Добавил:
Вуз:
Предмет:
Файл:
Курсовой проект Web-Сайт / venv / Lib / site-packages / pip / _vendor / pyparsing / core
.py#
# core.py
#
from collections import deque
import os
import typing
from typing import (
Any,
Callable,
Generator,
List,
NamedTuple,
Sequence,
Set,
TextIO,
Tuple,
Union,
cast,
)
from abc import ABC, abstractmethod
from enum import Enum
import string
import copy
import warnings
import re
import sys
from collections.abc import Iterable
import traceback
import types
from operator import itemgetter
from functools import wraps
from threading import RLock
from pathlib import Path
from .util import (
_FifoCache,
_UnboundedCache,
__config_flags,
_collapse_string_to_ranges,
_escape_regex_range_chars,
_bslash,
_flatten,
LRUMemo as _LRUMemo,
UnboundedMemo as _UnboundedMemo,
replaced_by_pep8,
)
from .exceptions import *
from .actions import *
from .results import ParseResults, _ParseResultsWithOffset
from .unicode import pyparsing_unicode
_MAX_INT = sys.maxsize
str_type: Tuple[type, ...] = (str, bytes)
#
# Copyright (c) 2003-2022 Paul T. McGuire
#
# Permission is hereby granted, free of charge, to any person obtaining
# a copy of this software and associated documentation files (the
# "Software"), to deal in the Software without restriction, including
# without limitation the rights to use, copy, modify, merge, publish,
# distribute, sublicense, and/or sell copies of the Software, and to
# permit persons to whom the Software is furnished to do so, subject to
# the following conditions:
#
# The above copyright notice and this permission notice shall be
# included in all copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
# EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
# MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
# IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
# CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
# TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
# SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
#
if sys.version_info >= (3, 8):
from functools import cached_property
else:
class cached_property:
def __init__(self, func):
self._func = func
def __get__(self, instance, owner=None):
ret = instance.__dict__[self._func.__name__] = self._func(instance)
return ret
class __compat__(__config_flags):
"""
A cross-version compatibility configuration for pyparsing features that will be
released in a future version. By setting values in this configuration to True,
those features can be enabled in prior versions for compatibility development
and testing.
- ``collect_all_And_tokens`` - flag to enable fix for Issue #63 that fixes erroneous grouping
of results names when an :class:`And` expression is nested within an :class:`Or` or :class:`MatchFirst`;
maintained for compatibility, but setting to ``False`` no longer restores pre-2.3.1
behavior
"""
_type_desc = "compatibility"
collect_all_And_tokens = True
_all_names = [__ for __ in locals() if not __.startswith("_")]
_fixed_names = """
collect_all_And_tokens
""".split()
class __diag__(__config_flags):
_type_desc = "diagnostic"
warn_multiple_tokens_in_named_alternation = False
warn_ungrouped_named_tokens_in_collection = False
warn_name_set_on_empty_Forward = False
warn_on_parse_using_empty_Forward = False
warn_on_assignment_to_Forward = False
warn_on_multiple_string_args_to_oneof = False
warn_on_match_first_with_lshift_operator = False
enable_debug_on_named_expressions = False
_all_names = [__ for __ in locals() if not __.startswith("_")]
_warning_names = [name for name in _all_names if name.startswith("warn")]
_debug_names = [name for name in _all_names if name.startswith("enable_debug")]
@classmethod
def enable_all_warnings(cls) -> None:
for name in cls._warning_names:
cls.enable(name)
class Diagnostics(Enum):
"""
Diagnostic configuration (all default to disabled)
- ``warn_multiple_tokens_in_named_alternation`` - flag to enable warnings when a results
name is defined on a :class:`MatchFirst` or :class:`Or` expression with one or more :class:`And` subexpressions
- ``warn_ungrouped_named_tokens_in_collection`` - flag to enable warnings when a results
name is defined on a containing expression with ungrouped subexpressions that also
have results names
- ``warn_name_set_on_empty_Forward`` - flag to enable warnings when a :class:`Forward` is defined
with a results name, but has no contents defined
- ``warn_on_parse_using_empty_Forward`` - flag to enable warnings when a :class:`Forward` is
defined in a grammar but has never had an expression attached to it
- ``warn_on_assignment_to_Forward`` - flag to enable warnings when a :class:`Forward` is defined
but is overwritten by assigning using ``'='`` instead of ``'<<='`` or ``'<<'``
- ``warn_on_multiple_string_args_to_oneof`` - flag to enable warnings when :class:`one_of` is
incorrectly called with multiple str arguments
- ``enable_debug_on_named_expressions`` - flag to auto-enable debug on all subsequent
calls to :class:`ParserElement.set_name`
Diagnostics are enabled/disabled by calling :class:`enable_diag` and :class:`disable_diag`.
