mirror of
https://github.com/facebook/sapling.git
synced 2024-10-10 08:47:12 +03:00
311e30746f
Summary: Biggrep recently moved to a sharded system where fbsource is spread over three shards and can have three different corpus revs. Let's update hg's biggrep integration to handle this new logic. We basically just take the earliest corpus commit and diff that against the users commit. There's room for some incorrectness here, like if the users commit is based off a commit before the earliest corpus commit then it will receive changes from the alter corpus commits. Long term biggrep will work on unifying the various shards to use a consistent corpus rev, but for now this is probably acceptable. Reviewed By: quark-zju Differential Revision: D23057055 fbshipit-source-id: 745cafe772cea44bbe2d8eb1eee70dd317f255fe
88 lines
2.1 KiB
Python
Executable File
88 lines
2.1 KiB
Python
Executable File
#!/usr/bin/env python
|
|
# This is a terribly anemic fake implementation of the biggrep client
|
|
import argparse
|
|
import re
|
|
import subprocess
|
|
|
|
|
|
# The null commit
|
|
NULL = "0" * 40
|
|
|
|
# Escape sequences used by biggrep_client
|
|
MAGENTA = "\x1b[35m\x1b[K"
|
|
OFF = "\x1b[m\x1b[K"
|
|
BLUE = "\x1b[36m\x1b[K"
|
|
GREEN = "\x1b[32m\x1b[K"
|
|
|
|
|
|
parser = argparse.ArgumentParser()
|
|
parser.add_argument("--stripdir", action="store_true")
|
|
parser.add_argument("-r", action="store_true")
|
|
parser.add_argument("--color")
|
|
parser.add_argument("--expression")
|
|
parser.add_argument("-f")
|
|
parser.add_argument("tier")
|
|
parser.add_argument("corpus")
|
|
parser.add_argument("engine")
|
|
args = parser.parse_args()
|
|
|
|
|
|
def magenta(what):
|
|
if args.color:
|
|
return MAGENTA + what + OFF
|
|
return what
|
|
|
|
|
|
def blue(what):
|
|
if args.color:
|
|
return BLUE + what + OFF
|
|
return what
|
|
|
|
|
|
def green(what):
|
|
if args.color:
|
|
return GREEN + what + OFF
|
|
return what
|
|
|
|
|
|
def result_line(filename, line, col, context):
|
|
if args.f:
|
|
if not re.match(args.f, filename):
|
|
return
|
|
|
|
if not re.match(args.expression, context):
|
|
return
|
|
|
|
print(
|
|
magenta(filename)
|
|
+ blue(":")
|
|
+ green(str(line))
|
|
+ blue(":")
|
|
+ green(str(col))
|
|
+ blue(":")
|
|
+ context
|
|
# stick _bg on the end so we can tell that the result
|
|
# came from biggrep
|
|
+ "_bg"
|
|
)
|
|
|
|
|
|
# Report the current commit as the corpus revision so that `hg grep` doesn't
|
|
# then need to go and run grep for itself over the files
|
|
p = subprocess.Popen(["hg", "log", "-r", ".", "-T{node}"], stdout=subprocess.PIPE)
|
|
out, err = p.communicate()
|
|
rev = out.rstrip()
|
|
print("#fake=%s:0" % rev.decode("utf-8"))
|
|
|
|
# This list is coupled with the "Set up the repository with some simple files"
|
|
# section of eden/scm/tests/test-fb-hgext-tweakdefaults-grep.t
|
|
files = {
|
|
"grepdir/grepfile1": "foobarbaz",
|
|
"grepdir/grepfile2": "foobarboo",
|
|
"grepdir/subdir1/subfile1": "foobar_subdir",
|
|
"grepdir/subdir2/subfile2": "foobar_dirsub",
|
|
}
|
|
|
|
for (filename, context) in files.items():
|
|
result_line(filename, 1, 1, context)
|