twin/ripgrep
1
0
Fork 0
mirror of https://github.com/BurntSushi/ripgrep.git synced 2025-07-28 10:41:58 -07:00
Code Issues Packages Projects Releases Wiki Activity

Compare commits

base: twin:grep-regex-0.1.1
Branches Tags
twin:master twin:ag/rollup 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:ag/partial-bstr-migration
Branches Tags
twin:ag/rollup 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

108 Commits
grep-regex ... ag/partial

Author SHA1 Message Date
Andrew Gallant
a872d33714 deps: add bstr to Cargo.lock 2019-04-05 22:58:58 -04:00
Andrew Gallant
f08f274c5f regex: print out final regex in trace mode 
This is useful for debugging to see what regex is actually being run.
We put this as a trace since the regex can be quite gnarly. (It is not
pretty printed.)
 2019-04-05 22:58:58 -04:00
Andrew Gallant
db7e828989 regex: fix a perf bug when using -w flag 
When looking for an inner literal to speed up searches, if only a prefix
is found, then we generally give up doing inner literal optimizations since
the regex engine will generally handle it for us. Unfortunately, this
decision was being made *before* we wrap the regex in (^|\W)...($|\W) when
using the -w/--word-regexp flag, which would then defeat the literal
optimizations inside the regex engine.

We fix this with a bit of a hack that says, "if we're doing a word regexp,
then give me back any literal you find, even if it's a prefix."
 2019-04-05 22:58:58 -04:00
Andrew Gallant
fb6cad7152 globset: small perf improvements 
This tweaks the path handling functions slightly to make them a hair
faster. In particular, `file_name` is called on every path that ripgrep
visits, and it was possible to remove a few branches without changing
behavior.
 2019-04-05 22:58:58 -04:00
Andrew Gallant
8e1d40ed7d globset: use bstr 
This simplifies the various path related functions and pushed more platform
dependent code down into bstr. This likely also makes things a bit more
efficient on Windows, since we now only do a single UTF-8 check for each
file path.
 2019-04-05 22:58:58 -04:00
Andrew Gallant
b1c064d5af cli: use bstr 
This uses bstr in the unescaping logic. This lets us remove some platform
specific code, and also lets us remove a hacked UTF-8 decoder on raw
bytes.
 2019-04-05 20:42:33 -04:00
Andrew Gallant
26a83c6301 config: switch to using bstrs 
This lets us implement correct Unicode trimming and also simplifies the
parsing logic a bit. This also removes the last platform specific bits of
code in ripgrep core.
 2019-04-05 20:42:32 -04:00
Andrew Gallant
5e50a3c43c printer: use bstr 
This starts the usage of bstr in the printer. We don't use it too much
yet, but it comes in handy for implementing PrinterPath and lets us push
down some platform specific code into bstr.
 2019-04-05 20:42:25 -04:00
Andrew Gallant
85417e52e9 searcher: partially migrate to bstr 
This commit causes grep-searcher to use byte strings internally for its
line buffer support. We manage to remove a use of `unsafe` by doing this
(by pushing it down into `bstr`).

We stop short of using byte strings everywhere else because we rely
heavily on the `impl ops::Index<[u8]> for grep_matcher::Match` impl,
which isn't available for byte strings. (It is premature to make bstr a
public dep of a core crate like grep-matcher, but maybe some day.)
 2019-04-05 20:41:26 -04:00
Andrew Gallant
bef1f0e770 ci: switch to xenial (#1234) 
Rust is having problems with trusty, in particular, see this bug I
filed: https://github.com/rust-lang/rust/issues/59411

This was purpotedly fixed in
https://github.com/rust-lang/rust/pull/59468,
but it appears the issue is still occurring.

This commit tries to update to Ubuntu 16.04 in the hope that it will fix
this problem.
 2019-04-03 19:52:34 -04:00
Andrew Gallant
cd9815cb37 deps: update to aho-corasick 0.7 
We do the simplest possible change to migrate to the new version.

Fixes #1228
 2019-04-03 13:51:26 -04:00
Andrew Gallant
3f22c3a658 deps: update everything 
This updates all dependencies to their latest versions.

We tolerate a duplicative aho-corasick for now, which we will fix in the
next commit.
 2019-04-03 13:07:26 -04:00
Andrew Gallant
0913972104 deps: bump encoding_rs_io 
This brings in a new API for disabling BOM sniffing.

This is part of the work toward completing
https://github.com/BurntSushi/ripgrep/issues/1207
 2019-03-03 16:36:34 -05:00
Andrew Gallant
f19b84fb23 regex: bump regex dep to fix match bug 
See

* 661bf53d5b
* edf45e6f5f

for details on the bug fix, which was in the regex engine.

Fixes #1203
 2019-02-27 17:42:14 -05:00
Andrew Gallant
59fc583aeb readme: include details about filtering 
Despite the fact that we mention this in several places, people are
still surprised by ripgrep's "smart" filtering.
 2019-02-27 08:01:23 -05:00
Andrew Gallant
1c7c4e6640 deps: update tempfile 2019-02-21 16:32:17 -05:00
Andrew Gallant
69c5e3938d deps: bump smallvec 
This gets rid of the unmaintained crates `unreachable` and `void`. Yay!
 2019-02-21 16:31:48 -05:00
Andrew Gallant
d9cf05ad50 deps: update to aho-corasick 0.6.10 
This brings in a fix for this bug:
https://github.com/BurntSushi/aho-corasick/issues/37

Fixes #1079
 2019-02-16 11:39:33 -05:00
Andrew Gallant
af8b6caebb deps: update various dependencies 2019-02-16 09:39:42 -05:00
Andrew Gallant
c84cfb6756 grep-regex-0.1.2 2019-02-16 09:30:06 -05:00
Andrew Gallant
895e26a000 ci: don't do releases on all tags 
This attempts to make Appveyor more conservative in what tags it thinks
are releases. I don't know for sure, but it looks like the previous
regex could match anywhere, so we anchor it.

Fixes #1195
 2019-02-10 12:51:56 -05:00
Andrew Gallant
8c95290ff6 deps: miscellaneous updates 2019-02-10 07:45:08 -05:00
Andrew Gallant
d6feeb7ff2 grep-searcher-0.1.3 2019-02-10 07:42:37 -05:00
Andrew Gallant
626ed00c19 searcher: revert big-endian patch 
This undoes the patch to stop using bytecount on big-endian
architectures. In particular, we bump our bytecount dependency to the
latest release, which has a fix.

This reverts commit a4868b8835.

Fixes #1144 (again), Closes #1194
 2019-02-10 07:40:32 -05:00
Andrew Gallant
332ad18401 tests: use const constructor for atomics 
We did this in 05411b2b for core ripgrep, but didn't carry it over to
tests.
 2019-02-09 16:27:25 -05:00
Andrew Gallant
fc3cf41247 grep-searcher-0.1.2 2019-02-09 16:13:07 -05:00
Andrew Gallant
a4868b8835 searcher: use naive line counting on big-endian 
This patches out bytecount's "fast" vectorized algorithm on big-endian
machines, where it has been observed to fail. Going forward, bytecount
should probably fix this on their end, but for now, we take a small
performance hit on big-endian machines.

Fixes #1144
 2019-02-09 16:13:07 -05:00
John Schmidt
f99b991117 ignore/types: add zig 
PR #1191
 2019-02-08 08:12:40 -05:00
Andrew Gallant
de0bc78982 deps: bump encoding_rs to 0.8.16 
This brings in an updated `encoding_rs` crate that uses `packed_simd`,
which compiles on the latest nightly. Compilation times do appear to be
impacted significantly though.

Fixes #1175 (again)
 2019-02-07 17:05:14 -05:00
Steffen Banhardt
147e96914c ignore/types: *.dtx and *.ins added for tex 
PR #1182
 2019-01-31 09:06:19 -05:00
Andrew Gallant
0abc40c23c readme: bump MSRV 
We bumped it a while back in the CI configuration, but didn't update the
README.
 2019-01-29 13:10:43 -05:00
Andrew Gallant
f768796e4f deps: update other deps 2019-01-29 13:08:56 -05:00
Andrew Gallant
da0c0c4705 deps: update to crossbeam-channel 0.3.8 
This drops dependencies on parking_lot and rand from ripgrep.

(rand is still used for tests.)
 2019-01-29 13:07:37 -05:00
Andrew Gallant
05411b2b32 deprecated: remove use of ATOMIC_BOOL_INIT 
Our MSRV is high enough that we can use const functions now.
 2019-01-29 13:05:16 -05:00
Andrew Gallant
cc93db3b18 cargo: include auto-generated message 
This is going to be annoying for a while if one switches between the
latest nightly compiler and older compilers. Sigh.
 2019-01-29 13:04:40 -05:00
Alex Macleod
049354b766 readme: remove EOL Fedora install instructions 
Fedora 27 and below are past their EOL, so it can now be said that it's
supported regularly on Fedora.

PR #1177
 2019-01-28 08:15:36 -05:00
Andrew Gallant
386dd2806d changelog: BUG #916 
This was fixed by bumping the MSRV above Rust 1.28.

Fixes #916
 2019-01-27 13:15:17 -05:00
Andrew Gallant
5fe9a954e6 changelog: BUG #1154 2019-01-27 13:05:50 -05:00
Andrew Gallant
f158a42a71 ignore: correctly detect hidden files on Windows 
This commit fixes a bug where ripgrep only treated files beginning with
a `.` as hidden. On Windows, we continue this tradition, but
additionally check whether a file has the special Windows "hidden"
attribute set. If so, we treat it as a hidden file.

In order to make this work without an additional stat call, we had to
rearrange some of the plumbing from the directory traverser.

Fixes #1154
 2019-01-27 12:11:52 -05:00
Andrew Gallant
5724391d39 doc: small updates to the FAQ and GUIDE 
Notably, ripgrep can do multiline search now. We also update the
supported compression format list and replace deprecated flags like
`--sort-files` with `--sort path`.
 2019-01-26 16:19:09 -05:00
Andrew Gallant
0df71240ff search: fix -F and -f interaction bug 
This fixes what appears to be a pretty egregious regression where the
`-F/--fixed-strings` flag wasn't be applied to patterns supplied via
the `-f/--file` flag. The same bug existed for the `-x/--line-regexp`
flag as well, which we fix here.

Fixes #1176
 2019-01-26 16:01:52 -05:00
Andrew Gallant
f3164f2615 exit: tweak exit status logic 
This changes how ripgrep emit exit status codes. In particular, any error
that occurs while searching will now cause ripgrep to emit a `2` exit
code, where as it previously would emit either a `0` or a `1` code based
on whether it matched or not. That is, ripgrep would only emit a `2` exit
code for a catastrophic error.

This tweak includes additional logic that GNU grep adheres to, which seems
like good sense. Namely, if -q/--quiet is given, and an error occurs and
a match occurs, then ripgrep will emit a `0` exit code.

Closes #1159
 2019-01-26 15:44:49 -05:00
Andrew Gallant
31d3e24130 args: prevent panicking in 'rg -h | rg' 
Previously, we relied on clap to handle printing either an error
message, or --help/--version output, in addition to setting the exit
status code. Unfortunately, for --help/--version output, clap was
panicking if the write failed, which can happen in fairly common
scenarios via a broken pipe error. e.g., `rg -h | head`.

We fix this by using clap's "safe" API and doing the printing ourselves.
We also set the exit code to `2` when an invalid command has been given.

Fixes #1125 and partially addresses #1159
 2019-01-26 14:39:40 -05:00
Andrew Gallant
bf842dbc7f doc: add note about inverted flags 
Fixes #1091
 2019-01-26 14:13:06 -05:00
Andrew Gallant
6d5dba85bd doc: clarify automatic encoding detection 
Fixes #1103
 2019-01-26 13:55:47 -05:00
Andrew Gallant
afb89bcdad fmt: shorten --ignore-file-case-insensitive description 2019-01-26 13:45:02 -05:00
Andrew Gallant
332dc56372 changelog: BUG #1095 2019-01-26 13:40:59 -05:00
Andrew Gallant
12a6ca45f9 config: add --no-ignore-dot flag 
This flag causes ripgrep to ignore `.ignore` files.

Closes #1138
 2019-01-26 13:40:12 -05:00
Andrew Gallant
9d703110cf regex: make CRLF hack more robust 
This commit improves the CRLF hack to be more robust. In particular, in
addition to rewriting `$` as `(?:\r??$)`, we now strip `\r` from the end
of a match if and only if the regex has an ending line anchor required for
a match. This doesn't quite make the hack 100% correct, but should fix most
use cases in practice. An example of a regex that will still be incorrect
is `foo|bar$`, since the analysis isn't quite sophisticated enough to
determine that a `\r` can be safely stripped from any match. Even if we
fix that, regexes like `foo\r|bar$` still won't be handled correctly. Alas,
more work on this front should really be focused on enabling this in the
regex engine itself.

The specific cause of this bug was that grep-searcher was sneakily
stripping CRLF from matching lines when it really shouldn't have. We remove
that code now, and instead rely on better match semantics provided at a
lower level.

Fixes #1095
 2019-01-26 12:34:28 -05:00
Andrew Gallant
e99b6bda0e deps: bump regex-syntax to 0.6.5 
This is necessary for the use of the new is_line_anchored_{start,end}
APIs.
 2019-01-26 12:20:02 -05:00
Andrew Gallant
276e2c9b9a searcher: always strip BOM 
This fixes a bug where a BOM prefix was included. While this was somewhat
intentional in order to have a faithful "UTF8 passthru" option, in
practice, this causes problems such as breaking patterns like `^` in a
really non-obvious way.

The actual fix was to add a new API to encoding_rs_io, which this commit
brings in.

Fixes #1163
 2019-01-25 17:18:57 -05:00
Andrew Gallant
9a9f54d44c readme: encoding_rs's SIMD support is broken 
Add a note about it to the README.

Also, remove mention of the avx-accel feature since it no longer exists.
(bytecount now uses runtime detection to enable SIMD support.)

Fixes #1175
 2019-01-24 07:00:53 -05:00
Andrew Gallant
47833b9ce7 deps: update removal of grep devdeps 2019-01-23 20:14:37 -05:00
Awad Mackie
44a9e37737 ignore/types: add method for retrieving file type definition 
Fixes #1116, Closes #1120
 2019-01-23 20:08:48 -05:00
Andrew Gallant
8fd05cacee changelog: BUG #1121 2019-01-23 20:06:01 -05:00
Rob Lourens
4691d11034 ripgrep: don't skip stdout in --files mode 
Specifically, this avoids triggering Windows antimalware when in --files mode.

See also #600.

Fixes #1121
 2019-01-23 20:04:44 -05:00
Andrew Gallant
519a6b68af grep: remove unused dependencies 
We remove these for now, but we'll eventually add them back once the
examples get more fleshed out.

Closes #1043
 2019-01-23 20:01:32 -05:00
Andrew Gallant
9c940b45f4 globset: permit ** to appear anywhere 
Previously, `man gitignore` specified that `**` was invalid unless it
was used in one of a few specific circumstances, i.e., `**`, `a/**`,
`**/b` or `a/**/b`. That is, `**` always had to be surrounded by either
a path separator or the beginning/end of the pattern.

It turns out that git itself has treated `**` outside the above contexts
as valid for quite a while, so there was an inconsistency between the
spec `man gitignore` and the implementation, and it wasn't clear which
was actually correct.

@okdana filed a bug against git[1] and got this fixed. The spec was wrong,
which has now been fixed [2] and updated[2].

This commit brings ripgrep in line with git and treats `**` outside of
the above contexts as two consecutive `*` patterns. We deprecate the
`InvalidRecursive` error since it is no longer used.

Fixes #373, Fixes #1098

[1] - https://public-inbox.org/git/C16A9F17-0375-42F9-90A9-A92C9F3D8BBA@dana.is
[2] - 627186d020
[3] - https://git-scm.com/docs/gitignore
 2019-01-23 19:59:39 -05:00
Andrew Gallant
0a167021c3 changelog: BUG #1174 2019-01-23 19:19:26 -05:00
Andrew Gallant
aeaa5fc1b1 globset: fix repeated use of ** 
This fixes a bug where repeated use of ** didn't behave as it should. In
particular, each use of `**` added a new requirement directory depth
requirement. For example, something like `**/**/b` would match
`foo/bar/b`, but it wouldn't match `foo/b` even though it should. In
particular, `**` semantics demand "infinite" depth, so repeated uses of
`**` should just coalesce as if only one was given.

We do this coalescing in the parser. It's a little tricky because we
treat `**/a`, `a/**` and `a/**/b` as distinct tokens with their own
regex conversions. We also test the crap out of it.

Fixes #1174
 2019-01-23 19:15:02 -05:00
Andrew Gallant
7048a06c31 changelog: BUG #1173 2019-01-23 18:14:16 -05:00
Andrew Gallant
23be3cf850 ignore: fix handling of ** 
When deciding whether to add the `**/` prefix or not, we should choose
not to add it if the pattern is simply a bare `**`. Previously, we were
only not adding it if it was `**/`, which is correct, but we also need
to do it for `**` since `**` can already match anywhere.

There's likely a more principled solution to this, but this works for
now.

Fixes #1173
 2019-01-23 18:12:35 -05:00
Andrew Gallant
b48bbf527d changelog: PR #1093 2019-01-23 17:56:18 -05:00
dana
8eabe47b57 ignore: always use literal_separator for gitignore patterns (#1093) 
PR #1093
 2019-01-23 17:54:28 -05:00
Michele Bologna
ff712bfd9d readme: add instructions for openSUSE 15.0 
PR #1088
 2019-01-22 21:46:11 -05:00
Mika Dede
a7f2d48234 printer: fix path handling in summarizer 
This commit fixes a bug where both of the following commands always
reported an error:

    rg --files-with-matches foo file
    rg --files-without-match foo file

In particular, the printer was erroneously respecting the `path` option
even the the summary kind was `PathWithMatch` or `PathWithoutMatch`. The
documented behavior is that those summary kinds always require a path,
and thus, the `path` option has no effect. We fix this by correcting the
case analysis.

This also fixes a bug where the exit code for `--files-without-match`
was not set correctly. We update the printer's `has_match` method to
report the correct value.

Fixes #1106, Closes #1130
 2019-01-22 21:37:23 -05:00
Andrew Gallant
57500ad013 changelog: brotli/zstd addition 2019-01-22 20:57:28 -05:00
dana
0b04553aff grep-cli: support Brotli/Zstd decompression 
Fixes #1099
 2019-01-22 20:56:16 -05:00
dana
1ae121122f ignore/types: add/update brotli, bzip2, gzip, xz, zstd 2019-01-22 20:56:16 -05:00
Andrew Gallant
688003e51c ripgrep: ban rustfmt 2019-01-22 20:07:26 -05:00
David Torosyan
718a00f6f2 ripgrep: add --ignore-file-case-insensitive 
The --ignore-file-case-insensitive flag causes all
.gitignore/.rgignore/.ignore files to have their globs matched without
regard for case. Because this introduces a potentially significant
performance regression, this is always disabled by default. Users that
need case insensitive matching can enable it on a case by case basis.

Closes #1164, Closes #1170
 2019-01-22 20:03:59 -05:00
Andrew Gallant
7cbc535d70 edition: fix build.rs 2019-01-19 10:46:57 -05:00
Andrew Gallant
7a6a40bae1 edition: move core ripgrep to Rust 2018 2019-01-19 10:44:30 -05:00
Andrew Gallant
1e9ee2cc85 deps: update memmap 2019-01-19 10:44:30 -05:00
Andrew Gallant
968491f8e9 deps: update to bytecount 0.5 
bytecount now uses runtime dispatch for enabling SIMD, which means we can
no longer need the avx-accel features. We remove it from ripgrep since the
next release will be a minor version bump, but leave them as no-ops for
the crates that previously used it.
 2019-01-19 10:44:30 -05:00
Andrew Gallant
63b0f31a22 deps: update various dependencies 
We also increase the MSRV to 1.32, the current stable release, which sets
the stage for migrating to Rust 2018.
 2019-01-19 10:44:30 -05:00
P M
7ecee299a5 ignore/types: add QML 
PR #1165
 2019-01-18 06:48:47 -05:00
David Håsäther
dd396ff34e doc: fix typo 
PR #1161
 2019-01-14 06:50:30 -05:00
Andrew Gallant
fb0a82f3c3 grep-printer: add macro docs, redux 2019-01-11 09:18:09 -05:00
Andrew Gallant
dbc8ca9cc1 grep-searcher: add docs for assert_eq_printed 
Looks like the deny(missing_docs) lint got a bit stronger.
 2019-01-11 09:03:00 -05:00
Marco Hinz
c3db8db93d doc: fix typo 2019-01-05 11:18:05 -05:00
Andrew Gallant
17ef4c40f3 ignore-0.4.6 2018-12-30 08:46:09 -05:00
Andrew Gallant
a9e0477ea8 ignore: permit use of deprecated trim_right 2018-12-30 08:44:59 -05:00
Andrew Gallant
b3c5773266 deps: bump ignore 2018-12-30 08:43:18 -05:00
Andrew Gallant
118b950085 ignore-0.4.5 2018-12-15 08:44:10 -05:00
Andrew Gallant
b45b2f58ea deps: update most other dependencies 
This commit is the result of doing:

  $ cargo update
  $ cargo update -p encoding_rs --precise 0.8.10

where the latter line prevents encoding_rs from updating to 0.8.11 (or
newer). In particular, the 0.8.11 release increased the minimum Rust
version to 1.29, where as ripgrep 0.10.x is still on 1.28. We stay on an
older version for now until ripgrep is ready to move to 0.11.x.
 2018-12-15 08:42:14 -05:00
Andrew Gallant
662a9bc73d deps: update to crossbeam-channel 0.3 
This also requires corresponding updates to both rand and rand_core. Doing
an update of rand without doing an update of rand_core results in
compilation errors because two distinct versions of rand_core are included
in the build, and the traits they expose are distinct and incompatible.

We also switch over to using tempfile instead of tempdir, which drops the
last remaining thing keeping rand 0.4 in the build.

Fixes #1141, Fixes #1142
 2018-12-15 08:40:04 -05:00
Andrew Gallant
401add0a99 deps: update regex and regex-syntax 
This brings in some new Unicode properties, such as \p{Emoji}.

It is now also technically possible construct a regex that recognizes
grapheme clusters.
 2018-12-09 16:33:37 -05:00
Simon Morgan
f81b72721b ignore/types: add ASP 
PR #1134
 2018-12-07 16:19:33 -05:00
Antony Lee
1d4fccaadc ignore/types: add postscript 
Although postscript/encapsulated postscript is usually thought of as a
binary format, it's actually mostly ASCII, so ripgrep will not ignore
these files.

The situation is basically the same as for pdf, which is also already
present in the list of known filetypes.

PR #1118
 2018-11-23 09:46:11 -05:00
Matteo Bertini
09e464e674 ignore/types: add more Cython file types 
From the [Cython file types](https://cython.readthedocs.io/en/latest/src/userguide/language_basics.html?highlight=pxi#cython-file-types) paragraph on the official docs:

> There are three file types in Cython:
>    The implementation files, carrying a .py or .pyx suffix.
>    The definition files, carrying a .pxd suffix.
>    The include files, carrying a .pxi suffix.

PR #1113
 2018-11-19 07:37:00 -05:00
Jon Parise
31adff6f3c ignore/types: add Apache Thrift 
PR #1102
 2018-11-07 07:42:13 -05:00
Andrew Gallant
b41e596327 doc: escape braces in AsciiDoc 
This commit fixes a bug where AsciiDoc would drop any line containing a
'{foo}' because it interpreted it as an undefined attribute reference:

> Simple attribute references take the form {<name>}. If the attribute name
> is defined its text value is substituted otherwise the line containing the
> reference is dropped from the output.

See: https://www.methods.co.nz/asciidoc/chunked/ch30.html

We fix this by simply replacing all occurrences of '{' and '}' with
their escaped forms: '&#123;' and '&#125;'.

Fixes #1101
 2018-11-06 06:57:16 -05:00
Andrew Gallant
fb62266620 deps: update encoding_rs 
This commit bumps the version of encoding_rs to use the latest release.
This appears to fix a panic in UTF-16 decoding.

Fixes #1089
 2018-10-22 06:50:35 -04:00
Dave Lee
acf226c39d ignore/types: add BUILD.bazel to bazel file type 
PR #1074
 2018-10-02 18:00:04 -04:00
Mathieu Bridon
8299625e48 ignore/types: add buildstream 
BuildStream is a Free Software tool for building/integrating software stacks.: https://buildstream.gitlab.io/buildstream/

It uses recipes written in YAML, in files with the `.bst` extension.

PR #1071
 2018-09-28 08:32:24 -04:00
Andrew Gallant
db256c87eb ripgrep: suggest -U/--multiline 
When a "\n literal is not allowed" error is reported, ripgrep will now
suggest the use of the -U/--multiline flag, which enables matching
newlines.

Fixes #1055
 2018-09-25 16:56:04 -04:00
Andrew Gallant
ba533f390e grep-searcher: update to encoding_rs_io 0.1.3 
This update includes a work-around for a presumed bug in encoding_rs
that causes a panic:
https://github.com/hsivonen/encoding_rs/issues/34

Specifically, to reproduce this in ripgrep, one can run the following:

    $ curl -LO https://cache.ruby-lang.org/pub/ruby/2.5/ruby-2.5.1.tar.gz
    $ tar xf ruby-2.5.1.tar.gz
    $ rg ZZZZZ ruby-2.5.1/test/rexml/data/t63-2.svg
    thread 'main' panicked at 'index out of bounds: the len is 1 but the index is 1'

Fixes #1052
 2018-09-25 16:56:04 -04:00
Andrew Gallant
ba503eb677 grep-regex: fix inner literal detection 
It seems the inner literal detector fails spectacularly in cases of
concatenations that involve groups. The issue here is that if the prefix
of a group inside a concatenation can match the empty string, then any
literals generated to that point in the concatenation need to be cut
such that they are never extended. The detector isn't really built to
handle this case, so we just act conservative cut literals whenever we
see a sub-group. This may make some regexes slower, but the inner
literal detector already misses plenty of cases.

Literal detection (including in the regex engine) is a key component
that needs to be completely rethought at some point.

