def get_plaintext(parse_actions=False):
"""Get plaintext parser element.
plaintext = { ( any of a-zA-Z0-9 or '!"$%&()+,-./?@\^_`~',
any of "[]*#:;='", any Unicode character without "|[]*#:;<>='{}" ) |
any Unicode character without "|[]*#:;<>='{}" }-;
:returns: plaintext parser element
:rtype: ParserElement
"""
str0 = '!"$%&()+,-./?@\^_`~'
str1 = "[]*#:;='"
str2 = "|[]*#:;<>='{}"
plaintext = pyparsing.Combine(
pyparsing.OneOrMore(
(
pyparsing.oneOf(" ".join(pyparsing.alphanums+str0))
+ pyparsing.oneOf(" ".join(str1))
+ pyparsing.CharsNotIn(str2, max=1)
)
^ pyparsing.CharsNotIn(str2, max=1)
)
)
plaintext.setName("plaintext")
plaintext.parseWithTabs()
if parse_actions:
pass
return plaintext
class TestTransformStringUsingParseActions(PyparsingExpressionTestCase):
markup_convert_map = {
'*': 'B',
'_': 'U',
'/': 'I',
}
def markup_convert(t):
htmltag = TestTransformStringUsingParseActions.markup_convert_map[
t.markup_symbol]
return "<{0}>{1}</{2}>".format(htmltag, t.body, htmltag)
tests = [
PpTestSpec(
desc="Use transformString to convert simple markup to HTML",
expr=(pp.oneOf(markup_convert_map)('markup_symbol') + "(" +
pp.CharsNotIn(")")('body') +
")").addParseAction(markup_convert),
text="Show in *(bold), _(underscore), or /(italic) type",
expected_list=[
'Show in <B>bold</B>, <U>underscore</U>, or <I>italic</I> type'
],
parse_fn='transformString',
),
]
def _parse(cls, text, function):
'''Prep `text` into a list of strs and then pass it to `function`.
The given text should contain one or more {}s in it,
like "foo {bar} buzz", and this function will convert it
to ['foo', ['bar'], 'buzz'] and then pass that to `function`.
Args:
text (str):
The text to convert into a list (of lists) of strs.
function (callable[list[str or list[str]]]):
The function that takes the parsed `text` as input.
Returns:
str: The output of function.
'''
# 1. We wrap the entire text in {}s, to make it a nested expression
text = cls._wrap(text)
pyparsing.ParserElement.setDefaultWhitespaceChars('\n\t')
_content = pyparsing.CharsNotIn(['{', '}'])
_curlys = pyparsing.nestedExpr('{', '}', content=_content)
# 2. Since we made the expression nested a little while ago, lets unpack it
# by getting the 0th index
#
parsed_text = _curlys.parseString(text).asList()[0]
result = function(parsed_text)
# 3. The {}s that we added with `_wrap` need to be removed.
# So we just remove the first and last characters, [1:-1]
# from the final result
#
return result[1:-1]
def _get_simple_ref_parser():
string = pp.CharsNotIn(_EXCLUDES).setParseAction(_string)
ref_open = pp.Literal(_REF_OPEN).suppress()
ref_close = pp.Literal(_REF_CLOSE).suppress()
reference = (ref_open + pp.Group(string) + ref_close).setParseAction(_reference)
line = pp.StringStart() + pp.Optional(string) + reference + pp.Optional(string) + pp.StringEnd()
return line
class Parser(object):
comment_def = "--" + pyparsing.NotAny(
'-' + pyparsing.CaselessKeyword('begin')) + pyparsing.ZeroOrMore(
pyparsing.CharsNotIn("\n"))
def __init__(self, scanner, retainSeparator=True):
self.scanner = scanner
self.scanner.ignore(pyparsing.sglQuotedString)
self.scanner.ignore(pyparsing.dblQuotedString)
self.scanner.ignore(self.comment_def)
self.scanner.ignore(pyparsing.cStyleComment)
self.retainSeparator = retainSeparator
def separate(self, txt):
itms = []
for (sqlcommand, start, end) in self.scanner.scanString(txt):
if sqlcommand:
if type(sqlcommand[0]) == pyparsing.ParseResults:
if self.retainSeparator:
itms.append("".join(sqlcommand[0]))
