def test_actor_from_string(self):
self.assertEqual(Actor._from_string("name"), Actor("name", None))
self.assertEqual(Actor._from_string("name <>"), Actor("name", ""))
self.assertEqual(
Actor._from_string(
"name last another <*****@*****.**>"),
Actor("name last another", "*****@*****.**"))
def test_stats(self):
commit = self.rorepo.commit('33ebe7acec14b25c5f84f35a664803fcab2f7781')
stats = commit.stats
def check_entries(d):
assert isinstance(d, dict)
for key in ("insertions", "deletions", "lines"):
assert key in d
# END assertion helper
assert stats.files
assert stats.total
check_entries(stats.total)
assert "files" in stats.total
for filepath, d in stats.files.items():
check_entries(d)
# END for each stated file
# assure data is parsed properly
michael = Actor._from_string("Michael Trier <*****@*****.**>")
assert commit.author == michael
assert commit.committer == michael
assert commit.authored_date == 1210193388
assert commit.committed_date == 1210193388
assert commit.author_tz_offset == 14400, commit.author_tz_offset
assert commit.committer_tz_offset == 14400, commit.committer_tz_offset
assert commit.message == "initial project\n"
def from_line(cls, line):
""":return: New RefLogEntry instance from the given revlog line.
:param line: line bytes without trailing newline
:raise ValueError: If line could not be parsed"""
line = line.decode(defenc)
fields = line.split('\t', 1)
if len(fields) == 1:
info, msg = fields[0], None
elif len(fields) == 2:
info, msg = fields
else:
raise ValueError("Line must have up to two TAB-separated fields."
" Got %s" % repr(line))
# END handle first split
oldhexsha = info[:40]
newhexsha = info[41:81]
for hexsha in (oldhexsha, newhexsha):
if not cls._re_hexsha_only.match(hexsha):
raise ValueError("Invalid hexsha: %r" % (hexsha,))
# END if hexsha re doesn't match
# END for each hexsha
email_end = info.find('>', 82)
if email_end == -1:
raise ValueError("Missing token: >")
# END handle missing end brace
actor = Actor._from_string(info[82:email_end + 1])
time, tz_offset = parse_date(info[email_end + 2:])
return RefLogEntry((oldhexsha, newhexsha, actor, (time, tz_offset), msg))
def test_unicode_actor(self):
# assure we can parse unicode actors correctly
name = "Üäöß ÄußÉ".decode("utf-8")
assert len(name) == 9
special = Actor._from_string(u"%s <*****@*****.**>" % name)
assert special.name == name
assert isinstance(special.name, unicode)
def test_stats(self):
commit = self.rorepo.commit('33ebe7acec14b25c5f84f35a664803fcab2f7781')
stats = commit.stats
def check_entries(d):
assert isinstance(d, dict)
for key in ("insertions", "deletions", "lines"):
assert key in d
# END assertion helper
assert stats.files
assert stats.total
check_entries(stats.total)
assert "files" in stats.total
for filepath, d in stats.files.items():
check_entries(d)
# END for each stated file
# assure data is parsed properly
michael = Actor._from_string("Michael Trier <*****@*****.**>")
assert commit.author == michael
assert commit.committer == michael
assert commit.authored_date == 1210193388
assert commit.committed_date == 1210193388
assert commit.author_tz_offset == 14400, commit.author_tz_offset
assert commit.committer_tz_offset == 14400, commit.committer_tz_offset
assert commit.message == "initial project\n"
def from_line(cls, line):
""":return: New RefLogEntry instance from the given revlog line.
