def test_string_lines(self):
f = DEFAULT_STREAM_FACTORY
s = f.from_string('line 1\nline 2\nline 3\n')
(l, s) = s_line(s, False)
assert l == 'line 1\n', l
(l, _) = s_line(s, False)
assert l == 'line 2\n', repr(l)
locn = s_fmt(s, '{location}')
assert locn == 'line 2, character 1', locn
sl = s_stream(s, l)
(_, sl) = s_next(sl, count=2)
locn = s_fmt(sl, '{location}')
assert locn == 'line 2, character 3', locn
def test_string_lines(self):
f = DEFAULT_STREAM_FACTORY
s = f.from_string('line 1\nline 2\nline 3\n')
(l, s) = s_line(s, False)
assert l == 'line 1\n', l
(l, _) = s_line(s, False)
assert l == 'line 2\n', repr(l)
locn = s_fmt(s, '{location}')
assert locn == 'line 2, character 1', locn
sl = s_stream(s, l)
(_, sl) = s_next(sl, count=2)
locn = s_fmt(sl, '{location}')
assert locn == 'line 2, character 3', locn
def _tokens(self, stream, max):
'''
Generate tokens, on demand.
'''
try:
id_ = s_id(stream)
while not s_empty(stream):
# avoid conflicts between tokens
id_ += 1
try:
(terminals, match, next_stream) = \
self.t_regexp.match(stream)
self._debug(fmt('Token: {0!r} {1!r} {2!s}',
terminals, match, s_debug(stream)))
yield (terminals, s_stream(stream, match, max=max, id_=id_))
except TypeError:
(terminals, _size, next_stream) = \
self.s_regexp.size_match(stream)
self._debug(fmt('Space: {0!r} {1!s}',
terminals, s_debug(stream)))
stream = next_stream
except TypeError:
raise RuntimeLexerError(
s_fmt(stream,
'No token for {rest} at {location} of {text}.'))
def _tokens(self, stream, max):
'''
Generate tokens, on demand.
'''
try:
id_ = s_id(stream)
while not s_empty(stream):
# avoid conflicts between tokens
id_ += 1
try:
(terminals, match, next_stream) = \
self.t_regexp.match(stream)
self._debug(
fmt('Token: {0!r} {1!r} {2!s}', terminals, match,
s_debug(stream)))
yield (terminals, s_stream(stream, match, max=max,
id_=id_))
except TypeError:
(terminals, _size, next_stream) = \
self.s_regexp.size_match(stream)
self._debug(
fmt('Space: {0!r} {1!s}', terminals, s_debug(stream)))
stream = next_stream
except TypeError:
raise RuntimeLexerError(
s_fmt(stream, 'No token for {rest} at {location} of {text}.'))
def test_all(self):
lines = iter(['first line', 'second line', 'third line'])
f = DEFAULT_STREAM_FACTORY
s1 = f(lines)
# just created
assert not s_empty(s1)
# get first line
(l1, s2) = s_line(s1, False)
assert 'first line' == l1, l1
# get first character of next line
(c21, s21) = s_next(s2)
assert c21 == 's', c21
# and test fmtting
locn = s_fmt(s21, '{location}: {rest}')
assert locn == "line 2, character 2: 'econd line'", locn
# then get rest of second line
(c22, s3) = s_next(s21, count=len('econd line'))
assert c22 == 'econd line', c22
d = s_debug(s21)
assert d == "1:'e'", d
# and move on to third line
(c31, s31) = s_next(s3)
assert c31 == 't', c31
(c32, s32) = s_next(s31)
assert c32 == 'h', c32
# now try branching (think tokens) at line 1
s10 = s_stream(s2, l1)
(l1, s20) = s_line(s10, False)
assert l1 == 'first line', l1
assert not s_empty(s20)
(c1, s11) = s_next(s10)
assert c1 == 'f', c1
d = s_debug(s11)
assert d == "1:'i'", d
# finally look at max depth (which was after 'h' in third line)
m = s_deepest(s1)
locn = s_fmt(m, '{location}: {rest}')
assert locn == "line 3, character 3: 'ird line'", locn
def _tokens(self, stream, max):
'''
Generate tokens, on demand.