Fixes #1064
 2018-09-25 16:56:04 -04:00
Andrew Gallant
f72c2dfd90 readme: touch up README 
Make the wording consistent.
 2018-09-14 11:33:56 -04:00
Sylvestre Ledru
c0aa58b4f7 Ripgrep is also available in Ubuntu (from Cosmic) 2018-09-14 08:41:05 +02:00
ykgmfq
184ee4c328 deb: add section info 
Put it in the same section as
https://packages.debian.org/stretch/grep

PR #1051
 2018-09-13 08:17:24 -04:00
Gabe Berke-Williams
e82fbf2c46 doc: fix typo 
"cretion" -> "creation"

PR #1045
 2018-09-10 06:49:48 -04:00
Andrew Gallant
eb18da0450 pcre2: use jit_if_available 
This will allow PCRE2 to fall back to non-JIT matching when running on
platforms without JIT support.

ref https://github.com/BurntSushi/rust-pcre2/issues/3
 2018-09-08 17:12:14 -04:00
Andrew Gallant
0f7494216f readme: update dpkg version 2018-09-08 10:46:40 -04:00
Andrew Chin
442a278635 readme: fancy regexes are not supported by default 
PR #1042
 2018-09-07 17:43:24 -04:00
Andrew Gallant
7ebed3ace6 pkg: update brew tap to 0.10.0 2018-09-07 14:43:59 -04:00
Andrew Gallant
8a7db1a918 ci: tweak deployment conditions 2018-09-07 14:07:52 -04:00
71 changed files with 1872 additions and 1103 deletions
Expand all files Collapse all files

8
.travis.yml
View File

@@ -1,4 +1,5 @@
language: rust
dist: xenial
env:
global:
- PROJECT_NAME: ripgrep
@@ -62,13 +63,13 @@ matrix:
# Minimum Rust supported channel. We enable these to make sure ripgrep
# continues to work on the advertised minimum Rust version.
- os: linux
rust: 1.28.0
rust: 1.32.0
env: TARGET=x86_64-unknown-linux-gnu
- os: linux
rust: 1.28.0
rust: 1.32.0
env: TARGET=x86_64-unknown-linux-musl
- os: linux
rust: 1.28.0
rust: 1.32.0
env: TARGET=arm-unknown-linux-gnueabihf GCC_VERSION=4.8
addons:
apt:
@@ -93,6 +94,7 @@ deploy:
skip_cleanup: true
on:
condition: $TRAVIS_RUST_VERSION = nightly
branch: master # i guess we do need this after all?
tags: true
api_key:
secure: "IbSnsbGkxSydR/sozOf1/SRvHplzwRUHzcTjM7BKnr7GccL86gRPUrsrvD103KjQUGWIc1TnK1YTq5M0Onswg/ORDjqa1JEJPkPdPnVh9ipbF7M2De/7IlB4X4qXLKoApn8+bx2x/mfYXu4G+G1/2QdbaKK2yfXZKyjz0YFx+6CNrVCT2Nk8q7aHvOOzAL58vsG8iPDpupuhxlMDDn/UhyOWVInmPPQ0iJR1ZUJN8xJwXvKvBbfp3AhaBiAzkhXHNLgBR8QC5noWWMXnuVDMY3k4f3ic0V+p/qGUCN/nhptuceLxKFicMCYObSZeUzE5RAI0/OBW7l3z2iCoc+TbAnn+JrX/ObJCfzgAOXAU3tLaBFMiqQPGFKjKg1ltSYXomOFP/F7zALjpvFp4lYTBajRR+O3dqaxA9UQuRjw27vOeUpMcga4ZzL4VXFHzrxZKBHN//XIGjYAVhJ1NSSeGpeJV5/+jYzzWKfwSagRxQyVCzMooYFFXzn8Yxdm3PJlmp3GaAogNkdB9qKcrEvRINCelalzALPi0hD/HUDi8DD2PNTCLLMo6VSYtvc685Zbe+KgNzDV1YyTrRCUW6JotrS0r2ULLwnsh40hSB//nNv3XmwNmC/CmW5QAnIGj8cBMF4S2t6ohADIndojdAfNiptmaZOIT6owK7bWMgPMyopo="

62
CHANGELOG.md
View File

@@ -1,3 +1,65 @@
0.11.0 (TBD)
============
TODO.
**BREAKING CHANGES**:
* ripgrep has tweaked its exit status codes to be more like GNU grep's. Namely,
if a non-fatal error occurs during a search, then ripgrep will now always
emit a `2` exit status code, regardless of whether a match is found or not.
Previously, ripgrep would only emit a `2` exit status code for a catastrophic
error (e.g., regex syntax error). One exception to this is if ripgrep is run
with `-q/--quiet`. In that case, if an error occurs and a match is found,
then ripgrep will exit with a `0` exit status code.
* The `avx-accel` feature of ripgrep has been removed since it is no longer
necessary. All uses of AVX in ripgrep are now enabled automatically via
runtime CPU feature detection. The `simd-accel` feature does remain
available, however, it does increase compilation times substantially at the
moment.
Feature enhancements:
* [FEATURE #1099](https://github.com/BurntSushi/ripgrep/pull/1099):
Add support for Brotli and Zstd to the `-z/--search-zip` flag.
* [FEATURE #1138](https://github.com/BurntSushi/ripgrep/pull/1138):
Add `--no-ignore-dot` flag for ignoring `.ignore` files.
* [FEATURE #1159](https://github.com/BurntSushi/ripgrep/pull/1159):
ripgrep's exit status logic should now match GNU grep. See updated man page.
* [FEATURE #1170](https://github.com/BurntSushi/ripgrep/pull/1170):
Add `--ignore-file-case-insensitive` for case insensitive .ignore globs.
Bug fixes:
* [BUG #373](https://github.com/BurntSushi/ripgrep/issues/373),
[BUG #1098](https://github.com/BurntSushi/ripgrep/issues/1098):
`**` is now accepted as valid syntax anywhere in a glob.
* [BUG #916](https://github.com/BurntSushi/ripgrep/issues/916):
ripgrep no longer hangs when searching `/proc` with a zombie process present.
* [BUG #1091](https://github.com/BurntSushi/ripgrep/issues/1091):
Add note about inverted flags to the man page.
* [BUG #1095](https://github.com/BurntSushi/ripgrep/issues/1095):
Fix corner cases involving the `--crlf` flag.
* [BUG #1103](https://github.com/BurntSushi/ripgrep/issues/1103):
Clarify what `--encoding auto` does.
* [BUG #1106](https://github.com/BurntSushi/ripgrep/issues/1106):
`--files-with-matches` and `--files-without-match` work with one file.
* [BUG #1093](https://github.com/BurntSushi/ripgrep/pull/1093):
Fix handling of literal slashes in gitignore patterns.
* [BUG #1121](https://github.com/BurntSushi/ripgrep/issues/1121):
Fix bug that was triggering Windows antimalware when using the --files flag.
* [BUG #1125](https://github.com/BurntSushi/ripgrep/issues/1125),
[BUG #1159](https://github.com/BurntSushi/ripgrep/issues/1159):
ripgrep shouldn't panic for `rg -h | rg` and should emit correct exit status.
* [BUG #1154](https://github.com/BurntSushi/ripgrep/issues/1154):
Windows files with "hidden" attribute are now treated as hidden.
* [BUG #1173](https://github.com/BurntSushi/ripgrep/issues/1173):
Fix handling of `**` patterns in gitignore files.
* [BUG #1174](https://github.com/BurntSushi/ripgrep/issues/1174):
Fix handling of repeated `**` patterns in gitignore files.
* [BUG #1176](https://github.com/BurntSushi/ripgrep/issues/1176):
Fix bug where `-F`/`-x` weren't applied to patterns given via `-f`.
0.10.0 (2018-09-07)
===================
This is a new minor version release of ripgrep that contains some major new

664
Cargo.lock generated
View File

File diff suppressed because it is too large Load Diff

6
Cargo.toml
View File

@@ -17,6 +17,7 @@ license = "Unlicense OR MIT"
exclude = ["HomebrewFormula"]
build = "build.rs"
autotests = false
edition = "2018"
[badges]
travis-ci = { repository = "BurntSushi/ripgrep" }
@@ -45,7 +46,8 @@ members = [
]
[dependencies]
grep = { version = "0.2.2", path = "grep" }
bstr = "0.1.2"
grep = { version = "0.2.3", path = "grep" }
ignore = { version = "0.4.4", path = "ignore" }
lazy_static = "1.1.0"
log = "0.4.5"
@@ -72,7 +74,6 @@ serde = "1.0.77"
serde_derive = "1.0.77"
[features]
avx-accel = ["grep/avx-accel"]
simd-accel = ["grep/simd-accel"]
pcre2 = ["grep/pcre2"]
@@ -81,6 +82,7 @@ debug = 1
[package.metadata.deb]
features = ["pcre2"]
section = "utils"
assets = [
["target/release/rg", "usr/bin/", "755"],
["COPYING", "usr/share/doc/ripgrep/", "644"],

15
FAQ.md
View File

@@ -118,7 +118,7 @@ from run to run of ripgrep.
The only way to make the order of results consistent is to ask ripgrep to
sort the output. Currently, this will disable all parallelism. (On smaller
repositories, you might not notice much of a performance difference!) You
can achieve this with the `--sort-files` flag.
can achieve this with the `--sort path` flag.
There is more discussion on this topic here:
https://github.com/BurntSushi/ripgrep/issues/152
@@ -136,10 +136,10 @@ How do I search compressed files?
</h3>
ripgrep's `-z/--search-zip` flag will cause it to search compressed files
automatically. Currently, this supports gzip, bzip2, lzma, lz4 and xz only and
requires the corresponding `gzip`, `bzip2` and `xz` binaries to be installed on
your system. (That is, ripgrep does decompression by shelling out to another
process.)
automatically. Currently, this supports gzip, bzip2, xz, lzma, lz4, Brotli and
Zstd. Each of these requires requires the corresponding `gzip`, `bzip2`, `xz`,
`lz4`, `brotli` and `zstd` binaries to be installed on your system. (That is,
ripgrep does decompression by shelling out to another process.)
ripgrep currently does not search archive formats, so `*.tar.gz` files, for
example, are skipped.
@@ -149,9 +149,8 @@ example, are skipped.
How do I search over multiple lines?
</h3>
This isn't currently possible. ripgrep is fundamentally a line-oriented search
tool. With that said,
[multiline search is a planned opt-in feature](https://github.com/BurntSushi/ripgrep/issues/176).
The `-U/--multiline` flag enables ripgrep to report results that span over
multiple lines.
<h3 name="fancy">

20
GUIDE.md
View File

@@ -235,6 +235,11 @@ Like `.gitignore`, a `.ignore` file can be placed in any directory. Its rules
will be processed with respect to the directory it resides in, just like
`.gitignore`.
To process `.gitignore` and `.ignore` files case insensitively, use the flag
`--ignore-file-case-insensitive`. This is especially useful on case insensitive
file systems like those on Windows and macOS. Note though that this can come
with a significant performance penalty, and is therefore disabled by default.
For a more in depth description of how glob patterns in a `.gitignore` file
are interpreted, please see `man gitignore`.
@@ -520,9 +525,9 @@ config file. Once the environment variable is set, open the file and just type
in the flags you want set automatically. There are only two rules for
describing the format of the config file:
1. Every line is a shell argument, after trimming ASCII whitespace.
2. Lines starting with `#` (optionally preceded by any amount of
ASCII whitespace) are ignored.
1. Every line is a shell argument, after trimming whitespace.
2. Lines starting with `#` (optionally preceded by any amount of whitespace)
are ignored.
In particular, there is no escaping. Each line is given to ripgrep as a single
command line argument verbatim.
@@ -604,7 +609,8 @@ topic, but we can try to summarize its relevancy to ripgrep:
the most popular encodings likely consist of ASCII, latin1 or UTF-8. As
a special exception, UTF-16 is prevalent in Windows environments
In light of the above, here is how ripgrep behaves:
In light of the above, here is how ripgrep behaves when `--encoding auto` is
given, which is the default:
* All input is assumed to be ASCII compatible (which means every byte that
corresponds to an ASCII codepoint actually is an ASCII codepoint). This
@@ -675,10 +681,10 @@ used options that will likely impact how you use ripgrep on a regular basis.
* `--files`: Print the files that ripgrep *would* search, but don't actually
search them.
* `-a/--text`: Search binary files as if they were plain text.
* `-z/--search-zip`: Search compressed files (gzip, bzip2, lzma, xz). This is
disabled by default.
* `-z/--search-zip`: Search compressed files (gzip, bzip2, lzma, xz, lz4,
brotli, zstd). This is disabled by default.
* `-C/--context`: Show the lines surrounding a match.
* `--sort-files`: Force ripgrep to sort its output by file name. (This disables
* `--sort path`: Force ripgrep to sort its output by file name. (This disables
parallelism, so it might be slower.)
* `-L/--follow`: Follow symbolic links while recursively searching.
* `-M/--max-columns`: Limit the length of lines printed by ripgrep.

