_from..end] and update our various
// pointers. N.B. This could be done with the unsafe ptr::copy, but
// I haven't been able to produce a benchmark that notices a difference
// in performance. (Invariably, ptr::copy is also clearer IMO.)
// pointers. N.B. This could be done with the ptr::copy, but I haven't
// been able to produce a benchmark that notices a difference in
// performance. (Invariably, ptr::copy is seems clearer IMO, but it is
// not safe.)
self.tmp.clear();
self.tmp.extend_from_slice(&self.buf[keep_from..self.end]);
self.buf[0..self.tmp.len()].copy_from_slice(&self.tmp);
@@ -582,89 +621,9 @@ pub fn is_binary(buf: &[u8]) -> bool {
}
/// Count the number of lines in the given buffer.
#[inline(never)]
#[inline(never)]
pub fn count_lines(buf: &[u8], eol: u8) -> u64 {
// This was adapted from code in the memchr crate. The specific benefit
// here is that we can avoid a branch in the inner loop because all we're
// doing is counting.
// The technique to count EOL bytes was adapted from:
// http://bits.stephan-brumme.com/null.html
const LO_U64: u64 = 0x0101010101010101;
const HI_U64: u64 = 0x8080808080808080;
// use truncation
const LO_USIZE: usize = LO_U64 as usize;
const HI_USIZE: usize = HI_U64 as usize;
#[cfg(target_pointer_width = "32")]
const USIZE_BYTES: usize = 4;
#[cfg(target_pointer_width = "64")]
const USIZE_BYTES: usize = 8;
fn count_eol(eol: usize) -> u64 {
// Ideally, this would compile down to a POPCNT instruction, but
// it looks like you need to set RUSTFLAGS="-C target-cpu=native"
// (or target-feature=+popcnt) to get that to work. Bummer.
(eol.wrapping_sub(LO_USIZE) & !eol & HI_USIZE).count_ones() as u64
}
#[cfg(target_pointer_width = "32")]
fn repeat_byte(b: u8) -> usize {
let mut rep = (b as usize) << 8 | b as usize;
rep = rep << 16 | rep;
rep
}
#[cfg(target_pointer_width = "64")]
fn repeat_byte(b: u8) -> usize {
let mut rep = (b as usize) << 8 | b as usize;
rep = rep << 16 | rep;
rep = rep << 32 | rep;
rep
}
fn count_lines_slow(mut buf: &[u8], eol: u8) -> u64 {
let mut count = 0;
while let Some(pos) = memchr(eol, buf) {
count += 1;
buf = &buf[pos + 1..];
}
count
}
let len = buf.len();
let ptr = buf.as_ptr();
let mut count = 0;
// Search up to an aligned boundary...
let align = (ptr as usize) & (USIZE_BYTES - 1);
let mut i = 0;
if align > 0 {
i = cmp::min(USIZE_BYTES - align, len);
count += count_lines_slow(&buf[..i], eol);
}
// ... and search the rest.
let repeated_eol = repeat_byte(eol);
if len >= 2 * USIZE_BYTES {
while i <= len - (2 * USIZE_BYTES) {
unsafe {
let u = *(ptr.offset(i as isize) as *const usize);
let v = *(ptr.offset((i + USIZE_BYTES) as isize)
as *const usize);
count += count_eol(u ^ repeated_eol);
count += count_eol(v ^ repeated_eol);
}
i += USIZE_BYTES * 2;
}
}
count += count_lines_slow(&buf[i..], eol);
count
bytecount::count(buf, eol) as u64
}
/// Replaces a with b in buf.
@@ -804,10 +763,8 @@ mod tests {
use std::path::Path;
use grep::GrepBuilder;
use term::{Terminal, TerminfoTerminal};
use out::ColoredTerminal;
use printer::Printer;
use termcolor;
use super::{InputBuffer, Searcher, start_of_previous_lines};
@@ -847,7 +804,7 @@ fn main() {
type TestSearcher<'a> = Searcher<
'a,
io::Cursor<Vec<u8>>,
ColoredTerminal<TerminfoTerminal<Vec<u8>>>,
termcolor::NoColor<Vec<u8>>,
>;
fn search_smallcap<F: FnMut(TestSearcher) -> TestSearcher>(
@@ -856,7 +813,7 @@ fn main() {
mut map: F,
) -> (u64, String) {
let mut inp = InputBuffer::with_capacity(1);
let outbuf = ColoredTerminal::NoColor(vec![]);
let outbuf = termcolor::NoColor::new(vec![]);
let mut pp = Printer::new(outbuf).with_filename(true);
let grep = GrepBuilder::new(pat).build().unwrap();
let count = {
@@ -873,7 +830,7 @@ fn main() {
mut map: F,
) -> (u64, String) {
let mut inp = InputBuffer::with_capacity(4096);
let outbuf = ColoredTerminal::NoColor(vec![]);
let outbuf = termcolor::NoColor::new(vec![]);
let mut pp = Printer::new(outbuf).with_filename(true);
let grep = GrepBuilder::new(pat).build().unwrap();
let count = {
@@ -1040,6 +997,34 @@ fn main() {
assert_eq!(out, "/baz.rs\n");
}
#[test]
fn files_without_matches() {
let (count, out) = search_smallcap(
"zzzz", SHERLOCK, |s| s.files_without_matches(true));
assert_eq!(0, count);
assert_eq!(out, "/baz.rs\n");
}
#[test]
fn max_count() {
let (count, out) = search_smallcap(
"Sherlock", SHERLOCK, |s| s.max_count(Some(1)));
assert_eq!(1, count);
assert_eq!(out, "\
/baz.rs:For the Doctor Watsons of this world, as opposed to the Sherlock
");
}
#[test]
fn invert_match_max_count() {
let (count, out) = search(
"zzzz", SHERLOCK, |s| s.invert_match(true).max_count(Some(1)));
assert_eq!(1, count);
assert_eq!(out, "\
/baz.rs:For the Doctor Watsons of this world, as opposed to the Sherlock
");
}
#[test]
fn invert_match() {
let (count, out) = search_smallcap(