:param line: line without trailing newline
:raise ValueError: If line could not be parsed"""
try:
info, msg = line.split('\t', 2)
except ValueError:
raise ValueError("line is missing tab separator")
#END handle first plit
oldhexsha = info[:40]
newhexsha = info[41:81]
for hexsha in (oldhexsha, newhexsha):
if not cls._re_hexsha_only.match(hexsha):
raise ValueError("Invalid hexsha: %s" % hexsha)
# END if hexsha re doesn't match
#END for each hexsha
email_end = info.find('>', 82)
if email_end == -1:
raise ValueError("Missing token: >")
#END handle missing end brace
actor = Actor._from_string(info[82:email_end + 1])
time, tz_offset = parse_date(info[email_end + 2:])
return RefLogEntry(
(oldhexsha, newhexsha, actor, (time, tz_offset), msg))
def from_line(cls, line):
""":return: New RefLogEntry instance from the given revlog line.
:param line: line without trailing newline
:raise ValueError: If line could not be parsed"""
try:
info, msg = line.split('\t', 2)
except ValueError:
raise ValueError("line is missing tab separator")
# END handle first plit
oldhexsha = info[:40]
newhexsha = info[41:81]
for hexsha in (oldhexsha, newhexsha):
if not cls._re_hexsha_only.match(hexsha):
raise ValueError("Invalid hexsha: %s" % hexsha)
# END if hexsha re doesn't match
# END for each hexsha
email_end = info.find('>', 82)
if email_end == -1:
raise ValueError("Missing token: >")
# END handle missing end brace
actor = Actor._from_string(info[82:email_end + 1])
time, tz_offset = parse_date(info[email_end + 2:])
return RefLogEntry((oldhexsha, newhexsha, actor, (time, tz_offset), msg))
def test_unicode_actor(self):
# assure we can parse unicode actors correctly
name = "Üäöß ÄußÉ".decode("utf-8")
assert len(name) == 9
special = Actor._from_string(u"%s <*****@*****.**>" % name)
assert special.name == name
assert isinstance(special.name, unicode)
def parse_actor_and_date(line):
"""Parse out the actor (author or committer) info from a line like::
author Tom Preston-Werner <*****@*****.**> 1191999972 -0700
:return: [Actor, int_seconds_since_epoch, int_timezone_offset]"""
m = _re_actor_epoch.search(line)
actor, epoch, offset = m.groups()
return (Actor._from_string(actor), int(epoch), utctz_to_altz(offset))
def parse_actor_and_date(line):
"""Parse out the actor (author or committer) info from a line like::
author Tom Preston-Werner <*****@*****.**> 1191999972 -0700
:return: [Actor, int_seconds_since_epoch, int_timezone_offset]"""
m = _re_actor_epoch.search(line)
actor, epoch, offset = m.groups()
return (Actor._from_string(actor), int(epoch), utctz_to_altz(offset))
def test_from_string_should_separate_name_and_email(self):
a = Actor._from_string("Michael Trier <*****@*****.**>")
assert "Michael Trier" == a.name
assert "*****@*****.**" == a.email
# base type capabilities
assert a == a
assert not (a != a)
m = set()
m.add(a)
m.add(a)
assert len(m) == 1
def test_from_string_should_separate_name_and_email(self):
a = Actor._from_string("Michael Trier <*****@*****.**>")
assert "Michael Trier" == a.name
assert "*****@*****.**" == a.email
# base type capabilities
assert a == a
assert not (a != a)
m = set()
m.add(a)
m.add(a)
assert len(m) == 1
def parse_actor_and_date(line: str) -> Tuple[Actor, int, int]:
"""Parse out the actor (author or committer) info from a line like::
author Tom Preston-Werner <*****@*****.**> 1191999972 -0700
:return: [Actor, int_seconds_since_epoch, int_timezone_offset]"""
actor, epoch, offset = '', '0', '0'
m = _re_actor_epoch.search(line)
if m:
actor, epoch, offset = m.groups()
else:
m = _re_only_actor.search(line)
actor = m.group(1) if m else line or ''
return (Actor._from_string(actor), int(epoch), utctz_to_altz(offset))
def blame(self, rev, file):
"""The blame information for the given file at the given revision.