All warnings can be enabled by calling :class:`enable_all_warnings`.
"""
warn_multiple_tokens_in_named_alternation = 0
warn_ungrouped_named_tokens_in_collection = 1
warn_name_set_on_empty_Forward = 2
warn_on_parse_using_empty_Forward = 3
warn_on_assignment_to_Forward = 4
warn_on_multiple_string_args_to_oneof = 5
warn_on_match_first_with_lshift_operator = 6
enable_debug_on_named_expressions = 7
def enable_diag(diag_enum: Diagnostics) -> None:
"""
Enable a global pyparsing diagnostic flag (see :class:`Diagnostics`).
"""
__diag__.enable(diag_enum.name)
def disable_diag(diag_enum: Diagnostics) -> None:
"""
Disable a global pyparsing diagnostic flag (see :class:`Diagnostics`).
"""
__diag__.disable(diag_enum.name)
def enable_all_warnings() -> None:
"""
Enable all global pyparsing diagnostic warnings (see :class:`Diagnostics`).
"""
__diag__.enable_all_warnings()
# hide abstract class
del __config_flags
def _should_enable_warnings(
cmd_line_warn_options: typing.Iterable[str], warn_env_var: typing.Optional[str]
) -> bool:
enable = bool(warn_env_var)
for warn_opt in cmd_line_warn_options:
w_action, w_message, w_category, w_module, w_line = (warn_opt + "::::").split(
":"
)[:5]
if not w_action.lower().startswith("i") and (
not (w_message or w_category or w_module) or w_module == "pyparsing"
):
enable = True
elif w_action.lower().startswith("i") and w_module in ("pyparsing", ""):
enable = False
return enable
if _should_enable_warnings(
sys.warnoptions, os.environ.get("PYPARSINGENABLEALLWARNINGS")
):
enable_all_warnings()
# build list of single arg builtins, that can be used as parse actions
_single_arg_builtins = {
sum,
len,
sorted,
reversed,
list,
tuple,
set,
any,
all,
min,
max,
}
_generatorType = types.GeneratorType
ParseImplReturnType = Tuple[int, Any]
PostParseReturnType = Union[ParseResults, Sequence[ParseResults]]
ParseAction = Union[
Callable[[], Any],
Callable[[ParseResults], Any],
Callable[[int, ParseResults], Any],
Callable[[str, int, ParseResults], Any],
]
ParseCondition = Union[
Callable[[], bool],
Callable[[ParseResults], bool],
Callable[[int, ParseResults], bool],
Callable[[str, int, ParseResults], bool],
]
ParseFailAction = Callable[[str, int, "ParserElement", Exception], None]
DebugStartAction = Callable[[str, int, "ParserElement", bool], None]
DebugSuccessAction = Callable[
[str, int, int, "ParserElement", ParseResults, bool], None
]
DebugExceptionAction = Callable[[str, int, "ParserElement", Exception, bool], None]
alphas = string.ascii_uppercase + string.ascii_lowercase
identchars = pyparsing_unicode.Latin1.identchars
identbodychars = pyparsing_unicode.Latin1.identbodychars
nums = "0123456789"
hexnums = nums + "ABCDEFabcdef"
alphanums = alphas + nums
printables = "".join([c for c in string.printable if c not in string.whitespace])
_trim_arity_call_line: traceback.StackSummary = None # type: ignore[assignment]
def _trim_arity(func, max_limit=3):
"""decorator to trim function calls to match the arity of the target"""
global _trim_arity_call_line
if func in _single_arg_builtins:
return lambda s, l, t: func(t)
