def tokens(self, text):
xhtmlsaxhandler = XHTMLSaxHandler(styles=self.styles, tags=self.tags)
xhtmlsaxerrhandler = XHTMLSaxErrorHandler(self, self.ignore)
# next conversion seems bug in python (why need bytes, not str?!)
if type(text) is str:
parsetext = core.strtobytes23(text)
else:
parsetext = text
sax.parseString(parsetext, xhtmlsaxhandler, xhtmlsaxerrhandler)
text = ''.join(xhtmlsaxhandler.text)
#_inlcode_re = r'@start_inline@(.*?)@end_inline@'
_blkcode_re = r'@start_block@(.*?)@end_block@'
#inlcode_re = re.compile(_inlcode_re, re.MULTILINE)
blkcode_re = re.compile(_blkcode_re, re.DOTALL | re.MULTILINE)
cmds = core.Cmd.syntax.findtokens('cmddef', text)
#inlcodes = inlcode_re.finditer(text)
blkcodes = blkcode_re.finditer(text)
tokens = []
for m in cmds:
token = core.CmdToken(
XHTMLSaxHandler._decode_spec_chars(m.group(1)), m.start(0),
m.end(0))
tokens.append(token)
#for m in inlcodes:
# # XXX break line in HTML is coded with paragraph styling and this linearizing
# # does not help
# tokentext = XHTMLSaxHandler._decode_spec_chars(m.group(1))
# tokentext = core.InlCodeToken.linearize(tokentext)
# token = core.InlCodeToken(tokentext, m.start(0), m.end(0))
# tokens.append(token)
for m in blkcodes:
tokentext = m.group(1).lstrip('\n').rstrip('\n ')
tokentext = core.deltextindent(tokentext)
token = core.BlkCodeToken(
XHTMLSaxHandler._decode_spec_chars(tokentext), m.start(0),
m.end(0))
tokens.append(token)
tokens.sort(key=lambda tok: tok.start)
tokens.append(core.EndToken(None))
return tokens
def tokens(self, text):
xhtmlsaxhandler = XHTMLSaxHandler(styles=self.styles, tags=self.tags)
xhtmlsaxerrhandler = XHTMLSaxErrorHandler(self, self.ignore)
# next conversion seems bug in python (why need bytes, not str?!)
if type(text) is str:
parsetext = core.strtobytes23(text)
else:
parsetext = text
sax.parseString(parsetext, xhtmlsaxhandler, xhtmlsaxerrhandler)
text = ''.join(xhtmlsaxhandler.text)
#_inlcode_re = r'@start_inline@(.*?)@end_inline@'
_blkcode_re = r'@start_block@(.*?)@end_block@'
#inlcode_re = re.compile(_inlcode_re, re.MULTILINE)
blkcode_re = re.compile(_blkcode_re, re.DOTALL|re.MULTILINE)
cmds = core.Cmd.syntax.findtokens('cmddef', text)
#inlcodes = inlcode_re.finditer(text)
blkcodes = blkcode_re.finditer(text)
tokens = []
for m in cmds:
token = core.CmdToken(XHTMLSaxHandler._decode_spec_chars(m.group(1)), m.start(0), m.end(0))
tokens.append(token)
#for m in inlcodes:
# # XXX break line in HTML is coded with paragraph styling and this linearizing
# # does not help
# tokentext = XHTMLSaxHandler._decode_spec_chars(m.group(1))
# tokentext = core.InlCodeToken.linearize(tokentext)
# token = core.InlCodeToken(tokentext, m.start(0), m.end(0))
# tokens.append(token)
for m in blkcodes:
tokentext = m.group(1).lstrip('\n').rstrip('\n ')
tokentext = core.deltextindent(tokentext)
token = core.BlkCodeToken(XHTMLSaxHandler._decode_spec_chars(tokentext), m.start(0), m.end(0))
tokens.append(token)
tokens.sort(key=lambda tok: tok.start)
tokens.append(core.EndToken(None))
return tokens
def tokens(self, text):
"""Returns tokens
>>> text = \
'\\\\documentclass[12pt]{article}\\n' \
'\\\\usepackage{amsmath}\\n' \
'\\\\title{\\\\LaTeX}\\n' \
'\\\\date{}\\n' \
'\\\\begin{document}\\n' \