176
src/terminal_win.rs
View File

@@ -1,176 +0,0 @@
/*!
This module contains a Windows-only *in-memory* implementation of the
`term::Terminal` trait.
This particular implementation is a bit idiosyncratic, and the "in-memory"
specification is to blame. In particular, on Windows, coloring requires
communicating with the console synchronously as data is written to stdout.
This is anathema to how ripgrep fundamentally works: by writing search results
to intermediate thread local buffers in order to maximize parallelism.
Eliminating parallelism on Windows isn't an option, because that would negate
a tremendous performance benefit just for coloring.
We've worked around this by providing an implementation of `term::Terminal`
that records precisely where a color or a reset should be invoked, according
to a byte offset in the in memory buffer. When the buffer is actually printed,
we copy the bytes from the buffer to stdout incrementally while invoking the
corresponding console APIs for coloring at the right location.
(Another approach would be to do ANSI coloring unconditionally, then parse that
and translate it to console commands. The advantage of that approach is that
it doesn't require any additional memory for storing offsets. In practice
though, coloring is only used in the terminal, which tends to correspond to
searches that produce very few results with respect to the corpus searched.
Therefore, this is an acceptable trade off. Namely, we do not pay for it when
coloring is disabled.
*/
use std::io;
use term::{self, Terminal};
use term::color::Color;
/// An in-memory buffer that provides Windows console coloring.
#[derive(Clone, Debug)]
pub struct WindowsBuffer {
buf: Vec<u8>,
pos: usize,
colors: Vec<WindowsColor>,
}
/// A color associated with a particular location in a buffer.
#[derive(Clone, Debug)]
struct WindowsColor {
pos: usize,
opt: WindowsOption,
}
/// A color or reset directive that can be translated into an instruction to
/// the Windows console.
#[derive(Clone, Debug)]
enum WindowsOption {
Foreground(Color),
Background(Color),
Reset,
}
impl WindowsBuffer {
/// Create a new empty buffer for Windows console coloring.
pub fn new() -> WindowsBuffer {
WindowsBuffer {
buf: vec![],
pos: 0,
colors: vec![],
}
}
fn push(&mut self, opt: WindowsOption) {
let pos = self.pos;
self.colors.push(WindowsColor { pos: pos, opt: opt });
}
/// Print the contents to the given terminal.
pub fn print_stdout<T: Terminal + Send>(&self, tt: &mut T) {
if !tt.supports_color() {
let _ = tt.write_all(&self.buf);
let _ = tt.flush();
return;
}
let mut last = 0;
for col in &self.colors {
let _ = tt.write_all(&self.buf[last..col.pos]);
match col.opt {
WindowsOption::Foreground(c) => {
let _ = tt.fg(c);
}
WindowsOption::Background(c) => {
let _ = tt.bg(c);
}
WindowsOption::Reset => {
let _ = tt.reset();
}
}
last = col.pos;
}
let _ = tt.write_all(&self.buf[last..]);
let _ = tt.flush();
}
/// Clear the buffer.
pub fn clear(&mut self) {
self.buf.clear();
self.colors.clear();
self.pos = 0;
}
}
impl io::Write for WindowsBuffer {
fn write(&mut self, buf: &[u8]) -> io::Result<usize> {
let n = try!(self.buf.write(buf));
self.pos += n;
Ok(n)
}
fn flush(&mut self) -> io::Result<()> {
Ok(())
}
}
impl Terminal for WindowsBuffer {
type Output = Vec<u8>;
fn fg(&mut self, fg: Color) -> term::Result<()> {
self.push(WindowsOption::Foreground(fg));
Ok(())
}
fn bg(&mut self, bg: Color) -> term::Result<()> {
self.push(WindowsOption::Background(bg));
Ok(())
}
fn attr(&mut self, _attr: term::Attr) -> term::Result<()> {
Err(term::Error::NotSupported)
}
fn supports_attr(&self, _attr: term::Attr) -> bool {
false
}
fn reset(&mut self) -> term::Result<()> {
self.push(WindowsOption::Reset);
Ok(())
}
fn supports_reset(&self) -> bool {
true
}
fn supports_color(&self) -> bool {
true
}
fn cursor_up(&mut self) -> term::Result<()> {
Err(term::Error::NotSupported)
}
fn delete_line(&mut self) -> term::Result<()> {
Err(term::Error::NotSupported)
}
fn carriage_return(&mut self) -> term::Result<()> {
Err(term::Error::NotSupported)
}
fn get_ref(&self) -> &Vec<u8> {
&self.buf
}
fn get_mut(&mut self) -> &mut Vec<u8> {
&mut self.buf
}
fn into_inner(self) -> Vec<u8> {
self.buf
}
}