limit = 0
found_arity = False
# synthesize what would be returned by traceback.extract_stack at the call to
# user's parse action 'func', so that we don't incur call penalty at parse time
# fmt: off
LINE_DIFF = 7
# IF ANY CODE CHANGES, EVEN JUST COMMENTS OR BLANK LINES, BETWEEN THE NEXT LINE AND
# THE CALL TO FUNC INSIDE WRAPPER, LINE_DIFF MUST BE MODIFIED!!!!
_trim_arity_call_line = (_trim_arity_call_line or traceback.extract_stack(limit=2)[-1])
pa_call_line_synth = (_trim_arity_call_line[0], _trim_arity_call_line[1] + LINE_DIFF)
def wrapper(*args):
nonlocal found_arity, limit
while 1:
try:
ret = func(*args[limit:])
found_arity = True
return ret
except TypeError as te:
# re-raise TypeErrors if they did not come from our arity testing
if found_arity:
raise
else:
tb = te.__traceback__
frames = traceback.extract_tb(tb, limit=2)
frame_summary = frames[-1]
trim_arity_type_error = (
[frame_summary[:2]][-1][:2] == pa_call_line_synth
)
del tb
if trim_arity_type_error:
if limit < max_limit:
limit += 1
continue
raise
# fmt: on
# copy func name to wrapper for sensible debug output
# (can't use functools.wraps, since that messes with function signature)
func_name = getattr(func, "__name__", getattr(func, "__class__").__name__)
wrapper.__name__ = func_name
wrapper.__doc__ = func.__doc__
return wrapper
def condition_as_parse_action(
fn: ParseCondition, message: typing.Optional[str] = None, fatal: bool = False
) -> ParseAction:
"""
Function to convert a simple predicate function that returns ``True`` or ``False``
into a parse action. Can be used in places when a parse action is required
and :class:`ParserElement.add_condition` cannot be used (such as when adding a condition
to an operator level in :class:`infix_notation`).
Optional keyword arguments:
- ``message`` - define a custom message to be used in the raised exception
- ``fatal`` - if True, will raise :class:`ParseFatalException` to stop parsing immediately;
otherwise will raise :class:`ParseException`
"""
msg = message if message is not None else "failed user-defined condition"
exc_type = ParseFatalException if fatal else ParseException
fn = _trim_arity(fn)
@wraps(fn)
def pa(s, l, t):
if not bool(fn(s, l, t)):
raise exc_type(s, l, msg)
return pa
def _default_start_debug_action(
instring: str, loc: int, expr: "ParserElement", cache_hit: bool = False
):
cache_hit_str = "*" if cache_hit else ""
print(
(
f"{cache_hit_str}Match {expr} at loc {loc}({lineno(loc, instring)},{col(loc, instring)})\n"
f" {line(loc, instring)}\n"
f" {' ' * (col(loc, instring) - 1)}^"
)
)
def _default_success_debug_action(
instring: str,
startloc: int,
endloc: int,
expr: "ParserElement",
toks: ParseResults,
cache_hit: bool = False,
):
cache_hit_str = "*" if cache_hit else ""
print(f"{cache_hit_str}Matched {expr} -> {toks.as_list()}")
def _default_exception_debug_action(
instring: str,
loc: int,
expr: "ParserElement",
exc: Exception,
cache_hit: bool = False,
):
cache_hit_str = "*" if cache_hit else ""
print(f"{cache_hit_str}Match {expr} failed, {type(exc).__name__} raised: {exc}")
def null_debug_action(*args):
"""'Do-nothing' debug action, to suppress debugging output during parsing."""
class ParserElement(ABC):
"""Abstract base level parser element class."""