' \\\\maketitle\\n' \
' \\\\LaTeX{} is a document preparation system for the \\\\TeX{}\\n' \
'\\n' \
' Testing of negative value <<isneg>>: \\\\verb#if a < 0#. Signature will\\n' \
' be <<fn.abs.decl>>: \\\\verb!int abs(int a)!. And now:\\n' \
' function absolute <<fn.abs>>:\\n' \
'\\n' \
' \\\\begin{verbatim}\\n' \
' <<=fn.abs.decl>>\\n' \
' if (<<=isneg, x:a>>) return -a;\\n' \
' else return a;\\n' \
' }\\n' \
' \\\\end{verbatim}\\n' \
'\\n' \
' % This is a comment; it will not be shown in the final output.\\n' \
' % The following shows a little of the typesetting power of LaTeX:\\n' \
' \\\\begin{align}\\n' \
' m &= \\\\frac{m_0}{\\\\sqrt{1-\\\\frac{v^2}{c^2}}}\\n' \
' \\\\end{align}\\n' \
'\\\\end{document}'
>>> p = TeXParser()
>>> toks = p.tokens(text)
>>> [tok.__class__.__name__ for tok in toks]
['CmdToken', 'InlCodeToken', 'CmdToken', 'InlCodeToken', 'CmdToken', 'BlkCodeToken', 'EndToken']
>>> toks[0].text
'isneg'
>>> toks[1].text
'if a < 0'
>>> toks[2].text
'fn.abs.decl'
>>> toks[3].text == 'int abs(int a)'
True
>>> toks[5].text.startswith('<<=fn.abs.decl>>')
True
>>> toks[5].text.endswith('}')
True
"""
_inlcmds = '|'.join(re.escape(s) for s in TeXParser.inlcmds)
# text inside inline chunk
_inltext = r'{(?P<code1>.+?)}|(?P<s>[+#!])(?P<code2>.+?)(?P=s)'
_blkcmds = '|'.join(re.escape(s) for s in TeXParser.blkcmds)
_inlcode_re = r'\\(%s)%s' % (_inlcmds, _inltext)
_blkcode_re = r'\\begin(\[.+?\])?{(%s)}(?P<code>.*?)\\end{(%s)}' % (
_blkcmds, _blkcmds)
inlcode_re = re.compile(_inlcode_re)
blkcode_re = re.compile(_blkcode_re, re.DOTALL)
cmds = core.Cmd.syntax.findtokens('cmddef', text)
inlcodes = inlcode_re.finditer(text)
blkcodes = blkcode_re.finditer(text)
tokens = []
for m in cmds:
token = core.CmdToken(m.group(1), m.start(0), m.end(0))
tokens.append(token)
for m in inlcodes:
groups = m.groupdict()
g = 'code1' if groups['code1'] else 'code2'
tokentext = core.InlCodeToken.linearize(m.group(g))
token = core.InlCodeToken(tokentext, m.start(0), m.end(0))
tokens.append(token)
for m in blkcodes:
# find each block and replace first left indent with ''
tokentext = m.group('code').lstrip('\n').rstrip('\n ')
tokentext = core.deltextindent(tokentext)
token = core.BlkCodeToken(tokentext, m.start(0), m.end(0))
tokens.append(token)
tokens.sort(key=lambda tok: tok.start)
tokens.append(core.EndToken(None))
return tokens
def tokens(self, text):
"""Returns tokens
>>> text = \
'Example of Literate Programming in Asciidoc\\n' \
'===========================================\\n\\n' \
\
'Code 1\\n' \
'------\\n\\n' \
\
'Test if variable is negative looks like <<c.isneg>>: +if a < 0+.\\n' \
'So, we can write absolute function <<c.fun>>\\n' \
'[source, python]\\n' \
'----\\n' \
' def fun(x):\\n' \
' <<=c.isneg,a:v>>:\\n' \
' a += 100\\n' \
' return -a\\n\\n' \
'----\\n\\n' \
'And <<c.sum>>\\n\\n' \
'----\\n' \
' def sum(x, y):\\n\\n' \
\
' return x+y\\n\\n' \
' x += 1\\n' \
'----'
>>> p = AsciiDocParser()
>>> toks = p.tokens(text)
>>> [tok.__class__.__name__ for tok in toks]
['CmdToken', 'InlCodeToken', 'CmdToken', 'BlkCodeToken', 'CmdToken', 'BlkCodeToken', 'EndToken']
>>> toks[0].text
'c.isneg'
>>> toks[1].text
'if a < 0'
>>> toks[2].text
'c.fun'