:parm rev: revision specifier, see git-rev-parse for viable options.
:return:
list: [git.Commit, list: [<line>]]
A list of tuples associating a Commit object with a list of lines that
changed within the given commit. The Commit objects will be given in order
of appearance."""
data = self.git.blame(rev, '--', file, p=True, stdout_as_string=False)
commits = dict()
blames = list()
info = None
keepends = True
for line in data.splitlines(keepends):
try:
line = line.rstrip().decode(defenc)
except UnicodeDecodeError:
firstpart = ''
is_binary = True
else:
# As we don't have an idea when the binary data ends, as it could contain multiple newlines
# in the process. So we rely on being able to decode to tell us what is is.
# This can absolutely fail even on text files, but even if it does, we should be fine treating it
# as binary instead
parts = self.re_whitespace.split(line, 1)
firstpart = parts[0]
is_binary = False
# end handle decode of line
if self.re_hexsha_only.search(firstpart):
# handles
# 634396b2f541a9f2d58b00be1a07f0c358b999b3 1 1 7 - indicates blame-data start
# 634396b2f541a9f2d58b00be1a07f0c358b999b3 2 2 - indicates
# another line of blame with the same data
digits = parts[-1].split(" ")
if len(digits) == 3:
info = {'id': firstpart}
blames.append([None, []])
elif info['id'] != firstpart:
info = {'id': firstpart}
blames.append([commits.get(firstpart), []])
# END blame data initialization
else:
m = self.re_author_committer_start.search(firstpart)
if m:
# handles:
# author Tom Preston-Werner
# author-mail <*****@*****.**>
# author-time 1192271832
# author-tz -0700
# committer Tom Preston-Werner
# committer-mail <*****@*****.**>
# committer-time 1192271832
# committer-tz -0700 - IGNORED BY US
role = m.group(0)
if firstpart.endswith('-mail'):
info["%s_email" % role] = parts[-1]
elif firstpart.endswith('-time'):
info["%s_date" % role] = int(parts[-1])
elif role == firstpart:
info[role] = parts[-1]
# END distinguish mail,time,name
else:
# handle
# filename lib/grit.rb
# summary add Blob
# <and rest>
if firstpart.startswith('filename'):
info['filename'] = parts[-1]
elif firstpart.startswith('summary'):
info['summary'] = parts[-1]
elif firstpart == '':
if info:
sha = info['id']
c = commits.get(sha)
if c is None:
c = Commit(self, hex_to_bin(sha),
author=Actor._from_string(info['author'] + ' ' + info['author_email']),
authored_date=info['author_date'],
committer=Actor._from_string(
info['committer'] + ' ' + info['committer_email']),
committed_date=info['committer_date'],
message=info['summary'])
commits[sha] = c
# END if commit objects needs initial creation
if not is_binary:
if line and line[0] == '\t':
line = line[1:]
else:
# NOTE: We are actually parsing lines out of binary data, which can lead to the
# binary being split up along the newline separator. We will append this to the blame
# we are currently looking at, even though it should be concatenated with the last line
# we have seen.
pass
# end handle line contents
blames[-1][0] = c
blames[-1][1].append(line)
info = {'id': sha}
# END if we collected commit info
# END distinguish filename,summary,rest
# END distinguish author|committer vs filename,summary,rest
# END distinguish hexsha vs other information
return blames
def blame(self, rev, file):
"""The blame information for the given file at the given revision.
:parm rev: revision specifier, see git-rev-parse for viable options.