DEFAULT_WHITE_CHARS: str = " \n\t\r"
verbose_stacktrace: bool = False
_literalStringClass: type = None # type: ignore[assignment]
@staticmethod
def set_default_whitespace_chars(chars: str) -> None:
r"""
Overrides the default whitespace chars
Example::
# default whitespace chars are space, <TAB> and newline
Word(alphas)[1, ...].parse_string("abc def\nghi jkl") # -> ['abc', 'def', 'ghi', 'jkl']
# change to just treat newline as significant
ParserElement.set_default_whitespace_chars(" \t")
Word(alphas)[1, ...].parse_string("abc def\nghi jkl") # -> ['abc', 'def']
"""
ParserElement.DEFAULT_WHITE_CHARS = chars
# update whitespace all parse expressions defined in this module
for expr in _builtin_exprs:
if expr.copyDefaultWhiteChars:
expr.whiteChars = set(chars)
@staticmethod
def inline_literals_using(cls: type) -> None:
"""
Set class to be used for inclusion of string literals into a parser.
Example::
# default literal class used is Literal
integer = Word(nums)
date_str = integer("year") + '/' + integer("month") + '/' + integer("day")
date_str.parse_string("1999/12/31") # -> ['1999', '/', '12', '/', '31']
# change to Suppress
ParserElement.inline_literals_using(Suppress)
date_str = integer("year") + '/' + integer("month") + '/' + integer("day")
date_str.parse_string("1999/12/31") # -> ['1999', '12', '31']
"""
ParserElement._literalStringClass = cls
@classmethod
def using_each(cls, seq, **class_kwargs):
"""
Yields a sequence of class(obj, **class_kwargs) for obj in seq.
Example::
LPAR, RPAR, LBRACE, RBRACE, SEMI = Suppress.using_each("(){};")
"""
yield from (cls(obj, **class_kwargs) for obj in seq)
class DebugActions(NamedTuple):
debug_try: typing.Optional[DebugStartAction]
debug_match: typing.Optional[DebugSuccessAction]
debug_fail: typing.Optional[DebugExceptionAction]
def __init__(self, savelist: bool = False):
self.parseAction: List[ParseAction] = list()
self.failAction: typing.Optional[ParseFailAction] = None
self.customName: str = None # type: ignore[assignment]
self._defaultName: typing.Optional[str] = None
self.resultsName: str = None # type: ignore[assignment]
self.saveAsList = savelist
self.skipWhitespace = True
self.whiteChars = set(ParserElement.DEFAULT_WHITE_CHARS)
self.copyDefaultWhiteChars = True
# used when checking for left-recursion
self.mayReturnEmpty = False
self.keepTabs = False
self.ignoreExprs: List["ParserElement"] = list()
self.debug = False
self.streamlined = False
# optimize exception handling for subclasses that don't advance parse index
self.mayIndexError = True
self.errmsg = ""
# mark results names as modal (report only last) or cumulative (list all)
self.modalResults = True
# custom debug actions
self.debugActions = self.DebugActions(None, None, None)
# avoid redundant calls to preParse
self.callPreparse = True
self.callDuringTry = False
self.suppress_warnings_: List[Diagnostics] = []
def suppress_warning(self, warning_type: Diagnostics) -> "ParserElement":
"""
Suppress warnings emitted for a particular diagnostic on this expression.
Example::
base = pp.Forward()
base.suppress_warning(Diagnostics.warn_on_parse_using_empty_Forward)
# statement would normally raise a warning, but is now suppressed
print(base.parse_string("x"))
"""
self.suppress_warnings_.append(warning_type)
return self
def visit_all(self):
"""General-purpose method to yield all expressions and sub-expressions
in a grammar. Typically just for internal use.
"""
to_visit = deque([self])
seen = set()
while to_visit:
cur = to_visit.popleft()
# guard against looping forever through recursive grammars
if cur in seen:
continue
seen.add(cur)
to_visit.extend(cur.recurse())
yield cur
def copy(self) -> "ParserElement":
"""
Make a copy of this :class:`ParserElement`. Useful for defining
different parse actions for the same parsing pattern, using copies of
the original parse element.