else:
itms.append(sqlcommand[0][0])
else:
if sqlcommand[0]:
itms.append(sqlcommand[0])
return itms
def parse_input(self, s):
function = (p.oneOf(accepted_functions) + "(" + p.CharsNotIn(")") +
")" + p.White()).setParseAction(self.function_action)
block = ("{block " + p.Word(p.nums) + "}" +
p.ZeroOrMore(function)).setParseAction(self.block_action)
cf2text = p.OneOrMore(block)
cf2text.parseString(s)
class TestTransformStringUsingParseActions(PyparsingExpressionTestCase):
markup_convert_map = {
"*": "B",
"_": "U",
"/": "I",
}
def markup_convert(t):
htmltag = TestTransformStringUsingParseActions.markup_convert_map[
t.markup_symbol]
return "<{}>{}</{}>".format(htmltag, t.body, htmltag)
tests = [
PpTestSpec(
desc="Use transformString to convert simple markup to HTML",
expr=(pp.oneOf(markup_convert_map)("markup_symbol") + "(" +
pp.CharsNotIn(")")("body") +
")").addParseAction(markup_convert),
text="Show in *(bold), _(underscore), or /(italic) type",
expected_list=[
"Show in <B>bold</B>, <U>underscore</U>, or <I>italic</I> type"
],
parse_fn="transformString",
),
]
def get_parser():
"Construct and return the parser."
left = pp.Suppress("(")
right = pp.Suppress(")")
value = pp.QuotedString(quote_char='"', esc_char="\\") | pp.CharsNotIn(")")
identifier = pp.Word(pp.alphas, pp.alphanums).set_parse_action(_Identifier)
label = (pp.Keyword("label") + left + value +
right).set_parse_action(_Label)
year = (pp.Keyword("year") + left + value + right).set_parse_action(_Year)
author = (pp.Keyword("author") + left + value +
right).set_parse_action(_Author)
orcid = (pp.Keyword("orcid") + left + value +
right).set_parse_action(_Orcid)
issn = (pp.Keyword("issn") + left + value + right).set_parse_action(_Issn)
published = (pp.Keyword("published") + left + value +
right).set_parse_action(_Published)
first = (pp.Keyword("first") + left + value +
right).set_parse_action(_First)
online = (pp.Keyword("online") + left + value +
right).set_parse_action(_Online)
modified = (pp.Keyword("modified") + left + value +
right).set_parse_action(_Modified)
no_pmid = (pp.Keyword("no_pmid") + left + right).set_parse_action(_NoPmid)
no_doi = (pp.Keyword("no_doi") + left + right).set_parse_action(_NoDoi)
no_label = (pp.Keyword("no_label") + left +
right).set_parse_action(_NoLabel)
function = (label
| year
| author
| orcid
| issn
| published
| first
| online
| modified
| no_pmid
| no_doi
| no_label)
if settings["TEMPORAL_LABELS"]:
current = (pp.Keyword("active") + left +
right).set_parse_action(_Active)
active = (pp.Keyword("active") + left + value +
right).set_parse_action(_Active)
function = function | current | active
union = pp.Literal("+").set_parse_action(_Union)
symdifference = pp.Literal("^").set_parse_action(_Symdifference)
intersection = pp.Literal("#").set_parse_action(_Intersection)
difference = pp.Literal("-").set_parse_action(_Difference)
operator = union | symdifference | difference | intersection
expression = pp.Forward()
atom = function | identifier | pp.Group(left + expression + right)
expression <<= atom + (operator + atom)[...]
expression.set_parse_action(_Expression)
expression.ignore("!" + pp.rest_of_line)
return expression
class constant_def:
zero = pp.Char("0")
nonzero_digit = pp.Char("123456789")
digit = pp.Char(pp.nums)
octal_digit = pp.Char("01234567")
hexadecimal_digit = pp.Char(pp.hexnums)
# (6.4.4.1) decimal-constant:
decimal_constant = nonzero_digit + digit.copy()[...]
# (6.4.4.1) octal-constant:
octal_constant = zero + octal_digit.copy()[...]