'''
id_ = s_id(stream)
try:
while not s_empty(stream):
# caches for different tokens with same contents differ
id_ += 1
(line, next_stream) = s_line(stream, False)
line_stream = s_stream(stream, line)
size = 0
# if we use blocks, match leading space
if self.blocks:
try:
(_, size, _) = self.s_regexp.size_match(line_stream)
except TypeError:
pass
# this will be empty (size=0) if blocks unused
(indent, next_line_stream) = s_next(line_stream, count=size)
indent = indent.replace('\t', self._tab)
yield ((START,),
s_stream(line_stream, indent, id_=id_, max=max))
line_stream = next_line_stream
while not s_empty(line_stream):
id_ += 1
try:
(terminals, match, next_line_stream) = \
self.t_regexp.match(line_stream)
yield (terminals, s_stream(line_stream, match,
max=max, id_=id_))
except TypeError:
(terminals, _size, next_line_stream) = \
self.s_regexp.size_match(line_stream)
line_stream = next_line_stream
id_ += 1
yield ((END,),
s_stream(line_stream, '', max=max, id_=id_))
stream = next_stream
except TypeError:
raise RuntimeLexerError(
s_fmt(stream,
'No token for {rest} at {location} of {text}.'))
def match(support, stream):
# we increment id so that different strings (which might overlap or
# be contiguous) don't affect each other's memoisation (the hash key
# is based on offset and ('one past the') end of one column can have
# the same offset as the start of the next).
id_ = s_id(stream)
# extract a line
(line, next_stream) = s_line(stream, False)
line_stream = s_stream(stream, line)
results = []
for ((left, right), matcher) in zip(indices, matchers):
id_ += 1
# extract the location in the line
(_, left_aligned_line_stream) = s_next(line_stream, count=left)
(word, _) = s_next(left_aligned_line_stream, count=right-left)
support._debug(fmt('Columns {0}-{1} {2!r}', left, right, word))
word_stream = s_stream(left_aligned_line_stream, word, id_=id_)
# do the match
support._debug(s_fmt(word_stream, 'matching {rest}'))
(result, _) = yield matcher._match(word_stream)
results.extend(result)
support._debug(repr(results))
yield (results, next_stream)
def match(support, stream):
# we increment id so that different strings (which might overlap or
# be contiguous) don't affect each other's memoisation (the hash key
# is based on offset and ('one past the') end of one column can have
# the same offset as the start of the next).
id_ = s_id(stream)
# extract a line
(line, next_stream) = s_line(stream, False)
line_stream = s_stream(stream, line)
results = []
for ((left, right), matcher) in zip(indices, matchers):
id_ += 1
# extract the location in the line
(_, left_aligned_line_stream) = s_next(line_stream, count=left)
(word, _) = s_next(left_aligned_line_stream, count=right - left)
support._debug(fmt('Columns {0}-{1} {2!r}', left, right, word))
word_stream = s_stream(left_aligned_line_stream, word, id_=id_)
# do the match
support._debug(s_fmt(word_stream, 'matching {rest}'))
(result, _) = yield matcher._match(word_stream)
results.extend(result)
support._debug(repr(results))
yield (results, next_stream)
def __init__(self, stream):
super(FullFirstMatchException, self).__init__(
s_fmt(s_deepest(stream),
'The match failed in {filename} at {rest} ({location}).'))
def fmt(self, state, template, prefix='', kargs=None):
line_stream = state_to_line_stream(state)
return s_fmt(line_stream, template, prefix=prefix, kargs=kargs)
def fmt(self, state, template, prefix='', kargs=None):
line_stream = state_to_line_stream(state)
return s_fmt(line_stream, template, prefix=prefix, kargs=kargs)
def __init__(self, stream):
super(FullFirstMatchException, self).__init__(
s_fmt(s_deepest(stream),
'The match failed in {filename} at {rest} ({location}).'))