128
src/unescape.rs Normal file
View File

@@ -0,0 +1,128 @@
/// A single state in the state machine used by `unescape`.
#[derive(Clone, Copy, Eq, PartialEq)]
enum State {
/// The state after seeing a `\`.
Escape,
/// The state after seeing a `\x`.
HexFirst,
/// The state after seeing a `\x[0-9A-Fa-f]`.
HexSecond(char),
/// Default state.
Literal,
}
/// Unescapes a string given on the command line. It supports a limited set of
/// escape sequences:
///
/// * \t, \r and \n are mapped to their corresponding ASCII bytes.
/// * \xZZ hexadecimal escapes are mapped to their byte.
pub fn unescape(s: &str) -> Vec<u8> {
use self::State::*;
let mut bytes = vec![];
let mut state = Literal;
for c in s.chars() {
match state {
Escape => {
match c {
'n' => { bytes.push(b'\n'); state = Literal; }
'r' => { bytes.push(b'\r'); state = Literal; }
't' => { bytes.push(b'\t'); state = Literal; }
'x' => { state = HexFirst; }
c => {
bytes.extend(&format!(r"\{}", c).into_bytes());
state = Literal;
}
}
}
HexFirst => {
match c {
'0'...'9' | 'A'...'F' | 'a'...'f' => {
state = HexSecond(c);
}
c => {
bytes.extend(&format!(r"\x{}", c).into_bytes());
state = Literal;
}
}
}
HexSecond(first) => {
match c {
'0'...'9' | 'A'...'F' | 'a'...'f' => {
let ordinal = format!("{}{}", first, c);
let byte = u8::from_str_radix(&ordinal, 16).unwrap();
bytes.push(byte);
state = Literal;
}
c => {
let original = format!(r"\x{}{}", first, c);
bytes.extend(&original.into_bytes());
state = Literal;
}
}
}
Literal => {
match c {
'\\' => { state = Escape; }
c => { bytes.extend(c.to_string().as_bytes()); }
}
}
}
}
match state {
Escape => bytes.push(b'\\'),
HexFirst => bytes.extend(b"\\x"),
HexSecond(c) => bytes.extend(&format!("\\x{}", c).into_bytes()),
Literal => {}
}
bytes
}
#[cfg(test)]
mod tests {
use super::unescape;
fn b(bytes: &'static [u8]) -> Vec<u8> {
bytes.to_vec()
}
#[test]
fn unescape_nul() {
assert_eq!(b(b"\x00"), unescape(r"\x00"));
}
#[test]
fn unescape_nl() {
assert_eq!(b(b"\n"), unescape(r"\n"));
}
#[test]
fn unescape_tab() {
assert_eq!(b(b"\t"), unescape(r"\t"));
}
#[test]
fn unescape_carriage() {
assert_eq!(b(b"\r"), unescape(r"\r"));
}
#[test]
fn unescape_nothing_simple() {
assert_eq!(b(b"\\a"), unescape(r"\a"));
}
#[test]
fn unescape_nothing_hex0() {
assert_eq!(b(b"\\x"), unescape(r"\x"));
}
#[test]
fn unescape_nothing_hex1() {
assert_eq!(b(b"\\xz"), unescape(r"\xz"));
}
#[test]
fn unescape_nothing_hex2() {
assert_eq!(b(b"\\xzz"), unescape(r"\xzz"));
}
}