# (6.4.4.1) hexadecimal-constant:
hexadecimal_prefix = pp.Word("0x") | pp.Word("0X")
hexadecimal_constant = hexadecimal_prefix + hexadecimal_digit.copy()[1, ...]
# (6.4.4.1) integer-suffix:
unsigned_suffix = pp.Char("uU")
long_suffix = pp.Char("lL")
longlong_suffix = pp.Word("ll") | pp.Word("LL")
integer_suffix = (unsigned_suffix + pp.Optional(long_suffix | longlong_suffix)) | ((long_suffix | longlong_suffix) + pp.Optional(unsigned_suffix))
# (6.4.4.1) integer-constant:
integer_constant = (decimal_constant | octal_constant | hexadecimal_constant) + pp.Optional(integer_suffix)
# (6.4.4.2) floating-constant:
# 略
# (6.4.4.3) enumeration-constant:
enumeration_constant = identifier
# (6.4.4.4) character-constant:
simple_escape_sequence = pp.Char("\\") + pp.Char("\\'\"?abfnrtv")
octal_escape_sequence = pp.Char("\\") + octal_digit.copy()[1, ...]
hexadecimal_escape_sequence = pp.Word("\\x") + hexadecimal_digit.copy()[1, ...]
escape_sequence = simple_escape_sequence | octal_escape_sequence | hexadecimal_escape_sequence | identifier_def.universal_character_name
c_char = pp.CharsNotIn("'\\\r\n") | escape_sequence
character_constant = pp.Optional(pp.Char("L")) + pp.Char("'") + c_char[1,...] + pp.Char("'")
def get_param(wiki_markup, parse_actions=False):
"""Get parameter parser element.
param = "{{{", { any Unicode character without "|={}" }-, [ default ],
"}}}";
default = "|", { any Unicode character };
:returns: parameter parser element
:rtype: ParserElement
"""
name = pyparsing.CharsNotIn("|={}").setResultsName("name")
name.setName("name")
name.parseWithTabs()
value = _get_value(wiki_markup, parse_actions=parse_actions)
param = pyparsing.nestedExpr(
opener="{{{",
closer="}}}",
content=(name + pyparsing.Optional(
pyparsing.Literal("|") + pyparsing.Optional(value))),
ignoreExpr=None)
param.setName("param")
param.parseWithTabs()
if parse_actions:
pass
return param
class Tokens(object):
# shared tokens
delim_chars = '[]{},'
pause = pp.FollowedBy(pp.Word(delim_chars) | pp.StringEnd())
number = (pp.pyparsing_common.number + pause)
quoted_string = pp.QuotedString('"', escChar='\\')
true = (pp.Regex(r'(True|true|yes|on)') + pause). \
setParseAction(lambda _: True)
false = (pp.Regex(r'(False|false|no|off)') + pause). \
setParseAction(lambda _: False)
unquoted_string = pp.CharsNotIn(delim_chars). \
setParseAction(lambda toks: toks[0].strip())
empty_value = pp.Empty(). \
setParseAction(lambda _: '')
# tokens for configs
identifier = pp.pyparsing_common.identifier.copy()
comma = pp.Literal(',').suppress()
assign = pp.Literal('=').suppress()
config_value = (
number | true | false | quoted_string | unquoted_string | empty_value)
key_value_pair = (identifier + assign + config_value). \
setParseAction(lambda toks: (toks[0], toks[1]))
key_value_pair_list = pp.Optional(
key_value_pair + pp.ZeroOrMore(comma + key_value_pair))
# tokens for tags
tag = quoted_string | unquoted_string | pp.Empty().suppress()
tag_list = pp.Optional(tag + pp.ZeroOrMore(comma + tag))
def _make_valid_state_name(self, state_name):
"""Transform the input state_name into a valid state in XMLBIF.
XMLBIF states must start with a letter an only contain letters,
numbers and underscores.