>>> toks[3].text.startswith('def fun')
True
>>> toks[3].text.endswith(' return -a')
True
>>> toks[5].text.startswith('def sum')
True
>>> toks[5].text.endswith('x += 1')
True
"""
_inlcode_re = r'\+([^\n\+]+)\+'
_blkcode_re = r'(?:\[source[^]]*?]\n|\n\n)-{4,}(?P<code>.*?)\n-{4,}'
inlcode_re = re.compile(_inlcode_re)
blkcode_re = re.compile(_blkcode_re, re.DOTALL)
cmds = core.Cmd.syntax.findtokens('cmddef', text)
inlcodes = inlcode_re.finditer(text)
blkcodes = blkcode_re.finditer(text)
tokens = []
for m in cmds:
token = core.CmdToken(m.group(1), m.start(0), m.end(0))
tokens.append(token)
for m in inlcodes:
tokentext = core.InlCodeToken.linearize(m.group(1))
token = core.InlCodeToken(tokentext, m.start(0), m.end(0))
tokens.append(token)
for m in blkcodes:
# find each block and replace first left indent with ''
tokentext = m.group('code').strip('\n')
tokentext = core.deltextindent(tokentext)
token = core.BlkCodeToken(tokentext, m.start(0), m.end(0))
tokens.append(token)
tokens.sort(key=lambda tok: tok.start)
tokens.append(core.EndToken(None))
return tokens
def tokens(self, text):
"""Returns tokens
>>> text = \
'= Example of Literate Programming in Txt2Tags =\\n' \
\
'== Code 1 ==\\n' \
\
'Test if variable is negative looks like <<c.isneg>>: ``if a < 0``.\\n' \
'So, we can write absolute function <<c.fun>>\\n' \
'```\\n' \
' def fun(x):\\n' \
' <<=c.isneg,a:v>>:\\n' \
' a += 100\\n' \
' return -a\\n\\n' \
'```\\n' \
'And <<c.sum>>\\n' \
'```\\n' \
' def sum(x, y):\\n\\n' \
\
' return x+y\\n\\n' \
' x += 1\\n' \
'```\\n' \
'Last chunk is <<c.run>>\\n' \
'``` $ ls -F1'
>>> p = Txt2TagsParser()
>>> toks = p.tokens(text)
>>> [tok.__class__.__name__ for tok in toks]
['CmdToken', 'InlCodeToken', 'CmdToken', 'BlkCodeToken', 'CmdToken', 'BlkCodeToken', 'CmdToken', 'BlkCodeToken', 'EndToken']
>>> toks[0].text
'c.isneg'
>>> toks[1].text
'if a < 0'
>>> toks[2].text
'c.fun'
>>> toks[3].text.startswith('def fun')
True
>>> toks[3].text.endswith(' return -a')
True
>>> toks[5].text.startswith('def sum')
True
>>> toks[5].text.endswith('x += 1')
True
>>> toks[7].text
'$ ls -F1'
"""
_inlcode_re = r'``([^\n`]+)``'
_blkcode_re1 = r'(?:\n|^)```(?P<code>[^\n]+)(?:\n|$)'
_blkcode_re2 = r'(?:\n|^)```\n+(?P<code>.*?)\n```(?:\n|$)'
inlcode_re = re.compile(_inlcode_re)
blkcode_re1 = re.compile(_blkcode_re1, re.DOTALL)
blkcode_re2 = re.compile(_blkcode_re2, re.DOTALL)
cmds = core.Cmd.syntax.findtokens('cmddef', text)
inlcodes = inlcode_re.finditer(text)
blkcodes = itertools.chain(blkcode_re1.finditer(text),
blkcode_re2.finditer(text))
tokens = []
for m in cmds:
token = core.CmdToken(m.group(1), m.start(0), m.end(0))
tokens.append(token)
for m in inlcodes:
tokentext = core.InlCodeToken.linearize(m.group(1))
token = core.InlCodeToken(tokentext, m.start(0), m.end(0))
tokens.append(token)
for m in blkcodes:
# find each block and replace first left indent with ''
tokentext = m.group('code').strip('\n')
tokentext = core.deltextindent(tokentext)
token = core.BlkCodeToken(tokentext, m.start(0), m.end(0))
tokens.append(token)
tokens.sort(key=lambda tok: tok.start)
tokens.append(core.EndToken(None))
return tokens
def tokens(self, text):
"""Returns tokens
>>> text = \
'\\\\documentclass[12pt]{article}\\n' \