291
src/worker.rs Normal file
View File

@@ -0,0 +1,291 @@
use std::fs::File;
use std::io;
use std::path::Path;
use grep::Grep;
use ignore::DirEntry;
use memmap::{Mmap, Protection};
use termcolor::WriteColor;
use pathutil::strip_prefix;
use printer::Printer;
use search_buffer::BufferSearcher;
use search_stream::{InputBuffer, Searcher};
use Result;
pub enum Work {
Stdin,
DirEntry(DirEntry),
}
pub struct WorkerBuilder {
grep: Grep,
opts: Options,
}
#[derive(Clone, Debug)]
struct Options {
mmap: bool,
after_context: usize,
before_context: usize,
count: bool,
files_with_matches: bool,
files_without_matches: bool,
eol: u8,
invert_match: bool,
line_number: bool,
max_count: Option<u64>,
no_messages: bool,
quiet: bool,
text: bool,
}
impl Default for Options {
fn default() -> Options {
Options {
mmap: false,
after_context: 0,
before_context: 0,
count: false,
files_with_matches: false,
files_without_matches: false,
eol: b'\n',
invert_match: false,
line_number: false,
max_count: None,
no_messages: false,
quiet: false,
text: false,
}
}
}
impl WorkerBuilder {
/// Create a new builder for a worker.
///
/// A reusable input buffer and a grep matcher are required, but there
/// are numerous additional options that can be configured on this builder.
pub fn new(grep: Grep) -> WorkerBuilder {
WorkerBuilder {
grep: grep,
opts: Options::default(),
}
}
/// Create the worker from this builder.
pub fn build(self) -> Worker {
let mut inpbuf = InputBuffer::new();
inpbuf.eol(self.opts.eol);
Worker {
grep: self.grep,
inpbuf: inpbuf,
opts: self.opts,
}
}
/// The number of contextual lines to show after each match. The default
/// is zero.
pub fn after_context(mut self, count: usize) -> Self {
self.opts.after_context = count;
self
}
/// The number of contextual lines to show before each match. The default
/// is zero.
pub fn before_context(mut self, count: usize) -> Self {
self.opts.before_context = count;
self
}
/// If enabled, searching will print a count instead of each match.
///
/// Disabled by default.
pub fn count(mut self, yes: bool) -> Self {
self.opts.count = yes;
self
}
/// If enabled, searching will print the path instead of each match.
///
/// Disabled by default.
pub fn files_with_matches(mut self, yes: bool) -> Self {
self.opts.files_with_matches = yes;
self
}
/// If enabled, searching will print the path of files without any matches.
///
/// Disabled by default.
pub fn files_without_matches(mut self, yes: bool) -> Self {
self.opts.files_without_matches = yes;
self
}
/// Set the end-of-line byte used by this searcher.
pub fn eol(mut self, eol: u8) -> Self {
self.opts.eol = eol;
self
}
/// If enabled, matching is inverted so that lines that *don't* match the
/// given pattern are treated as matches.
pub fn invert_match(mut self, yes: bool) -> Self {
self.opts.invert_match = yes;
self
}
/// If enabled, compute line numbers and prefix each line of output with
/// them.
pub fn line_number(mut self, yes: bool) -> Self {
self.opts.line_number = yes;
self
}
/// Limit the number of matches to the given count.
///
/// The default is None, which corresponds to no limit.
pub fn max_count(mut self, count: Option<u64>) -> Self {
self.opts.max_count = count;
self
}
/// If enabled, try to use memory maps for searching if possible.
pub fn mmap(mut self, yes: bool) -> Self {
self.opts.mmap = yes;
self
}
/// If enabled, error messages are suppressed.
///
/// This is disabled by default.
pub fn no_messages(mut self, yes: bool) -> Self {
self.opts.no_messages = yes;
self
}
/// If enabled, don't show any output and quit searching after the first
/// match is found.
pub fn quiet(mut self, yes: bool) -> Self {
self.opts.quiet = yes;
self
}
/// If enabled, search binary files as if they were text.
pub fn text(mut self, yes: bool) -> Self {
self.opts.text = yes;
self
}
}
/// Worker is responsible for executing searches on file paths, while choosing
/// streaming search or memory map search as appropriate.
pub struct Worker {
inpbuf: InputBuffer,
grep: Grep,
opts: Options,
}
impl Worker {
/// Execute the worker with the given printer and work item.
///
/// A work item can either be stdin or a file path.
pub fn run<W: WriteColor>(
&mut self,
printer: &mut Printer<W>,
work: Work,
) -> u64 {
let result = match work {
Work::Stdin => {
let stdin = io::stdin();
let stdin = stdin.lock();
self.search(printer, &Path::new("<stdin>"), stdin)
}
Work::DirEntry(dent) => {
let mut path = dent.path();
let file = match File::open(path) {
Ok(file) => file,
Err(err) => {
if !self.opts.no_messages {
eprintln!("{}: {}", path.display(), err);
}
return 0;
}
};
if let Some(p) = strip_prefix("./", path) {
path = p;
}
if self.opts.mmap {
self.search_mmap(printer, path, &file)
} else {
self.search(printer, path, file)
}
}
};
match result {
Ok(count) => {
count
}
Err(err) => {
if !self.opts.no_messages {
eprintln!("{}", err);
}
0
}
}
}
fn search<R: io::Read, W: WriteColor>(
&mut self,
printer: &mut Printer<W>,
path: &Path,
rdr: R,
) -> Result<u64> {
let searcher = Searcher::new(
&mut self.inpbuf, printer, &self.grep, path, rdr);
searcher
.after_context(self.opts.after_context)
.before_context(self.opts.before_context)
.count(self.opts.count)
.files_with_matches(self.opts.files_with_matches)
.files_without_matches(self.opts.files_without_matches)
.eol(self.opts.eol)
.line_number(self.opts.line_number)
.invert_match(self.opts.invert_match)
.max_count(self.opts.max_count)
.quiet(self.opts.quiet)
.text(self.opts.text)
.run()
.map_err(From::from)
}
fn search_mmap<W: WriteColor>(
&mut self,
printer: &mut Printer<W>,
path: &Path,
file: &File,
) -> Result<u64> {
if try!(file.metadata()).len() == 0 {
// Opening a memory map with an empty file results in an error.
// However, this may not actually be an empty file! For example,
// /proc/cpuinfo reports itself as an empty file, but it can
// produce data when it's read from. Therefore, we fall back to
// regular read calls.
return self.search(printer, path, file);
}
let mmap = try!(Mmap::open(file, Protection::Read));
let searcher = BufferSearcher::new(
printer, &self.grep, path, unsafe { mmap.as_slice() });
Ok(searcher
.count(self.opts.count)
.files_with_matches(self.opts.files_with_matches)
.files_without_matches(self.opts.files_without_matches)
.eol(self.opts.eol)
.line_number(self.opts.line_number)
.invert_match(self.opts.invert_match)
.max_count(self.opts.max_count)
.quiet(self.opts.quiet)
.text(self.opts.text)
.run())
}
}

20
termcolor/Cargo.toml Normal file
View File

@@ -0,0 +1,20 @@
[package]
name = "termcolor"
version = "0.1.0" #:version
authors = ["Andrew Gallant <jamslam@gmail.com>"]
description = """
A simple cross platform library for writing colored text to a terminal.
"""
documentation = "https://docs.rs/termcolor"
homepage = "https://github.com/BurntSushi/ripgrep/tree/master/termcolor"
repository = "https://github.com/BurntSushi/ripgrep/tree/master/termcolor"
readme = "README.md"
keywords = ["windows", "win", "color", "ansi", "console"]
license = "Unlicense/MIT"
[lib]
name = "termcolor"
bench = false
[target.'cfg(windows)'.dependencies]
wincolor = { version = "0.1.0", path = "../wincolor" }

