pre-commit/pre-commit-hooks
1
0
Fork 0
mirror of https://github.com/pre-commit/pre-commit-hooks.git synced 2026-04-09 04:54:16 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Refactor _detect_line_ending

Browse source
This commit is contained in:
Morgan Courbet 2017-07-10 21:27:09 +02:00
parent 8bc4af45ba
commit 4fc9624b6a
No known key found for this signature in database
GPG key ID: 467299D324A21B24
1 changed files with 27 additions and 15 deletions
Download patch file Download diff file Expand all files Collapse all files

42
pre_commit_hooks/mixed_line_ending.py
View file

@ -8,7 +8,7 @@ from enum import Enum
class LineEnding(Enum): class LineEnding(Enum):
CR = b'\r', '\\r', 'cr', re.compile(b'\r', re.DOTALL) CR = b'\r', '\\r', 'cr', re.compile(b'\r(?!\n)', re.DOTALL)
CRLF = b'\r\n', '\\r\\n', 'crlf', re.compile(b'\r\n', re.DOTALL) CRLF = b'\r\n', '\\r\\n', 'crlf', re.compile(b'\r\n', re.DOTALL)
LF = b'\n', '\\n', 'lf', re.compile(b'(?<!\r)\n', re.DOTALL) LF = b'\n', '\\n', 'lf', re.compile(b'(?<!\r)\n', re.DOTALL)
@ -132,26 +132,38 @@ def _detect_line_ending(filename):
with open(filename, 'rb') as f: with open(filename, 'rb') as f:
buf = f.read() buf = f.read()
crlf_nb = len(LineEnding.CRLF.regex.findall(buf)) le_counts = {}
lf_nb = len(LineEnding.LF.regex.findall(buf)) for le_enum in LineEnding:
le_counts[le_enum] = len(le_enum.regex.findall(buf))
crlf_found = crlf_nb > 0 logging.debug('_detect_line_ending: le_counts = ' + str(le_counts))
lf_found = lf_nb > 0
logging.debug('_detect_line_ending: crlf_nb = %d, lf_nb = %d, ' mixed = False
'crlf_found = %s, lf_found = %s', le_found_previously = False
crlf_nb, lf_nb, crlf_found, lf_found) most_le = None
max_le_count = 0
if crlf_nb == lf_nb: for le, le_count in le_counts.items():
return MixedLineDetection.UNKNOWN le_found_cur = le_count > 0
if crlf_found ^ lf_found: mixed |= le_found_previously and le_found_cur
le_found_previously |= le_found_cur
if le_count == max_le_count:
most_le = None
elif le_count > max_le_count:
max_le_count = le_count
most_le = le
if not mixed:
return MixedLineDetection.NOT_MIXED return MixedLineDetection.NOT_MIXED
if crlf_nb > lf_nb: for mld in MixedLineDetection:
return MixedLineDetection.MIXED_MOSTLY_CRLF if mld.line_ending_enum is not None \
else: and mld.line_ending_enum == most_le:
return MixedLineDetection.MIXED_MOSTLY_LF return mld
return MixedLineDetection.UNKNOWN
def _process_no_fix(filenames): def _process_no_fix(filenames):