'\\\\usepackage{amsmath}\\n' \
'\\\\title{\\\\LaTeX}\\n' \
'\\\\date{}\\n' \
'\\\\begin{document}\\n' \
' \\\\maketitle\\n' \
' \\\\LaTeX{} is a document preparation system for the \\\\TeX{}\\n' \
'\\n' \
' Testing of negative value <<isneg>>: \\\\verb#if a < 0#. Signature will\\n' \
' be <<fn.abs.decl>>: \\\\verb!int abs(int a)!. And now:\\n' \
' function absolute <<fn.abs>>:\\n' \
'\\n' \
' \\\\begin{verbatim}\\n' \
' <<=fn.abs.decl>>\\n' \
' if (<<=isneg, x:a>>) return -a;\\n' \
' else return a;\\n' \
' }\\n' \
' \\\\end{verbatim}\\n' \
'\\n' \
' % This is a comment; it will not be shown in the final output.\\n' \
' % The following shows a little of the typesetting power of LaTeX:\\n' \
' \\\\begin{align}\\n' \
' m &= \\\\frac{m_0}{\\\\sqrt{1-\\\\frac{v^2}{c^2}}}\\n' \
' \\\\end{align}\\n' \
'\\\\end{document}'
>>> p = TeXParser()
>>> toks = p.tokens(text)
>>> [tok.__class__.__name__ for tok in toks]
['CmdToken', 'InlCodeToken', 'CmdToken', 'InlCodeToken', 'CmdToken', 'BlkCodeToken', 'EndToken']
>>> toks[0].text
'isneg'
>>> toks[1].text
'if a < 0'
>>> toks[2].text
'fn.abs.decl'
>>> toks[3].text == 'int abs(int a)'
True
>>> toks[5].text.startswith('<<=fn.abs.decl>>')
True
>>> toks[5].text.endswith('}')
True
"""
_inlcmds = '|'.join(re.escape(s) for s in TeXParser.inlcmds)
# text inside inline chunk
_inltext = r'{(?P<code1>.+?)}|(?P<s>[+#!])(?P<code2>.+?)(?P=s)'
_blkcmds = '|'.join(re.escape(s) for s in TeXParser.blkcmds)
_inlcode_re = r'\\(%s)%s'%(_inlcmds, _inltext)
_blkcode_re = r'\\begin(\[.+?\])?{(%s)}(?P<code>.*?)\\end{(%s)}'%(_blkcmds, _blkcmds)
inlcode_re = re.compile(_inlcode_re)
blkcode_re = re.compile(_blkcode_re, re.DOTALL)
cmds = core.Cmd.syntax.findtokens('cmddef', text)
inlcodes = inlcode_re.finditer(text)
blkcodes = blkcode_re.finditer(text)
tokens = []
for m in cmds:
token = core.CmdToken(m.group(1), m.start(0), m.end(0))
tokens.append(token)
for m in inlcodes:
groups = m.groupdict()
g = 'code1' if groups['code1'] else 'code2'
tokentext = core.InlCodeToken.linearize(m.group(g))
token = core.InlCodeToken(tokentext, m.start(0), m.end(0))
tokens.append(token)
for m in blkcodes:
# find each block and replace first left indent with ''
tokentext = m.group('code').lstrip('\n').rstrip('\n ')
tokentext = core.deltextindent(tokentext)
token = core.BlkCodeToken(tokentext, m.start(0), m.end(0))
tokens.append(token)
tokens.sort(key=lambda tok: tok.start)
tokens.append(core.EndToken(None))
return tokens
def tokens(self, text):
"""Returns tokens
>>> text = \
'Example of Literate Programming in Asciidoc\\n' \
'===========================================\\n\\n' \
\
'Code 1\\n' \
'------\\n\\n' \
\
'Test if variable is negative looks like <<c.isneg>>: +if a < 0+.\\n' \
'So, we can write absolute function <<c.fun>>\\n' \
'[source, python]\\n' \
'----\\n' \
' def fun(x):\\n' \
' <<=c.isneg,a:v>>:\\n' \
' a += 100\\n' \
' return -a\\n\\n' \
'----\\n\\n' \
'And <<c.sum>>\\n\\n' \
'----\\n' \
' def sum(x, y):\\n\\n' \
\
' return x+y\\n\\n' \
' x += 1\\n' \
'----'
>>> p = AsciiDocParser()
>>> toks = p.tokens(text)
>>> [tok.__class__.__name__ for tok in toks]
['CmdToken', 'InlCodeToken', 'CmdToken', 'BlkCodeToken', 'CmdToken', 'BlkCodeToken', 'EndToken']