"""
s = str(state_name)
s_fixed = pp.CharsNotIn(pp.alphanums + "_").setParseAction(pp.replaceWith("_")).transformString(s)
if not s_fixed[0].isalpha():
s_fixed = "state" + s_fixed
return s_fixed
def _get_parser(self):
assign = pp.oneOf(['=', '?=', ':=', '::=', '+='])('assign')
var_name = pp.Word(pp.alphas + '_', pp.alphanums + '_')('var')
enclosed = pp.Forward()
nested_parents = pp.nestedExpr('$(', ')', content=enclosed)
nested_brackets = pp.nestedExpr('${', '}', content=enclosed)
enclosed <<= (nested_parents | nested_brackets |
pp.CharsNotIn('$(){}\n')).leaveWhitespace()
return pp.lineStart + var_name + assign + pp.ZeroOrMore(pp.White()) + pp.ZeroOrMore(enclosed)('value')
def _make_parser():
word = pp.CharsNotIn(''.join(whitespace))
word.skipWhitespace = True
value = pp.MatchFirst([
pp.dblQuotedString.copy().setParseAction(pp.removeQuotes),
pp.sglQuotedString.copy().setParseAction(pp.removeQuotes),
pp.Empty() + pp.CharsNotIn(''.join(whitespace)),
])
expressions = []
for field in named_fields:
exp = pp.Suppress(pp.CaselessLiteral(field) + ':') + \
value.copy().setParseAction(_decorate_match(field))
expressions.append(exp)
any_ = value.copy().setParseAction(_decorate_match('any'))
expressions.append(any_)
return pp.ZeroOrMore(pp.MatchFirst(expressions))
def _make_parser():
word = pp.CharsNotIn("".join(whitespace))
word.skipWhitespace = True
value = pp.MatchFirst([
pp.dbl_quoted_string.copy().set_parse_action(pp.remove_quotes),
pp.sgl_quoted_string.copy().set_parse_action(pp.remove_quotes),
pp.Empty() + pp.CharsNotIn("".join(whitespace)),
])
expressions = []
for field in named_fields:
exp = pp.Suppress(pp.CaselessLiteral(field) +
":") + value.copy().set_parse_action(
_decorate_match(field))
expressions.append(exp)
any_ = value.copy().set_parse_action(_decorate_match("any"))
expressions.append(any_)
return pp.ZeroOrMore(pp.MatchFirst(expressions))
def _make():
# Order is important - multi-char expressions need to come before narrow
# ones.
parts = []
for cls in filter_unary:
f = pp.Literal(f"~{cls.code}") + pp.WordEnd()
f.setParseAction(cls.make)
parts.append(f)
# This is a bit of a hack to simulate Word(pyparsing_unicode.printables),
# which has a horrible performance with len(pyparsing.pyparsing_unicode.printables) == 1114060
unicode_words = pp.CharsNotIn("()~'\"" + pp.ParserElement.DEFAULT_WHITE_CHARS)
unicode_words.skipWhitespace = True
regex = (
unicode_words
| pp.QuotedString('"', escChar='\\')
| pp.QuotedString("'", escChar='\\')
)
for cls in filter_rex:
f = pp.Literal(f"~{cls.code}") + pp.WordEnd() + regex.copy()
f.setParseAction(cls.make)
parts.append(f)
for cls in filter_int:
f = pp.Literal(f"~{cls.code}") + pp.WordEnd() + pp.Word(pp.nums)
f.setParseAction(cls.make)
parts.append(f)
# A naked rex is a URL rex:
f = regex.copy()
f.setParseAction(FUrl.make)
parts.append(f)
atom = pp.MatchFirst(parts)
expr = pp.infixNotation(
atom,
[(pp.Literal("!").suppress(),
1,
pp.opAssoc.RIGHT,
lambda x: FNot(*x)),
(pp.Literal("&").suppress(),
2,
pp.opAssoc.LEFT,
lambda x: FAnd(*x)),
(pp.Literal("|").suppress(),
2,
pp.opAssoc.LEFT,
lambda x: FOr(*x)),
])
expr = pp.OneOrMore(expr)
return expr.setParseAction(lambda x: FAnd(x) if len(x) != 1 else x)
def parser_factory(styler):
"""Builds the S-expression parser."""