88
termcolor/README.md Normal file
View File

@@ -0,0 +1,88 @@
termcolor
=========
A simple cross platform library for writing colored text to a terminal. This
library writes colored text either using standard ANSI escape sequences or
by interacting with the Windows console. Several convenient abstractions
are provided for use in single-threaded or multi-threaded command line
applications.
[![Windows build status](https://ci.appveyor.com/api/projects/status/github/BurntSushi/ripgrep?svg=true)](https://ci.appveyor.com/project/BurntSushi/ripgrep)
[![](https://img.shields.io/crates/v/wincolor.svg)](https://crates.io/crates/wincolor)
[![Linux build status](https://api.travis-ci.org/BurntSushi/ripgrep.png)](https://travis-ci.org/BurntSushi/ripgrep)
[![Windows build status](https://ci.appveyor.com/api/projects/status/github/BurntSushi/ripgrep?svg=true)](https://ci.appveyor.com/project/BurntSushi/ripgrep)
[![](https://img.shields.io/crates/v/termcolor.svg)](https://crates.io/crates/termcolor)
Dual-licensed under MIT or the [UNLICENSE](http://unlicense.org).
### Documentation
[https://docs.rs/termcolor](https://docs.rs/termcolor)
### Usage
Add this to your `Cargo.toml`:
```toml
[dependencies]
termcolor = "0.1"
```
and this to your crate root:
```rust
extern crate termcolor;
```
### Organization
The `WriteColor` trait extends the `io::Write` trait with methods for setting
colors or resetting them.
`Stdout` and `StdoutLock` both satisfy `WriteColor` and are analogous to
`std::io::Stdout` and `std::io::StdoutLock`.
`Buffer` is an in memory buffer that supports colored text. In a parallel
program, each thread might write to its own buffer. A buffer can be printed
to stdout using a `BufferWriter`. The advantage of this design is that
each thread can work in parallel on a buffer without having to synchronize
access to global resources such as the Windows console. Moreover, this design
also prevents interleaving of buffer output.
`Ansi` and `NoColor` both satisfy `WriteColor` for arbitrary implementors of
`io::Write`. These types are useful when you know exactly what you need. An
analogous type for the Windows console is not provided since it cannot exist.
### Example: using `Stdout`
The `Stdout` type in this crate works similarly to `std::io::Stdout`, except
it is augmented with methods for coloring by the `WriteColor` trait. For
example, to write some green text:
```rust
use std::io::Write;
use termcolor::{Color, ColorChoice, ColorSpec, Stdout, WriteColor};
let mut stdout = Stdout::new(ColorChoice::Always);
try!(stdout.set_color(ColorSpec::new().set_fg(Some(Color::Green))));
try!(writeln!(&mut stdout, "green text!"));
```
### Example: using `BufferWriter`
A `BufferWriter` can create buffers and write buffers to stdout. It does *not*
implement `io::Write` or `WriteColor` itself. Instead, `Buffer` implements
`io::Write` and `io::WriteColor`.
This example shows how to print some green text to stdout.
```rust
use std::io::Write;
use termcolor::{BufferWriter, Color, ColorChoice, ColorSpec, WriteColor};
let mut bufwtr = BufferWriter::stdout(ColorChoice::Always);
let mut buffer = bufwtr.buffer();
try!(buffer.set_color(ColorSpec::new().set_fg(Some(Color::Green))));
try!(writeln!(&mut buffer, "green text!"));
try!(bufwtr.print(&buffer));
```