59
README.md
View File

@@ -1,11 +1,12 @@
ripgrep (rg)
------------
ripgrep is a line-oriented search tool that recursively searches your current
directory for a regex pattern while respecting your gitignore rules. ripgrep
directory for a regex pattern. By default, ripgrep will respect your .gitignore
and automatically skip hidden files/directories and binary files. ripgrep
has first class support on Windows, macOS and Linux, with binary downloads
available for [every release](https://github.com/BurntSushi/ripgrep/releases).
ripgrep is similar to other popular search tools like The Silver Searcher,
ack and grep.
ripgrep is similar to other popular search tools like The Silver Searcher, ack
and grep.
[![Linux build status](https://travis-ci.org/BurntSushi/ripgrep.svg)](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)
@@ -105,7 +106,7 @@ increases the times to `2.640s` for ripgrep and `10.277s` for GNU grep.
supporting Unicode (which is always on).
* ripgrep has optional support for switching its regex engine to use PCRE2.
Among other things, this makes it possible to use look-around and
backreferences in your patterns, which are supported in ripgrep's default
backreferences in your patterns, which are not supported in ripgrep's default
regex engine. PCRE2 support is enabled with `-P`.
* ripgrep supports searching files in text encodings other than UTF-8, such
as UTF-16, latin-1, GBK, EUC-JP, Shift_JIS and more. (Some support for
@@ -248,21 +249,22 @@ If you're a **Gentoo** user, you can install ripgrep from the
$ emerge sys-apps/ripgrep
```
If you're a **Fedora 27+** user, you can install ripgrep from official
If you're a **Fedora** user, you can install ripgrep from official
repositories.
```
$ sudo dnf install ripgrep
```
If you're a **Fedora 24+** user, you can install ripgrep from
[copr](https://copr.fedorainfracloud.org/coprs/carlwgeorge/ripgrep/):
If you're an **openSUSE Leap 15.0** user, you can install ripgrep from the
[utilities repo](https://build.opensuse.org/package/show/utilities/ripgrep):
```
$ sudo dnf copr enable carlwgeorge/ripgrep
$ sudo dnf install ripgrep
$ sudo zypper ar https://download.opensuse.org/repositories/utilities/openSUSE_Leap_15.0/utilities.repo
$ sudo zypper install ripgrep
```
If you're an **openSUSE Tumbleweed** user, you can install ripgrep from the
[official repo](http://software.opensuse.org/package/ripgrep):
@@ -288,12 +290,11 @@ $ # (Or using the attribute name, which is also ripgrep.)
If you're a **Debian** user (or a user of a Debian derivative like **Ubuntu**),
then ripgrep can be installed using a binary `.deb` file provided in each
[ripgrep release](https://github.com/BurntSushi/ripgrep/releases). Note that
ripgrep is not in the official Debian or Ubuntu repositories.
[ripgrep release](https://github.com/BurntSushi/ripgrep/releases).
```
$ curl -LO https://github.com/BurntSushi/ripgrep/releases/download/0.9.0/ripgrep_0.9.0_amd64.deb
$ sudo dpkg -i ripgrep_0.9.0_amd64.deb
$ curl -LO https://github.com/BurntSushi/ripgrep/releases/download/0.10.0/ripgrep_0.10.0_amd64.deb
$ sudo dpkg -i ripgrep_0.10.0_amd64.deb
```
If you run Debian Buster (currently Debian testing) or Debian sid, ripgrep is
@@ -302,6 +303,14 @@ If you run Debian Buster (currently Debian testing) or Debian sid, ripgrep is
$ sudo apt-get install ripgrep
```
If you're an **Ubuntu Cosmic (18.10)** (or newer) user, ripgrep is
[available](https://launchpad.net/ubuntu/+source/rust-ripgrep) using the same
packaging as Debian:
```
$ sudo apt-get install ripgrep
```
(N.B. Various snaps for ripgrep on Ubuntu are also available, but none of them
seem to work right and generate a number of very strange bug reports that I
don't know how to fix and don't have the time to fix. Therefore, it is no
@@ -330,7 +339,7 @@ If you're a **NetBSD** user, then you can install ripgrep from
If you're a **Rust programmer**, ripgrep can be installed with `cargo`.
* Note that the minimum supported version of Rust for ripgrep is **1.28.0**,
* Note that the minimum supported version of Rust for ripgrep is **1.32.0**,
although ripgrep may work with older versions.
* Note that the binary may be bigger than expected because it contains debug
symbols. This is intentional. To remove debug symbols and therefore reduce
@@ -351,7 +360,7 @@ ripgrep isn't currently in any other package repositories.
ripgrep is written in Rust, so you'll need to grab a
[Rust installation](https://www.rust-lang.org/) in order to compile it.
ripgrep compiles with Rust 1.28.0 (stable) or newer. In general, ripgrep tracks
ripgrep compiles with Rust 1.32.0 (stable) or newer. In general, ripgrep tracks
the latest stable release of the Rust compiler.
To build ripgrep:
@@ -368,18 +377,14 @@ If you have a Rust nightly compiler and a recent Intel CPU, then you can enable
additional optional SIMD acceleration like so:
```
RUSTFLAGS="-C target-cpu=native" cargo build --release --features 'simd-accel avx-accel'
RUSTFLAGS="-C target-cpu=native" cargo build --release --features 'simd-accel'
```
If your machine doesn't support AVX instructions, then simply remove
`avx-accel` from the features list. Similarly for SIMD (which corresponds
roughly to SSE instructions).
The `simd-accel` and `avx-accel` features enable SIMD support in certain
ripgrep dependencies (responsible for counting lines and transcoding). They
are not necessary to get SIMD optimizations for search; those are enabled
automatically. Hopefully, some day, the `simd-accel` and `avx-accel` features
will similarly become unnecessary.
The `simd-accel` feature enables SIMD support in certain ripgrep dependencies
(responsible for transcoding). They are not necessary to get SIMD optimizations
for search; those are enabled automatically. Hopefully, some day, the
`simd-accel` feature will similarly become unnecessary. **WARNING:** Currently,
enabling this option can increase compilation times dramatically.
Finally, optional PCRE2 support can be built with ripgrep by enabling the
`pcre2` feature:
@@ -388,8 +393,8 @@ Finally, optional PCRE2 support can be built with ripgrep by enabling the
$ cargo build --release --features 'pcre2'
```
(Tip: use `--features 'pcre2 simd-accel avx-accel'` to also include compile
time SIMD optimizations, which will only work with a nightly compiler.)
(Tip: use `--features 'pcre2 simd-accel'` to also include compile time SIMD
optimizations, which will only work with a nightly compiler.)
Enabling the PCRE2 feature works with a stable Rust compiler and will
attempt to automatically find and link with your system's PCRE2 library via

2
appveyor.yml
View File

@@ -73,7 +73,7 @@ deploy:
# deploy when a new tag is pushed and only on the stable channel
on:
CHANNEL: stable
APPVEYOR_REPO_TAG: true
appveyor_repo_tag: true
branches:
only:

12
build.rs
View File

@@ -1,8 +1,3 @@
#[macro_use]
extern crate clap;
#[macro_use]
extern crate lazy_static;
use std::env;
use std::fs::{self, File};
use std::io::{self, Read, Write};
@@ -168,7 +163,12 @@ fn formatted_arg(arg: &RGArg) -> io::Result<String> {
}
fn formatted_doc_txt(arg: &RGArg) -> io::Result<String> {
let paragraphs: Vec<&str> = arg.doc_long.split("\n\n").collect();
let paragraphs: Vec<String> = arg.doc_long
.replace("{", "&#123;")
.replace("}", r"&#125;")
.split("\n\n")
.map(|s| s.to_string())
.collect();
if paragraphs.is_empty() {
return Err(ioerr(format!("missing docs for --{}", arg.name)));
}

12
complete/_rg
View File

@@ -112,8 +112,12 @@ _rg() {
$no"--no-hidden[don't search hidden files and directories]"
+ '(ignore)' # Ignore-file options
"(--no-ignore-global --no-ignore-parent --no-ignore-vcs)--no-ignore[don't respect ignore files]"
$no'(--ignore-global --ignore-parent --ignore-vcs)--ignore[respect ignore files]'
"(--no-ignore-global --no-ignore-parent --no-ignore-vcs --no-ignore-dot)--no-ignore[don't respect ignore files]"
$no'(--ignore-global --ignore-parent --ignore-vcs --ignore-dot)--ignore[respect ignore files]'
+ '(ignore-file-case-insensitive)' # Ignore-file case sensitivity options
'--ignore-file-case-insensitive[process ignore files case insensitively]'
$no'--no-ignore-file-case-insensitive[process ignore files case sensitively]'
+ '(ignore-global)' # Global ignore-file options
"--no-ignore-global[don't respect global ignore files]"
@@ -127,6 +131,10 @@ _rg() {
"--no-ignore-vcs[don't respect version control ignore files]"
$no'--ignore-vcs[respect version control ignore files]'
+ '(ignore-dot)' # .ignore-file options
"--no-ignore-dot[don't respect .ignore files]"
$no'--ignore-dot[respect .ignore files]'
+ '(json)' # JSON options
'--json[output results in JSON Lines format]'
$no"--no-json[don't output results in JSON Lines format]"

61
doc/rg.1.txt.tpl
View File

@@ -34,12 +34,12 @@ files/directories and binary files.
ripgrep's default regex engine uses finite automata and guarantees linear
time searching. Because of this, features like backreferences and arbitrary
look-around are not supported. However, if ripgrep is built with PCRE2, then
the --pcre2 flag can be used to enable backreferences and look-around.
the *--pcre2* flag can be used to enable backreferences and look-around.
ripgrep supports configuration files. Set RIPGREP_CONFIG_PATH to a
ripgrep supports configuration files. Set *RIPGREP_CONFIG_PATH* to a
configuration file. The file can specify one shell argument per line. Lines
starting with '#' are ignored. For more details, see the man page or the
README.
starting with *#* are ignored. For more details, see the man page or the
*README*.
REGEX SYNTAX
@@ -52,10 +52,10 @@ https://docs.rs/regex/*/regex/bytes/index.html#syntax
To a first approximation, ripgrep uses Perl-like regexes without look-around or
backreferences. This makes them very similar to the "extended" (ERE) regular
expressions supported by `egrep`, but with a few additional features like
expressions supported by *egrep*, but with a few additional features like
Unicode character classes.
If you're using ripgrep with the --pcre2 flag, then please consult
If you're using ripgrep with the *--pcre2* flag, then please consult
https://www.pcre.org or the PCRE2 man pages for documentation on the supported
syntax.
@@ -68,18 +68,37 @@ _PATTERN_::
_PATH_::
A file or directory to search. Directories are searched recursively. Paths
specified expicitly on the command line override glob and ignore rules.
specified explicitly on the command line override glob and ignore rules.
OPTIONS
-------
Note that for many options, there exist flags to disable them. In some cases,
those flags are not listed in a first class way below. For example, the
*--column* flag (listed below) enables column numbers in ripgrep's output, but
the *--no-column* flag (not listed below) disables them. The reverse can also
exist. For example, the *--no-ignore* flag (listed below) disables ripgrep's
*gitignore* logic, but the *--ignore* flag (not listed below) enables it. These
flags are useful for overriding a ripgrep configuration file on the command
line. Each flag's documentation notes whether an inverted flag exists. In all
cases, the flag specified last takes precedence.
{OPTIONS}
EXIT STATUS
-----------
If ripgrep finds a match, then the exit status of the program is 0. If no match
could be found, then the exit status is non-zero.
could be found, then the exit status is 1. If an error occurred, then the exit
status is always 2 unless ripgrep was run with the *--quiet* flag and a match
was found. In summary:
* `0` exit status occurs only when at least one match was found, and if
no error occurred, unless *--quiet* was given.
* `1` exit status occurs only when no match was found and no error occurred.
* `2` exit status occurs when an error occurred. This is true for both
catastrophic errors (e.g., a regex syntax error) and for soft errors (e.g.,
unable to read a file).
CONFIGURATION FILES
@@ -88,12 +107,12 @@ ripgrep supports reading configuration files that change ripgrep's default
behavior. The format of the configuration file is an "rc" style and is very
simple. It is defined by two rules:
1. Every line is a shell argument, after trimming ASCII whitespace.
2. Lines starting with _#_ (optionally preceded by any amount of
ASCII whitespace) are ignored.
1. Every line is a shell argument, after trimming whitespace.
2. Lines starting with *#* (optionally preceded by any amount of
whitespace) are ignored.
ripgrep will look for a single configuration file if and only if the
_RIPGREP_CONFIG_PATH_ environment variable is set and is non-empty.
*RIPGREP_CONFIG_PATH* environment variable is set and is non-empty.
ripgrep will parse shell arguments from this file on startup and will
behave as if the arguments in this file were prepended to any explicit
arguments given to ripgrep on the command line.
@@ -155,20 +174,20 @@ SHELL COMPLETION
Shell completion files are included in the release tarball for Bash, Fish, Zsh
and PowerShell.
For *bash*, move `rg.bash` to `$XDG_CONFIG_HOME/bash_completion`
or `/etc/bash_completion.d/`.
For *bash*, move *rg.bash* to *$XDG_CONFIG_HOME/bash_completion*
or */etc/bash_completion.d/*.
For *fish*, move `rg.fish` to `$HOME/.config/fish/completions`.
For *fish*, move *rg.fish* to *$HOME/.config/fish/completions*.
For *zsh*, move `_rg` to one of your `$fpath` directories.
For *zsh*, move *_rg* to one of your *$fpath* directories.
CAVEATS
-------
ripgrep may abort unexpectedly when using default settings if it searches a
file that is simultaneously truncated. This behavior can be avoided by passing
the --no-mmap flag which will forcefully disable the use of memory maps in all
cases.
the *--no-mmap* flag which will forcefully disable the use of memory maps in
all cases.
VERSION
@@ -180,7 +199,11 @@ HOMEPAGE
--------
https://github.com/BurntSushi/ripgrep
Please report bugs and feature requests in the issue tracker.
Please report bugs and feature requests in the issue tracker. Please do your
best to provide a reproducible test case for bugs. This should include the
corpus being searched, the *rg* command, the actual output and the expected
output. Please also include the output of running the same *rg* command but
with the *--debug* flag.
AUTHORS

6
globset/Cargo.toml
View File

@@ -19,11 +19,11 @@ name = "globset"
bench = false
[dependencies]
aho-corasick = "0.6.8"
aho-corasick = "0.7.3"
bstr = { version = "0.1.2", default-features = false, features = ["std"] }
fnv = "1.0.6"
log = "0.4.5"
memchr = "2.0.2"
regex = "1.0.5"
regex = "1.1.5"
[dev-dependencies]
glob = "0.2.11"

103
globset/src/glob.rs
View File

@@ -120,7 +120,7 @@ impl GlobMatcher {
/// Tests whether the given path matches this pattern or not.
pub fn is_match_candidate(&self, path: &Candidate) -> bool {
self.re.is_match(&path.path)
self.re.is_match(path.path.as_bytes())
}
}
@@ -145,7 +145,7 @@ impl GlobStrategic {
/// Tests whether the given path matches this pattern or not.
fn is_match_candidate(&self, candidate: &Candidate) -> bool {
let byte_path = &*candidate.path;
let byte_path = candidate.path.as_bytes();
match self.strategy {
MatchStrategy::Literal(ref lit) => lit.as_bytes() == byte_path,
@@ -837,40 +837,66 @@ impl<'a> Parser<'a> {
fn parse_star(&mut self) -> Result<(), Error> {
let prev = self.prev;
if self.chars.peek() != Some(&'*') {
if self.peek() != Some('*') {
self.push_token(Token::ZeroOrMore)?;
return Ok(());
}
assert!(self.bump() == Some('*'));
if !self.have_tokens()? {
self.push_token(Token::RecursivePrefix)?;
let next = self.bump();
if !next.map(is_separator).unwrap_or(true) {
return Err(self.error(ErrorKind::InvalidRecursive));
if !self.peek().map_or(true, is_separator) {
self.push_token(Token::ZeroOrMore)?;
self.push_token(Token::ZeroOrMore)?;
} else {
self.push_token(Token::RecursivePrefix)?;
assert!(self.bump().map_or(true, is_separator));
}
return Ok(());
}
self.pop_token()?;
if !prev.map(is_separator).unwrap_or(false) {
if self.stack.len() <= 1
|| (prev != Some(',') && prev != Some('{')) {
return Err(self.error(ErrorKind::InvalidRecursive));
|| (prev != Some(',') && prev != Some('{'))
{
self.push_token(Token::ZeroOrMore)?;
self.push_token(Token::ZeroOrMore)?;
return Ok(());
}
}
match self.chars.peek() {
None => {
assert!(self.bump().is_none());
self.push_token(Token::RecursiveSuffix)
let is_suffix =
match self.peek() {
None => {
assert!(self.bump().is_none());
true
}
Some(',') | Some('}') if self.stack.len() >= 2 => {
true
}
Some(c) if is_separator(c) => {
assert!(self.bump().map(is_separator).unwrap_or(false));
false
}
_ => {
self.push_token(Token::ZeroOrMore)?;
self.push_token(Token::ZeroOrMore)?;
return Ok(());
}
};
match self.pop_token()? {
Token::RecursivePrefix => {
self.push_token(Token::RecursivePrefix)?;
}
Some(&',') | Some(&'}') if self.stack.len() >= 2 => {
self.push_token(Token::RecursiveSuffix)
Token::RecursiveSuffix => {
self.push_token(Token::RecursiveSuffix)?;
}
Some(&c) if is_separator(c) => {
assert!(self.bump().map(is_separator).unwrap_or(false));
self.push_token(Token::RecursiveZeroOrMore)
_ => {
if is_suffix {
self.push_token(Token::RecursiveSuffix)?;
} else {
self.push_token(Token::RecursiveZeroOrMore)?;
}
}
_ => Err(self.error(ErrorKind::InvalidRecursive)),
}
Ok(())
}
fn parse_class(&mut self) -> Result<(), Error> {
@@ -959,6 +985,10 @@ impl<'a> Parser<'a> {
self.cur = self.chars.next();
self.cur
}
fn peek(&mut self) -> Option<char> {
self.chars.peek().map(|&ch| ch)
}
}
#[cfg(test)]
@@ -1144,13 +1174,6 @@ mod tests {
syntax!(cls20, "[^a]", vec![classn('a', 'a')]);
syntax!(cls21, "[^a-z]", vec![classn('a', 'z')]);
syntaxerr!(err_rseq1, "a**", ErrorKind::InvalidRecursive);
syntaxerr!(err_rseq2, "**a", ErrorKind::InvalidRecursive);
syntaxerr!(err_rseq3, "a**b", ErrorKind::InvalidRecursive);
syntaxerr!(err_rseq4, "***", ErrorKind::InvalidRecursive);
syntaxerr!(err_rseq5, "/a**", ErrorKind::InvalidRecursive);
syntaxerr!(err_rseq6, "/**a", ErrorKind::InvalidRecursive);
syntaxerr!(err_rseq7, "/a**b", ErrorKind::InvalidRecursive);
syntaxerr!(err_unclosed1, "[", ErrorKind::UnclosedClass);
syntaxerr!(err_unclosed2, "[]", ErrorKind::UnclosedClass);
syntaxerr!(err_unclosed3, "[!", ErrorKind::UnclosedClass);
@@ -1194,8 +1217,30 @@ mod tests {
toregex!(re8, "[*]", r"^[\*]$");
toregex!(re9, "[+]", r"^[\+]$");
toregex!(re10, "+", r"^\+$");
toregex!(re11, "**", r"^.*$");
toregex!(re12, "☃", r"^\xe2\x98\x83$");
toregex!(re11, "☃", r"^\xe2\x98\x83$");
toregex!(re12, "**", r"^.*$");
toregex!(re13, "**/", r"^.*$");
toregex!(re14, "**/*", r"^(?:/?|.*/).*$");
toregex!(re15, "**/**", r"^.*$");
toregex!(re16, "**/**/*", r"^(?:/?|.*/).*$");
toregex!(re17, "**/**/**", r"^.*$");
toregex!(re18, "**/**/**/*", r"^(?:/?|.*/).*$");
toregex!(re19, "a/**", r"^a(?:/?|/.*)$");
toregex!(re20, "a/**/**", r"^a(?:/?|/.*)$");
toregex!(re21, "a/**/**/**", r"^a(?:/?|/.*)$");
toregex!(re22, "a/**/b", r"^a(?:/|/.*/)b$");
toregex!(re23, "a/**/**/b", r"^a(?:/|/.*/)b$");
toregex!(re24, "a/**/**/**/b", r"^a(?:/|/.*/)b$");
toregex!(re25, "**/b", r"^(?:/?|.*/)b$");
toregex!(re26, "**/**/b", r"^(?:/?|.*/)b$");
toregex!(re27, "**/**/**/b", r"^(?:/?|.*/)b$");
toregex!(re28, "a**", r"^a.*.*$");
toregex!(re29, "**a", r"^.*.*a$");
toregex!(re30, "a**b", r"^a.*.*b$");
toregex!(re31, "***", r"^.*.*.*$");
toregex!(re32, "/a**", r"^/a.*.*$");
toregex!(re33, "/**a", r"^/.*.*a$");
toregex!(re34, "/a**b", r"^/a.*.*b$");
matches!(match1, "a", "a");
matches!(match2, "a*b", "a_b");

97
globset/src/lib.rs
View File

@@ -104,27 +104,25 @@ or to enable case insensitive matching.
#![deny(missing_docs)]
extern crate aho_corasick;
extern crate bstr;
extern crate fnv;
#[macro_use]
extern crate log;
extern crate memchr;
extern crate regex;
use std::borrow::Cow;
use std::collections::{BTreeMap, HashMap};
use std::error::Error as StdError;
use std::ffi::OsStr;
use std::fmt;
use std::hash;
use std::path::Path;
use std::str;
use aho_corasick::{Automaton, AcAutomaton, FullAcAutomaton};
use aho_corasick::AhoCorasick;
use bstr::{B, BStr, BString};
use regex::bytes::{Regex, RegexBuilder, RegexSet};
use pathutil::{
file_name, file_name_ext, normalize_path, os_str_bytes, path_bytes,
};
use pathutil::{file_name, file_name_ext, normalize_path};
use glob::MatchStrategy;
pub use glob::{Glob, GlobBuilder, GlobMatcher};
@@ -143,8 +141,13 @@ pub struct Error {
/// The kind of error that can occur when parsing a glob pattern.
#[derive(Clone, Debug, Eq, PartialEq)]
pub enum ErrorKind {
/// Occurs when a use of `**` is invalid. Namely, `**` can only appear
/// adjacent to a path separator, or the beginning/end of a glob.
/// **DEPRECATED**.
///
/// This error used to occur for consistency with git's glob specification,
/// but the specification now accepts all uses of `**`. When `**` does not
/// appear adjacent to a path separator or at the beginning/end of a glob,
/// it is now treated as two consecutive `*` patterns. As such, this error
/// is no longer used.
InvalidRecursive,
/// Occurs when a character class (e.g., `[abc]`) is not closed.
UnclosedClass,
@@ -289,6 +292,7 @@ pub struct GlobSet {
impl GlobSet {
/// Create an empty `GlobSet`. An empty set matches nothing.
#[inline]
pub fn empty() -> GlobSet {
GlobSet {
len: 0,
@@ -297,11 +301,13 @@ impl GlobSet {
}
/// Returns true if this set is empty, and therefore matches nothing.
#[inline]
pub fn is_empty(&self) -> bool {
self.len == 0
}
/// Returns the number of globs in this set.
#[inline]
pub fn len(&self) -> usize {
self.len
}
@@ -484,24 +490,25 @@ impl GlobSetBuilder {
/// path against multiple globs or sets of globs.
#[derive(Clone, Debug)]
pub struct Candidate<'a> {
path: Cow<'a, [u8]>,
basename: Cow<'a, [u8]>,
ext: Cow<'a, [u8]>,
path: Cow<'a, BStr>,
basename: Cow<'a, BStr>,
ext: Cow<'a, BStr>,
}
impl<'a> Candidate<'a> {
/// Create a new candidate for matching from the given path.
pub fn new<P: AsRef<Path> + ?Sized>(path: &'a P) -> Candidate<'a> {
let path = path.as_ref();
let basename = file_name(path).unwrap_or(OsStr::new(""));
let path = normalize_path(BString::from_path_lossy(path.as_ref()));
let basename = file_name(&path).unwrap_or(Cow::Borrowed(B("")));
let ext = file_name_ext(&basename).unwrap_or(Cow::Borrowed(B("")));
Candidate {
path: normalize_path(path_bytes(path)),
basename: os_str_bytes(basename),
ext: file_name_ext(basename).unwrap_or(Cow::Borrowed(b"")),
path: path,
basename: basename,
ext: ext,
}
}
fn path_prefix(&self, max: usize) -> &[u8] {
fn path_prefix(&self, max: usize) -> &BStr {
if self.path.len() <= max {
&*self.path
} else {
@@ -509,7 +516,7 @@ impl<'a> Candidate<'a> {
}
}
fn path_suffix(&self, max: usize) -> &[u8] {
fn path_suffix(&self, max: usize) -> &BStr {
if self.path.len() <= max {
&*self.path
} else {
@@ -570,12 +577,12 @@ impl LiteralStrategy {
}
fn is_match(&self, candidate: &Candidate) -> bool {
self.0.contains_key(&*candidate.path)
self.0.contains_key(candidate.path.as_bytes())
}
#[inline(never)]
fn matches_into(&self, candidate: &Candidate, matches: &mut Vec<usize>) {
if let Some(hits) = self.0.get(&*candidate.path) {
if let Some(hits) = self.0.get(candidate.path.as_bytes()) {
matches.extend(hits);
}
}
@@ -597,7 +604,7 @@ impl BasenameLiteralStrategy {
if candidate.basename.is_empty() {
return false;
}
self.0.contains_key(&*candidate.basename)
self.0.contains_key(candidate.basename.as_bytes())
}
#[inline(never)]
@@ -605,7 +612,7 @@ impl BasenameLiteralStrategy {
if candidate.basename.is_empty() {
return;
}
if let Some(hits) = self.0.get(&*candidate.basename) {
if let Some(hits) = self.0.get(candidate.basename.as_bytes()) {
matches.extend(hits);
}
}
@@ -627,7 +634,7 @@ impl ExtensionStrategy {
if candidate.ext.is_empty() {
return false;
}
self.0.contains_key(&*candidate.ext)
self.0.contains_key(candidate.ext.as_bytes())
}
#[inline(never)]
@@ -635,7 +642,7 @@ impl ExtensionStrategy {
if candidate.ext.is_empty() {
return;
}
if let Some(hits) = self.0.get(&*candidate.ext) {