def cond_optional(expr):
return pp.Optional(expr) if styler else expr
LPAR, RPAR, SQUO, DQUO = map(pp.Suppress, '()\'"')
form_first = pp.Forward()
form = pp.Forward()
nil = pp.CaselessKeyword('nil').addParseAction(pp.replaceWith([]))
t = pp.CaselessKeyword('t').addParseAction(pp.replaceWith(True))
constant = styler('class:constant', nil | t)
number = styler('class:number', ppc.number).setName('number')
control_chars = ''.join(map(chr, range(0, 32))) + '\x7f'
symbol = pp.CharsNotIn(control_chars + '\'"`;,()[]{} ')
symbol = styler('class:symbol', symbol).setName('symbol')
symbol.addParseAction(lambda t: Symbol(t[0]))
call = styler('class:call', symbol)
string = DQUO + pp.Combine(pp.Optional(
pp.CharsNotIn('"'))) + cond_optional(DQUO)
string = styler('class:string', string).setName('string')
forms = (form_first + pp.ZeroOrMore(form)).setName('one or more forms')
sexp = (LPAR + pp.Optional(forms) +
cond_optional(RPAR)).setName('s-expression')
sexp.addParseAction(lambda t: [list(t)])
quote = (styler('class:quote', SQUO) + form).setName('quoted form')
quote.addParseAction(lambda t: Quote(t[0]))
form_first <<= constant | number ^ call | string | sexp | quote
form <<= constant | number ^ symbol | string | sexp | quote
return form
def get_pagename(parse_actions=False):
"""Get pagename parser element.
pagename = { any Unicode character without "|[]#<>{}" }-;
:returns: pagename parser element
:rtype: ParserElement
"""
pagename = pyparsing.CharsNotIn("|[]#<>{}").setResultsName("pagename")
pagename.setName("pagename")
pagename.parseWithTabs()
if parse_actions:
pass
return pagename
def _get_heading(parse_actions=False):
"""Get heading parser element.
heading = { any Unicode character without "|[]" }-;
:returns: heading parser element
:rtype: ParserElement
"""
heading = pyparsing.CharsNotIn("|[]")
heading.setName("heading")
heading.parseWithTabs()
if parse_actions:
pass
return heading
def _get_label(parse_actions=False):
"""Get label parser element.
label = { any Unicode character without "|[]" }-;
:returns: label parser element
:rtype: ParserElement
"""
label = pyparsing.CharsNotIn("|[]").setResultsName("label")
label.setName("label")
label.parseWithTabs()
if parse_actions:
pass
return label
def _get_anchor(parse_actions=False):
"""Get anchor parser element.
anchor = { any Unicode character without "[]" }-;
:returns: anchor parser element
:rtype: ParserElement
"""
anchor = pyparsing.CharsNotIn("[]").setResultsName("anchor")
anchor.setName("anchor")
anchor.parseWithTabs()
if parse_actions:
pass
return anchor
def _get_parser():
double_escape = pp.Combine(pp.Literal(_DOUBLE_ESCAPE) + pp.MatchFirst([pp.FollowedBy(_REF_OPEN), pp.FollowedBy(_REF_CLOSE),
pp.FollowedBy(_INV_OPEN), pp.FollowedBy(_INV_CLOSE)])).setParseAction(pp.replaceWith(_ESCAPE))
ref_open = pp.Literal(_REF_OPEN).suppress()
ref_close = pp.Literal(_REF_CLOSE).suppress()
ref_not_open = ~pp.Literal(_REF_OPEN) + ~pp.Literal(_REF_ESCAPE_OPEN) + ~pp.Literal(_REF_DOUBLE_ESCAPE_OPEN)
ref_not_close = ~pp.Literal(_REF_CLOSE) + ~pp.Literal(_REF_ESCAPE_CLOSE) + ~pp.Literal(_REF_DOUBLE_ESCAPE_CLOSE)
ref_escape_open = pp.Literal(_REF_ESCAPE_OPEN).setParseAction(pp.replaceWith(_REF_OPEN))
ref_escape_close = pp.Literal(_REF_ESCAPE_CLOSE).setParseAction(pp.replaceWith(_REF_CLOSE))
ref_text = pp.CharsNotIn(_REF_EXCLUDES) | pp.CharsNotIn(_REF_CLOSE_FIRST, exact=1)
ref_content = pp.Combine(pp.OneOrMore(ref_not_open + ref_not_close + ref_text))
ref_string = pp.MatchFirst([double_escape, ref_escape_open, ref_escape_close, ref_content]).setParseAction(_string)