1071
termcolor/src/lib.rs Normal file
View File

File diff suppressed because it is too large Load Diff

112
tests/tests.rs
View File

@@ -339,6 +339,15 @@ sherlock!(files_with_matches, "Sherlock", ".", |wd: WorkDir, mut cmd: Command| {
assert_eq!(lines, expected);
});
sherlock!(files_without_matches, "Sherlock", ".",
|wd: WorkDir, mut cmd: Command| {
wd.create("file.py", "foo");
cmd.arg("--files-without-match");
let lines: String = wd.stdout(&mut cmd);
let expected = "file.py\n";
assert_eq!(lines, expected);
});
sherlock!(after_context, |wd: WorkDir, mut cmd: Command| {
cmd.arg("-A").arg("1");
let lines: String = wd.stdout(&mut cmd);
@@ -552,6 +561,7 @@ sherlock:be, to a very large extent, the result of luck. Sherlock Holmes
assert_eq!(lines, expected);
});
#[cfg(not(windows))]
sherlock!(symlink_nofollow, "Sherlock", ".", |wd: WorkDir, mut cmd: Command| {
wd.remove("sherlock");
wd.create_dir("foo");
@@ -563,6 +573,7 @@ sherlock!(symlink_nofollow, "Sherlock", ".", |wd: WorkDir, mut cmd: Command| {
wd.assert_err(&mut cmd);
});
#[cfg(not(windows))]
sherlock!(symlink_follow, "Sherlock", ".", |wd: WorkDir, mut cmd: Command| {
wd.remove("sherlock");
wd.create_dir("foo");
@@ -789,6 +800,15 @@ clean!(regression_127, "Sherlock", ".", |wd: WorkDir, mut cmd: Command| {
assert_eq!(lines, expected);
});
// See: https://github.com/BurntSushi/ripgrep/issues/128
clean!(regression_128, "x", ".", |wd: WorkDir, mut cmd: Command| {
wd.create_bytes("foo", b"01234567\x0b\n\x0b\n\x0b\n\x0b\nx");
cmd.arg("-n");
let lines: String = wd.stdout(&mut cmd);
assert_eq!(lines, "foo:5:x\n");
});
// See: https://github.com/BurntSushi/ripgrep/issues/131
//
// TODO(burntsushi): Darwin doesn't like this test for some reason.
@@ -865,6 +885,73 @@ clean!(regression_184, "test", ".", |wd: WorkDir, mut cmd: Command| {
assert_eq!(lines, "baz:test\n");
});
// See: https://github.com/BurntSushi/ripgrep/issues/199
clean!(regression_199, r"\btest\b", ".", |wd: WorkDir, mut cmd: Command| {
wd.create("foo", "tEsT");
cmd.arg("--smart-case");
let lines: String = wd.stdout(&mut cmd);
assert_eq!(lines, "foo:tEsT\n");
});
// See: https://github.com/BurntSushi/ripgrep/issues/206
clean!(regression_206, "test", ".", |wd: WorkDir, mut cmd: Command| {
wd.create_dir("foo");
wd.create("foo/bar.txt", "test");
cmd.arg("-g").arg("*.txt");
let lines: String = wd.stdout(&mut cmd);
assert_eq!(lines, format!("{}:test\n", path("foo/bar.txt")));
});
// See: https://github.com/BurntSushi/ripgrep/issues/210
#[cfg(unix)]
#[test]
fn regression_210() {
use std::ffi::OsStr;
use std::os::unix::ffi::OsStrExt;
let badutf8 = OsStr::from_bytes(&b"foo\xffbar"[..]);
let wd = WorkDir::new("regression_210");
let mut cmd = wd.command();
wd.create(badutf8, "test");
cmd.arg("-H").arg("test").arg(badutf8);
let out = wd.output(&mut cmd);
assert_eq!(out.stdout, b"foo\xffbar:test\n".to_vec());
}
// See: https://github.com/BurntSushi/ripgrep/issues/228
clean!(regression_228, "test", ".", |wd: WorkDir, mut cmd: Command| {
wd.create_dir("foo");
cmd.arg("--ignore-file").arg("foo");
wd.assert_err(&mut cmd);
});
// See: https://github.com/BurntSushi/ripgrep/issues/7
sherlock!(feature_7, "-fpat", "sherlock", |wd: WorkDir, mut cmd: Command| {
wd.create("pat", "Sherlock\nHolmes");
let lines: String = wd.stdout(&mut cmd);
let expected = "\
For the Doctor Watsons of this world, as opposed to the Sherlock
Holmeses, success in the province of detective work must always
be, to a very large extent, the result of luck. Sherlock Holmes
";
assert_eq!(lines, expected);
});
// See: https://github.com/BurntSushi/ripgrep/issues/7
sherlock!(feature_7_dash, "-f-", ".", |wd: WorkDir, mut cmd: Command| {
let output = wd.pipe(&mut cmd, "Sherlock");
let lines = String::from_utf8_lossy(&output.stdout);
let expected = "\
sherlock:For the Doctor Watsons of this world, as opposed to the Sherlock
sherlock:be, to a very large extent, the result of luck. Sherlock Holmes
";
assert_eq!(lines, expected);
});
// See: https://github.com/BurntSushi/ripgrep/issues/20
sherlock!(feature_20_no_filename, "Sherlock", ".",
|wd: WorkDir, mut cmd: Command| {
@@ -980,6 +1067,16 @@ sherlock!(feature_89_files_with_matches, "Sherlock", ".",
assert_eq!(lines, "sherlock\x00");
});
// See: https://github.com/BurntSushi/ripgrep/issues/89
sherlock!(feature_89_files_without_matches, "Sherlock", ".",
|wd: WorkDir, mut cmd: Command| {
wd.create("file.py", "foo");
cmd.arg("--null").arg("--files-without-match");
let lines: String = wd.stdout(&mut cmd);
assert_eq!(lines, "file.py\x00");
});
// See: https://github.com/BurntSushi/ripgrep/issues/89
sherlock!(feature_89_count, "Sherlock", ".",
|wd: WorkDir, mut cmd: Command| {
@@ -1043,6 +1140,21 @@ clean!(feature_109_case_sensitive_part2, "test", ".",
wd.assert_err(&mut cmd);
});
// See: https://github.com/BurntSushi/ripgrep/issues/159
clean!(feature_159_works, "test", ".", |wd: WorkDir, mut cmd: Command| {
wd.create("foo", "test\ntest");
cmd.arg("-m1");
let lines: String = wd.stdout(&mut cmd);
assert_eq!(lines, "foo:test\n");
});
// See: https://github.com/BurntSushi/ripgrep/issues/159
clean!(feature_159_zero_max, "test", ".", |wd: WorkDir, mut cmd: Command| {
wd.create("foo", "test\ntest");
cmd.arg("-m0");
wd.assert_err(&mut cmd);
});
#[test]
fn binary_nosearch() {
let wd = WorkDir::new("binary_nosearch");