if let Some(hits) = self.0.get(candidate.ext.as_bytes()) {
matches.extend(hits);
}
}
@@ -643,7 +650,7 @@ impl ExtensionStrategy {
#[derive(Clone, Debug)]
struct PrefixStrategy {
matcher: FullAcAutomaton<Vec<u8>>,
matcher: AhoCorasick,
map: Vec<usize>,
longest: usize,
}
@@ -651,8 +658,8 @@ struct PrefixStrategy {
impl PrefixStrategy {
fn is_match(&self, candidate: &Candidate) -> bool {
let path = candidate.path_prefix(self.longest);
for m in self.matcher.find_overlapping(path) {
if m.start == 0 {
for m in self.matcher.find_overlapping_iter(path) {
if m.start() == 0 {
return true;
}
}
@@ -661,9 +668,9 @@ impl PrefixStrategy {
fn matches_into(&self, candidate: &Candidate, matches: &mut Vec<usize>) {
let path = candidate.path_prefix(self.longest);
for m in self.matcher.find_overlapping(path) {
if m.start == 0 {
matches.push(self.map[m.pati]);
for m in self.matcher.find_overlapping_iter(path) {
if m.start() == 0 {
matches.push(self.map[m.pattern()]);
}
}
}
@@ -671,7 +678,7 @@ impl PrefixStrategy {
#[derive(Clone, Debug)]
struct SuffixStrategy {
matcher: FullAcAutomaton<Vec<u8>>,
matcher: AhoCorasick,
map: Vec<usize>,
longest: usize,
}
@@ -679,8 +686,8 @@ struct SuffixStrategy {
impl SuffixStrategy {
fn is_match(&self, candidate: &Candidate) -> bool {
let path = candidate.path_suffix(self.longest);
for m in self.matcher.find_overlapping(path) {
if m.end == path.len() {
for m in self.matcher.find_overlapping_iter(path) {
if m.end() == path.len() {
return true;
}
}
@@ -689,9 +696,9 @@ impl SuffixStrategy {
fn matches_into(&self, candidate: &Candidate, matches: &mut Vec<usize>) {
let path = candidate.path_suffix(self.longest);
for m in self.matcher.find_overlapping(path) {
if m.end == path.len() {
matches.push(self.map[m.pati]);
for m in self.matcher.find_overlapping_iter(path) {
if m.end() == path.len() {
matches.push(self.map[m.pattern()]);
}
}
}
@@ -705,11 +712,11 @@ impl RequiredExtensionStrategy {
if candidate.ext.is_empty() {
return false;
}
match self.0.get(&*candidate.ext) {
match self.0.get(candidate.ext.as_bytes()) {
None => false,
Some(regexes) => {
for &(_, ref re) in regexes {
if re.is_match(&*candidate.path) {
if re.is_match(candidate.path.as_bytes()) {
return true;
}
}
@@ -723,9 +730,9 @@ impl RequiredExtensionStrategy {
if candidate.ext.is_empty() {
return;
}
if let Some(regexes) = self.0.get(&*candidate.ext) {
if let Some(regexes) = self.0.get(candidate.ext.as_bytes()) {
for &(global_index, ref re) in regexes {
if re.is_match(&*candidate.path) {
if re.is_match(candidate.path.as_bytes()) {
matches.push(global_index);
}
}
@@ -741,11 +748,11 @@ struct RegexSetStrategy {
impl RegexSetStrategy {
fn is_match(&self, candidate: &Candidate) -> bool {
self.matcher.is_match(&*candidate.path)
self.matcher.is_match(candidate.path.as_bytes())
}
fn matches_into(&self, candidate: &Candidate, matches: &mut Vec<usize>) {
for i in self.matcher.matches(&*candidate.path) {
for i in self.matcher.matches(candidate.path.as_bytes()) {
matches.push(self.map[i]);
}
}
@@ -776,18 +783,16 @@ impl MultiStrategyBuilder {
}
fn prefix(self) -> PrefixStrategy {
let it = self.literals.into_iter().map(|s| s.into_bytes());
PrefixStrategy {
matcher: AcAutomaton::new(it).into_full(),
matcher: AhoCorasick::new_auto_configured(&self.literals),
map: self.map,
longest: self.longest,
}
}
fn suffix(self) -> SuffixStrategy {
let it = self.literals.into_iter().map(|s| s.into_bytes());
SuffixStrategy {
matcher: AcAutomaton::new(it).into_full(),
matcher: AhoCorasick::new_auto_configured(&self.literals),
map: self.map,
longest: self.longest,
}

103
globset/src/pathutil.rs
View File

@@ -1,41 +1,26 @@
use std::borrow::Cow;
use std::ffi::OsStr;
use std::path::Path;
use bstr::BStr;
/// The final component of the path, if it is a normal file.
///
/// If the path terminates in ., .., or consists solely of a root of prefix,
/// file_name will return None.
#[cfg(unix)]
pub fn file_name<'a, P: AsRef<Path> + ?Sized>(
path: &'a P,
) -> Option<&'a OsStr> {
use std::os::unix::ffi::OsStrExt;
use memchr::memrchr;
let path = path.as_ref().as_os_str().as_bytes();
pub fn file_name<'a>(path: &Cow<'a, BStr>) -> Option<Cow<'a, BStr>> {
if path.is_empty() {
return None;
} else if path.len() == 1 && path[0] == b'.' {
return None;
} else if path.last() == Some(&b'.') {
return None;
} else if path.len() >= 2 && &path[path.len() - 2..] == &b".."[..] {
} else if path.last() == Some(b'.') {
return None;
}
let last_slash = memrchr(b'/', path).map(|i| i + 1).unwrap_or(0);
Some(OsStr::from_bytes(&path[last_slash..]))
}
/// The final component of the path, if it is a normal file.
///
/// If the path terminates in ., .., or consists solely of a root of prefix,
/// file_name will return None.
#[cfg(not(unix))]
pub fn file_name<'a, P: AsRef<Path> + ?Sized>(
path: &'a P,
) -> Option<&'a OsStr> {
path.as_ref().file_name()
let last_slash = path.rfind_byte(b'/').map(|i| i + 1).unwrap_or(0);
Some(match *path {
Cow::Borrowed(path) => Cow::Borrowed(&path[last_slash..]),
Cow::Owned(ref path) => {
let mut path = path.clone();
path.drain_bytes(..last_slash);
Cow::Owned(path)
}
})
}
/// Return a file extension given a path's file name.
@@ -54,59 +39,28 @@ pub fn file_name<'a, P: AsRef<Path> + ?Sized>(
/// a pattern like `*.rs` is obviously trying to match files with a `rs`
/// extension, but it also matches files like `.rs`, which doesn't have an
/// extension according to std::path::Path::extension.
pub fn file_name_ext(name: &OsStr) -> Option<Cow<[u8]>> {
pub fn file_name_ext<'a>(name: &Cow<'a, BStr>) -> Option<Cow<'a, BStr>> {
if name.is_empty() {
return None;
}
let name = os_str_bytes(name);
let last_dot_at = {
let result = name
.iter().enumerate().rev()
.find(|&(_, &b)| b == b'.')
.map(|(i, _)| i);
match result {
None => return None,
Some(i) => i,
}
let last_dot_at = match name.rfind_byte(b'.') {
None => return None,
Some(i) => i,
};
Some(match name {
Some(match *name {
Cow::Borrowed(name) => Cow::Borrowed(&name[last_dot_at..]),
Cow::Owned(mut name) => {
name.drain(..last_dot_at);
Cow::Owned(ref name) => {
let mut name = name.clone();
name.drain_bytes(..last_dot_at);
Cow::Owned(name)
}
})
}
/// Return raw bytes of a path, transcoded to UTF-8 if necessary.
pub fn path_bytes(path: &Path) -> Cow<[u8]> {
os_str_bytes(path.as_os_str())
}
/// Return the raw bytes of the given OS string, possibly transcoded to UTF-8.
#[cfg(unix)]
pub fn os_str_bytes(s: &OsStr) -> Cow<[u8]> {
use std::os::unix::ffi::OsStrExt;
Cow::Borrowed(s.as_bytes())
}
/// Return the raw bytes of the given OS string, possibly transcoded to UTF-8.
#[cfg(not(unix))]
pub fn os_str_bytes(s: &OsStr) -> Cow<[u8]> {
// TODO(burntsushi): On Windows, OS strings are WTF-8, which is a superset
// of UTF-8, so even if we could get at the raw bytes, they wouldn't
// be useful. We *must* convert to UTF-8 before doing path matching.
// Unfortunate, but necessary.
match s.to_string_lossy() {
Cow::Owned(s) => Cow::Owned(s.into_bytes()),
Cow::Borrowed(s) => Cow::Borrowed(s.as_bytes()),
}
}
/// Normalizes a path to use `/` as a separator everywhere, even on platforms
/// that recognize other characters as separators.
#[cfg(unix)]
pub fn normalize_path(path: Cow<[u8]>) -> Cow<[u8]> {
pub fn normalize_path(path: Cow<BStr>) -> Cow<BStr> {
// UNIX only uses /, so we're good.
path
}
@@ -114,7 +68,7 @@ pub fn normalize_path(path: Cow<[u8]>) -> Cow<[u8]> {
/// Normalizes a path to use `/` as a separator everywhere, even on platforms
/// that recognize other characters as separators.
#[cfg(not(unix))]
pub fn normalize_path(mut path: Cow<[u8]>) -> Cow<[u8]> {
pub fn normalize_path(mut path: Cow<BStr>) -> Cow<BStr> {
use std::path::is_separator;
for i in 0..path.len() {
@@ -129,7 +83,8 @@ pub fn normalize_path(mut path: Cow<[u8]>) -> Cow<[u8]> {
#[cfg(test)]
mod tests {
use std::borrow::Cow;
use std::ffi::OsStr;
use bstr::{B, BString};
use super::{file_name_ext, normalize_path};
@@ -137,8 +92,9 @@ mod tests {
($name:ident, $file_name:expr, $ext:expr) => {
#[test]
fn $name() {
let got = file_name_ext(OsStr::new($file_name));
assert_eq!($ext.map(|s| Cow::Borrowed(s.as_bytes())), got);
let bs = BString::from($file_name);
let got = file_name_ext(&Cow::Owned(bs));
assert_eq!($ext.map(|s| Cow::Borrowed(B(s))), got);
}
};
}
@@ -153,7 +109,8 @@ mod tests {
($name:ident, $path:expr, $expected:expr) => {
#[test]
fn $name() {
let got = normalize_path(Cow::Owned($path.to_vec()));
let bs = BString::from_slice($path);
let got = normalize_path(Cow::Owned(bs));
assert_eq!($expected.to_vec(), got.into_owned());
}
};

7
grep-cli/Cargo.toml
View File

@@ -14,12 +14,13 @@ license = "Unlicense/MIT"
[dependencies]
atty = "0.2.11"
bstr = "0.1.2"
globset = { version = "0.4.2", path = "../globset" }
lazy_static = "1.1.0"
log = "0.4.5"
regex = "1.0.5"
same-file = "1.0.3"
termcolor = "1.0.3"
regex = "1.1"
same-file = "1.0.4"
termcolor = "1.0.4"
[target.'cfg(windows)'.dependencies.winapi-util]
version = "0.1.1"

9
grep-cli/src/decompress.rs
View File

@@ -352,6 +352,8 @@ fn default_decompression_commands() -> Vec<DecompressionCommand> {
const ARGS_XZ: &[&str] = &["xz", "-d", "-c"];
const ARGS_LZ4: &[&str] = &["lz4", "-d", "-c"];
const ARGS_LZMA: &[&str] = &["xz", "--format=lzma", "-d", "-c"];
const ARGS_BROTLI: &[&str] = &["brotli", "-d", "-c"];
const ARGS_ZSTD: &[&str] = &["zstd", "-q", "-d", "-c"];
fn cmd(glob: &str, args: &[&str]) -> DecompressionCommand {
DecompressionCommand {
@@ -367,15 +369,14 @@ fn default_decompression_commands() -> Vec<DecompressionCommand> {
vec![
cmd("*.gz", ARGS_GZIP),
cmd("*.tgz", ARGS_GZIP),
cmd("*.bz2", ARGS_BZIP),
cmd("*.tbz2", ARGS_BZIP),
cmd("*.xz", ARGS_XZ),
cmd("*.txz", ARGS_XZ),
cmd("*.lz4", ARGS_LZ4),
cmd("*.lzma", ARGS_LZMA),
cmd("*.br", ARGS_BROTLI),
cmd("*.zst", ARGS_ZSTD),
cmd("*.zstd", ARGS_ZSTD),
]
}

74
grep-cli/src/escape.rs
View File

@@ -1,6 +1,8 @@
use std::ffi::OsStr;
use std::str;
use bstr::{BStr, BString};
/// A single state in the state machine used by `unescape`.
#[derive(Clone, Copy, Eq, PartialEq)]
enum State {
@@ -35,18 +37,16 @@ enum State {
///
/// assert_eq!(r"foo\nbar\xFFbaz", escape(b"foo\nbar\xFFbaz"));
/// ```
pub fn escape(mut bytes: &[u8]) -> String {
pub fn escape(bytes: &[u8]) -> String {
let bytes = BStr::new(bytes);
let mut escaped = String::new();
while let Some(result) = decode_utf8(bytes) {
match result {
Ok(cp) => {
escape_char(cp, &mut escaped);
bytes = &bytes[cp.len_utf8()..];
}
Err(byte) => {
escape_byte(byte, &mut escaped);
bytes = &bytes[1..];
for (s, e, ch) in bytes.char_indices() {
if ch == '\u{FFFD}' {
for b in bytes[s..e].bytes() {
escape_byte(b, &mut escaped);
}
} else {
escape_char(ch, &mut escaped);
}
}
escaped
@@ -56,19 +56,7 @@ pub fn escape(mut bytes: &[u8]) -> String {
///
/// This is like [`escape`](fn.escape.html), but accepts an OS string.
pub fn escape_os(string: &OsStr) -> String {
#[cfg(unix)]
fn imp(string: &OsStr) -> String {
use std::os::unix::ffi::OsStrExt;
escape(string.as_bytes())
}
#[cfg(not(unix))]
fn imp(string: &OsStr) -> String {
escape(string.to_string_lossy().as_bytes())
}
imp(string)
escape(BString::from_os_str_lossy(string).as_bytes())
}
/// Unescapes a string.
@@ -195,46 +183,6 @@ fn escape_byte(byte: u8, into: &mut String) {
}
}
/// Decodes the next UTF-8 encoded codepoint from the given byte slice.
///
/// If no valid encoding of a codepoint exists at the beginning of the given
/// byte slice, then the first byte is returned instead.
///
/// This returns `None` if and only if `bytes` is empty.
fn decode_utf8(bytes: &[u8]) -> Option<Result<char, u8>> {
if bytes.is_empty() {
return None;
}
let len = match utf8_len(bytes[0]) {
None => return Some(Err(bytes[0])),
Some(len) if len > bytes.len() => return Some(Err(bytes[0])),
Some(len) => len,
};
match str::from_utf8(&bytes[..len]) {
Ok(s) => Some(Ok(s.chars().next().unwrap())),
Err(_) => Some(Err(bytes[0])),
}
}
/// Given a UTF-8 leading byte, this returns the total number of code units
/// in the following encoded codepoint.
///
/// If the given byte is not a valid UTF-8 leading byte, then this returns
/// `None`.
fn utf8_len(byte: u8) -> Option<usize> {
if byte <= 0x7F {
Some(1)
} else if byte <= 0b110_11111 {
Some(2)
} else if byte <= 0b1110_1111 {
Some(3)
} else if byte <= 0b1111_0111 {
Some(4)
} else {
None
}
}
#[cfg(test)]
mod tests {
use super::{escape, unescape};

1
grep-cli/src/lib.rs
View File

@@ -159,6 +159,7 @@ error message is crafted that typically tells the user how to fix the problem.
#![deny(missing_docs)]
extern crate atty;
extern crate bstr;
extern crate globset;
#[macro_use]
extern crate lazy_static;

4
grep-matcher/Cargo.toml
View File

@@ -14,10 +14,10 @@ license = "Unlicense/MIT"
autotests = false
[dependencies]
memchr = "2.0.2"
memchr = "2.1"
[dev-dependencies]
regex = "1.0.5"
regex = "1.1"
[[test]]
name = "integration"

4
grep-pcre2/Cargo.toml
View File

@@ -1,6 +1,6 @@
[package]
name = "grep-pcre2"
version = "0.1.1" #:version
version = "0.1.2" #:version
authors = ["Andrew Gallant <jamslam@gmail.com>"]
description = """
Use PCRE2 with the 'grep' crate.
@@ -14,4 +14,4 @@ license = "Unlicense/MIT"
[dependencies]
grep-matcher = { version = "0.1.1", path = "../grep-matcher" }
pcre2 = "0.1.0"
pcre2 = "0.1.1"

22
grep-pcre2/src/matcher.rs
View File

@@ -199,16 +199,34 @@ impl RegexMatcherBuilder {
self
}
/// Enable PCRE2's JIT.
/// Enable PCRE2's JIT and return an error if it's not available.
///
/// This generally speeds up matching quite a bit. The downside is that it
/// can increase the time it takes to compile a pattern.
///
/// This is disabled by default.
/// If the JIT isn't available or if JIT compilation returns an error, then
/// regex compilation will fail with the corresponding error.
///
/// This is disabled by default, and always overrides `jit_if_available`.
pub fn jit(&mut self, yes: bool) -> &mut RegexMatcherBuilder {
self.builder.jit(yes);
self
}
/// Enable PCRE2's JIT if it's available.
///
/// This generally speeds up matching quite a bit. The downside is that it
/// can increase the time it takes to compile a pattern.
///
/// If the JIT isn't available or if JIT compilation returns an error,
/// then a debug message with the error will be emitted and the regex will
/// otherwise silently fall back to non-JIT matching.
///
/// This is disabled by default, and always overrides `jit`.
pub fn jit_if_available(&mut self, yes: bool) -> &mut RegexMatcherBuilder {
self.builder.jit_if_available(yes);
self
}
}
/// An implementation of the `Matcher` trait using PCRE2.

5
grep-printer/Cargo.toml
View File

@@ -18,10 +18,11 @@ default = ["serde1"]
serde1 = ["base64", "serde", "serde_derive", "serde_json"]
[dependencies]
base64 = { version = "0.9.2", optional = true }
base64 = { version = "0.10.0", optional = true }
bstr = "0.1.2"
grep-matcher = { version = "0.1.1", path = "../grep-matcher" }
grep-searcher = { version = "0.1.1", path = "../grep-searcher" }
termcolor = "1.0.3"
termcolor = "1.0.4"
serde = { version = "1.0.77", optional = true }
serde_derive = { version = "1.0.77", optional = true }
serde_json = { version = "1.0.27", optional = true }

44
grep-printer/src/json.rs
View File

@@ -817,7 +817,8 @@ impl<'a> SubMatches<'a> {
#[cfg(test)]
mod tests {
use grep_regex::RegexMatcher;
use grep_regex::{RegexMatcher, RegexMatcherBuilder};
use grep_matcher::LineTerminator;
use grep_searcher::SearcherBuilder;
use super::{JSON, JSONBuilder};
@@ -918,4 +919,45 @@ and exhibited clearly, with a label attached.\
assert_eq!(got.lines().count(), 2);
assert!(got.contains("begin") && got.contains("end"));
}
#[test]
fn missing_crlf() {
let haystack = "test\r\n".as_bytes();
let matcher = RegexMatcherBuilder::new()
.build("test")
.unwrap();
let mut printer = JSONBuilder::new()
.build(vec![]);
SearcherBuilder::new()
.build()
.search_reader(&matcher, haystack, printer.sink(&matcher))
.unwrap();
let got = printer_contents(&mut printer);
assert_eq!(got.lines().count(), 3);
assert!(
got.lines().nth(1).unwrap().contains(r"test\r\n"),
r"missing 'test\r\n' in '{}'",
got.lines().nth(1).unwrap(),
);
let matcher = RegexMatcherBuilder::new()
.crlf(true)
.build("test")
.unwrap();
let mut printer = JSONBuilder::new()
.build(vec![]);
SearcherBuilder::new()
.line_terminator(LineTerminator::crlf())
.build()
.search_reader(&matcher, haystack, printer.sink(&matcher))
.unwrap();
let got = printer_contents(&mut printer);
assert_eq!(got.lines().count(), 3);
assert!(
got.lines().nth(1).unwrap().contains(r"test\r\n"),
r"missing 'test\r\n' in '{}'",
got.lines().nth(1).unwrap(),
);
}
}

1
grep-printer/src/lib.rs
View File

@@ -70,6 +70,7 @@ fn example() -> Result<(), Box<Error>> {
#[cfg(feature = "serde1")]
extern crate base64;
extern crate bstr;
extern crate grep_matcher;
#[cfg(test)]
extern crate grep_regex;

1
grep-printer/src/macros.rs
View File

@@ -1,3 +1,4 @@
/// Like assert_eq, but nicer output for long strings.
#[cfg(test)]
#[macro_export]
macro_rules! assert_eq_printed {

7
grep-printer/src/summary.rs
View File

@@ -403,7 +403,7 @@ impl<W: WriteColor> Summary<W> {
where M: Matcher,
P: ?Sized + AsRef<Path>,
{
if !self.config.path {
if !self.config.path && !self.config.kind.requires_path() {
return self.sink(matcher);
}
let stats =
@@ -477,7 +477,10 @@ impl<'p, 's, M: Matcher, W: WriteColor> SummarySink<'p, 's, M, W> {
/// This is unaffected by the result of searches before the previous
/// search.
pub fn has_match(&self) -> bool {
self.match_count > 0
match self.summary.config.kind {
SummaryKind::PathWithoutMatch => self.match_count == 0,
_ => self.match_count > 0,
}
}
/// If binary data was found in the previous search, this returns the

23
grep-printer/src/util.rs
View File

@@ -4,6 +4,7 @@ use std::io;
use std::path::Path;
use std::time;
use bstr::{BStr, BString};
use grep_matcher::{Captures, LineTerminator, Match, Matcher};
use grep_searcher::{
LineIter,
@@ -262,26 +263,12 @@ impl<'a> Sunk<'a> {
/// portability with a small cost: on Windows, paths that are not valid UTF-16
/// will not roundtrip correctly.
#[derive(Clone, Debug)]
pub struct PrinterPath<'a>(Cow<'a, [u8]>);
pub struct PrinterPath<'a>(Cow<'a, BStr>);
impl<'a> PrinterPath<'a> {
/// Create a new path suitable for printing.
pub fn new(path: &'a Path) -> PrinterPath<'a> {
PrinterPath::new_impl(path)
}
#[cfg(unix)]
fn new_impl(path: &'a Path) -> PrinterPath<'a> {
use std::os::unix::ffi::OsStrExt;
PrinterPath(Cow::Borrowed(path.as_os_str().as_bytes()))
}
#[cfg(not(unix))]
fn new_impl(path: &'a Path) -> PrinterPath<'a> {
PrinterPath(match path.to_string_lossy() {
Cow::Owned(path) => Cow::Owned(path.into_bytes()),
Cow::Borrowed(path) => Cow::Borrowed(path.as_bytes()),
})
PrinterPath(BString::from_path_lossy(path))
}
/// Create a new printer path from the given path which can be efficiently
@@ -302,7 +289,7 @@ impl<'a> PrinterPath<'a> {
/// path separators that are both replaced by `new_sep`. In all other
/// environments, only `/` is treated as a path separator.
fn replace_separator(&mut self, new_sep: u8) {
let transformed_path: Vec<_> = self.as_bytes().iter().map(|&b| {
let transformed_path: BString = self.0.bytes().map(|b| {
if b == b'/' || (cfg!(windows) && b == b'\\') {
new_sep
} else {
@@ -314,7 +301,7 @@ impl<'a> PrinterPath<'a> {
/// Return the raw bytes for this path.
pub fn as_bytes(&self) -> &[u8] {
&*self.0
self.0.as_bytes()
}
}

6
grep-regex/Cargo.toml
View File

@@ -1,6 +1,6 @@
[package]
name = "grep-regex"
version = "0.1.1" #:version
version = "0.1.2" #:version
authors = ["Andrew Gallant <jamslam@gmail.com>"]
description = """
Use Rust's regex library with the 'grep' crate.
@@ -15,7 +15,7 @@ license = "Unlicense/MIT"
[dependencies]
log = "0.4.5"
grep-matcher = { version = "0.1.1", path = "../grep-matcher" }
regex = "1.0.5"
regex-syntax = "0.6.2"
regex = "1.1"
regex-syntax = "0.6.5"
thread_local = "0.3.6"
utf8-ranges = "1.0.1"

10
grep-regex/src/config.rs
View File

@@ -160,6 +160,14 @@ impl ConfiguredHIR {
non_matching_bytes(&self.expr)
}
/// Returns true if and only if this regex needs to have its match offsets
/// tweaked because of CRLF support. Specifically, this occurs when the
/// CRLF hack is enabled and the regex is line anchored at the end. In
/// this case, matches that end with a `\r` have the `\r` stripped.
pub fn needs_crlf_stripped(&self) -> bool {
self.config.crlf && self.expr.is_line_anchored_end()
}
/// Builds a regular expression from this HIR expression.
pub fn regex(&self) -> Result<Regex, Error> {
self.pattern_to_regex(&self.expr.to_string())
@@ -199,7 +207,7 @@ impl ConfiguredHIR {
if self.config.line_terminator.is_none() {
return Ok(None);
}
match LiteralSets::new(&self.expr).one_regex() {
match LiteralSets::new(&self.expr).one_regex(self.config.word) {
None => Ok(None),
Some(pattern) => self.pattern_to_regex(&pattern).map(Some),
}

105
grep-regex/src/crlf.rs
View File

@@ -1,5 +1,110 @@
use std::collections::HashMap;
use grep_matcher::{Match, Matcher, NoError};
use regex::bytes::Regex;
use regex_syntax::hir::{self, Hir, HirKind};
use config::ConfiguredHIR;
use error::Error;
use matcher::RegexCaptures;
/// A matcher for implementing "word match" semantics.
#[derive(Clone, Debug)]
pub struct CRLFMatcher {
/// The regex.
regex: Regex,
/// A map from capture group name to capture group index.
names: HashMap<String, usize>,
}
impl CRLFMatcher {
/// Create a new matcher from the given pattern that strips `\r` from the
/// end of every match.
///
/// This panics if the given expression doesn't need its CRLF stripped.
pub fn new(expr: &ConfiguredHIR) -> Result<CRLFMatcher, Error> {
assert!(expr.needs_crlf_stripped());
let regex = expr.regex()?;
let mut names = HashMap::new();
for (i, optional_name) in regex.capture_names().enumerate() {
if let Some(name) = optional_name {
names.insert(name.to_string(), i.checked_sub(1).unwrap());
}
}
Ok(CRLFMatcher { regex, names })
}
/// Return the underlying regex used by this matcher.
pub fn regex(&self) -> &Regex {
&self.regex
}
}
impl Matcher for CRLFMatcher {
type Captures = RegexCaptures;
type Error = NoError;
fn find_at(
&self,
haystack: &[u8],
at: usize,
) -> Result<Option<Match>, NoError> {
let m = match self.regex.find_at(haystack, at) {
None => return Ok(None),
Some(m) => Match::new(m.start(), m.end()),
};
Ok(Some(adjust_match(haystack, m)))
}
fn new_captures(&self) -> Result<RegexCaptures, NoError> {
Ok(RegexCaptures::new(self.regex.capture_locations()))
}
fn capture_count(&self) -> usize {
self.regex.captures_len().checked_sub(1).unwrap()
}
fn capture_index(&self, name: &str) -> Option<usize> {
self.names.get(name).map(|i| *i)
}
fn captures_at(
&self,
haystack: &[u8],
at: usize,
caps: &mut RegexCaptures,
) -> Result<bool, NoError> {
caps.strip_crlf(false);
let r = self.regex.captures_read_at(caps.locations(), haystack, at);
if !r.is_some() {
return Ok(false);
}
// If the end of our match includes a `\r`, then strip it from all
// capture groups ending at the same location.
let end = caps.locations().get(0).unwrap().1;
if end > 0 && haystack.get(end - 1) == Some(&b'\r') {
caps.strip_crlf(true);
}
Ok(true)
}
// We specifically do not implement other methods like find_iter or
// captures_iter. Namely, the iter methods are guaranteed to be correct
// by virtue of implementing find_at and captures_at above.
}
/// If the given match ends with a `\r`, then return a new match that ends
/// immediately before the `\r`.
pub fn adjust_match(haystack: &[u8], m: Match) -> Match {
if m.end() > 0 && haystack.get(m.end() - 1) == Some(&b'\r') {
m.with_end(m.end() - 1)
} else {
m
}
}
/// Substitutes all occurrences of multi-line enabled `$` with `(?:\r?$)`.
///
/// This does not preserve the exact semantics of the given expression,

45
grep-regex/src/literal.rs
View File

@@ -47,18 +47,23 @@ impl LiteralSets {
/// generated these literal sets. The idea here is that the pattern
/// returned by this method is much cheaper to search for. i.e., It is