:return:
list: [git.Commit, list: [<line>]]
A list of tuples associating a Commit object with a list of lines that
changed within the given commit. The Commit objects will be given in order
of appearance."""
data = self.git.blame(rev, '--', file, p=True)
commits = dict()
blames = list()
info = None
for line in data.splitlines(False):
parts = self.re_whitespace.split(line, 1)
firstpart = parts[0]
if self.re_hexsha_only.search(firstpart):
# handles
# 634396b2f541a9f2d58b00be1a07f0c358b999b3 1 1 7 - indicates blame-data start
# 634396b2f541a9f2d58b00be1a07f0c358b999b3 2 2
digits = parts[-1].split(" ")
if len(digits) == 3:
info = {'id': firstpart}
blames.append([None, []])
# END blame data initialization
else:
m = self.re_author_committer_start.search(firstpart)
if m:
# handles:
# author Tom Preston-Werner
# author-mail <*****@*****.**>
# author-time 1192271832
# author-tz -0700
# committer Tom Preston-Werner
# committer-mail <*****@*****.**>
# committer-time 1192271832
# committer-tz -0700 - IGNORED BY US
role = m.group(0)
if firstpart.endswith('-mail'):
info["%s_email" % role] = parts[-1]
elif firstpart.endswith('-time'):
info["%s_date" % role] = int(parts[-1])
elif role == firstpart:
info[role] = parts[-1]
# END distinguish mail,time,name
else:
# handle
# filename lib/grit.rb
# summary add Blob
# <and rest>
if firstpart.startswith('filename'):
info['filename'] = parts[-1]
elif firstpart.startswith('summary'):
info['summary'] = parts[-1]
elif firstpart == '':
if info:
sha = info['id']
c = commits.get(sha)
if c is None:
c = Commit( self, hex_to_bin(sha),
author=Actor._from_string(info['author'] + ' ' + info['author_email']),
authored_date=info['author_date'],
committer=Actor._from_string(info['committer'] + ' ' + info['committer_email']),
committed_date=info['committer_date'],
message=info['summary'])
commits[sha] = c
# END if commit objects needs initial creation
m = self.re_tab_full_line.search(line)
text, = m.groups()
blames[-1][0] = c
blames[-1][1].append( text )
info = None
# END if we collected commit info
# END distinguish filename,summary,rest
# END distinguish author|committer vs filename,summary,rest
# END distinguish hexsha vs other information
return blames
def blame(repo, start_commit, end_commit, filename):
data = repo.git.blame('%s^..%s' % (start_commit, end_commit), '--', filename, p=True)
commits = dict()
blames = list()
info = None
for line in data.splitlines(False):
parts = repo.re_whitespace.split(line, 1)
firstpart = parts[0]
if repo.re_hexsha_only.search(firstpart):
# handles
# 634396b2f541a9f2d58b00be1a07f0c358b999b3 1 1 7 - indicates blame-data start
# 634396b2f541a9f2d58b00be1a07f0c358b999b3 2 2
digits = parts[-1].split(" ")
if len(digits) == 3:
info = {'id': firstpart}
blames.append([None, []])
elif info['id'] != firstpart:
info = {'id': firstpart}
blames.append([commits.get(firstpart), []])
# END blame data initialization
else:
m = repo.re_author_committer_start.search(firstpart)
if m:
# handles:
# author Tom Preston-Werner
# author-mail <*****@*****.**>
# author-time 1192271832
# author-tz -0700
# committer Tom Preston-Werner
# committer-mail <*****@*****.**>
# committer-time 1192271832
# committer-tz -0700 - IGNORED BY US
role = m.group(0)
if firstpart.endswith('-mail'):
info["%s_email" % role] = parts[-1]
elif firstpart.endswith('-time'):
info["%s_date" % role] = int(parts[-1])
elif role == firstpart:
info[role] = parts[-1]
# END distinguish mail,time,name
else:
# handle
# filename lib/grit.rb
# summary add Blob
# <and rest>
if firstpart.startswith('filename'):
info['filename'] = parts[-1]
elif firstpart.startswith('summary'):
info['summary'] = parts[-1]
elif firstpart.startswith('boundary'):
info['boundary'] = True
elif firstpart == '':
if info:
sha = info['id']
c = commits.get(sha)
if c is None:
if info.get('boundary'):
commits[sha] = False
else:
c = repo.CommitCls(
repo,
hex_to_bin(sha),
author=Actor._from_string(info['author'] + ' ' + info['author_email']),
