Skip to content

Commit

Permalink
Fix some private functions
Browse files Browse the repository at this point in the history 
- Fix some cmark emphasis detection and removal functions
  See issue #25
  • Loading branch information
@frnmst
frnmst committed Dec 18, 2022
1 parent d941bf5 commit 4b758ec
Show file tree
Hide file tree
Showing 2 changed files with 140 additions and 28 deletions.
152 changes: 129 additions & 23 deletions md_toc/cmark/inlines_c.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -191,7 +191,7 @@ def __init__(self):
def scroll(self):
r"""Print the list."""
print(self.start)
print("===")
print('===')
x = self.start
while x is not None:
print(x)
Expand All @@ -212,9 +212,9 @@ def _cmark_make_literal(subj: _cmarkSubject, t: int, start_column: int,
e.mem = copy.deepcopy(subj.mem)
e.type = t
e.start_line = e.end_line = subj.line
# columns are (NOT) 1 based. FIXME
e.start_column: int = start_column + subj.column_offset + subj.block_offset
e.end_column: int = end_column + subj.column_offset + subj.block_offset
# columns are 1 based.
e.start_column: int = start_column + 1 + subj.column_offset + subj.block_offset
e.end_column: int = end_column + 1 + subj.column_offset + subj.block_offset
return e


Expand Down Expand Up @@ -406,14 +406,23 @@ def _cmark_advance(subj: _cmarkSubject):
subj.pos += 1


def _cmark_skip_spaces(subj: _cmarkSubject) -> bool:
skipped: bool = False
while chr(_cmark_peek_char(subj)) == ' ' or chr(
_cmark_peek_char(subj)) == '\t':
_cmark_advance(subj)
skipped = True
return skipped


# 0.29, 0.30
def _cmark_skip_line_end(subj: _cmarkSubject) -> bool:
seen_line_end_char: bool = False

if _cmark_peek_char(subj) == '\r':
if chr(_cmark_peek_char(subj)) == '\r':
_cmark_advance(subj)
seen_line_end_char = True
if _cmark_peek_char(subj) == '\n':
if chr(_cmark_peek_char(subj)) == '\n':
_cmark_advance(subj)
seen_line_end_char = True
return seen_line_end_char or _cmark_is_eof(subj)
Expand Down Expand Up @@ -781,6 +790,66 @@ def _cmark_handle_delim(subj: _cmarkSubject,
return inl_text


# 0.30
# Assumes we have a hyphen at the current position.
def _cmark_handle_hyphen(subj: _cmarkSubject, smart: bool) -> _cmarkCmarkNode:
startpos: int = subj.pos

_cmark_advance(subj)

if not smart or chr(_cmark_peek_char(subj)) != '-':
return _cmark_make_str(subj, subj.pos - 1, subj.pos - 1,
_cmark_cmark_chunk_literal('-'))

while smart and chr(_cmark_peek_char(subj)) == '-':
_cmark_advance(subj)

numhyphens: int = subj.pos - startpos
en_count: int = 0
em_count: int = 0
i: int
buf: _cmarkCmarkStrbuf = _cmark_CMARK_BUF_INIT(subj.mem)

if numhyphens % 3 == 0: # if divisible by 3, use all em dashes
em_count = numhyphens / 3
elif numhyphens % 2 == 0: # if divisible by 2, use all en dashes
en_count = numhyphens / 2
elif numhyphens % 3 == 2: # use one en dash at end
en_count = 1
em_count = (numhyphens - 2) / 3
else: # use two en dashes at the end
en_count = 2
em_count = (numhyphens - 4) / 3

for i in range(em_count, 0, step=-1):
_cmark_cmark_strbuf_puts(buf, md_parser['cmark']['generic']['EMDASH'])

for i in range(en_count, 0, step=-1):
_cmark_cmark_strbuf_puts(buf, md_parser['cmark']['generic']['ENDASH'])

return _cmark_make_str_from_buf(subj, startpos, subj.pos - 1, buf)


# 0.30
# Assumes we have a period at the current position.
def _cmark_handle_period(subj: _cmarkSubject, smart: bool) -> _cmarkCmarkNode:
_cmark_advance(subj)
if smart and chr(_cmark_peek_char(subj)) == '.':
_cmark_advance(subj)
if chr(_cmark_peek_char(subj)) == '.':
_cmark_advance(subj)
return _cmark_make_str(
subj, subj.pos - 3, subj.pos - 1,
_cmark_cmark_chunk_literal(
md_parser['cmark']['generic']['ELLIPSES']))
else:
return _cmark_make_str(subj, subj.pos - 2, subj.pos - 1,
_cmark_cmark_chunk_literal('..'))
else:
return _cmark_make_str(subj, subj.pos - 1, subj.pos - 1,
_cmark_cmark_chunk_literal('.'))