66
tests/workdir.rs
View File

@@ -43,9 +43,14 @@ impl WorkDir {
/// Create a new file with the given name and contents in this directory.
pub fn create<P: AsRef<Path>>(&self, name: P, contents: &str) {
self.create_bytes(name, contents.as_bytes());
}
/// Create a new file with the given name and contents in this directory.
pub fn create_bytes<P: AsRef<Path>>(&self, name: P, contents: &[u8]) {
let path = self.dir.join(name);
let mut file = nice_err(&path, File::create(&path));
nice_err(&path, file.write_all(contents.as_bytes()));
nice_err(&path, file.write_all(contents));
nice_err(&path, file.flush());
}
@@ -71,8 +76,29 @@ impl WorkDir {
}
/// Returns the path to the ripgrep executable.
#[cfg(not(windows))]
pub fn bin(&self) -> PathBuf {
self.root.join("rg")
let path = self.root.join("rg");
if !path.is_file() {
// Looks like a recent version of Cargo changed the cwd or the
// location of the test executable.
self.root.join("../rg")
} else {
path
}
}
/// Returns the path to the ripgrep executable.
#[cfg(windows)]
pub fn bin(&self) -> PathBuf {
let path = self.root.join("rg.exe");
if !path.is_file() {
// Looks like a recent version of Cargo changed the cwd or the
// location of the test executable.
self.root.join("../rg.exe")
} else {
path
}
}
/// Returns the path to this directory.
@@ -148,7 +174,41 @@ impl WorkDir {
/// Gets the output of a command. If the command failed, then this panics.
pub fn output(&self, cmd: &mut process::Command) -> process::Output {
let o = cmd.output().unwrap();
let output = cmd.output().unwrap();
self.expect_success(cmd, output)
}
/// Pipe `input` to a command, and collect the output.
pub fn pipe(
&self,
cmd: &mut process::Command,
input: &str
) -> process::Output {
cmd.stdin(process::Stdio::piped());
cmd.stdout(process::Stdio::piped());
cmd.stderr(process::Stdio::piped());
let mut child = cmd.spawn().unwrap();
// Pipe input to child process using a separate thread to avoid
// risk of deadlock between parent and child process.
let mut stdin = child.stdin.take().expect("expected standard input");
let input = input.to_owned();
let worker = thread::spawn(move || {
write!(stdin, "{}", input)
});
let output = self.expect_success(cmd, child.wait_with_output().unwrap());
worker.join().unwrap().unwrap();
output
}
/// If `o` is not the output of a successful process run
fn expect_success(
&self,
cmd: &process::Command,
o: process::Output
) -> process::Output {
if !o.status.success() {
let suggest =
if o.stderr.is_empty() {

21
wincolor/Cargo.toml Normal file
View File

@@ -0,0 +1,21 @@
[package]
name = "wincolor"
version = "0.1.0" #:version
authors = ["Andrew Gallant <jamslam@gmail.com>"]
description = """
A simple Windows specific API for controlling text color in a Windows console.
"""
documentation = "https://docs.rs/wincolor"
homepage = "https://github.com/BurntSushi/ripgrep/tree/master/wincolor"
repository = "https://github.com/BurntSushi/ripgrep/tree/master/wincolor"
readme = "README.md"
keywords = ["windows", "win", "color", "ansi", "console"]
license = "Unlicense/MIT"
[lib]
name = "wincolor"
bench = false
[dependencies]
kernel32-sys = "0.2.2"
winapi = "0.2.8"

44
wincolor/README.md Normal file
View File

@@ -0,0 +1,44 @@
wincolor
========
A simple Windows specific API for controlling text color in a Windows console.
The purpose of this crate is to expose the full inflexibility of the Windows
console without any platform independent abstraction.
[![Windows build status](https://ci.appveyor.com/api/projects/status/github/BurntSushi/ripgrep?svg=true)](https://ci.appveyor.com/project/BurntSushi/ripgrep)
[![](https://img.shields.io/crates/v/wincolor.svg)](https://crates.io/crates/wincolor)
Dual-licensed under MIT or the [UNLICENSE](http://unlicense.org).
### Documentation
[https://docs.rs/wincolor](https://docs.rs/wincolor)
### Usage
Add this to your `Cargo.toml`:
```toml
[dependencies]
wincolor = "0.1"
```
and this to your crate root:
```rust
extern crate wincolor;
```
### Example
This is a simple example that shows how to write text with a foreground color
of cyan and the intense attribute set:
```rust
use wincolor::{Console, Color, Intense};
let mut con = Console::stdout().unwrap();
con.fg(Intense::Yes, Color::Cyan).unwrap();
println!("This text will be intense cyan.");
con.reset().unwrap();
println!("This text will be normal.");
```

29
wincolor/src/lib.rs Normal file
View File

@@ -0,0 +1,29 @@
/*!
This crate provides a safe and simple Windows specific API to control
text attributes in the Windows console. Text attributes are limited to
foreground/background colors, as well as whether to make colors intense or not.
Note that on non-Windows platforms, this crate is empty but will compile.
# Example
```no_run
use wincolor::{Console, Color, Intense};
let mut con = Console::stdout().unwrap();
con.fg(Intense::Yes, Color::Cyan).unwrap();
println!("This text will be intense cyan.");
con.reset().unwrap();
println!("This text will be normal.");
```
*/
#[cfg(windows)]
extern crate kernel32;
#[cfg(windows)]
extern crate winapi;
#[cfg(windows)]
pub use win::*;
#[cfg(windows)]
mod win;