/// usually a single literal or an alternation of literals.
pub fn one_regex(&self) -> Option<String> {
pub fn one_regex(&self, word: bool) -> Option<String> {
// TODO: The logic in this function is basically inscrutable. It grew
// organically in the old grep 0.1 crate. Ideally, it would be
// re-worked. In fact, the entire inner literal extraction should be
// re-worked. Actually, most of regex-syntax's literal extraction
// should also be re-worked. Alas... only so much time in the day.
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,
// so we don't need to return anything.
return None;
if !word {
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,
// so we don't need to return anything. But we only do this
// if we aren't doing a word regex, since a word regex adds
// a `(?:\W|^)` to the beginning of the regex, thereby
// defeating the regex engine's literal detection.
return None;
}
}
// Out of inner required literals, prefixes and suffixes, which one
@@ -166,10 +171,10 @@ fn union_required(expr: &Hir, lits: &mut Literals) {
lits.cut();
continue;
}
if lits2.contains_empty() {
if lits2.contains_empty() || !is_simple(&e) {
lits.cut();
}
if !lits.cross_product(&lits2) {
if !lits.cross_product(&lits2) || !lits2.any_complete() {
// If this expression couldn't yield any literal that
// could be extended, then we need to quit. Since we're
// short-circuiting, we also need to freeze every member.
@@ -250,6 +255,20 @@ fn alternate_literals<F: FnMut(&Hir, &mut Literals)>(
}
}
fn is_simple(expr: &Hir) -> bool {
match *expr.kind() {
HirKind::Empty
| HirKind::Literal(_)
| HirKind::Class(_)
| HirKind::Repetition(_)
| HirKind::Concat(_)
| HirKind::Alternation(_) => true,
HirKind::Anchor(_)
| HirKind::WordBoundary(_)
| HirKind::Group(_) => false,
}
}
/// Return the number of characters in the given class.
fn count_unicode_class(cls: &hir::ClassUnicode) -> u32 {
cls.iter().map(|r| 1 + (r.end() as u32 - r.start() as u32)).sum()
@@ -271,7 +290,7 @@ mod tests {
}
fn one_regex(pattern: &str) -> Option<String> {
sets(pattern).one_regex()
sets(pattern).one_regex(false)
}
// Put a pattern into the same format as the one returned by `one_regex`.
@@ -301,4 +320,12 @@ mod tests {
// assert_eq!(one_regex(r"\w(foo|bar|baz)"), pat("foo|bar|baz"));
// assert_eq!(one_regex(r"\w(foo|bar|baz)\w"), pat("foo|bar|baz"));
}
#[test]
fn regression_1064() {
// Regression from:
// https://github.com/BurntSushi/ripgrep/issues/1064
// assert_eq!(one_regex(r"a.*c"), pat("a"));
assert_eq!(one_regex(r"a(.*c)"), pat("a"));
}
}

71
grep-regex/src/matcher.rs
View File

@@ -6,6 +6,7 @@ use grep_matcher::{
use regex::bytes::{CaptureLocations, Regex};
use config::{Config, ConfiguredHIR};
use crlf::CRLFMatcher;
use error::Error;
use word::WordMatcher;
@@ -49,9 +50,12 @@ impl RegexMatcherBuilder {
if let Some(ref re) = fast_line_regex {
trace!("extracted fast line regex: {:?}", re);
}
let matcher = RegexMatcherImpl::new(&chir)?;
trace!("final regex: {:?}", matcher.regex());
Ok(RegexMatcher {
config: self.config.clone(),
matcher: RegexMatcherImpl::new(&chir)?,
matcher: matcher,
fast_line_regex: fast_line_regex,
non_matching_bytes: non_matching_bytes,
})
@@ -344,6 +348,11 @@ impl RegexMatcher {
enum RegexMatcherImpl {
/// The standard matcher used for all regular expressions.
Standard(StandardMatcher),
/// A matcher that strips `\r` from the end of matches.
///
/// This is only used when the CRLF hack is enabled and the regex is line
/// anchored at the end.
CRLF(CRLFMatcher),
/// A matcher that only matches at word boundaries. This transforms the
/// regex to `(^|\W)(...)($|\W)` instead of the more intuitive `\b(...)\b`.
/// Because of this, the WordMatcher provides its own implementation of
@@ -358,10 +367,21 @@ impl RegexMatcherImpl {
fn new(expr: &ConfiguredHIR) -> Result<RegexMatcherImpl, Error> {
if expr.config().word {
Ok(RegexMatcherImpl::Word(WordMatcher::new(expr)?))
} else if expr.needs_crlf_stripped() {
Ok(RegexMatcherImpl::CRLF(CRLFMatcher::new(expr)?))
} else {
Ok(RegexMatcherImpl::Standard(StandardMatcher::new(expr)?))
}
}
/// Return the underlying regex object used.
fn regex(&self) -> &Regex {
match *self {
RegexMatcherImpl::Word(ref x) => x.regex(),
RegexMatcherImpl::CRLF(ref x) => x.regex(),
RegexMatcherImpl::Standard(ref x) => &x.regex,
}
}
}
// This implementation just dispatches on the internal matcher impl except
@@ -379,6 +399,7 @@ impl Matcher for RegexMatcher {
use self::RegexMatcherImpl::*;
match self.matcher {
Standard(ref m) => m.find_at(haystack, at),
CRLF(ref m) => m.find_at(haystack, at),
Word(ref m) => m.find_at(haystack, at),
}
}
@@ -387,6 +408,7 @@ impl Matcher for RegexMatcher {
use self::RegexMatcherImpl::*;
match self.matcher {
Standard(ref m) => m.new_captures(),
CRLF(ref m) => m.new_captures(),
Word(ref m) => m.new_captures(),
}
}
@@ -395,6 +417,7 @@ impl Matcher for RegexMatcher {
use self::RegexMatcherImpl::*;
match self.matcher {
Standard(ref m) => m.capture_count(),
CRLF(ref m) => m.capture_count(),
Word(ref m) => m.capture_count(),
}
}
@@ -403,6 +426,7 @@ impl Matcher for RegexMatcher {
use self::RegexMatcherImpl::*;
match self.matcher {
Standard(ref m) => m.capture_index(name),
CRLF(ref m) => m.capture_index(name),
Word(ref m) => m.capture_index(name),
}
}
@@ -411,6 +435,7 @@ impl Matcher for RegexMatcher {
use self::RegexMatcherImpl::*;
match self.matcher {
Standard(ref m) => m.find(haystack),
CRLF(ref m) => m.find(haystack),
Word(ref m) => m.find(haystack),
}
}
@@ -425,6 +450,7 @@ impl Matcher for RegexMatcher {
use self::RegexMatcherImpl::*;
match self.matcher {
Standard(ref m) => m.find_iter(haystack, matched),
CRLF(ref m) => m.find_iter(haystack, matched),
Word(ref m) => m.find_iter(haystack, matched),
}
}
@@ -439,6 +465,7 @@ impl Matcher for RegexMatcher {
use self::RegexMatcherImpl::*;
match self.matcher {
Standard(ref m) => m.try_find_iter(haystack, matched),
CRLF(ref m) => m.try_find_iter(haystack, matched),
Word(ref m) => m.try_find_iter(haystack, matched),
}
}
@@ -451,6 +478,7 @@ impl Matcher for RegexMatcher {
use self::RegexMatcherImpl::*;
match self.matcher {
Standard(ref m) => m.captures(haystack, caps),
CRLF(ref m) => m.captures(haystack, caps),
Word(ref m) => m.captures(haystack, caps),
}
}
@@ -466,6 +494,7 @@ impl Matcher for RegexMatcher {
use self::RegexMatcherImpl::*;
match self.matcher {
Standard(ref m) => m.captures_iter(haystack, caps, matched),
CRLF(ref m) => m.captures_iter(haystack, caps, matched),
Word(ref m) => m.captures_iter(haystack, caps, matched),
}
}
@@ -481,6 +510,7 @@ impl Matcher for RegexMatcher {
use self::RegexMatcherImpl::*;
match self.matcher {
Standard(ref m) => m.try_captures_iter(haystack, caps, matched),
CRLF(ref m) => m.try_captures_iter(haystack, caps, matched),
Word(ref m) => m.try_captures_iter(haystack, caps, matched),
}
}
@@ -494,6 +524,7 @@ impl Matcher for RegexMatcher {
use self::RegexMatcherImpl::*;
match self.matcher {
Standard(ref m) => m.captures_at(haystack, at, caps),
CRLF(ref m) => m.captures_at(haystack, at, caps),
Word(ref m) => m.captures_at(haystack, at, caps),
}
}
@@ -509,6 +540,7 @@ impl Matcher for RegexMatcher {
use self::RegexMatcherImpl::*;
match self.matcher {
Standard(ref m) => m.replace(haystack, dst, append),
CRLF(ref m) => m.replace(haystack, dst, append),
Word(ref m) => m.replace(haystack, dst, append),
}
}
@@ -527,6 +559,9 @@ impl Matcher for RegexMatcher {
Standard(ref m) => {
m.replace_with_captures(haystack, caps, dst, append)
}
CRLF(ref m) => {
m.replace_with_captures(haystack, caps, dst, append)
}
Word(ref m) => {
m.replace_with_captures(haystack, caps, dst, append)
}
@@ -537,6 +572,7 @@ impl Matcher for RegexMatcher {
use self::RegexMatcherImpl::*;
match self.matcher {
Standard(ref m) => m.is_match(haystack),
CRLF(ref m) => m.is_match(haystack),
Word(ref m) => m.is_match(haystack),
}
}
@@ -549,6 +585,7 @@ impl Matcher for RegexMatcher {
use self::RegexMatcherImpl::*;
match self.matcher {
Standard(ref m) => m.is_match_at(haystack, at),
CRLF(ref m) => m.is_match_at(haystack, at),
Word(ref m) => m.is_match_at(haystack, at),
}
}
@@ -560,6 +597,7 @@ impl Matcher for RegexMatcher {
use self::RegexMatcherImpl::*;
match self.matcher {
Standard(ref m) => m.shortest_match(haystack),
CRLF(ref m) => m.shortest_match(haystack),
Word(ref m) => m.shortest_match(haystack),
}
}
@@ -572,6 +610,7 @@ impl Matcher for RegexMatcher {
use self::RegexMatcherImpl::*;
match self.matcher {
Standard(ref m) => m.shortest_match_at(haystack, at),
CRLF(ref m) => m.shortest_match_at(haystack, at),
Word(ref m) => m.shortest_match_at(haystack, at),
}
}
@@ -712,6 +751,9 @@ pub struct RegexCaptures {
/// and the capturing groups must behave as if `(re)` is the `0`th capture
/// group.
offset: usize,
/// When enable, the end of a match has `\r` stripped from it, if one
/// exists.
strip_crlf: bool,
}
impl Captures for RegexCaptures {
@@ -720,8 +762,25 @@ impl Captures for RegexCaptures {
}
fn get(&self, i: usize) -> Option<Match> {
let actual = i.checked_add(self.offset).unwrap();
self.locs.pos(actual).map(|(s, e)| Match::new(s, e))
if !self.strip_crlf {
let actual = i.checked_add(self.offset).unwrap();
return self.locs.pos(actual).map(|(s, e)| Match::new(s, e));
}
// currently don't support capture offsetting with CRLF stripping
assert_eq!(self.offset, 0);
let m = match self.locs.pos(i).map(|(s, e)| Match::new(s, e)) {
None => return None,
Some(m) => m,
};
// If the end position of this match corresponds to the end position
// of the overall match, then we apply our CRLF stripping. Otherwise,
// we cannot assume stripping is correct.
if i == 0 || m.end() == self.locs.pos(0).unwrap().1 {
Some(m.with_end(m.end() - 1))
} else {
Some(m)
}
}
}
@@ -734,12 +793,16 @@ impl RegexCaptures {
locs: CaptureLocations,
offset: usize,
) -> RegexCaptures {
RegexCaptures { locs, offset }
RegexCaptures { locs, offset, strip_crlf: false }
}
pub(crate) fn locations(&mut self) -> &mut CaptureLocations {
&mut self.locs
}
pub(crate) fn strip_crlf(&mut self, yes: bool) {
self.strip_crlf = yes;
}
}
#[cfg(test)]

5
grep-regex/src/word.rs
View File

@@ -55,6 +55,11 @@ impl WordMatcher {
}
Ok(WordMatcher { regex, names, locs })
}
/// Return the underlying regex used by this matcher.
pub fn regex(&self) -> &Regex {
&self.regex
}
}
impl Matcher for WordMatcher {

26
grep-searcher/Cargo.toml
View File

@@ -1,6 +1,6 @@
[package]
name = "grep-searcher"
version = "0.1.1" #:version
version = "0.1.3" #:version
authors = ["Andrew Gallant <jamslam@gmail.com>"]
description = """
Fast line oriented regex searching as a library.
@@ -13,23 +13,21 @@ keywords = ["regex", "grep", "egrep", "search", "pattern"]
license = "Unlicense/MIT"
[dependencies]
bytecount = "0.3.2"
encoding_rs = "0.8.6"
encoding_rs_io = "0.1.2"
bstr = { version = "0.1.2", default-features = false, features = ["std"] }
bytecount = "0.5"
encoding_rs = "0.8.14"
encoding_rs_io = "0.1.4"
grep-matcher = { version = "0.1.1", path = "../grep-matcher" }
log = "0.4.5"
memchr = "2.0.2"
memmap = "0.6.2"
memmap = "0.7"
[dev-dependencies]
grep-regex = { version = "0.1.1", path = "../grep-regex" }
regex = "1.0.5"
regex = "1.1"
[features]
avx-accel = [
"bytecount/avx-accel",
]
simd-accel = [
"bytecount/simd-accel",
"encoding_rs/simd-accel",
]
default = ["bytecount/runtime-dispatch-simd"]
simd-accel = ["encoding_rs/simd-accel"]
# This feature is DEPRECATED. Runtime dispatch is used for SIMD now.
avx-accel = []

2
grep-searcher/src/lib.rs
View File

@@ -99,13 +99,13 @@ searches stdin.
#![deny(missing_docs)]
extern crate bstr;
extern crate bytecount;
extern crate encoding_rs;
extern crate encoding_rs_io;
extern crate grep_matcher;
#[macro_use]
extern crate log;
extern crate memchr;
extern crate memmap;
#[cfg(test)]
extern crate regex;

125
grep-searcher/src/line_buffer.rs
View File

@@ -1,8 +1,7 @@
use std::cmp;
use std::io;
use std::ptr;
use memchr::{memchr, memrchr};
use bstr::{BStr, BString};
/// The default buffer capacity that we use for the line buffer.
pub(crate) const DEFAULT_BUFFER_CAPACITY: usize = 8 * (1<<10); // 8 KB
@@ -123,7 +122,7 @@ impl LineBufferBuilder {
pub fn build(&self) -> LineBuffer {
LineBuffer {
config: self.config,
buf: vec![0; self.config.capacity],
buf: BString::from(vec![0; self.config.capacity]),
pos: 0,
last_lineterm: 0,
end: 0,
@@ -255,6 +254,12 @@ impl<'b, R: io::Read> LineBufferReader<'b, R> {
/// Return the contents of this buffer.
pub fn buffer(&self) -> &[u8] {
self.line_buffer.buffer().as_bytes()
}
/// Return the underlying buffer as a byte string. Used for tests only.
#[cfg(test)]
fn bstr(&self) -> &BStr {
self.line_buffer.buffer()
}
@@ -284,7 +289,7 @@ pub struct LineBuffer {
/// The configuration of this buffer.
config: Config,
/// The primary buffer with which to hold data.
buf: Vec<u8>,
buf: BString,
/// The current position of this buffer. This is always a valid sliceable
/// index into `buf`, and its maximum value is the length of `buf`.
pos: usize,
@@ -339,13 +344,13 @@ impl LineBuffer {
}
/// Return the contents of this buffer.
fn buffer(&self) -> &[u8] {
fn buffer(&self) -> &BStr {
&self.buf[self.pos..self.last_lineterm]
}
/// Return the contents of the free space beyond the end of the buffer as
/// a mutable slice.
fn free_buffer(&mut self) -> &mut [u8] {
fn free_buffer(&mut self) -> &mut BStr {
&mut self.buf[self.end..]
}
@@ -396,7 +401,7 @@ impl LineBuffer {
assert_eq!(self.pos, 0);
loop {
self.ensure_capacity()?;
let readlen = rdr.read(self.free_buffer())?;
let readlen = rdr.read(self.free_buffer().as_bytes_mut())?;
if readlen == 0 {
// We're only done reading for good once the caller has
// consumed everything.
@@ -416,7 +421,7 @@ impl LineBuffer {
match self.config.binary {
BinaryDetection::None => {} // nothing to do
BinaryDetection::Quit(byte) => {
if let Some(i) = memchr(byte, newbytes) {
if let Some(i) = newbytes.find_byte(byte) {
self.end = oldend + i;
self.last_lineterm = self.end;
self.binary_byte_offset =
@@ -444,7 +449,7 @@ impl LineBuffer {
}
// Update our `last_lineterm` positions if we read one.
if let Some(i) = memrchr(self.config.lineterm, newbytes) {
if let Some(i) = newbytes.rfind_byte(self.config.lineterm) {
self.last_lineterm = oldend + i + 1;
return Ok(true);
}
@@ -467,40 +472,8 @@ impl LineBuffer {
return;
}
assert!(self.pos < self.end && self.end <= self.buf.len());
let roll_len = self.end - self.pos;
unsafe {
// SAFETY: A buffer contains Copy data, so there's no problem
// moving it around. Safety also depends on our indices being
// in bounds, which they should always be, and we enforce with
// an assert above.
//
// It seems like it should be possible to do this in safe code that
// results in the same codegen. I tried the obvious:
//
// for (src, dst) in (self.pos..self.end).zip(0..) {
// self.buf[dst] = self.buf[src];
// }
//
// But the above does not work, and in fact compiles down to a slow
// byte-by-byte loop. I tried a few other minor variations, but
// alas, better minds might prevail.
//
// Overall, this doesn't save us *too* much. It mostly matters when
// the number of bytes we're copying is large, which can happen
// if the searcher is asked to produce a lot of context. We could
// decide this isn't worth it, but it does make an appreciable
// impact at or around the context=30 range on my machine.
//
// We could also use a temporary buffer that compiles down to two
// memcpys and is faster than the byte-at-a-time loop, but it
// complicates our options for limiting memory allocation a bit.
ptr::copy(
self.buf[self.pos..].as_ptr(),
self.buf.as_mut_ptr(),
roll_len,
);
}
self.buf.copy_within(self.pos.., 0);
self.pos = 0;
self.last_lineterm = roll_len;
self.end = roll_len;
@@ -536,14 +509,15 @@ impl LineBuffer {
}
}
/// Replaces `src` with `replacement` in bytes.
fn replace_bytes(bytes: &mut [u8], src: u8, replacement: u8) -> Option<usize> {
/// Replaces `src` with `replacement` in bytes, and return the offset of the
/// first replacement, if one exists.
fn replace_bytes(bytes: &mut BStr, src: u8, replacement: u8) -> Option<usize> {
if src == replacement {
return None;
}
let mut first_pos = None;
let mut pos = 0;
while let Some(i) = memchr(src, &bytes[pos..]).map(|i| pos + i) {
while let Some(i) = bytes[pos..].find_byte(src).map(|i| pos + i) {
if first_pos.is_none() {
first_pos = Some(i);
}
@@ -560,6 +534,7 @@ fn replace_bytes(bytes: &mut [u8], src: u8, replacement: u8) -> Option<usize> {
#[cfg(test)]
mod tests {
use std::str;
use bstr::BString;
use super::*;
const SHERLOCK: &'static str = "\
@@ -575,18 +550,14 @@ and exhibited clearly, with a label attached.\
slice.to_string()
}
fn btos(slice: &[u8]) -> &str {
str::from_utf8(slice).unwrap()
}
fn replace_str(
slice: &str,
src: u8,
replacement: u8,
) -> (String, Option<usize>) {
let mut dst = slice.to_string().into_bytes();
let mut dst = BString::from(slice);
let result = replace_bytes(&mut dst, src, replacement);
(String::from_utf8(dst).unwrap(), result)
(dst.into_string().unwrap(), result)
}
#[test]
@@ -607,7 +578,7 @@ and exhibited clearly, with a label attached.\
assert!(rdr.buffer().is_empty());
assert!(rdr.fill().unwrap());
assert_eq!(btos(rdr.buffer()), "homer\nlisa\n");
assert_eq!(rdr.bstr(), "homer\nlisa\n");
assert_eq!(rdr.absolute_byte_offset(), 0);
rdr.consume(5);
assert_eq!(rdr.absolute_byte_offset(), 5);
@@ -615,7 +586,7 @@ and exhibited clearly, with a label attached.\
assert_eq!(rdr.absolute_byte_offset(), 11);
assert!(rdr.fill().unwrap());
assert_eq!(btos(rdr.buffer()), "maggie");
assert_eq!(rdr.bstr(), "maggie");
rdr.consume_all();
assert!(!rdr.fill().unwrap());
@@ -630,7 +601,7 @@ and exhibited clearly, with a label attached.\
let mut rdr = LineBufferReader::new(bytes.as_bytes(), &mut linebuf);
assert!(rdr.fill().unwrap());
assert_eq!(btos(rdr.buffer()), "homer\nlisa\nmaggie\n");
assert_eq!(rdr.bstr(), "homer\nlisa\nmaggie\n");
rdr.consume_all();
assert!(!rdr.fill().unwrap());
@@ -645,7 +616,7 @@ and exhibited clearly, with a label attached.\
let mut rdr = LineBufferReader::new(bytes.as_bytes(), &mut linebuf);
assert!(rdr.fill().unwrap());
assert_eq!(btos(rdr.buffer()), "\n");
assert_eq!(rdr.bstr(), "\n");
rdr.consume_all();
assert!(!rdr.fill().unwrap());
@@ -660,7 +631,7 @@ and exhibited clearly, with a label attached.\
let mut rdr = LineBufferReader::new(bytes.as_bytes(), &mut linebuf);
assert!(rdr.fill().unwrap());
assert_eq!(btos(rdr.buffer()), "\n\n");
assert_eq!(rdr.bstr(), "\n\n");
rdr.consume_all();
assert!(!rdr.fill().unwrap());
@@ -698,12 +669,12 @@ and exhibited clearly, with a label attached.\
let mut linebuf = LineBufferBuilder::new().capacity(1).build();
let mut rdr = LineBufferReader::new(bytes.as_bytes(), &mut linebuf);
let mut got = vec![];
let mut got = BString::new();
while rdr.fill().unwrap() {
got.extend(rdr.buffer());
got.push(rdr.buffer());
rdr.consume_all();
}
assert_eq!(bytes, btos(&got));
assert_eq!(bytes, got);
assert_eq!(rdr.absolute_byte_offset(), bytes.len() as u64);
assert_eq!(rdr.binary_byte_offset(), None);
}
@@ -718,11 +689,11 @@ and exhibited clearly, with a label attached.\
let mut rdr = LineBufferReader::new(bytes.as_bytes(), &mut linebuf);
assert!(rdr.fill().unwrap());
assert_eq!(btos(rdr.buffer()), "homer\n");
assert_eq!(rdr.bstr(), "homer\n");
rdr.consume_all();
assert!(rdr.fill().unwrap());
assert_eq!(btos(rdr.buffer()), "lisa\n");
assert_eq!(rdr.bstr(), "lisa\n");
rdr.consume_all();
// This returns an error because while we have just enough room to
@@ -732,11 +703,11 @@ and exhibited clearly, with a label attached.\
assert!(rdr.fill().is_err());
// We can mush on though!
assert_eq!(btos(rdr.buffer()), "m");
assert_eq!(rdr.bstr(), "m");
rdr.consume_all();
assert!(rdr.fill().unwrap());
assert_eq!(btos(rdr.buffer()), "aggie");
assert_eq!(rdr.bstr(), "aggie");
rdr.consume_all();
assert!(!rdr.fill().unwrap());
@@ -752,16 +723,16 @@ and exhibited clearly, with a label attached.\
let mut rdr = LineBufferReader::new(bytes.as_bytes(), &mut linebuf);
assert!(rdr.fill().unwrap());
assert_eq!(btos(rdr.buffer()), "homer\n");
assert_eq!(rdr.bstr(), "homer\n");
rdr.consume_all();
assert!(rdr.fill().unwrap());
assert_eq!(btos(rdr.buffer()), "lisa\n");
assert_eq!(rdr.bstr(), "lisa\n");
rdr.consume_all();
// We have just enough space.
assert!(rdr.fill().unwrap());
assert_eq!(btos(rdr.buffer()), "maggie");
assert_eq!(rdr.bstr(), "maggie");
rdr.consume_all();
assert!(!rdr.fill().unwrap());
@@ -777,7 +748,7 @@ and exhibited clearly, with a label attached.\
let mut rdr = LineBufferReader::new(bytes.as_bytes(), &mut linebuf);
assert!(rdr.fill().is_err());
assert_eq!(btos(rdr.buffer()), "");
assert_eq!(rdr.bstr(), "");
}
#[test]
@@ -789,7 +760,7 @@ and exhibited clearly, with a label attached.\
assert!(rdr.buffer().is_empty());
assert!(rdr.fill().unwrap());
assert_eq!(btos(rdr.buffer()), "homer\nli\x00sa\nmaggie\n");
assert_eq!(rdr.bstr(), "homer\nli\x00sa\nmaggie\n");
rdr.consume_all();
assert!(!rdr.fill().unwrap());
@@ -808,7 +779,7 @@ and exhibited clearly, with a label attached.\
assert!(rdr.buffer().is_empty());
assert!(rdr.fill().unwrap());
assert_eq!(btos(rdr.buffer()), "homer\nli");
assert_eq!(rdr.bstr(), "homer\nli");
rdr.consume_all();
assert!(!rdr.fill().unwrap());
@@ -825,7 +796,7 @@ and exhibited clearly, with a label attached.\
let mut rdr = LineBufferReader::new(bytes.as_bytes(), &mut linebuf);
assert!(!rdr.fill().unwrap());
assert_eq!(btos(rdr.buffer()), "");
assert_eq!(rdr.bstr(), "");
assert_eq!(rdr.absolute_byte_offset(), 0);
assert_eq!(rdr.binary_byte_offset(), Some(0));
}
@@ -841,7 +812,7 @@ and exhibited clearly, with a label attached.\
assert!(rdr.buffer().is_empty());
assert!(rdr.fill().unwrap());
assert_eq!(btos(rdr.buffer()), "homer\nlisa\nmaggie\n");
assert_eq!(rdr.bstr(), "homer\nlisa\nmaggie\n");
rdr.consume_all();
assert!(!rdr.fill().unwrap());
@@ -860,7 +831,7 @@ and exhibited clearly, with a label attached.\
assert!(rdr.buffer().is_empty());
assert!(rdr.fill().unwrap());
assert_eq!(btos(rdr.buffer()), "homer\nlisa\nmaggie");
assert_eq!(rdr.bstr(), "homer\nlisa\nmaggie");
rdr.consume_all();
assert!(!rdr.fill().unwrap());
@@ -878,7 +849,7 @@ and exhibited clearly, with a label attached.\
assert!(rdr.buffer().is_empty());
assert!(rdr.fill().unwrap());
assert_eq!(btos(rdr.buffer()), "\
assert_eq!(rdr.bstr(), "\
For the Doctor Watsons of this world, as opposed to the Sherlock
Holmeses, s\
");
@@ -901,7 +872,7 @@ Holmeses, s\
assert!(rdr.buffer().is_empty());
assert!(rdr.fill().unwrap());
assert_eq!(btos(rdr.buffer()), "homer\nli\nsa\nmaggie\n");
assert_eq!(rdr.bstr(), "homer\nli\nsa\nmaggie\n");
rdr.consume_all();
assert!(!rdr.fill().unwrap());
@@ -920,7 +891,7 @@ Holmeses, s\
assert!(rdr.buffer().is_empty());
assert!(rdr.fill().unwrap());
assert_eq!(btos(rdr.buffer()), "\nhomer\nlisa\nmaggie\n");
assert_eq!(rdr.bstr(), "\nhomer\nlisa\nmaggie\n");
rdr.consume_all();
assert!(!rdr.fill().unwrap());
@@ -939,7 +910,7 @@ Holmeses, s\
assert!(rdr.buffer().is_empty());
assert!(rdr.fill().unwrap());
assert_eq!(btos(rdr.buffer()), "homer\nlisa\nmaggie\n\n");
assert_eq!(rdr.bstr(), "homer\nlisa\nmaggie\n\n");
rdr.consume_all();
assert!(!rdr.fill().unwrap());
@@ -958,7 +929,7 @@ Holmeses, s\
assert!(rdr.buffer().is_empty());
assert!(rdr.fill().unwrap());
assert_eq!(btos(rdr.buffer()), "homer\nlisa\nmaggie\n\n");
assert_eq!(rdr.bstr(), "homer\nlisa\nmaggie\n\n");
rdr.consume_all();
assert!(!rdr.fill().unwrap());

14
grep-searcher/src/lines.rs
View File

@@ -2,8 +2,8 @@
A collection of routines for performing operations on lines.
*/
use bstr::B;
use bytecount;
use memchr::{memchr, memrchr};
use grep_matcher::{LineTerminator, Match};
/// An iterator over lines in a particular slice of bytes.
@@ -85,7 +85,7 @@ impl LineStep {
#[inline(always)]
fn next_impl(&mut self, mut bytes: &[u8]) -> Option<(usize, usize)> {
bytes = &bytes[..self.end];