authored_date=info['author_date'],
committer=Actor._from_string(info['committer'] + ' ' + info['committer_email']),
committed_date=info['committer_date'],
message=info['summary']
)
commits[sha] = c
if c is not False:
# END if commit objects needs initial creation
m = repo.re_tab_full_line.search(line)
text, = m.groups()
blames[-1][0] = c
blames[-1][1].append(text)
info = { 'id' : sha }
# END if we collected commit info
# END distinguish filename,summary,rest
# END distinguish author|committer vs filename,summary,rest
# END distinguish hexsha vs other information
for commit, lines in blames:
if commit is not None:
yield commit, lines
def test_str_should_alias_name(self):
a = Actor._from_string("Michael Trier <*****@*****.**>")
assert a.name == str(a)
def test_should_display_representation(self):
a = Actor._from_string("Michael Trier <*****@*****.**>")
assert '<git.Actor "Michael Trier <*****@*****.**>">' == repr(a)
def test_from_string_should_handle_just_name(self):
a = Actor._from_string("Michael Trier")
assert "Michael Trier" == a.name
assert None == a.email
def test_str_should_alias_name(self):
a = Actor._from_string("Michael Trier <*****@*****.**>")
assert a.name == str(a)
def blame(self, rev, file):
"""The blame information for the given file at the given revision.
:parm rev: revision specifier, see git-rev-parse for viable options.
:return:
list: [git.Commit, list: [<line>]]
A list of tuples associating a Commit object with a list of lines that
changed within the given commit. The Commit objects will be given in order
of appearance."""
data = self.git.blame(rev, '--', file, p=True, stdout_as_string=False)
commits = dict()
blames = list()
info = None
keepends = True
for line in data.splitlines(keepends):
try:
line = line.rstrip().decode(defenc)
except UnicodeDecodeError:
firstpart = ''
is_binary = True
else:
# As we don't have an idea when the binary data ends, as it could contain multiple newlines
# in the process. So we rely on being able to decode to tell us what is is.
# This can absolutely fail even on text files, but even if it does, we should be fine treating it
# as binary instead
parts = self.re_whitespace.split(line, 1)
firstpart = parts[0]
is_binary = False
# end handle decode of line
if self.re_hexsha_only.search(firstpart):
# handles
# 634396b2f541a9f2d58b00be1a07f0c358b999b3 1 1 7 - indicates blame-data start
# 634396b2f541a9f2d58b00be1a07f0c358b999b3 2 2 - indicates
# another line of blame with the same data
digits = parts[-1].split(" ")
if len(digits) == 3:
info = {'id': firstpart}
blames.append([None, []])
elif info['id'] != firstpart:
info = {'id': firstpart}
blames.append([commits.get(firstpart), []])
# END blame data initialization
else:
m = self.re_author_committer_start.search(firstpart)
if m:
# handles:
# author Tom Preston-Werner
# author-mail <*****@*****.**>
# author-time 1192271832
# author-tz -0700
# committer Tom Preston-Werner
# committer-mail <*****@*****.**>
# committer-time 1192271832
# committer-tz -0700 - IGNORED BY US
role = m.group(0)
if firstpart.endswith('-mail'):
info["%s_email" % role] = parts[-1]
elif firstpart.endswith('-time'):
info["%s_date" % role] = int(parts[-1])
elif role == firstpart:
info[role] = parts[-1]
# END distinguish mail,time,name
else:
# handle
# filename lib/grit.rb
# summary add Blob
# <and rest>
if firstpart.startswith('filename'):
info['filename'] = parts[-1]
elif firstpart.startswith('summary'):
info['summary'] = parts[-1]
elif firstpart == '':
if info:
sha = info['id']
c = commits.get(sha)
if c is None:
c = Commit(
self,
hex_to_bin(sha),
author=Actor._from_string(
info['author'] + ' ' +
info['author_email']),
authored_date=info['author_date'],
committer=Actor._from_string(
info['committer'] + ' ' +
info['committer_email']),
committed_date=info['committer_date'],
message=info['summary'])
commits[sha] = c
# END if commit objects needs initial creation
if not is_binary:
if line and line[0] == '\t':
line = line[1:]
else:
# NOTE: We are actually parsing lines out of binary data, which can lead to the
# binary being split up along the newline separator. We will append this to the blame
# we are currently looking at, even though it should be concatenated with the last line
# we have seen.
pass
# end handle line contents
blames[-1][0] = c
blames[-1][1].append(line)
info = {'id': sha}
# END if we collected commit info
# END distinguish filename,summary,rest
# END distinguish author|committer vs filename,summary,rest
# END distinguish hexsha vs other information
return blames
def test_from_string_should_handle_just_name(self):
a = Actor._from_string("Michael Trier")
assert "Michael Trier" == a.name
assert None == a.email
def blame(self, rev, file):
"""The blame information for the given file at the given revision.
:parm rev: revision specifier, see git-rev-parse for viable options.
:return:
list: [git.Commit, list: [<line>]]
A list of tuples associating a Commit object with a list of lines that
changed within the given commit. The Commit objects will be given in order
of appearance."""
data = self.git.blame(rev, '--', file, p=True)
commits = dict()
blames = list()
info = None
for line in data.splitlines(False):
parts = self.re_whitespace.split(line, 1)
firstpart = parts[0]
if self.re_hexsha_only.search(firstpart):
# handles
# 634396b2f541a9f2d58b00be1a07f0c358b999b3 1 1 7 - indicates blame-data start
# 634396b2f541a9f2d58b00be1a07f0c358b999b3 2 2
digits = parts[-1].split(" ")
if len(digits) == 3:
info = {'id': firstpart}
blames.append([None, []])
# END blame data initialization
else:
m = self.re_author_committer_start.search(firstpart)
if m:
# handles:
# author Tom Preston-Werner
# author-mail <*****@*****.**>
# author-time 1192271832
# author-tz -0700
# committer Tom Preston-Werner
# committer-mail <*****@*****.**>
# committer-time 1192271832
# committer-tz -0700 - IGNORED BY US
role = m.group(0)
if firstpart.endswith('-mail'):
info["%s_email" % role] = parts[-1]
elif firstpart.endswith('-time'):
info["%s_date" % role] = int(parts[-1])
elif role == firstpart:
info[role] = parts[-1]
# END distinguish mail,time,name
else:
# handle
# filename lib/grit.rb
# summary add Blob
# <and rest>
if firstpart.startswith('filename'):
info['filename'] = parts[-1]
elif firstpart.startswith('summary'):
info['summary'] = parts[-1]
elif firstpart == '':
if info:
sha = info['id']
c = commits.get(sha)
if c is None:
c = Commit(
self,
hex_to_bin(sha),
author=Actor._from_string(
info['author'] + ' ' +
info['author_email']),
authored_date=info['author_date'],
committer=Actor._from_string(
info['committer'] + ' ' +
info['committer_email']),
committed_date=info['committer_date'],
message=info['summary'])
commits[sha] = c
# END if commit objects needs initial creation
m = self.re_tab_full_line.search(line)
text, = m.groups()
blames[-1][0] = c
blames[-1][1].append(text)
info = None
# END if we collected commit info
# END distinguish filename,summary,rest
# END distinguish author|committer vs filename,summary,rest
# END distinguish hexsha vs other information
return blames
def test_should_display_representation(self):
a = Actor._from_string("Michael Trier <*****@*****.**>")
assert '<git.Actor "Michael Trier <*****@*****.**>">' == repr(a)