>>> toks[0].text
'c.isneg'
>>> toks[1].text
'if a < 0'
>>> toks[2].text
'c.fun'
>>> toks[3].text.startswith('def fun')
True
>>> toks[3].text.endswith(' return -a')
True
>>> toks[5].text.startswith('def sum')
True
>>> toks[5].text.endswith('x += 1')
True
"""
_inlcode_re = r'\+([^\n\+]+)\+'
_blkcode_re = r'(?:\[source[^]]*?]\n|\n\n)-{4,}(?P<code>.*?)\n-{4,}'
inlcode_re = re.compile(_inlcode_re)
blkcode_re = re.compile(_blkcode_re, re.DOTALL)
cmds = core.Cmd.syntax.findtokens('cmddef', text)
inlcodes = inlcode_re.finditer(text)
blkcodes = blkcode_re.finditer(text)
tokens = []
for m in cmds:
token = core.CmdToken(m.group(1), m.start(0), m.end(0))
tokens.append(token)
for m in inlcodes:
tokentext = core.InlCodeToken.linearize(m.group(1))
token = core.InlCodeToken(tokentext, m.start(0), m.end(0))
tokens.append(token)
for m in blkcodes:
# find each block and replace first left indent with ''
tokentext = m.group('code').strip('\n')
tokentext = core.deltextindent(tokentext)
token = core.BlkCodeToken(tokentext, m.start(0), m.end(0))
tokens.append(token)
tokens.sort(key=lambda tok: tok.start)
tokens.append(core.EndToken(None))
return tokens
def tokens(self, text):
"""Returns tokens
>>> text = \
'= Example of Literate Programming in Txt2Tags =\\n' \
\
'== Code 1 ==\\n' \
\
'Test if variable is negative looks like <<c.isneg>>: ``if a < 0``.\\n' \
'So, we can write absolute function <<c.fun>>\\n' \
'```\\n' \
' def fun(x):\\n' \
' <<=c.isneg,a:v>>:\\n' \
' a += 100\\n' \
' return -a\\n\\n' \
'```\\n' \
'And <<c.sum>>\\n' \
'```\\n' \
' def sum(x, y):\\n\\n' \
\
' return x+y\\n\\n' \
' x += 1\\n' \
'```\\n' \
'Last chunk is <<c.run>>\\n' \
'``` $ ls -F1'
>>> p = Txt2TagsParser()
>>> toks = p.tokens(text)
>>> [tok.__class__.__name__ for tok in toks]
['CmdToken', 'InlCodeToken', 'CmdToken', 'BlkCodeToken', 'CmdToken', 'BlkCodeToken', 'CmdToken', 'BlkCodeToken', 'EndToken']
>>> toks[0].text
'c.isneg'
>>> toks[1].text
'if a < 0'
>>> toks[2].text
'c.fun'
>>> toks[3].text.startswith('def fun')
True
>>> toks[3].text.endswith(' return -a')
True
>>> toks[5].text.startswith('def sum')
True
>>> toks[5].text.endswith('x += 1')
True
>>> toks[7].text
'$ ls -F1'
"""
_inlcode_re = r'``([^\n`]+)``'
_blkcode_re1 = r'(?:\n|^)```(?P<code>[^\n]+)(?:\n|$)'
_blkcode_re2 = r'(?:\n|^)```\n+(?P<code>.*?)\n```(?:\n|$)'
inlcode_re = re.compile(_inlcode_re)
blkcode_re1 = re.compile(_blkcode_re1, re.DOTALL)
blkcode_re2 = re.compile(_blkcode_re2, re.DOTALL)
cmds = core.Cmd.syntax.findtokens('cmddef', text)
inlcodes = inlcode_re.finditer(text)
blkcodes = itertools.chain(blkcode_re1.finditer(text), blkcode_re2.finditer(text))
tokens = []
for m in cmds:
token = core.CmdToken(m.group(1), m.start(0), m.end(0))
tokens.append(token)
for m in inlcodes:
tokentext = core.InlCodeToken.linearize(m.group(1))
token = core.InlCodeToken(tokentext, m.start(0), m.end(0))
tokens.append(token)
for m in blkcodes:
# find each block and replace first left indent with ''
tokentext = m.group('code').strip('\n')
tokentext = core.deltextindent(tokentext)
token = core.BlkCodeToken(tokentext, m.start(0), m.end(0))
tokens.append(token)
tokens.sort(key=lambda tok: tok.start)
tokens.append(core.EndToken(None))
return tokens