ref_item = pp.Forward()
ref_items = pp.OneOrMore(ref_item)
reference = (ref_open + pp.Group(ref_items) + ref_close).setParseAction(_reference)
ref_item << (reference | ref_string)
inv_open = pp.Literal(_INV_OPEN).suppress()
inv_close = pp.Literal(_INV_CLOSE).suppress()
inv_not_open = ~pp.Literal(_INV_OPEN) + ~pp.Literal(_INV_ESCAPE_OPEN) + ~pp.Literal(_INV_DOUBLE_ESCAPE_OPEN)
inv_not_close = ~pp.Literal(_INV_CLOSE) + ~pp.Literal(_INV_ESCAPE_CLOSE) + ~pp.Literal(_INV_DOUBLE_ESCAPE_CLOSE)
inv_escape_open = pp.Literal(_INV_ESCAPE_OPEN).setParseAction(pp.replaceWith(_INV_OPEN))
inv_escape_close = pp.Literal(_INV_ESCAPE_CLOSE).setParseAction(pp.replaceWith(_INV_CLOSE))
inv_text = pp.CharsNotIn(_INV_CLOSE_FIRST)
inv_content = pp.Combine(pp.OneOrMore(inv_not_close + inv_text))
inv_string = pp.MatchFirst([double_escape, inv_escape_open, inv_escape_close, inv_content]).setParseAction(_string)
inv_items = pp.OneOrMore(inv_string)
export = (inv_open + pp.Group(inv_items) + inv_close).setParseAction(_invquery)
text = pp.CharsNotIn(_EXCLUDES) | pp.CharsNotIn('', exact=1)
content = pp.Combine(pp.OneOrMore(ref_not_open + inv_not_open + text))
string = pp.MatchFirst([double_escape, ref_escape_open, inv_escape_open, content]).setParseAction(_string)
item = reference | export | string
line = pp.OneOrMore(item) + pp.StringEnd()
return line
def get_simple_ref_parser(settings):
ESCAPE = settings.escape_character
REF_OPEN, REF_CLOSE = settings.reference_sentinels
INV_OPEN, INV_CLOSE = settings.export_sentinels
EXCLUDES = ESCAPE + REF_OPEN + REF_CLOSE + INV_OPEN + INV_CLOSE
string = pp.CharsNotIn(EXCLUDES).setParseAction(_tag_with(tags.STR))
ref_open = pp.Literal(REF_OPEN).suppress()
ref_close = pp.Literal(REF_CLOSE).suppress()
reference = (ref_open + pp.Group(string) + ref_close).setParseAction(
_tag_with(tags.REF))
line = pp.StringStart() + pp.Optional(string) + reference + pp.Optional(
string) + s_end
return line.leaveWhitespace()
def parser_factory(styler):
"""Builds the repr() parser."""
squo = styler('class:string', "'")
dquo = styler('class:string', '"')
esc_single = pp.oneOf(r'\\ \' \" \n \r \t')
esc_hex = pp.Literal(r'\x') + pp.Word(pp.hexnums, exact=2)
escs = styler('class:escape', esc_single | esc_hex)
control_chars = ''.join(map(chr, range(32))) + '\x7f'
normal_chars_squo = pp.CharsNotIn(control_chars + r"\'")
chars_squo = styler('class:string', normal_chars_squo) | escs
normal_chars_dquo = pp.CharsNotIn(control_chars + r'\"')
chars_dquo = styler('class:string', normal_chars_dquo) | escs
skip_white = pp.Optional(pp.White())
bytes_prefix = pp.Optional(styler('class:string_prefix', 'b'))
string_squo = skip_white + bytes_prefix + squo - pp.ZeroOrMore(
chars_squo) + squo
string_dquo = skip_white + bytes_prefix + dquo - pp.ZeroOrMore(
chars_dquo) + dquo
string = string_squo | string_dquo
string.leaveWhitespace()
address = styler('class:address', '0x' + pp.Word(pp.hexnums))
number = styler('class:number', ppc.number)
const = pp.oneOf('True False None NotImplemented Ellipsis ...')
const = styler('class:constant', const)
kwarg = styler('class:kwarg', ppc.identifier) + styler(
'class:operator', '=')
call = styler('class:call', ppc.identifier) + pp.FollowedBy('(')
magic = styler('class:magic', pp.Regex(r'__[a-zA-Z0-9_]+__'))
token = string | address | number | const | kwarg | call | magic
token.parseWithTabs()
return pp.originalTextFor(token)
def parser_factory(styler):
"""Builds the JSON parser."""