# 0.30
def _cmark_process_emphasis(subj: _cmarkSubject, stack_bottom: _cmarkDelimiter,
ignore: list) -> list:
Expand Down Expand Up @@ -884,7 +953,8 @@ def _cmark_process_emphasis(subj: _cmarkSubject, stack_bottom: _cmarkDelimiter,
# 0.29, 0.30
def _cmark_remove_emph(subj: _cmarkSubject, opener: _cmarkDelimiter,
closer: _cmarkDelimiter, ignore: list):
# This function refers to S_insert_emph()
# This function replaces to S_insert_emph(). Some code is common
# to that function.
delim: _cmarkDelimiter
tmp_delim: _cmarkDelimiter
use_delims: int
Expand Down Expand Up @@ -945,10 +1015,10 @@ def _cmark_remove_emph(subj: _cmarkSubject, opener: _cmarkDelimiter,
#############

# Custom variables and computations.
opener_relative_start = opener_inl.end_column - use_delims + 1 - opener.offset
opener_relative_end = opener_inl.end_column + 1 - opener.offset
closer_relative_start = closer_inl.start_column + closer.offset
closer_relative_end = closer_inl.start_column + use_delims + closer.offset
opener_relative_start = opener_inl.end_column - use_delims - opener.offset
opener_relative_end = opener_inl.end_column - opener.offset
closer_relative_start = closer_inl.start_column + closer.offset - 1
closer_relative_end = closer_inl.start_column + use_delims + closer.offset - 1

ignore.append(range(opener_relative_start, opener_relative_end))
ignore.append(range(closer_relative_start, closer_relative_end))
Expand Down Expand Up @@ -1138,12 +1208,12 @@ def _cmark_link_label(subj: _cmarkSubject, raw_label: _cmarkCmarkChunk) -> int:
no_match: bool = False

# advance past [
if _cmark_peek_char(subj) == '[':
if chr(_cmark_peek_char(subj)) == '[':
_cmark_advance(subj)
else:
return 0

c = _cmark_peek_char(subj)
c = chr(_cmark_peek_char(subj))
while c and c != '[' and c != ']':
if c == '\\':
_cmark_advance(subj)
Expand All @@ -1161,7 +1231,7 @@ def _cmark_link_label(subj: _cmarkSubject, raw_label: _cmarkCmarkChunk) -> int:
if length > md_parser['cmark']['link']['max chars label'] + 1:
no_match = True

c = _cmark_peek_char(subj)
c = chr(_cmark_peek_char(subj))

if no_match:
subj.pos = startpos # rewind
Expand Down Expand Up @@ -1294,7 +1364,7 @@ def _cmark_handle_close_bracket(subj: _cmarkSubject,

in_inline_link: bool = False
# First, look for an inline link.
if _cmark_peek_char(subj) == ord('('):
if chr(_cmark_peek_char(subj)) == '(':
sps = _cmark_scan_spacechars(subj.input, subj.pos + 1)
if sps > -1:
n, url_chunk = _cmark_manual_scan_link_url(subj.input,
Expand All @@ -1316,7 +1386,7 @@ def _cmark_handle_close_bracket(subj: _cmarkSubject,

endall = endtitle + _cmark_scan_spacechars(subj.input, endtitle)

if _cmark_peek_at(subj, endall) == ord(')'):
if chr(_cmark_peek_at(subj, endall)) == ')':
subj.pos = endall + 1

title_chunk = _cmark_cmark_chunk_dup(subj.input, starttitle,
Expand Down Expand Up @@ -1409,6 +1479,31 @@ def _cmark_handle_close_bracket(subj: _cmarkSubject,
_cmark_cmark_chunk_literal(']'))


# 0.30
# Parse a hard or soft linebreak, returning an inline.
# Assumes the subject has a cr or newline at the current position.
def _cmark_handle_newline(subj: _cmarkSubject) -> _cmarkCmarkNode:
nlpos: int = subj.pos
# skip over cr, crlf, or lf:
if chr(_cmark_peek_at(subj, subj.pos)) == '\r':
_cmark_advance(subj)
if chr(_cmark_peek_at(subj, subj.pos)) == '\n':
_cmark_advance(subj)

subj.line += 1

# FIXME TODO: The next instruction messes up some tests!
# subj.column_offset = - subj.pos

# skip spaces at beginning of line
_cmark_skip_spaces(subj)
if (nlpos > 1 and chr(_cmark_peek_at(subj, nlpos - 1)) == ' '
and chr(_cmark_peek_at(subj, nlpos - 2)) == ' '):
return _cmark_make_linebreak(subj.mem)
else:
return _cmark_make_softbreak(subj.mem)