match memchr(self.line_term, &bytes[self.pos..]) {
match B(&bytes[self.pos..]).find_byte(self.line_term) {
None => {
if self.pos < bytes.len() {
let m = (self.pos, bytes.len());
@@ -135,14 +135,16 @@ pub fn locate(
line_term: u8,
range: Match,
) -> Match {
let line_start = memrchr(line_term, &bytes[0..range.start()])
let line_start = B(&bytes[..range.start()])
.rfind_byte(line_term)
.map_or(0, |i| i + 1);
let line_end =
if range.end() > line_start && bytes[range.end() - 1] == line_term {
range.end()
} else {
memchr(line_term, &bytes[range.end()..])
.map_or(bytes.len(), |i| range.end() + i + 1)
B(&bytes[range.end()..])
.find_byte(line_term)
.map_or(bytes.len(), |i| range.end() + i + 1)
};
Match::new(line_start, line_end)
}
@@ -180,7 +182,7 @@ fn preceding_by_pos(
pos -= 1;
}
loop {
match memrchr(line_term, &bytes[..pos]) {
match B(&bytes[..pos]).rfind_byte(line_term) {
None => {
return 0;
}

1
grep-searcher/src/macros.rs
View File

@@ -1,3 +1,4 @@
/// Like assert_eq, but nicer output for long strings.
#[cfg(test)]
#[macro_export]
macro_rules! assert_eq_printed {

15
grep-searcher/src/searcher/core.rs
View File

@@ -1,6 +1,6 @@
use std::cmp;
use memchr::memchr;
use bstr::B;
use grep_matcher::{LineMatchKind, Matcher};
use lines::{self, LineStep};
@@ -149,7 +149,7 @@ impl<'s, M: Matcher, S: Sink> Core<'s, M, S> {
BinaryDetection::Quit(b) => b,
_ => return false,
};
if let Some(i) = memchr(binary_byte, &buf[*range]) {
if let Some(i) = B(&buf[*range]).find_byte(binary_byte) {
self.binary_byte_offset = Some(range.start() + i);
true
} else {
@@ -424,16 +424,7 @@ impl<'s, M: Matcher, S: Sink> Core<'s, M, S> {
}
self.count_lines(buf, range.start());
let offset = self.absolute_byte_offset + range.start() as u64;
let linebuf =
if self.config.line_term.is_crlf() {
// Normally, a line terminator is never part of a match, but
// if the line terminator is CRLF, then it's possible for `\r`
// to end up in the match, which we generally don't want. So
// we strip it here.
lines::without_terminator(&buf[*range], self.config.line_term)
} else {
&buf[*range]
};
let linebuf = &buf[*range];
let keepgoing = self.sink.matched(
&self.searcher,
&SinkMatch {

1
grep-searcher/src/searcher/mod.rs
View File

@@ -307,6 +307,7 @@ impl SearcherBuilder {
decode_builder
.encoding(self.config.encoding.as_ref().map(|e| e.0))
.utf8_passthru(true)
.strip_bom(true)
.bom_override(true);
Searcher {
config: config,

5
grep-searcher/src/testutil.rs
View File

@@ -1,10 +1,10 @@
use std::io::{self, Write};
use std::str;
use bstr::B;
use grep_matcher::{
LineMatchKind, LineTerminator, Match, Matcher, NoCaptures, NoError,
};
use memchr::memchr;
use regex::bytes::{Regex, RegexBuilder};
use searcher::{BinaryDetection, Searcher, SearcherBuilder};
@@ -94,7 +94,8 @@ impl Matcher for RegexMatcher {
}
// Make it interesting and return the last byte in the current
// line.
let i = memchr(self.line_term.unwrap().as_byte(), haystack)
let i = B(haystack)
.find_byte(self.line_term.unwrap().as_byte())
.map(|i| i)
.unwrap_or(haystack.len() - 1);
Ok(Some(LineMatchKind::Candidate(i)))

19
grep/Cargo.toml
View File

@@ -1,6 +1,6 @@
[package]
name = "grep"
version = "0.2.2" #:version
version = "0.2.3" #:version
authors = ["Andrew Gallant <jamslam@gmail.com>"]
description = """
Fast line oriented regex searching as a library.
@@ -15,23 +15,18 @@ license = "Unlicense/MIT"
[dependencies]
grep-cli = { version = "0.1.1", path = "../grep-cli" }
grep-matcher = { version = "0.1.1", path = "../grep-matcher" }
grep-pcre2 = { version = "0.1.1", path = "../grep-pcre2", optional = true }
grep-pcre2 = { version = "0.1.2", path = "../grep-pcre2", optional = true }
grep-printer = { version = "0.1.1", path = "../grep-printer" }
grep-regex = { version = "0.1.1", path = "../grep-regex" }
grep-searcher = { version = "0.1.1", path = "../grep-searcher" }
[dev-dependencies]
atty = "0.2.11"
regex = "1"
termcolor = "1"
walkdir = "2.2.2"
[dev-dependencies.clap]
version = "2.32.0"
default-features = false
features = ["suggestions"]
termcolor = "1.0.4"
walkdir = "2.2.7"
[features]
avx-accel = ["grep-searcher/avx-accel"]
simd-accel = ["grep-searcher/simd-accel"]
pcre2 = ["grep-pcre2"]
# This feature is DEPRECATED. Runtime dispatch is used for SIMD now.
avx-accel = []

18
ignore/Cargo.toml
View File

@@ -1,6 +1,6 @@
[package]
name = "ignore"
version = "0.4.4" #:version
version = "0.4.6" #:version
authors = ["Andrew Gallant <jamslam@gmail.com>"]
description = """
A fast library for efficiently matching ignore files such as `.gitignore`
@@ -18,21 +18,21 @@ name = "ignore"
bench = false
[dependencies]
crossbeam-channel = "0.2.4"
crossbeam-channel = "0.3.6"
globset = { version = "0.4.2", path = "../globset" }
lazy_static = "1.1.0"
lazy_static = "1.1"
log = "0.4.5"
memchr = "2.0.2"
regex = "1.0.5"
same-file = "1.0.3"
memchr = "2.1"
regex = "1.1"
same-file = "1.0.4"
thread_local = "0.3.6"
walkdir = "2.2.5"
walkdir = "2.2.7"
[target.'cfg(windows)'.dependencies.winapi-util]
version = "0.1.1"
version = "0.1.2"
[dev-dependencies]
tempdir = "0.3.7"
tempfile = "3.0.5"
[features]
simd-accel = ["globset/simd-accel"]

6
ignore/examples/walk.rs
View File

@@ -37,19 +37,19 @@ fn main() {
Box::new(move |result| {
use ignore::WalkState::*;
tx.send(DirEntry::Y(result.unwrap()));
tx.send(DirEntry::Y(result.unwrap())).unwrap();
Continue
})
});
} else if simple {
let walker = WalkDir::new(path);
for result in walker {
tx.send(DirEntry::X(result.unwrap()));
tx.send(DirEntry::X(result.unwrap())).unwrap();
}
} else {
let walker = WalkBuilder::new(path).build();
for result in walker {
tx.send(DirEntry::Y(result.unwrap()));
tx.send(DirEntry::Y(result.unwrap())).unwrap();
}
}
drop(tx);

119
ignore/src/dir.rs
View File

@@ -22,6 +22,7 @@ use gitignore::{self, Gitignore, GitignoreBuilder};
use pathutil::{is_hidden, strip_prefix};
use overrides::{self, Override};
use types::{self, Types};
use walk::DirEntry;
use {Error, Match, PartialErrorBuilder};
/// IgnoreMatch represents information about where a match came from when using
@@ -73,6 +74,8 @@ struct IgnoreOptions {
git_ignore: bool,
/// Whether to read .git/info/exclude files.
git_exclude: bool,
/// Whether to ignore files case insensitively
ignore_case_insensitive: bool,
}
/// Ignore is a matcher useful for recursively walking one or more directories.
@@ -225,7 +228,11 @@ impl Ignore {
Gitignore::empty()
} else {
let (m, err) =
create_gitignore(&dir, &self.0.custom_ignore_filenames);
create_gitignore(
&dir,
&self.0.custom_ignore_filenames,
self.0.opts.ignore_case_insensitive,
);
errs.maybe_push(err);
m
};
@@ -233,7 +240,12 @@ impl Ignore {
if !self.0.opts.ignore {
Gitignore::empty()
} else {
let (m, err) = create_gitignore(&dir, &[".ignore"]);
let (m, err) =
create_gitignore(
&dir,
&[".ignore"],
self.0.opts.ignore_case_insensitive,
);
errs.maybe_push(err);
m
};
@@ -241,7 +253,12 @@ impl Ignore {
if !self.0.opts.git_ignore {
Gitignore::empty()
} else {
let (m, err) = create_gitignore(&dir, &[".gitignore"]);
let (m, err) =
create_gitignore(
&dir,
&[".gitignore"],
self.0.opts.ignore_case_insensitive,
);
errs.maybe_push(err);
m
};
@@ -249,7 +266,12 @@ impl Ignore {
if !self.0.opts.git_exclude {
Gitignore::empty()
} else {
let (m, err) = create_gitignore(&dir, &[".git/info/exclude"]);
let (m, err) =
create_gitignore(
&dir,
&[".git/info/exclude"],
self.0.opts.ignore_case_insensitive,
);
errs.maybe_push(err);
m
};
@@ -285,11 +307,23 @@ impl Ignore {
|| has_explicit_ignores
}
/// Like `matched`, but works with a directory entry instead.
pub fn matched_dir_entry<'a>(
&'a self,
dent: &DirEntry,
) -> Match<IgnoreMatch<'a>> {
let m = self.matched(dent.path(), dent.is_dir());
if m.is_none() && self.0.opts.hidden && is_hidden(dent) {
return Match::Ignore(IgnoreMatch::hidden());
}
m
}
/// Returns a match indicating whether the given file path should be
/// ignored or not.
///
/// The match contains information about its origin.
pub fn matched<'a, P: AsRef<Path>>(
fn matched<'a, P: AsRef<Path>>(
&'a self,
path: P,
is_dir: bool,
@@ -330,9 +364,6 @@ impl Ignore {
whitelisted = mat;
}
}
if whitelisted.is_none() && self.0.opts.hidden && is_hidden(path) {
return Match::Ignore(IgnoreMatch::hidden());
}
whitelisted
}
@@ -483,6 +514,7 @@ impl IgnoreBuilder {
git_global: true,
git_ignore: true,
git_exclude: true,
ignore_case_insensitive: false,
},
}
}
@@ -496,7 +528,11 @@ impl IgnoreBuilder {
if !self.opts.git_global {
Gitignore::empty()
} else {
let (gi, err) = Gitignore::global();
let mut builder = GitignoreBuilder::new("");
builder
.case_insensitive(self.opts.ignore_case_insensitive)
.unwrap();
let (gi, err) = builder.build_global();
if let Some(err) = err {
debug!("{}", err);
}
@@ -627,6 +663,17 @@ impl IgnoreBuilder {
self.opts.git_exclude = yes;
self
}
/// Process ignore files case insensitively
///
/// This is disabled by default.
pub fn ignore_case_insensitive(
&mut self,
yes: bool,
) -> &mut IgnoreBuilder {
self.opts.ignore_case_insensitive = yes;
self
}
}
/// Creates a new gitignore matcher for the directory given.
@@ -638,9 +685,11 @@ impl IgnoreBuilder {
pub fn create_gitignore<T: AsRef<OsStr>>(
dir: &Path,
names: &[T],
case_insensitive: bool,
) -> (Gitignore, Option<Error>) {
let mut builder = GitignoreBuilder::new(dir);
let mut errs = PartialErrorBuilder::default();
builder.case_insensitive(case_insensitive).unwrap();
for name in names {
let gipath = dir.join(name.as_ref());
errs.maybe_push_ignore_io(builder.add(gipath));
@@ -661,7 +710,7 @@ mod tests {
use std::io::Write;
use std::path::Path;
use tempdir::TempDir;
use tempfile::{self, TempDir};
use dir::IgnoreBuilder;
use gitignore::Gitignore;
@@ -683,9 +732,13 @@ mod tests {
}
}
fn tmpdir(prefix: &str) -> TempDir {
tempfile::Builder::new().prefix(prefix).tempdir().unwrap()
}
#[test]
fn explicit_ignore() {
let td = TempDir::new("ignore-test-").unwrap();
let td = tmpdir("ignore-test-");
wfile(td.path().join("not-an-ignore"), "foo\n!bar");
let (gi, err) = Gitignore::new(td.path().join("not-an-ignore"));
@@ -700,7 +753,7 @@ mod tests {
#[test]
fn git_exclude() {
let td = TempDir::new("ignore-test-").unwrap();
let td = tmpdir("ignore-test-");
mkdirp(td.path().join(".git/info"));
wfile(td.path().join(".git/info/exclude"), "foo\n!bar");
@@ -713,7 +766,7 @@ mod tests {
#[test]
fn gitignore() {
let td = TempDir::new("ignore-test-").unwrap();
let td = tmpdir("ignore-test-");
mkdirp(td.path().join(".git"));
wfile(td.path().join(".gitignore"), "foo\n!bar");
@@ -726,7 +779,7 @@ mod tests {
#[test]
fn gitignore_no_git() {
let td = TempDir::new("ignore-test-").unwrap();
let td = tmpdir("ignore-test-");
wfile(td.path().join(".gitignore"), "foo\n!bar");
let (ig, err) = IgnoreBuilder::new().build().add_child(td.path());
@@ -738,7 +791,7 @@ mod tests {
#[test]
fn ignore() {
let td = TempDir::new("ignore-test-").unwrap();
let td = tmpdir("ignore-test-");
wfile(td.path().join(".ignore"), "foo\n!bar");
let (ig, err) = IgnoreBuilder::new().build().add_child(td.path());
@@ -750,7 +803,7 @@ mod tests {
#[test]
fn custom_ignore() {
let td = TempDir::new("ignore-test-").unwrap();
let td = tmpdir("ignore-test-");
let custom_ignore = ".customignore";
wfile(td.path().join(custom_ignore), "foo\n!bar");
@@ -766,7 +819,7 @@ mod tests {
// Tests that a custom ignore file will override an .ignore.
#[test]
fn custom_ignore_over_ignore() {
let td = TempDir::new("ignore-test-").unwrap();
let td = tmpdir("ignore-test-");
let custom_ignore = ".customignore";
wfile(td.path().join(".ignore"), "foo");
wfile(td.path().join(custom_ignore), "!foo");
@@ -781,7 +834,7 @@ mod tests {
// Tests that earlier custom ignore files have lower precedence than later.
#[test]
fn custom_ignore_precedence() {
let td = TempDir::new("ignore-test-").unwrap();
let td = tmpdir("ignore-test-");
let custom_ignore1 = ".customignore1";
let custom_ignore2 = ".customignore2";
wfile(td.path().join(custom_ignore1), "foo");
@@ -798,7 +851,7 @@ mod tests {
// Tests that an .ignore will override a .gitignore.
#[test]
fn ignore_over_gitignore() {
let td = TempDir::new("ignore-test-").unwrap();
let td = tmpdir("ignore-test-");
wfile(td.path().join(".gitignore"), "foo");
wfile(td.path().join(".ignore"), "!foo");
@@ -810,7 +863,7 @@ mod tests {
// Tests that exclude has lower precedent than both .ignore and .gitignore.
#[test]
fn exclude_lowest() {
let td = TempDir::new("ignore-test-").unwrap();
let td = tmpdir("ignore-test-");
wfile(td.path().join(".gitignore"), "!foo");
wfile(td.path().join(".ignore"), "!bar");
mkdirp(td.path().join(".git/info"));
@@ -825,8 +878,8 @@ mod tests {
#[test]
fn errored() {
let td = TempDir::new("ignore-test-").unwrap();
wfile(td.path().join(".gitignore"), "f**oo");
let td = tmpdir("ignore-test-");
wfile(td.path().join(".gitignore"), "{foo");
let (_, err) = IgnoreBuilder::new().build().add_child(td.path());
assert!(err.is_some());
@@ -834,9 +887,9 @@ mod tests {
#[test]
fn errored_both() {
let td = TempDir::new("ignore-test-").unwrap();
wfile(td.path().join(".gitignore"), "f**oo");
wfile(td.path().join(".ignore"), "fo**o");
let td = tmpdir("ignore-test-");
wfile(td.path().join(".gitignore"), "{foo");
wfile(td.path().join(".ignore"), "{bar");
let (_, err) = IgnoreBuilder::new().build().add_child(td.path());
assert_eq!(2, partial(err.expect("an error")).len());
@@ -844,9 +897,9 @@ mod tests {
#[test]
fn errored_partial() {
let td = TempDir::new("ignore-test-").unwrap();
let td = tmpdir("ignore-test-");
mkdirp(td.path().join(".git"));
wfile(td.path().join(".gitignore"), "f**oo\nbar");
wfile(td.path().join(".gitignore"), "{foo\nbar");
let (ig, err) = IgnoreBuilder::new().build().add_child(td.path());
assert!(err.is_some());
@@ -855,8 +908,8 @@ mod tests {
#[test]
fn errored_partial_and_ignore() {
let td = TempDir::new("ignore-test-").unwrap();
wfile(td.path().join(".gitignore"), "f**oo\nbar");
let td = tmpdir("ignore-test-");
wfile(td.path().join(".gitignore"), "{foo\nbar");
wfile(td.path().join(".ignore"), "!bar");
let (ig, err) = IgnoreBuilder::new().build().add_child(td.path());
@@ -866,7 +919,7 @@ mod tests {
#[test]
fn not_present_empty() {
let td = TempDir::new("ignore-test-").unwrap();
let td = tmpdir("ignore-test-");
let (_, err) = IgnoreBuilder::new().build().add_child(td.path());
assert!(err.is_none());
@@ -876,7 +929,7 @@ mod tests {
fn stops_at_git_dir() {
// This tests that .gitignore files beyond a .git barrier aren't
// matched, but .ignore files are.
let td = TempDir::new("ignore-test-").unwrap();
let td = tmpdir("ignore-test-");
mkdirp(td.path().join(".git"));
mkdirp(td.path().join("foo/.git"));
wfile(td.path().join(".gitignore"), "foo");
@@ -897,7 +950,7 @@ mod tests {
#[test]
fn absolute_parent() {
let td = TempDir::new("ignore-test-").unwrap();
let td = tmpdir("ignore-test-");
mkdirp(td.path().join(".git"));
mkdirp(td.path().join("foo"));
wfile(td.path().join(".gitignore"), "bar");
@@ -920,7 +973,7 @@ mod tests {
#[test]
fn absolute_parent_anchored() {
let td = TempDir::new("ignore-test-").unwrap();
let td = tmpdir("ignore-test-");
mkdirp(td.path().join(".git"));
mkdirp(td.path().join("src/llvm"));
wfile(td.path().join(".gitignore"), "/llvm/\nfoo");

75
ignore/src/gitignore.rs
View File

@@ -69,8 +69,7 @@ impl Glob {
/// Returns true if and only if this glob has a `**/` prefix.
fn has_doublestar_prefix(&self) -> bool {
self.actual.starts_with("**/")
|| (self.actual == "**" && self.is_only_dir)
self.actual.starts_with("**/") || self.actual == "**"
}
}
@@ -127,16 +126,7 @@ impl Gitignore {
/// `$XDG_CONFIG_HOME/git/ignore` is read. If `$XDG_CONFIG_HOME` is not
/// set or is empty, then `$HOME/.config/git/ignore` is used instead.
pub fn global() -> (Gitignore, Option<Error>) {
match gitconfig_excludes_path() {
None => (Gitignore::empty(), None),
Some(path) => {
if !path.is_file() {
(Gitignore::empty(), None)
} else {
Gitignore::new(path)
}
}
}
GitignoreBuilder::new("").build_global()
}
/// Creates a new empty gitignore matcher that never matches anything.
@@ -359,6 +349,36 @@ impl GitignoreBuilder {
})
}
/// Build a global gitignore matcher using the configuration in this
/// builder.
///
/// This consumes ownership of the builder unlike `build` because it
/// must mutate the builder to add the global gitignore globs.
///
/// Note that this ignores the path given to this builder's constructor
/// and instead derives the path automatically from git's global
/// configuration.
pub fn build_global(mut self) -> (Gitignore, Option<Error>) {
match gitconfig_excludes_path() {
None => (Gitignore::empty(), None),
Some(path) => {
if !path.is_file() {
(Gitignore::empty(), None)
} else {
let mut errs = PartialErrorBuilder::default();
errs.maybe_push_ignore_io(self.add(path));
match self.build() {
Ok(gi) => (gi, errs.into_error_option()),
Err(err) => {
errs.push(err);
(Gitignore::empty(), errs.into_error_option())
}
}
}
}
}
}
/// Add each glob from the file path given.
///
/// The file given should be formatted as a `gitignore` file.
@@ -419,6 +439,8 @@ impl GitignoreBuilder {
from: Option<PathBuf>,
mut line: &str,
) -> Result<&mut GitignoreBuilder, Error> {
#![allow(deprecated)]
if line.starts_with("#") {
return Ok(self);
}
@@ -435,7 +457,6 @@ impl GitignoreBuilder {
is_whitelist: false,
is_only_dir: false,
};
let mut literal_separator = false;
let mut is_absolute = false;
if line.starts_with("\\!") || line.starts_with("\\#") {
line = &line[1..];
@@ -450,7 +471,6 @@ impl GitignoreBuilder {
// then the glob can only match the beginning of a path
// (relative to the location of gitignore). We achieve this by
// simply banning wildcards from matching /.
literal_separator = true;
line = &line[1..];
is_absolute = true;
}
@@ -463,16 +483,11 @@ impl GitignoreBuilder {
line = &line[..i];
}
}
// If there is a literal slash, then we note that so that globbing
// doesn't let wildcards match slashes.
glob.actual = line.to_string();
if is_absolute || line.chars().any(|c| c == '/') {
literal_separator = true;
}
// If there was a slash, then this is a glob that must match the entire
// path name. Otherwise, we should let it match anywhere, so use a **/
// prefix.
if !literal_separator {
// If there is a literal slash, then this is a glob that must match the
// entire path name. Otherwise, we should let it match anywhere, so use
// a **/ prefix.
if !is_absolute && !line.chars().any(|c| c == '/') {
// ... but only if we don't already have a **/ prefix.
if !glob.has_doublestar_prefix() {
glob.actual = format!("**/{}", glob.actual);
@@ -486,7 +501,7 @@ impl GitignoreBuilder {
}
let parsed =
GlobBuilder::new(&glob.actual)
.literal_separator(literal_separator)
.literal_separator(true)
.case_insensitive(self.case_insensitive)
.backslash_escape(true)
.build()
@@ -503,12 +518,16 @@ impl GitignoreBuilder {
/// Toggle whether the globs should be matched case insensitively or not.
///
/// When this option is changed, only globs added after the change will be affected.
/// When this option is changed, only globs added after the change will be
/// affected.
///
/// This is disabled by default.
pub fn case_insensitive(
&mut self, yes: bool
&mut self,
yes: bool,
) -> Result<&mut GitignoreBuilder, Error> {
// TODO: This should not return a `Result`. Fix this in the next semver
// release.
self.case_insensitive = yes;
Ok(self)
}
@@ -689,6 +708,9 @@ mod tests {
ignored!(ig39, ROOT, "\\?", "?");
ignored!(ig40, ROOT, "\\*", "*");
ignored!(ig41, ROOT, "\\a", "a");
ignored!(ig42, ROOT, "s*.rs", "sfoo.rs");
ignored!(ig43, ROOT, "**", "foo.rs");
ignored!(ig44, ROOT, "**/**/*", "a/foo.rs");
not_ignored!(ignot1, ROOT, "amonths", "months");
not_ignored!(ignot2, ROOT, "monthsa", "months");
@@ -710,6 +732,7 @@ mod tests {
not_ignored!(ignot16, ROOT, "*\n!**/", "foo", true);
not_ignored!(ignot17, ROOT, "src/*.rs", "src/grep/src/main.rs");
not_ignored!(ignot18, ROOT, "path1/*", "path2/path1/foo");
not_ignored!(ignot19, ROOT, "s*.rs", "src/foo.rs");
fn bytes(s: &str) -> Vec<u8> {
s.to_string().into_bytes()

2
ignore/src/lib.rs
View File

@@ -56,7 +56,7 @@ extern crate memchr;
extern crate regex;
extern crate same_file;
#[cfg(test)]
extern crate tempdir;
extern crate tempfile;
extern crate thread_local;
extern crate walkdir;
#[cfg(windows)]

7
ignore/src/overrides.rs
View File

@@ -139,13 +139,16 @@ impl OverrideBuilder {
}
/// Toggle whether the globs should be matched case insensitively or not.
///
///
/// When this option is changed, only globs added after the change will be affected.
///
/// This is disabled by default.
pub fn case_insensitive(
&mut self, yes: bool
&mut self,
yes: bool,
) -> Result<&mut OverrideBuilder, Error> {
// TODO: This should not return a `Result`. Fix this in the next semver
// release.
self.builder.case_insensitive(yes)?;
Ok(self)
}

48
ignore/src/pathutil.rs
View File

@@ -1,22 +1,56 @@
use std::ffi::OsStr;
use std::path::Path;
/// Returns true if and only if this file path is considered to be hidden.
use walk::DirEntry;
/// Returns true if and only if this entry is considered to be hidden.
///
/// This only returns true if the base name of the path starts with a `.`.
///
/// On Unix, this implements a more optimized check.
#[cfg(unix)]
pub fn is_hidden<P: AsRef<Path>>(path: P) -> bool {
pub fn is_hidden(dent: &DirEntry) -> bool {
use std::os::unix::ffi::OsStrExt;
if let Some(name) = file_name(path.as_ref()) {
if let Some(name) = file_name(dent.path()) {
name.as_bytes().get(0) == Some(&b'.')
} else {
false
}
}
/// Returns true if and only if this file path is considered to be hidden.
#[cfg(not(unix))]
pub fn is_hidden<P: AsRef<Path>>(path: P) -> bool {
if let Some(name) = file_name(path.as_ref()) {
/// Returns true if and only if this entry is considered to be hidden.
///
/// On Windows, this returns true if one of the following is true:
///
/// * The base name of the path starts with a `.`.
/// * The file attributes have the `HIDDEN` property set.
#[cfg(windows)]
pub fn is_hidden(dent: &DirEntry) -> bool {
use std::os::windows::fs::MetadataExt;
use winapi_util::file;
// This looks like we're doing an extra stat call, but on Windows, the
// directory traverser reuses the metadata retrieved from each directory
// entry and stores it on the DirEntry itself. So this is "free."
if let Ok(md) = dent.metadata() {
if file::is_hidden(md.file_attributes() as u64) {
return true;
}
}
if let Some(name) = file_name(dent.path()) {
name.to_str().map(|s| s.starts_with(".")).unwrap_or(false)
} else {
false
}
}
/// Returns true if and only if this entry is considered to be hidden.
///
/// This only returns true if the base name of the path starts with a `.`.
#[cfg(not(any(unix, windows)))]
pub fn is_hidden(dent: &DirEntry) -> bool {
if let Some(name) = file_name(dent.path()) {
name.to_str().map(|s| s.starts_with(".")).unwrap_or(false)
} else {
false

32
ignore/src/types.rs
View File

@@ -103,11 +103,14 @@ const DEFAULT_TYPES: &'static [(&'static str, &'static [&'static str])] = &[
("amake", &["*.mk", "*.bp"]),
("asciidoc", &["*.adoc", "*.asc", "*.asciidoc"]),
("asm", &["*.asm", "*.s", "*.S"]),
("asp", &["*.aspx", "*.aspx.cs", "*.aspx.cs", "*.ascx", "*.ascx.cs", "*.ascx.vb"]),
("avro", &["*.avdl", "*.avpr", "*.avsc"]),
("awk", &["*.awk"]),
("bazel", &["*.bzl", "WORKSPACE", "BUILD"]),
("bazel", &["*.bzl", "WORKSPACE", "BUILD", "BUILD.bazel"]),
("bitbake", &["*.bb", "*.bbappend", "*.bbclass", "*.conf", "*.inc"]),
("bzip2", &["*.bz2"]),
("brotli", &["*.br"]),
("buildstream", &["*.bst"]),
("bzip2", &["*.bz2", "*.tbz2"]),
("c", &["*.c", "*.h", "*.H", "*.cats"]),
("cabal", &["*.cabal"]),
("cbor", &["*.cbor"]),
@@ -127,7 +130,7 @@ const DEFAULT_TYPES: &'static [(&'static str, &'static [&'static str])] = &[
("cshtml", &["*.cshtml"]),
("css", &["*.css", "*.scss"]),
("csv", &["*.csv"]),
("cython", &["*.pyx"]),
("cython", &["*.pyx", "*.pxi", "*.pxd"]),
("dart", &["*.dart"]),
("d", &["*.d"]),
("dhall", &["*.dhall"]),