LBRK, RBRK, LBRC, RBRC, COLON, DQUO = map(pp.Suppress, '[]{}:"')
DQUO = styler('class:string', DQUO)
control_chars = ''.join(map(chr, range(32))) + '\x7f'
normal_chars = pp.CharsNotIn(control_chars + '\\"')
s_quo = pp.Literal('\\"').addParseAction(pp.replaceWith('"'))
s_sol = pp.Literal('\\/').addParseAction(pp.replaceWith('/'))
s_rsol = pp.Literal('\\\\').addParseAction(pp.replaceWith('\\'))
s_back = pp.Literal('\\b').addParseAction(pp.replaceWith('\b'))
s_form = pp.Literal('\\f').addParseAction(pp.replaceWith('\f'))
s_nl = pp.Literal('\\n').addParseAction(pp.replaceWith('\n'))
s_ret = pp.Literal('\\r').addParseAction(pp.replaceWith('\r'))
s_tab = pp.Literal('\\t').addParseAction(pp.replaceWith('\t'))
s_unicode = pp.Suppress('\\u') + pp.Word(pp.hexnums, exact=4)
s_unicode.addParseAction(lambda t: chr(int(t[0], 16)))
escape_seqs = s_quo | s_sol | s_rsol | s_back | s_form | s_nl | s_ret | s_tab | s_unicode
chars = styler('class:string', normal_chars) | styler(
'class:escape', escape_seqs)
skip_white = pp.Optional(pp.Suppress(pp.White()))
string = skip_white + DQUO - pp.Combine(pp.ZeroOrMore(chars)) + DQUO
string.leaveWhitespace()
string.setName('string')
value = pp.Forward()
pair = string + COLON + value
pair.addParseAction(tuple)
obj = LBRC - pp.Optional(pp.delimitedList(pair)) + pp.NotAny(',') + RBRC
obj.addParseAction(lambda t: {k: v for k, v in t})
obj.setName('object')
array = LBRK - pp.Optional(pp.delimitedList(value)) + pp.NotAny(',') + RBRK
array.addParseAction(lambda t: [list(t)])
array.setName('array')
true = pp.Literal('true').addParseAction(pp.replaceWith(True))
false = pp.Literal('false').addParseAction(pp.replaceWith(False))
null = pp.Literal('null').addParseAction(pp.replaceWith(None))
constant = styler('class:constant', true | false | null)
value <<= obj | array | string | styler('class:number',
ppc.number) | constant
value.parseWithTabs()
value.setName('JSON value')
return value
def _get_parser(self):
""" Initialize the pyparsing parser for Makefile
"""
assign = pp.oneOf(['=', '?=', ':=', '::=', '+='])('assign')
var_name = pp.Word(pp.alphas + '_', pp.alphanums + '_')('var')
enclosed = pp.Forward()
nested_parents = pp.nestedExpr('$(', ')', content=enclosed)
nested_brackets = pp.nestedExpr('${', '}', content=enclosed)
enclosed <<= (nested_parents | nested_brackets
| pp.CharsNotIn('$(){}#\n')).leaveWhitespace()
return pp.lineStart + var_name + pp.ZeroOrMore(
pp.White()) + assign + pp.ZeroOrMore(
pp.White()) + pp.ZeroOrMore(enclosed)('value') + pp.Optional(
pp.pythonStyleComment)('comment')
def get_simple_ref_parser(escape_character, reference_sentinels,
export_sentinels):
_ESCAPE = escape_character
_REF_OPEN = reference_sentinels[0]
_REF_CLOSE = reference_sentinels[1]
_INV_OPEN = export_sentinels[0]
_INV_CLOSE = export_sentinels[1]
_EXCLUDES = _ESCAPE + _REF_OPEN + _REF_CLOSE + _INV_OPEN + _INV_CLOSE
string = pp.CharsNotIn(_EXCLUDES).setParseAction(_string)
ref_open = pp.Literal(_REF_OPEN).suppress()
ref_close = pp.Literal(_REF_CLOSE).suppress()
reference = (ref_open + pp.Group(string) +