# 0.29, 0.30
def _cmark_subject_find_special_char(subj: _cmarkSubject, options: int) -> int:
# "\r\n\\`&_*[]<!"
Expand Down Expand Up @@ -1972,24 +2067,28 @@ def _cmark_parse_inline(subj: _cmarkSubject,
if c == 0:
return 0
elif chr(c) in ['\r', '\n']:
_cmark_advance(subj)
# TODO
new_inl = _cmark_handle_newline(subj)
elif chr(c) in ['`']:
new_inl = _cmark_handle_backticks(subj, options)
elif chr(c) in ['\\']:
new_inl = _cmark_handle_backslash(subj)
elif chr(c) in ['&']:
_cmark_advance(subj)
# TODO
elif chr(c) in ['<']:
new_inl = _cmark_handle_pointy_brace(subj, options)
elif chr(c) in ['*', '_', '\'', '"']:
new_inl = _cmark_handle_delim(
subj, chr(c),
(options & md_parser['cmark']['generic']['CMARK_OPT_SMART']) != 0)
elif chr(c) in ['-']:
_cmark_advance(subj)
# TODO
new_inl = _cmark_handle_hyphen(
subj,
(options & md_parser['cmark']['generic']['CMARK_OPT_SMART']) != 0)
elif chr(c) in ['.']:
_cmark_advance(subj)
# TODO
new_inl = _cmark_handle_period(
subj,
(options & md_parser['cmark']['generic']['CMARK_OPT_SMART']) != 0)
elif chr(c) in ['[']:
_cmark_advance(subj)
new_inl = _cmark_make_str(subj, subj.pos - 1, subj.pos - 1,
Expand All @@ -1999,7 +2098,14 @@ def _cmark_parse_inline(subj: _cmarkSubject,
_cmark_handle_close_bracket(subj, ignore)
elif chr(c) in ['!']:
_cmark_advance(subj)
# TODO
if _cmark_peek_char(subj) == '[':
_cmark_advance(subj)
new_inl = _cmark_make_str(subj, subj.pos - 2, subj.pos - 1,
_cmark_cmark_chunk_literal('!['))
_cmark_push_bracket(subj, True, new_inl)
else:
new_inl = _cmark_make_str(subj, subj.pos - 1, subj.pos - 1,
_cmark_cmark_chunk_literal('!'))
else:
endpos = _cmark_subject_find_special_char(subj, options)
contents = _cmark_cmark_chunk_dup(subj.input, subj.pos,
Expand Down
16 changes: 11 additions & 5 deletions md_toc/tests/tests.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -1940,7 +1940,8 @@ def test_remove_emphasis(self):
# self.assertEqual(api.remove_emphasis(r'[link *foo **bar** `#`*](/uri)'), r'[link foo bar `#`](/uri)')

# Example 516 [Commonmark 0.30].
# self.assertEqual(api.remove_emphasis(r'[![moon](moon.jpg)](/uri)'), r'[![moon](moon.jpg)](/uri)')
self.assertEqual(api.remove_emphasis(r'[![moon](moon.jpg)](/uri)'),
r'[![moon](moon.jpg)](/uri)')

# Example 517 [Commonmark 0.30].
self.assertEqual(api.remove_emphasis(r'*[foo* [bar](/uri)](/uri)'),
Expand All @@ -1950,7 +1951,8 @@ def test_remove_emphasis(self):
# self.assertEqual(api.remove_emphasis(r'[foo *[bar [baz](/uri)](/uri)*](/uri)'), r'[foo [bar [baz](/uri)](/uri)](/uri)')

# Example 519 [Commonmark 0.30].
# self.assertEqual(api.remove_emphasis(r'![[[foo](uri1)](uri2)](uri3)'), r'![[[foo](uri1)](uri2)](uri3)')
self.assertEqual(api.remove_emphasis(r'![[[foo](uri1)](uri2)](uri3)'),
r'![[[foo](uri1)](uri2)](uri3)')

# Example 520 [Commonmark 0.30].
self.assertEqual(api.remove_emphasis(r'*[foo*](/uri)'),
Expand All @@ -1965,13 +1967,17 @@ def test_remove_emphasis(self):
r'foo [bar baz]')

# Example 523 [Commonmark 0.30].
# self.assertEqual(api.remove_emphasis(r'[foo <bar attr="](baz)">'), r'[foo <bar attr="](baz)">')
self.assertEqual(api.remove_emphasis(r'[foo <bar attr="](baz)">'),
r'[foo <bar attr="](baz)">')

# Example 524 [Commonmark 0.30].
# self.assertEqual(api.remove_emphasis(r'[foo`](/uri)`'), r'[foo`](/uri)`')
self.assertEqual(api.remove_emphasis(r'[foo`](/uri)`'),
r'[foo`](/uri)`')

# Example 525 [Commonmark 0.30].
# self.assertEqual(api.remove_emphasis(r'[foo<http://example.com/?search=](uri)>'), r'[foo<http://example.com/?search=](uri)>')
self.assertEqual(
api.remove_emphasis(r'[foo<http://example.com/?search=](uri)>'),
r'[foo<http://example.com/?search=](uri)>')

# Possibly all the way up to 592 [Commonmark 0.30]?

Expand Down

0 comments on commit 4b758ec

Please sign in to comment.