@@ -145,7 +148,7 @@ const DEFAULT_TYPES: &'static [(&'static str, &'static [&'static str])] = &[
("fsharp", &["*.fs", "*.fsx", "*.fsi"]),
("gn", &["*.gn", "*.gni"]),
("go", &["*.go"]),
("gzip", &["*.gz"]),
("gzip", &["*.gz", "*.tgz"]),
("groovy", &["*.groovy", "*.gradle"]),
("h", &["*.h", "*.hpp"]),
("hbs", &["*.hbs"]),
@@ -224,12 +227,14 @@ const DEFAULT_TYPES: &'static [(&'static str, &'static [&'static str])] = &[
("pdf", &["*.pdf"]),
("php", &["*.php", "*.php3", "*.php4", "*.php5", "*.phtml"]),
("pod", &["*.pod"]),
("postscript", &[".eps", ".ps"]),
("protobuf", &["*.proto"]),
("ps", &["*.cdxml", "*.ps1", "*.ps1xml", "*.psd1", "*.psm1"]),
("puppet", &["*.erb", "*.pp", "*.rb"]),
("purs", &["*.purs"]),
("py", &["*.py"]),
("qmake", &["*.pro", "*.pri", "*.prf"]),
("qml", &["*.qml"]),
("readme", &["README*", "*README"]),
("r", &["*.R", "*.r", "*.Rmd", "*.Rnw"]),
("rdoc", &["*.rdoc"]),
@@ -278,8 +283,9 @@ const DEFAULT_TYPES: &'static [(&'static str, &'static [&'static str])] = &[
]),
("taskpaper", &["*.taskpaper"]),
("tcl", &["*.tcl"]),
("tex", &["*.tex", "*.ltx", "*.cls", "*.sty", "*.bib"]),
("tex", &["*.tex", "*.ltx", "*.cls", "*.sty", "*.bib", "*.dtx", "*.ins"]),
("textile", &["*.textile"]),
("thrift", &["*.thrift"]),
("tf", &["*.tf"]),
("ts", &["*.ts", "*.tsx"]),
("txt", &["*.txt"]),
@@ -294,9 +300,10 @@ const DEFAULT_TYPES: &'static [(&'static str, &'static [&'static str])] = &[
("wiki", &["*.mediawiki", "*.wiki"]),
("webidl", &["*.idl", "*.webidl", "*.widl"]),
("xml", &["*.xml", "*.xml.dist"]),
("xz", &["*.xz"]),
("xz", &["*.xz", "*.txz"]),
("yacc", &["*.y"]),
("yaml", &["*.yaml", "*.yml"]),
("zig", &["*.zig"]),
("zsh", &[
".zshenv", "zshenv",
".zlogin", "zlogin",
@@ -305,6 +312,7 @@ const DEFAULT_TYPES: &'static [(&'static str, &'static [&'static str])] = &[
".zshrc", "zshrc",
"*.zsh",
]),
("zstd", &["*.zst", "*.zstd"]),
];
/// Glob represents a single glob in a set of file type definitions.
@@ -343,6 +351,18 @@ impl<'a> Glob<'a> {
fn unmatched() -> Glob<'a> {
Glob(GlobInner::UnmatchedIgnore)
}
/// Return the file type defintion that matched, if one exists. A file type
/// definition always exists when a specific definition matches a file
/// path.
pub fn file_type_def(&self) -> Option<&FileTypeDef> {
match self {
Glob(GlobInner::UnmatchedIgnore) => None,
Glob(GlobInner::Matched { def, .. }) => {
Some(def)
},
}
}
}
/// A single file type definition.

94
ignore/src/walk.rs
View File

@@ -99,7 +99,7 @@ impl DirEntry {
}
/// Returns true if and only if this entry points to a directory.
fn is_dir(&self) -> bool {
pub(crate) fn is_dir(&self) -> bool {
self.dent.is_dir()
}
@@ -764,6 +764,14 @@ impl WalkBuilder {
self
}
/// Process ignore files case insensitively
///
/// This is disabled by default.
pub fn ignore_case_insensitive(&mut self, yes: bool) -> &mut WalkBuilder {
self.ig_builder.ignore_case_insensitive(yes);
self
}
/// Set a function for sorting directory entries by their path.
///
/// If a compare function is set, the resulting iterator will return all
@@ -875,16 +883,17 @@ impl Walk {
return Ok(true);
}
}
let is_dir = ent.file_type().map_or(false, |ft| ft.is_dir());
let max_size = self.max_filesize;
let should_skip_path = skip_path(&self.ig, ent.path(), is_dir);
let should_skip_filesize = if !is_dir && max_size.is_some() {
skip_filesize(max_size.unwrap(), ent.path(), &ent.metadata().ok())
} else {
false
};
Ok(should_skip_path || should_skip_filesize)
if should_skip_entry(&self.ig, ent) {
return Ok(true);
}
if self.max_filesize.is_some() && !ent.is_dir() {
return Ok(skip_filesize(
self.max_filesize.unwrap(),
ent.path(),
&ent.metadata().ok(),
));
}
Ok(false)
}
}
@@ -1118,7 +1127,7 @@ impl WalkParallel {
dent: dent,
ignore: self.ig_root.clone(),
root_device: root_device,
}));
})).unwrap();
any_work = true;
}
// ... but there's no need to start workers if we don't need them.
@@ -1412,13 +1421,11 @@ impl Worker {
return WalkState::Continue;
}
}
let is_dir = dent.is_dir();
let max_size = self.max_filesize;
let should_skip_path = skip_path(ig, dent.path(), is_dir);
let should_skip_path = should_skip_entry(ig, &dent);
let should_skip_filesize =
if !is_dir && max_size.is_some() {
if self.max_filesize.is_some() && !dent.is_dir() {
skip_filesize(
max_size.unwrap(),
self.max_filesize.unwrap(),
dent.path(),
&dent.metadata().ok(),
)
@@ -1431,7 +1438,7 @@ impl Worker {
dent: dent,
ignore: ig.clone(),
root_device: root_device,
}));
})).unwrap();
}
WalkState::Continue
}
@@ -1446,12 +1453,12 @@ impl Worker {
return None;
}
match self.rx.try_recv() {
Some(Message::Work(work)) => {
Ok(Message::Work(work)) => {
self.waiting(false);
self.quitting(false);
return Some(work);
}
Some(Message::Quit) => {
Ok(Message::Quit) => {
// We can't just quit because a Message::Quit could be
// spurious. For example, it's possible to observe that
// all workers are waiting even if there's more work to
@@ -1482,12 +1489,12 @@ impl Worker {
// Otherwise, spin.
}
}
None => {
Err(_) => {
self.waiting(true);
self.quitting(false);
if self.num_waiting() == self.threads {
for _ in 0..self.threads {
self.tx.send(Message::Quit);
self.tx.send(Message::Quit).unwrap();
}
} else {
// You're right to consider this suspicious, but it's
@@ -1601,17 +1608,16 @@ fn skip_filesize(
}
}
fn skip_path(
fn should_skip_entry(
ig: &Ignore,
path: &Path,
is_dir: bool,
dent: &DirEntry,
) -> bool {
let m = ig.matched(path, is_dir);
let m = ig.matched_dir_entry(dent);
if m.is_ignore() {
debug!("ignoring {}: {:?}", path.display(), m);
debug!("ignoring {}: {:?}", dent.path().display(), m);
true
} else if m.is_whitelist() {
debug!("whitelisting {}: {:?}", path.display(), m);
debug!("whitelisting {}: {:?}", dent.path().display(), m);
false
} else {
false
@@ -1702,7 +1708,7 @@ mod tests {
use std::path::Path;
use std::sync::{Arc, Mutex};
use tempdir::TempDir;
use tempfile::{self, TempDir};
use super::{DirEntry, WalkBuilder, WalkState};
@@ -1789,6 +1795,10 @@ mod tests {
paths
}
fn tmpdir(prefix: &str) -> TempDir {
tempfile::Builder::new().prefix(prefix).tempdir().unwrap()
}
fn assert_paths(
prefix: &Path,
builder: &WalkBuilder,
@@ -1802,7 +1812,7 @@ mod tests {
#[test]
fn no_ignores() {
let td = TempDir::new("walk-test-").unwrap();
let td = tmpdir("walk-test-");
mkdirp(td.path().join("a/b/c"));
mkdirp(td.path().join("x/y"));
wfile(td.path().join("a/b/foo"), "");
@@ -1815,7 +1825,7 @@ mod tests {
#[test]
fn custom_ignore() {
let td = TempDir::new("walk-test-").unwrap();
let td = tmpdir("walk-test-");
let custom_ignore = ".customignore";
mkdirp(td.path().join("a"));
wfile(td.path().join(custom_ignore), "foo");
@@ -1831,7 +1841,7 @@ mod tests {
#[test]
fn custom_ignore_exclusive_use() {
let td = TempDir::new("walk-test-").unwrap();
let td = tmpdir("walk-test-");
let custom_ignore = ".customignore";
mkdirp(td.path().join("a"));
wfile(td.path().join(custom_ignore), "foo");
@@ -1851,7 +1861,7 @@ mod tests {
#[test]
fn gitignore() {
let td = TempDir::new("walk-test-").unwrap();
let td = tmpdir("walk-test-");
mkdirp(td.path().join(".git"));
mkdirp(td.path().join("a"));
wfile(td.path().join(".gitignore"), "foo");
@@ -1867,7 +1877,7 @@ mod tests {
#[test]
fn explicit_ignore() {
let td = TempDir::new("walk-test-").unwrap();
let td = tmpdir("walk-test-");
let igpath = td.path().join(".not-an-ignore");
mkdirp(td.path().join("a"));
wfile(&igpath, "foo");
@@ -1883,7 +1893,7 @@ mod tests {
#[test]
fn explicit_ignore_exclusive_use() {
let td = TempDir::new("walk-test-").unwrap();
let td = tmpdir("walk-test-");
let igpath = td.path().join(".not-an-ignore");
mkdirp(td.path().join("a"));
wfile(&igpath, "foo");
@@ -1901,7 +1911,7 @@ mod tests {
#[test]
fn gitignore_parent() {
let td = TempDir::new("walk-test-").unwrap();
let td = tmpdir("walk-test-");
mkdirp(td.path().join(".git"));
mkdirp(td.path().join("a"));
wfile(td.path().join(".gitignore"), "foo");
@@ -1914,7 +1924,7 @@ mod tests {
#[test]
fn max_depth() {
let td = TempDir::new("walk-test-").unwrap();
let td = tmpdir("walk-test-");
mkdirp(td.path().join("a/b/c"));
wfile(td.path().join("foo"), "");
wfile(td.path().join("a/foo"), "");
@@ -1934,7 +1944,7 @@ mod tests {
#[test]
fn max_filesize() {
let td = TempDir::new("walk-test-").unwrap();
let td = tmpdir("walk-test-");
mkdirp(td.path().join("a/b"));
wfile_size(td.path().join("foo"), 0);
wfile_size(td.path().join("bar"), 400);
@@ -1961,7 +1971,7 @@ mod tests {
#[cfg(unix)] // because symlinks on windows are weird
#[test]
fn symlinks() {
let td = TempDir::new("walk-test-").unwrap();
let td = tmpdir("walk-test-");
mkdirp(td.path().join("a/b"));
symlink(td.path().join("a/b"), td.path().join("z"));
wfile(td.path().join("a/b/foo"), "");
@@ -1978,7 +1988,7 @@ mod tests {
#[cfg(unix)] // because symlinks on windows are weird
#[test]
fn first_path_not_symlink() {
let td = TempDir::new("walk-test-").unwrap();
let td = tmpdir("walk-test-");
mkdirp(td.path().join("foo"));
let dents = WalkBuilder::new(td.path().join("foo"))
@@ -1999,7 +2009,7 @@ mod tests {
#[cfg(unix)] // because symlinks on windows are weird
#[test]
fn symlink_loop() {
let td = TempDir::new("walk-test-").unwrap();
let td = tmpdir("walk-test-");
mkdirp(td.path().join("a/b"));
symlink(td.path().join("a"), td.path().join("a/b/c"));
@@ -2029,7 +2039,7 @@ mod tests {
// If our test directory actually isn't a different volume from /sys,
// then this test is meaningless and we shouldn't run it.
let td = TempDir::new("walk-test-").unwrap();
let td = tmpdir("walk-test-");
if device_num(td.path()).unwrap() == device_num("/sys").unwrap() {
return;
}

6
pkg/brew/ripgrep-bin.rb
View File

@@ -1,14 +1,14 @@
class RipgrepBin < Formula
version '0.9.0'
version '0.10.0'
desc "Recursively search directories for a regex pattern."
homepage "https://github.com/BurntSushi/ripgrep"
if OS.mac?
url "https://github.com/BurntSushi/ripgrep/releases/download/#{version}/ripgrep-#{version}-x86_64-apple-darwin.tar.gz"
sha256 "36003ea8b62ad6274dc14140039f448cdf5026827d53cf24dad2d84005557a8c"
sha256 "32754b4173ac87a7bfffd436d601a49362676eb1841ab33440f2f49c002c8967"
elsif OS.linux?
url "https://github.com/BurntSushi/ripgrep/releases/download/#{version}/ripgrep-#{version}-x86_64-unknown-linux-musl.tar.gz"
sha256 "2eb4443e58f95051ff76ea036ed1faf940d5a04af4e7ff5a7dbd74576b907e99"
sha256 "c76080aa807a339b44139885d77d15ad60ab8cdd2c2fdaf345d0985625bc0f97"
end
conflicts_with "ripgrep"

1
rustfmt.toml Normal file
View File

@@ -0,0 +1 @@
disable_all_formatting = true

76
src/app.rs
View File

@@ -9,7 +9,8 @@
// is where we read clap's configuration from the end user's arguments and turn
// it into a ripgrep-specific configuration type that is not coupled with clap.
use clap::{self, App, AppSettings};
use clap::{self, App, AppSettings, crate_authors, crate_version};
use lazy_static::lazy_static;
const ABOUT: &str = "
ripgrep (rg) recursively searches your current directory for a regex pattern.
@@ -570,6 +571,7 @@ pub fn all_args_and_flags() -> Vec<RGArg> {
flag_iglob(&mut args);
flag_ignore_case(&mut args);
flag_ignore_file(&mut args);
flag_ignore_file_case_insensitive(&mut args);
flag_invert_match(&mut args);
flag_json(&mut args);
flag_line_buffered(&mut args);
@@ -584,6 +586,7 @@ pub fn all_args_and_flags() -> Vec<RGArg> {
flag_multiline_dotall(&mut args);
flag_no_config(&mut args);
flag_no_ignore(&mut args);
flag_no_ignore_dot(&mut args);
flag_no_ignore_global(&mut args);
flag_no_ignore_messages(&mut args);
flag_no_ignore_parent(&mut args);
@@ -788,17 +791,17 @@ to one of eight choices: red, blue, green, cyan, magenta, yellow, white and
black. Styles are limited to nobold, bold, nointense, intense, nounderline
or underline.
The format of the flag is `{type}:{attribute}:{value}`. `{type}` should be
one of path, line, column 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}`.
The format of the flag is '{type}:{attribute}:{value}'. '{type}' should be
one of path, line, column 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.
Extended colors can be used for `{value}` when the terminal supports ANSI color
Extended colors can be used for '{value}' when the terminal supports ANSI color
sequences. These are specified as either 'x' (256-color) or 'x,x,x' (24-bit
truecolor) where x is a number between 0 and 255 inclusive. x may be given as
a normal decimal number or a hexadecimal number, which is prefixed by `0x`.
@@ -979,10 +982,15 @@ fn flag_encoding(args: &mut Vec<RGArg>) {
const LONG: &str = long!("\
Specify the text encoding that ripgrep will use on all files searched. The
default value is 'auto', which will cause ripgrep to do a best effort automatic
detection of encoding on a per-file basis. Other supported values can be found
in the list of labels here:
detection of encoding on a per-file basis. Automatic detection in this case
only applies to files that begin with a UTF-8 or UTF-16 byte-order mark (BOM).
No other automatic detection is performend.
Other supported values can be found in the list of labels here:
https://encoding.spec.whatwg.org/#concept-encoding-get
For more details on encoding and how ripgrep deals with it, see GUIDE.md.
This flag can be disabled with --no-encoding.
");
let arg = RGArg::flag("encoding", "ENCODING").short("E")
@@ -1016,7 +1024,7 @@ fn flag_files(args: &mut Vec<RGArg>) {
const SHORT: &str = "Print each file that would be searched.";
const LONG: &str = long!("\
Print each file that would be searched without actually performing the search.
This is useful to determine whether a particular file is being search or not.
This is useful to determine whether a particular file is being searched or not.
");
let arg = RGArg::switch("files")
.help(SHORT).long_help(LONG)
@@ -1208,6 +1216,26 @@ directly on the command line, then used -g instead.
args.push(arg);
}
fn flag_ignore_file_case_insensitive(args: &mut Vec<RGArg>) {
const SHORT: &str = "Process ignore files case insensitively.";
const LONG: &str = long!("\
Process ignore files (.gitignore, .ignore, etc.) case insensitively. Note that
this comes with a performance penalty and is most useful on case insensitive
file systems (such as Windows).
This flag can be disabled with the --no-ignore-file-case-insensitive flag.
");
let arg = RGArg::switch("ignore-file-case-insensitive")
.help(SHORT).long_help(LONG)
.overrides("no-ignore-file-case-insensitive");
args.push(arg);
let arg = RGArg::switch("no-ignore-file-case-insensitive")
.hidden()
.overrides("ignore-file-case-insensitive");
args.push(arg);
}
fn flag_invert_match(args: &mut Vec<RGArg>) {
const SHORT: &str = "Invert matching.";
const LONG: &str = long!("\
@@ -1535,7 +1563,7 @@ fn flag_no_ignore(args: &mut Vec<RGArg>) {
const SHORT: &str = "Don't respect ignore files.";
const LONG: &str = long!("\
Don't respect ignore files (.gitignore, .ignore, etc.). This implies
--no-ignore-parent and --no-ignore-vcs.
--no-ignore-parent, --no-ignore-dot and --no-ignore-vcs.
This flag can be disabled with the --ignore flag.
");
@@ -1550,6 +1578,24 @@ This flag can be disabled with the --ignore flag.
args.push(arg);
}
fn flag_no_ignore_dot(args: &mut Vec<RGArg>) {
const SHORT: &str = "Don't respect .ignore files.";
const LONG: &str = long!("\
Don't respect .ignore files.
This flag can be disabled with the --ignore-dot flag.
");
let arg = RGArg::switch("no-ignore-dot")
.help(SHORT).long_help(LONG)
.overrides("ignore-dot");
args.push(arg);
let arg = RGArg::switch("ignore-dot")
.hidden()
.overrides("no-ignore-dot");
args.push(arg);
}
fn flag_no_ignore_global(args: &mut Vec<RGArg>) {
const SHORT: &str = "Don't respect global ignore files.";
const LONG: &str = long!("\
@@ -1998,9 +2044,9 @@ This flag can be used with the -o/--only-matching flag.
fn flag_search_zip(args: &mut Vec<RGArg>) {
const SHORT: &str = "Search in compressed files.";
const LONG: &str = long!("\
Search in compressed files. Currently gz, bz2, xz, lzma and lz4 files are
supported. This option expects the decompression binaries to be available in
your PATH.
Search in compressed files. Currently gzip, bzip2, xz, LZ4, LZMA, Brotli and
Zstd files are supported. This option expects the decompression binaries to be
available in your PATH.
This flag can be disabled with --no-search-zip.
");
@@ -2067,7 +2113,7 @@ for this flag are:
path Sort by file path.
modified Sort by the last modified time on a file.
accessed Sort by the last accessed time on a file.
created Sort by the cretion time on a file.
created Sort by the creation time on a file.
none Do not sort results.
If the sorting criteria isn't available on your system (for example, creation
@@ -2100,7 +2146,7 @@ for this flag are:
path Sort by file path.
modified Sort by the last modified time on a file.
accessed Sort by the last accessed time on a file.
created Sort by the cretion time on a file.
created Sort by the creation time on a file.
none Do not sort results.
If the sorting criteria isn't available on your system (for example, creation

141
src/args.rs
View File

@@ -1,9 +1,10 @@
use std::cmp;
use std::env;
use std::ffi::OsStr;
use std::ffi::{OsStr, OsString};
use std::fs;
use std::io;
use std::io::{self, Write};
use std::path::{Path, PathBuf};
use std::process;
use std::sync::Arc;
use std::time::SystemTime;
@@ -34,20 +35,22 @@ use ignore::types::{FileTypeDef, Types, TypesBuilder};
use ignore::{Walk, WalkBuilder, WalkParallel};
use log;
use num_cpus;
use path_printer::{PathPrinter, PathPrinterBuilder};
use regex;
use termcolor::{
WriteColor,
BufferWriter, ColorChoice,
};
use app;
use config;
use logger::Logger;
use messages::{set_messages, set_ignore_messages};
use search::{PatternMatcher, Printer, SearchWorker, SearchWorkerBuilder};
use subject::SubjectBuilder;
use Result;
use crate::app;
use crate::config;
use crate::logger::Logger;
use crate::messages::{set_messages, set_ignore_messages};
use crate::path_printer::{PathPrinter, PathPrinterBuilder};
use crate::search::{
PatternMatcher, Printer, SearchWorker, SearchWorkerBuilder,
};
use crate::subject::SubjectBuilder;
use crate::Result;
/// The command that ripgrep should execute based on the command line
/// configuration.
@@ -128,7 +131,7 @@ impl Args {
// trying to parse config files. If a config file exists and has
// arguments, then we re-parse argv, otherwise we just use the matches
// we have here.
let early_matches = ArgMatches::new(app::app().get_matches());
let early_matches = ArgMatches::new(clap_matches(env::args_os())?);
set_messages(!early_matches.is_present("no-messages"));
set_ignore_messages(!early_matches.is_present("no-ignore-messages"));
@@ -143,7 +146,7 @@ impl Args {
log::set_max_level(log::LevelFilter::Warn);
}
let matches = early_matches.reconfigure();
let matches = early_matches.reconfigure()?;
// The logging level may have changed if we brought in additional
// arguments from a configuration file, so recheck it and set the log
// level as appropriate.
@@ -265,6 +268,11 @@ impl Args {
Ok(builder.build(wtr))
}
/// Returns true if and only if ripgrep should be "quiet."
pub fn quiet(&self) -> bool {
self.matches().is_present("quiet")
}
/// Returns true if and only if the search should quit after finding the
/// first match.
pub fn quit_after_match(&self) -> Result<bool> {
@@ -488,25 +496,27 @@ impl ArgMatches {
///
/// If there are no additional arguments from the environment (e.g., a
/// config file), then the given matches are returned as is.
fn reconfigure(self) -> ArgMatches {
fn reconfigure(self) -> Result<ArgMatches> {
// If the end user says no config, then respect it.
if self.is_present("no-config") {
debug!("not reading config files because --no-config is present");
return self;
log::debug!(
"not reading config files because --no-config is present"
);
return Ok(self);
}
// If the user wants ripgrep to use a config file, then parse args
// from that first.
let mut args = config::args();
if args.is_empty() {
return self;
return Ok(self);
}
let mut cliargs = env::args_os();
if let Some(bin) = cliargs.next() {
args.insert(0, bin);
}
args.extend(cliargs);
debug!("final argv: {:?}", args);
ArgMatches::new(app::app().get_matches_from(args))
log::debug!("final argv: {:?}", args);
Ok(ArgMatches(clap_matches(args)?))
}
/// Convert the result of parsing CLI arguments into ripgrep's higher level
@@ -615,7 +625,10 @@ impl ArgMatches {
if let Some(limit) = self.dfa_size_limit()? {
builder.dfa_size_limit(limit);
}
Ok(builder.build(&patterns.join("|"))?)
match builder.build(&patterns.join("|")) {
Ok(m) => Ok(m),
Err(err) => Err(From::from(suggest_multiline(err.to_string()))),
}
}
/// Build a matcher using PCRE2.
@@ -633,7 +646,7 @@ impl ArgMatches {
// For whatever reason, the JIT craps out during regex compilation with
// a "no more memory" error on 32 bit systems. So don't use it there.
if !cfg!(target_pointer_width = "32") {
builder.jit(true);
builder.jit_if_available(true);
}
if self.pcre2_unicode() {
builder.utf(true).ucp(true);
@@ -779,18 +792,16 @@ impl ArgMatches {
.max_filesize(self.max_file_size()?)
.threads(self.threads()?)
.same_file_system(self.is_present("one-file-system"))
.skip_stdout(true)
.skip_stdout(!self.is_present("files"))
.overrides(self.overrides()?)
.types(self.types()?)
.hidden(!self.hidden())
.parents(!self.no_ignore_parent())
.ignore(!self.no_ignore())
.git_global(
!self.no_ignore()
&& !self.no_ignore_vcs()
&& !self.no_ignore_global())
.git_ignore(!self.no_ignore() && !self.no_ignore_vcs())
.git_exclude(!self.no_ignore() && !self.no_ignore_vcs());
.ignore(!self.no_ignore_dot())
.git_global(!self.no_ignore_vcs() && !self.no_ignore_global())
.git_ignore(!self.no_ignore_vcs())
.git_exclude(!self.no_ignore_vcs())
.ignore_case_insensitive(self.ignore_file_case_insensitive());
if !self.no_ignore() {
builder.add_custom_ignore_filename(".rgignore");
}
@@ -996,6 +1007,11 @@ impl ArgMatches {
self.is_present("hidden") || self.unrestricted_count() >= 2
}
/// Returns true if ignore files should be processed case insensitively.
fn ignore_file_case_insensitive(&self) -> bool {
self.is_present("ignore-file-case-insensitive")
}
/// Return all of the ignore file paths given on the command line.
fn ignore_paths(&self) -> Vec<PathBuf> {
let paths = match self.values_of_os("ignore-file") {
@@ -1090,6 +1106,11 @@ impl ArgMatches {
self.is_present("no-ignore") || self.unrestricted_count() >= 1
}
/// Returns true if .ignore files should be ignored.
fn no_ignore_dot(&self) -> bool {
self.is_present("no-ignore-dot") || self.no_ignore()
}
/// Returns true if global ignore files should be ignored.
fn no_ignore_global(&self) -> bool {
self.is_present("no-ignore-global") || self.no_ignore()
@@ -1136,7 +1157,7 @@ impl ArgMatches {
builder.add(&glob)?;
}
// This only enables case insensitivity for subsequent globs.
builder.case_insensitive(true)?;
builder.case_insensitive(true).unwrap();
for glob in self.values_of_lossy_vec("iglob") {
builder.add(&glob)?;
}
@@ -1250,9 +1271,15 @@ impl ArgMatches {
if let Some(paths) = self.values_of_os("file") {
for path in paths {
if path == "-" {
pats.extend(cli::patterns_from_stdin()?);
pats.extend(cli::patterns_from_stdin()?
.into_iter()
.map(|p| self.pattern_from_string(p))
);
} else {
pats.extend(cli::patterns_from_path(path)?);
pats.extend(cli::patterns_from_path(path)?
.into_iter()
.map(|p| self.pattern_from_string(p))
);
}
}
}
@@ -1281,13 +1308,17 @@ impl ArgMatches {
/// Converts a &str pattern to a String pattern. The pattern is escaped
/// if -F/--fixed-strings is set.
fn pattern_from_str(&self, pat: &str) -> String {
let litpat = self.pattern_literal(pat.to_string());
let s = self.pattern_line(litpat);
self.pattern_from_string(pat.to_string())
}
if s.is_empty() {
/// Applies additional processing on the given pattern if necessary
/// (such as escaping meta characters or turning it into a line regex).
fn pattern_from_string(&self, pat: String) -> String {
let pat = self.pattern_line(self.pattern_literal(pat));
if pat.is_empty() {
self.pattern_empty()
} else {
s
pat
}
}