ref_close).setParseAction(_reference)
line = pp.StringStart() + pp.Optional(string) + reference + pp.Optional(
string) + pp.StringEnd()
return line
def read(self, state):
if state is None:
state = {}
self.formatString = state.get(
'formatString', '/tmp/<1.artist>/<1.title>/<#> - <title>')
self.replaceChars = state.get('replaceChars', '\\:/')
self.replaceBy = state.get('replaceBy', '_.;')
self.removeChars = state.get('removeChars', '?*')
if len(self.replaceChars) != len(self.replaceBy):
raise ValueError("replaceChars and replaceBy must equal in length")
self.translation = str.maketrans(self.replaceChars, self.replaceBy,
self.removeChars)
oldDefaultWhitespaceChars = pyparsing.ParserElement.DEFAULT_WHITE_CHARS
pyparsing.ParserElement.setDefaultWhitespaceChars("\t\n")
#pyparsing.ParserElement.enablePackrat() does not work (buggy)
lbrace = Literal("<").suppress()
rbrace = Literal(">").suppress()
number = Word(nums)
levelDef = number + Literal(".").suppress()
levelDef.setParseAction(self.levelDefAction)
tagName = Word(alphas + "_", alphanums + "_:()/\\")
tagDefinition = Optional(levelDef("levelDef")) + (tagName | "#")("tag")
tagDefinition.setParseAction(self.tagDefinitionAction)
expression = Forward()
ifExpr = Literal("?").suppress() + expression
elseExpr = Literal("!").suppress() + expression
condition = lbrace + tagDefinition("tagDef") \
+ ((Optional(ifExpr("if")) + Optional(elseExpr("else"))) | (Optional(elseExpr("else")) + Optional(ifExpr("else"))))\
+ rbrace # i am sure this is possible in a nicer way ...
condition.setParseAction(self.conditionAction)
staticText = pyparsing.CharsNotIn(
"<>!?") #Word("".join(p for p in printables if p not in "<>"))
expression << OneOrMore(condition | staticText)
self.expression = expression
pyparsing.ParserElement.setDefaultWhitespaceChars(
oldDefaultWhitespaceChars)
def _ParseTags(cls, string : str) -> dict:
SINGLE_VALUE = pp.CharsNotIn(";$")
LIST_VALUE = pp.OneOrMore(pp.Word(pp.alphanums + "_.") + pp.Literal(",")) + pp.Word(pp.alphanums + "_.")
ANY_VALUE = pp.Group(LIST_VALUE("listVal") | SINGLE_VALUE("singleVal"))
TAGS = "$$" + pp.OneOrMore(
pp.Group(pp.Word(pp.alphas)("tag") + "=" + ANY_VALUE("value") + pp.Optional(";")))("tags") + "$$"
tags = {}
for t, s, e in TAGS.scanString(string):
for tag in t.get("tags"):
val = tag.get("value")
if "listVal" in val.keys():
val = "".join(val).split(",")
elif "singleVal" in val.keys():
val = val.get("singleVal").strip()
else:
raise Exception("Illegal Tag Format")
tags[tag.get("tag").lower()] = val
return tags
def _get_named_arg(wiki_markup, parse_actions=False):
"""Get named argument parser element.
:param ParserElement wiki_markup: wiki markup
named_arg = { any Unicode character without "|=" }, "=", value;
:returns: named argument parser element
:rtype: ParserElement
"""
name = pyparsing.CharsNotIn("|=").setResultsName("name")
name.setName("name")
name.parseWithTabs()
value = _get_value(wiki_markup)
named_arg = (pyparsing.Optional(name) + pyparsing.Literal("=") + value)
named_arg.setName("named_arg")
named_arg.parseWithTabs()
if parse_actions:
pass
return named_arg