223
wincolor/src/win.rs Normal file
View File

@@ -0,0 +1,223 @@
use std::io;
use std::mem;
use kernel32;
use winapi::{DWORD, HANDLE, WORD};
use winapi::winbase::STD_OUTPUT_HANDLE;
use winapi::wincon::{
FOREGROUND_BLUE as FG_BLUE,
FOREGROUND_GREEN as FG_GREEN,
FOREGROUND_RED as FG_RED,
FOREGROUND_INTENSITY as FG_INTENSITY,
};
const FG_CYAN: DWORD = FG_BLUE | FG_GREEN;
const FG_MAGENTA: DWORD = FG_BLUE | FG_RED;
const FG_YELLOW: DWORD = FG_GREEN | FG_RED;
const FG_WHITE: DWORD = FG_BLUE | FG_GREEN | FG_RED;
/// A Windows console.
///
/// This represents a very limited set of functionality available to a Windows
/// console. In particular, it can only change text attributes such as color
/// and intensity.
///
/// There is no way to "write" to this console. Simply write to
/// stdout or stderr instead, while interleaving instructions to the console
/// to change text attributes.
///
/// A common pitfall when using a console is to forget to flush writes to
/// stdout before setting new text attributes.
#[derive(Debug)]
pub struct Console {
handle: HANDLE,
start_attr: TextAttributes,
cur_attr: TextAttributes,
}
unsafe impl Send for Console {}
impl Drop for Console {
fn drop(&mut self) {
unsafe { kernel32::CloseHandle(self.handle); }
}
}
impl Console {
/// Create a new Console to stdout.
///
/// If there was a problem creating the console, then an error is returned.
pub fn stdout() -> io::Result<Console> {
let mut info = unsafe { mem::zeroed() };
let (handle, res) = unsafe {
let handle = kernel32::GetStdHandle(STD_OUTPUT_HANDLE);
(handle, kernel32::GetConsoleScreenBufferInfo(handle, &mut info))
};
if res == 0 {
return Err(io::Error::last_os_error());
}
let attr = TextAttributes::from_word(info.wAttributes);
Ok(Console {
handle: handle,
start_attr: attr,
cur_attr: attr,
})
}
/// Applies the current text attributes.
fn set(&mut self) -> io::Result<()> {
let attr = self.cur_attr.to_word();
let res = unsafe {
kernel32::SetConsoleTextAttribute(self.handle, attr)
};
if res == 0 {
return Err(io::Error::last_os_error());
}
Ok(())
}
/// Apply the given intensity and color attributes to the console
/// foreground.
///
/// If there was a problem setting attributes on the console, then an error
/// is returned.
pub fn fg(&mut self, intense: Intense, color: Color) -> io::Result<()> {
self.cur_attr.fg_color = color;
self.cur_attr.fg_intense = intense;
self.set()
}
/// Apply the given intensity and color attributes to the console
/// background.
///
/// If there was a problem setting attributes on the console, then an error
/// is returned.
pub fn bg(&mut self, intense: Intense, color: Color) -> io::Result<()> {
self.cur_attr.bg_color = color;
self.cur_attr.bg_intense = intense;
self.set()
}
/// Reset the console text attributes to their original settings.
///
/// The original settings correspond to the text attributes on the console
/// when this `Console` value was created.
///
/// If there was a problem setting attributes on the console, then an error
/// is returned.
pub fn reset(&mut self) -> io::Result<()> {
self.cur_attr = self.start_attr;
self.set()
}
}
/// A representation of text attributes for the Windows console.
#[derive(Copy, Clone, Debug, Eq, PartialEq)]
struct TextAttributes {
fg_color: Color,
fg_intense: Intense,
bg_color: Color,
bg_intense: Intense,
}
impl TextAttributes {
fn to_word(&self) -> WORD {
let mut w = 0;
w |= self.fg_color.to_fg();
w |= self.fg_intense.to_fg();
w |= self.bg_color.to_bg();
w |= self.bg_intense.to_bg();
w as WORD
}
fn from_word(word: WORD) -> TextAttributes {
let attr = word as DWORD;
TextAttributes {
fg_color: Color::from_fg(attr),
fg_intense: Intense::from_fg(attr),
bg_color: Color::from_bg(attr),
bg_intense: Intense::from_bg(attr),
}
}
}
/// Whether to use intense colors or not.
#[derive(Clone, Copy, Debug, Eq, PartialEq)]
pub enum Intense {
Yes,
No,
}
impl Intense {
fn to_bg(&self) -> DWORD {
self.to_fg() << 4
}
fn from_bg(word: DWORD) -> Intense {
Intense::from_fg(word >> 4)
}
fn to_fg(&self) -> DWORD {
match *self {
Intense::No => 0,
Intense::Yes => FG_INTENSITY,
}
}
fn from_fg(word: DWORD) -> Intense {
if word & FG_INTENSITY > 0 {
Intense::Yes
} else {
Intense::No
}
}
}
/// The set of available colors for use with a Windows console.
#[derive(Clone, Copy, Debug, Eq, PartialEq)]
pub enum Color {
Black,
Blue,
Green,
Red,
Cyan,
Magenta,
Yellow,
White,
}
impl Color {
fn to_bg(&self) -> DWORD {
self.to_fg() << 4
}
fn from_bg(word: DWORD) -> Color {
Color::from_fg(word >> 4)
}
fn to_fg(&self) -> DWORD {
match *self {
Color::Black => 0,
Color::Blue => FG_BLUE,
Color::Green => FG_GREEN,
Color::Red => FG_RED,
Color::Cyan => FG_CYAN,
Color::Magenta => FG_MAGENTA,
Color::Yellow => FG_YELLOW,
Color::White => FG_WHITE,
}
}
fn from_fg(word: DWORD) -> Color {
match word & 0b111 {
FG_BLUE => Color::Blue,
FG_GREEN => Color::Green,
FG_RED => Color::Red,
FG_CYAN => Color::Cyan,
FG_MAGENTA => Color::Magenta,
FG_YELLOW => Color::Yellow,
FG_WHITE => Color::White,
_ => Color::Black,
}
}
}