@@ -1546,6 +1577,17 @@ and look-around.", msg)
}
}
fn suggest_multiline(msg: String) -> String {
if msg.contains("the literal") && msg.contains("not allowed") {
format!("{}
Consider enabling multiline mode with the --multiline flag (or -U for short).
When multiline mode is enabled, new line characters can be matched.", msg)
} else {
msg
}
}
/// Convert the result of parsing a human readable file size to a `usize`,
/// failing if the type does not fit.
fn u64_to_usize(
@@ -1592,3 +1634,32 @@ where G: Fn(&fs::Metadata) -> io::Result<SystemTime>
t1.cmp(&t2)
}
}
/// Returns a clap matches object if the given arguments parse successfully.
///
/// Otherwise, if an error occurred, then it is returned unless the error
/// corresponds to a `--help` or `--version` request. In which case, the
/// corresponding output is printed and the current process is exited
/// successfully.
fn clap_matches<I, T>(
args: I,
) -> Result<clap::ArgMatches<'static>>
where I: IntoIterator<Item=T>,
T: Into<OsString> + Clone
{
let err = match app::app().get_matches_from_safe(args) {
Ok(matches) => return Ok(matches),
Err(err) => err,
};
if err.use_stderr() {
return Err(err.into());
}
// Explicitly ignore any error returned by writeln!. The most likely error
// at this point is a broken pipe error, in which case, we want to ignore
// it and exit quietly.
//
// (This is the point of this helper function. clap's functionality for
// doing this will panic on a broken pipe error.)
let _ = writeln!(io::stdout(), "{}", err);
process::exit(0);
}

60
src/config.rs
View File

@@ -5,11 +5,14 @@
use std::env;
use std::error::Error;
use std::fs::File;
use std::io::{self, BufRead};
use std::io;
use std::ffi::OsString;
use std::path::{Path, PathBuf};
use Result;
use bstr::io::BufReadExt;
use log;
use crate::Result;
/// Return a sequence of arguments derived from ripgrep rc configuration files.
pub fn args() -> Vec<OsString> {
@@ -34,7 +37,7 @@ pub fn args() -> Vec<OsString> {
message!("{}:{}", config_path.display(), err);
}
}
debug!(
log::debug!(
"{}: arguments loaded from config file: {:?}",
config_path.display(),
args
@@ -74,62 +77,29 @@ fn parse<P: AsRef<Path>>(
fn parse_reader<R: io::Read>(
rdr: R,
) -> Result<(Vec<OsString>, Vec<Box<Error>>)> {
let mut bufrdr = io::BufReader::new(rdr);
let bufrdr = io::BufReader::new(rdr);
let (mut args, mut errs) = (vec![], vec![]);
let mut line = vec![];
let mut line_number = 0;
while {
line.clear();
bufrdr.for_byte_line_with_terminator(|line| {
line_number += 1;
bufrdr.read_until(b'\n', &mut line)? > 0
} {
trim(&mut line);
let line = line.trim();
if line.is_empty() || line[0] == b'#' {
continue;
return Ok(true);
}
match bytes_to_os_string(&line) {
match line.to_os_str() {
Ok(osstr) => {
args.push(osstr);
args.push(osstr.to_os_string());
}
Err(err) => {
errs.push(format!("{}: {}", line_number, err).into());
}
}
}
Ok(true)
})?;
Ok((args, errs))
}
/// Trim the given bytes of whitespace according to the ASCII definition.
fn trim(x: &mut Vec<u8>) {
let upto = x.iter().take_while(|b| is_space(**b)).count();
x.drain(..upto);
let revto = x.len() - x.iter().rev().take_while(|b| is_space(**b)).count();
x.drain(revto..);
}
/// Returns true if and only if the given byte is an ASCII space character.
fn is_space(b: u8) -> bool {
b == b'\t'
|| b == b'\n'
|| b == b'\x0B'
|| b == b'\x0C'
|| b == b'\r'
|| b == b' '
}
/// On Unix, get an OsString from raw bytes.
#[cfg(unix)]
fn bytes_to_os_string(bytes: &[u8]) -> Result<OsString> {
use std::os::unix::ffi::OsStringExt;
Ok(OsString::from_vec(bytes.to_vec()))
}
/// On non-Unix (like Windows), require UTF-8.
#[cfg(not(unix))]
fn bytes_to_os_string(bytes: &[u8]) -> Result<OsString> {
String::from_utf8(bytes.to_vec()).map(OsString::from).map_err(From::from)
}
#[cfg(test)]
mod tests {
use std::ffi::OsString;

64
src/main.rs
View File

@@ -1,17 +1,3 @@
#[macro_use]
extern crate clap;
extern crate grep;
extern crate ignore;
#[macro_use]
extern crate lazy_static;
#[macro_use]
extern crate log;
extern crate num_cpus;
extern crate regex;
#[macro_use]
extern crate serde_json;
extern crate termcolor;
use std::io::{self, Write};
use std::process;
use std::sync::{Arc, Mutex};
@@ -36,33 +22,37 @@ mod subject;
type Result<T> = ::std::result::Result<T, Box<::std::error::Error>>;
fn main() {
match Args::parse().and_then(try_main) {
Ok(true) => process::exit(0),
Ok(false) => process::exit(1),
Err(err) => {
eprintln!("{}", err);
process::exit(2);
}
if let Err(err) = Args::parse().and_then(try_main) {
eprintln!("{}", err);
process::exit(2);
}
}
fn try_main(args: Args) -> Result<bool> {
fn try_main(args: Args) -> Result<()> {
use args::Command::*;
match args.command()? {
Search => search(args),
SearchParallel => search_parallel(args),
SearchNever => Ok(false),
Files => files(args),
FilesParallel => files_parallel(args),
Types => types(args),
let matched =
match args.command()? {
Search => search(&args),
SearchParallel => search_parallel(&args),
SearchNever => Ok(false),
Files => files(&args),
FilesParallel => files_parallel(&args),
Types => types(&args),
}?;
if matched && (args.quiet() || !messages::errored()) {
process::exit(0)
} else if messages::errored() {
process::exit(2)
} else {
process::exit(1)
}
}
/// The top-level entry point for single-threaded search. This recursively
/// steps through the file list (current directory by default) and searches
/// each file sequentially.
fn search(args: Args) -> Result<bool> {
fn search(args: &Args) -> Result<bool> {
let started_at = Instant::now();
let quit_after_match = args.quit_after_match()?;
let subject_builder = args.subject_builder();
@@ -82,7 +72,7 @@ fn search(args: Args) -> Result<bool> {
if err.kind() == io::ErrorKind::BrokenPipe {
break;
}
message!("{}: {}", subject.path().display(), err);
err_message!("{}: {}", subject.path().display(), err);
continue;
}
};
@@ -105,7 +95,7 @@ fn search(args: Args) -> Result<bool> {
/// The top-level entry point for multi-threaded search. The parallelism is
/// itself achieved by the recursive directory traversal. All we need to do is
/// feed it a worker for performing a search on each file.
fn search_parallel(args: Args) -> Result<bool> {
fn search_parallel(args: &Args) -> Result<bool> {
use std::sync::atomic::AtomicBool;
use std::sync::atomic::Ordering::SeqCst;
@@ -141,7 +131,7 @@ fn search_parallel(args: Args) -> Result<bool> {
let search_result = match searcher.search(&subject) {
Ok(search_result) => search_result,
Err(err) => {
message!("{}: {}", subject.path().display(), err);
err_message!("{}: {}", subject.path().display(), err);
return WalkState::Continue;
}
};
@@ -158,7 +148,7 @@ fn search_parallel(args: Args) -> Result<bool> {
return WalkState::Quit;
}
// Otherwise, we continue on our merry way.
message!("{}: {}", subject.path().display(), err);
err_message!("{}: {}", subject.path().display(), err);
}
if matched.load(SeqCst) && quit_after_match {
WalkState::Quit
@@ -183,7 +173,7 @@ fn search_parallel(args: Args) -> Result<bool> {
/// The top-level entry point for listing files without searching them. This
/// recursively steps through the file list (current directory by default) and
/// prints each path sequentially using a single thread.
fn files(args: Args) -> Result<bool> {
fn files(args: &Args) -> Result<bool> {
let quit_after_match = args.quit_after_match()?;
let subject_builder = args.subject_builder();
let mut matched = false;
@@ -213,7 +203,7 @@ fn files(args: Args) -> Result<bool> {
/// The top-level entry point for listing files without searching them. This
/// recursively steps through the file list (current directory by default) and
/// prints each path sequentially using multiple threads.
fn files_parallel(args: Args) -> Result<bool> {
fn files_parallel(args: &Args) -> Result<bool> {
use std::sync::atomic::AtomicBool;
use std::sync::atomic::Ordering::SeqCst;
use std::sync::mpsc;
@@ -265,7 +255,7 @@ fn files_parallel(args: Args) -> Result<bool> {
}
/// The top-level entry point for --type-list.
fn types(args: Args) -> Result<bool> {
fn types(args: &Args) -> Result<bool> {
let mut count = 0;
let mut stdout = args.stdout();
for def in args.type_defs()? {

34
src/messages.rs
View File

@@ -1,21 +1,35 @@
use std::sync::atomic::{ATOMIC_BOOL_INIT, AtomicBool, Ordering};
use std::sync::atomic::{AtomicBool, Ordering};
static MESSAGES: AtomicBool = ATOMIC_BOOL_INIT;
static IGNORE_MESSAGES: AtomicBool = ATOMIC_BOOL_INIT;
static MESSAGES: AtomicBool = AtomicBool::new(false);
static IGNORE_MESSAGES: AtomicBool = AtomicBool::new(false);
static ERRORED: AtomicBool = AtomicBool::new(false);
/// Emit a non-fatal error message, unless messages were disabled.
#[macro_export]
macro_rules! message {
($($tt:tt)*) => {
if ::messages::messages() {
if crate::messages::messages() {
eprintln!($($tt)*);
}
}
}
/// Like message, but sets ripgrep's "errored" flag, which controls the exit
/// status.
#[macro_export]
macro_rules! err_message {
($($tt:tt)*) => {
crate::messages::set_errored();
message!($($tt)*);
}
}
/// Emit a non-fatal ignore-related error message (like a parse error), unless
/// ignore-messages were disabled.
#[macro_export]
macro_rules! ignore_message {
($($tt:tt)*) => {
if ::messages::messages() && ::messages::ignore_messages() {
if crate::messages::messages() && crate::messages::ignore_messages() {
eprintln!($($tt)*);
}
}
@@ -48,3 +62,13 @@ pub fn ignore_messages() -> bool {
pub fn set_ignore_messages(yes: bool) {
IGNORE_MESSAGES.store(yes, Ordering::SeqCst)
}
/// Returns true if and only if ripgrep came across a non-fatal error.
pub fn errored() -> bool {
ERRORED.load(Ordering::SeqCst)
}
/// Indicate that ripgrep has come across a non-fatal error.
pub fn set_errored() {
ERRORED.store(true, Ordering::SeqCst);
}

3
src/search.rs
View File

@@ -13,9 +13,10 @@ use grep::regex::{RegexMatcher as RustRegexMatcher};
use grep::searcher::Searcher;
use ignore::overrides::Override;
use serde_json as json;
use serde_json::json;
use termcolor::WriteColor;
use subject::Subject;
use crate::subject::Subject;
/// The configuration for the search worker. Among a few other things, the
/// configuration primarily controls the way we show search results to users

19
src/subject.rs
View File

@@ -1,6 +1,7 @@
use std::path::Path;
use ignore::{self, DirEntry};
use log;
/// A configuration for describing how subjects should be built.
#[derive(Clone, Debug)]
@@ -40,7 +41,7 @@ impl SubjectBuilder {
match result {
Ok(dent) => self.build(dent),
Err(err) => {
message!("{}", err);
err_message!("{}", err);
None
}
}
@@ -79,7 +80,7 @@ impl SubjectBuilder {
// directory. Otherwise, emitting messages for directories is just
// noisy.
if !subj.is_dir() {
debug!(
log::debug!(
"ignoring {}: failed to pass subject filter: \
file type: {:?}, metadata: {:?}",
subj.dent.path().display(),
@@ -126,9 +127,19 @@ impl Subject {
self.dent.is_stdin()
}
/// Returns true if and only if this subject points to a directory.
/// Returns true if and only if this subject points to a directory after
/// following symbolic links.
fn is_dir(&self) -> bool {
self.dent.file_type().map_or(false, |ft| ft.is_dir())
let ft = match self.dent.file_type() {
None => return false,
Some(ft) => ft,
};
if ft.is_dir() {
return true;
}
// If this is a symlink, then we want to follow it to determine
// whether it's a directory or not.
self.dent.path_is_symlink() && self.dent.path().is_dir()
}
/// Returns true if and only if this subject points to a file.

2
tests/data/sherlock.br Normal file
View File

@@ -0,0 +1,2 @@
n<01><><EFBFBD>-_<>.<2E> <0C><><11><>cM<63><4D><EFBFBD><EFBFBD>Y<EFBFBD>4<08><><EFBFBD><EFBFBD><EFBFBD>Ya <0B>-L<>O(<28>8<EFBFBD>sn^Gwш!,<2C>
KD<EFBFBD><EFBFBD>/7<><37>th<74><1A> <0C><10>]j<02><><EFBFBD><EFBFBD><EFBFBD>E_;d<1E>rF<72>Qs<51>/:DIVB}<7D>T7<54><37>ѵ<04><16>H<EFBFBD>2<EFBFBD><32><EFBFBD>)<29><>M[u<><75><EFBFBD>i<EFBFBD><69><EFBFBD><0F>50ڮ<30>Y6<><36><EFBFBD><EFBFBD><17><><EFBFBD><EFBFBD><07>%<25>ר_<D7A8><5F>U by<62>4<EFBFBD><34>Ϡ<EFBFBD>!&<26>g<><15>#<23>

BIN
tests/data/sherlock.zst Normal file
View File

Binary file not shown.

20
tests/feature.rs
View File

@@ -1,5 +1,5 @@
use hay::{SHERLOCK, SHERLOCK_CRLF};
use util::{Dir, TestCommand, sort_lines};
use crate::hay::{SHERLOCK, SHERLOCK_CRLF};
use crate::util::{Dir, TestCommand, sort_lines};
// See: https://github.com/BurntSushi/ripgrep/issues/1
rgtest!(f1_sjis, |dir: Dir, mut cmd: TestCommand| {
@@ -629,3 +629,19 @@ rgtest!(f993_null_data, |dir: Dir, mut cmd: TestCommand| {
let expected = "foo\x00bar\x00baz\x00";
eqnice!(expected, cmd.stdout());
});
// See: https://github.com/BurntSushi/ripgrep/issues/1138
rgtest!(f1138_no_ignore_dot, |dir: Dir, mut cmd: TestCommand| {
dir.create_dir(".git");
dir.create(".gitignore", "foo");
dir.create(".ignore", "bar");
dir.create(".fzf-ignore", "quux");
dir.create("foo", "");
dir.create("bar", "");
dir.create("quux", "");
cmd.arg("--sort").arg("path").arg("--files");
eqnice!("quux\n", cmd.stdout());
eqnice!("bar\nquux\n", cmd.arg("--no-ignore-dot").stdout());
eqnice!("bar\n", cmd.arg("--ignore-file").arg(".fzf-ignore").stdout());
});

68
tests/json.rs
View File

@@ -1,9 +1,10 @@
use std::time;
use serde_derive::Deserialize;
use serde_json as json;
use hay::{SHERLOCK, SHERLOCK_CRLF};
use util::{Dir, TestCommand};
use crate::hay::{SHERLOCK, SHERLOCK_CRLF};
use crate::util::{Dir, TestCommand};
#[derive(Clone, Debug, Deserialize, PartialEq, Eq)]
#[serde(tag = "type", content = "data")]
@@ -152,7 +153,10 @@ rgtest!(basic, |dir: Dir, mut cmd: TestCommand| {
msgs[1].unwrap_context(),
Context {
path: Some(Data::text("sherlock")),
lines: Data::text("Holmeses, success in the province of detective work must always\n"),
lines: Data::text(
"Holmeses, success in the province of \
detective work must always\n",
),
line_number: Some(2),
absolute_offset: 65,
submatches: vec![],
@@ -162,7 +166,10 @@ rgtest!(basic, |dir: Dir, mut cmd: TestCommand| {
msgs[2].unwrap_match(),
Match {
path: Some(Data::text("sherlock")),
lines: Data::text("be, to a very large extent, the result of luck. Sherlock Holmes\n"),
lines: Data::text(
"be, to a very large extent, the result of luck. \
Sherlock Holmes\n",
),
line_number: Some(3),
absolute_offset: 129,
submatches: vec![
@@ -211,7 +218,9 @@ rgtest!(notutf8, |dir: Dir, mut cmd: TestCommand| {
let contents = &b"quux\xFFbaz"[..];
// APFS does not support creating files with invalid UTF-8 bytes, so just
// skip the test if we can't create our file.
// skip the test if we can't create our file. Presumably we don't need this
// check if we're already skipping it on macOS, but maybe other file
// systems won't like this test either?
if !dir.try_create_bytes(OsStr::from_bytes(name), contents).is_ok() {
return;
}
@@ -304,3 +313,52 @@ rgtest!(crlf, |dir: Dir, mut cmd: TestCommand| {
},
);
});
// See: https://github.com/BurntSushi/ripgrep/issues/1095
//
// This test checks that we don't drop the \r\n in a matching line when --crlf
// mode is enabled.
rgtest!(r1095_missing_crlf, |dir: Dir, mut cmd: TestCommand| {
dir.create("foo", "test\r\n");
// Check without --crlf flag.
let msgs = json_decode(&cmd.arg("--json").arg("test").stdout());
assert_eq!(msgs.len(), 4);
assert_eq!(msgs[1].unwrap_match().lines, Data::text("test\r\n"));
// Now check with --crlf flag.
let msgs = json_decode(&cmd.arg("--crlf").stdout());
assert_eq!(msgs.len(), 4);
assert_eq!(msgs[1].unwrap_match().lines, Data::text("test\r\n"));
});
// See: https://github.com/BurntSushi/ripgrep/issues/1095
//
// This test checks that we don't return empty submatches when matching a `\n`
// in CRLF mode.
rgtest!(r1095_crlf_empty_match, |dir: Dir, mut cmd: TestCommand| {
dir.create("foo", "test\r\n\n");
// Check without --crlf flag.
let msgs = json_decode(&cmd.arg("-U").arg("--json").arg("\n").stdout());
assert_eq!(msgs.len(), 5);
let m = msgs[1].unwrap_match();
assert_eq!(m.lines, Data::text("test\r\n"));
assert_eq!(m.submatches[0].m, Data::text("\n"));
let m = msgs[2].unwrap_match();
assert_eq!(m.lines, Data::text("\n"));
assert_eq!(m.submatches[0].m, Data::text("\n"));
// Now check with --crlf flag.
let msgs = json_decode(&cmd.arg("--crlf").stdout());
let m = msgs[1].unwrap_match();
assert_eq!(m.lines, Data::text("test\r\n"));
assert_eq!(m.submatches[0].m, Data::text("\n"));
let m = msgs[2].unwrap_match();
assert_eq!(m.lines, Data::text("\n"));
assert_eq!(m.submatches[0].m, Data::text("\n"));
});

4
tests/macros.rs
View File

@@ -3,11 +3,11 @@ macro_rules! rgtest {
($name:ident, $fun:expr) => {
#[test]
fn $name() {
let (dir, cmd) = ::util::setup(stringify!($name));
let (dir, cmd) = crate::util::setup(stringify!($name));
$fun(dir, cmd);
if cfg!(feature = "pcre2") {
let (dir, cmd) = ::util::setup_pcre2(stringify!($name));
let (dir, cmd) = crate::util::setup_pcre2(stringify!($name));
$fun(dir, cmd);
}
}

34
tests/misc.rs
View File

@@ -1,5 +1,5 @@
use hay::SHERLOCK;
use util::{Dir, TestCommand, cmd_exists, sort_lines};
use crate::hay::SHERLOCK;
use crate::util::{Dir, TestCommand, cmd_exists, sort_lines};
// This file contains "miscellaneous" tests that were either written before
// features were tracked more explicitly, or were simply written without
@@ -909,6 +909,36 @@ be, to a very large extent, the result of luck. Sherlock Holmes
eqnice!(expected, cmd.stdout());
});
rgtest!(compressed_brotli, |dir: Dir, mut cmd: TestCommand| {
if !cmd_exists("brotli") {
return;
}
dir.create_bytes("sherlock.br", include_bytes!("./data/sherlock.br"));
cmd.arg("-z").arg("Sherlock").arg("sherlock.br");
let expected = "\
For the Doctor Watsons of this world, as opposed to the Sherlock
be, to a very large extent, the result of luck. Sherlock Holmes
";
eqnice!(expected, cmd.stdout());
});
rgtest!(compressed_zstd, |dir: Dir, mut cmd: TestCommand| {
if !cmd_exists("zstd") {
return;
}
dir.create_bytes("sherlock.zst", include_bytes!("./data/sherlock.zst"));
cmd.arg("-z").arg("Sherlock").arg("sherlock.zst");
let expected = "\
For the Doctor Watsons of this world, as opposed to the Sherlock
be, to a very large extent, the result of luck. Sherlock Holmes
";
eqnice!(expected, cmd.stdout());
});
rgtest!(compressed_failing_gzip, |dir: Dir, mut cmd: TestCommand| {
if !cmd_exists("gzip") {
return;

4
tests/multiline.rs
View File

@@ -1,5 +1,5 @@
use hay::SHERLOCK;
use util::{Dir, TestCommand};
use crate::hay::SHERLOCK;
use crate::util::{Dir, TestCommand};
// This tests that multiline matches that span multiple lines, but where
// multiple matches may begin and end on the same line work correctly.

147
tests/regression.rs
View File

@@ -1,5 +1,5 @@
use hay::SHERLOCK;
use util::{Dir, TestCommand, sort_lines};
use crate::hay::SHERLOCK;
use crate::util::{Dir, TestCommand, sort_lines};
// See: https://github.com/BurntSushi/ripgrep/issues/16
rgtest!(r16, |dir: Dir, mut cmd: TestCommand| {
@@ -562,3 +562,146 @@ rgtest!(r900, |dir: Dir, mut cmd: TestCommand| {
cmd.arg("-fpat").arg("sherlock").assert_err();
});
// See: https://github.com/BurntSushi/ripgrep/issues/1064
rgtest!(r1064, |dir: Dir, mut cmd: TestCommand| {
dir.create("input", "abc");
eqnice!("input:abc\n", cmd.arg("a(.*c)").stdout());
});
// See: https://github.com/BurntSushi/ripgrep/issues/1174
rgtest!(r1098, |dir: Dir, mut cmd: TestCommand| {
dir.create_dir(".git");
dir.create(".gitignore", "a**b");
dir.create("afoob", "test");
cmd.arg("test").assert_err();
});
// See: https://github.com/BurntSushi/ripgrep/issues/1130
rgtest!(r1130, |dir: Dir, mut cmd: TestCommand| {
dir.create("foo", "test");
eqnice!(
"foo\n",
cmd.arg("--files-with-matches").arg("test").arg("foo").stdout()
);
let mut cmd = dir.command();
eqnice!(
"foo\n",
cmd.arg("--files-without-match").arg("nada").arg("foo").stdout()
);
});
// See: https://github.com/BurntSushi/ripgrep/issues/1159
rgtest!(r1159_invalid_flag, |_: Dir, mut cmd: TestCommand| {
cmd.arg("--wat").assert_exit_code(2);
});
// See: https://github.com/BurntSushi/ripgrep/issues/1159
rgtest!(r1159_exit_status, |dir: Dir, _: TestCommand| {
dir.create("foo", "test");
// search with a match gets 0 exit status.
let mut cmd = dir.command();
cmd.arg("test").assert_exit_code(0);
// search with --quiet and a match gets 0 exit status.
let mut cmd = dir.command();
cmd.arg("-q").arg("test").assert_exit_code(0);
// search with a match and an error gets 2 exit status.
let mut cmd = dir.command();
cmd.arg("test").arg("no-file").assert_exit_code(2);
// search with a match in --quiet mode and an error gets 0 exit status.
let mut cmd = dir.command();
cmd.arg("-q").arg("test").arg("foo").arg("no-file").assert_exit_code(0);
// search with no match gets 1 exit status.
let mut cmd = dir.command();
cmd.arg("nada").assert_exit_code(1);
// search with --quiet and no match gets 1 exit status.
let mut cmd = dir.command();
cmd.arg("-q").arg("nada").assert_exit_code(1);
// search with no match and an error gets 2 exit status.
let mut cmd = dir.command();
cmd.arg("nada").arg("no-file").assert_exit_code(2);
// search with no match in --quiet mode and an error gets 2 exit status.
let mut cmd = dir.command();
cmd.arg("-q").arg("nada").arg("foo").arg("no-file").assert_exit_code(2);
});
// See: https://github.com/BurntSushi/ripgrep/issues/1163
rgtest!(r1163, |dir: Dir, mut cmd: TestCommand| {
dir.create("bom.txt", "\u{FEFF}test123\ntest123");
eqnice!(
"bom.txt:test123\nbom.txt:test123\n",
cmd.arg("^test123").stdout()
);
});
// See: https://github.com/BurntSushi/ripgrep/issues/1164
rgtest!(r1164, |dir: Dir, mut cmd: TestCommand| {
dir.create_dir(".git");
dir.create(".gitignore", "myfile");
dir.create("MYFILE", "test");
cmd.arg("--ignore-file-case-insensitive").arg("test").assert_err();
eqnice!(
"MYFILE:test\n",
cmd.arg("--no-ignore-file-case-insensitive").stdout()
);
});
// See: https://github.com/BurntSushi/ripgrep/issues/1173
rgtest!(r1173, |dir: Dir, mut cmd: TestCommand| {
dir.create_dir(".git");
dir.create(".gitignore", "**");
dir.create("foo", "test");
cmd.arg("test").assert_err();
});
// See: https://github.com/BurntSushi/ripgrep/issues/1174
rgtest!(r1174, |dir: Dir, mut cmd: TestCommand| {
dir.create_dir(".git");
dir.create(".gitignore", "**/**/*");
dir.create_dir("a");
dir.create("a/foo", "test");
cmd.arg("test").assert_err();
});
// See: https://github.com/BurntSushi/ripgrep/issues/1176
rgtest!(r1176_literal_file, |dir: Dir, mut cmd: TestCommand| {
dir.create("patterns", "foo(bar\n");
dir.create("test", "foo(bar");
eqnice!(
"foo(bar\n",
cmd.arg("-F").arg("-f").arg("patterns").arg("test").stdout()
);
});
// See: https://github.com/BurntSushi/ripgrep/issues/1176
rgtest!(r1176_line_regex, |dir: Dir, mut cmd: TestCommand| {
dir.create("patterns", "foo\n");
dir.create("test", "foobar\nfoo\nbarfoo\n");
eqnice!(
"foo\n",
cmd.arg("-x").arg("-f").arg("patterns").arg("test").stdout()
);
});
// See: https://github.com/BurntSushi/ripgrep/issues/1203
rgtest!(r1203_reverse_suffix_literal, |dir: Dir, _: TestCommand| {
dir.create("test", "153.230000\n");
let mut cmd = dir.command();
eqnice!("153.230000\n", cmd.arg(r"\d\d\d00").arg("test").stdout());
let mut cmd = dir.command();
eqnice!("153.230000\n", cmd.arg(r"\d\d\d000").arg("test").stdout());
});

5
tests/tests.rs
View File

@@ -1,8 +1,3 @@
extern crate serde;
#[macro_use]
extern crate serde_derive;
extern crate serde_json;
// Macros useful for testing.
#[macro_use]
mod macros;

4
tests/util.rs
View File

@@ -5,12 +5,12 @@ use std::fs::{self, File};
use std::io::{self, Write};
use std::path::{Path, PathBuf};
use std::process::{self, Command};
use std::sync::atomic::{ATOMIC_USIZE_INIT, AtomicUsize, Ordering};
use std::sync::atomic::{AtomicUsize, Ordering};
use std::thread;
use std::time::Duration;
static TEST_DIR: &'static str = "ripgrep-tests";
static NEXT_ID: AtomicUsize = ATOMIC_USIZE_INIT;
static NEXT_ID: AtomicUsize = AtomicUsize::new(0);
/// Setup an empty work directory and return a command pointing to the ripgrep
/// executable whose CWD is set to the work directory.