Example::
integer = Word(nums).set_parse_action(lambda toks: int(toks[0]))
integerK = integer.copy().add_parse_action(lambda toks: toks[0] * 1024) + Suppress("K")
integerM = integer.copy().add_parse_action(lambda toks: toks[0] * 1024 * 1024) + Suppress("M")
print((integerK | integerM | integer)[1, ...].parse_string("5K 100 640K 256M"))
prints::
[5120, 100, 655360, 268435456]
Equivalent form of ``expr.copy()`` is just ``expr()``::
integerM = integer().add_parse_action(lambda toks: toks[0] * 1024 * 1024) + Suppress("M")
"""
cpy = copy.copy(self)
cpy.parseAction = self.parseAction[:]
cpy.ignoreExprs = self.ignoreExprs[:]
if self.copyDefaultWhiteChars:
cpy.whiteChars = set(ParserElement.DEFAULT_WHITE_CHARS)
return cpy
def set_results_name(
self, name: str, list_all_matches: bool = False, *, listAllMatches: bool = False
) -> "ParserElement":
"""
Define name for referencing matching tokens as a nested attribute
of the returned parse results.
Normally, results names are assigned as you would assign keys in a dict:
any existing value is overwritten by later values. If it is necessary to
keep all values captured for a particular results name, call ``set_results_name``
with ``list_all_matches`` = True.
NOTE: ``set_results_name`` returns a *copy* of the original :class:`ParserElement` object;
this is so that the client can define a basic element, such as an
integer, and reference it in multiple places with different names.
You can also set results names using the abbreviated syntax,
``expr("name")`` in place of ``expr.set_results_name("name")``
- see :class:`__call__`. If ``list_all_matches`` is required, use
``expr("name*")``.
Example::
date_str = (integer.set_results_name("year") + '/'
+ integer.set_results_name("month") + '/'
+ integer.set_results_name("day"))
# equivalent form:
date_str = integer("year") + '/' + integer("month") + '/' + integer("day")
"""
listAllMatches = listAllMatches or list_all_matches
return self._setResultsName(name, listAllMatches)
def _setResultsName(self, name, listAllMatches=False):
if name is None:
return self
newself = self.copy()
if name.endswith("*"):
name = name[:-1]
listAllMatches = True
newself.resultsName = name
newself.modalResults = not listAllMatches
return newself
def set_break(self, break_flag: bool = True) -> "ParserElement":
"""
Method to invoke the Python pdb debugger when this element is
about to be parsed. Set ``break_flag`` to ``True`` to enable, ``False`` to
disable.
"""
if break_flag:
_parseMethod = self._parse
def breaker(instring, loc, doActions=True, callPreParse=True):
import pdb
# this call to pdb.set_trace() is intentional, not a checkin error
pdb.set_trace()
return _parseMethod(instring, loc, doActions, callPreParse)
breaker._originalParseMethod = _parseMethod # type: ignore [attr-defined]
self._parse = breaker # type: ignore [assignment]
else:
if hasattr(self._parse, "_originalParseMethod"):
self._parse = self._parse._originalParseMethod # type: ignore [attr-defined, assignment]
return self
def set_parse_action(self, *fns: ParseAction, **kwargs) -> "ParserElement":
"""
Define one or more actions to perform when successfully matching parse element definition.
Parse actions can be called to perform data conversions, do extra validation,
update external data structures, or enhance or replace the parsed tokens.
Each parse action ``fn`` is a callable method with 0-3 arguments, called as
``fn(s, loc, toks)`` , ``fn(loc, toks)`` , ``fn(toks)`` , or just ``fn()`` , where:
- ``s`` = the original string being parsed (see note below)
- ``loc`` = the location of the matching substring
- ``toks`` = a list of the matched tokens, packaged as a :class:`ParseResults` object
The parsed tokens are passed to the parse action as ParseResults. They can be
modified in place using list-style append, extend, and pop operations to update
the parsed list elements; and with dictionary-style item set and del operations
to add, update, or remove any named results. If the tokens are modified in place,
it is not necessary to return them with a return statement.
Parse actions can also completely replace the given tokens, with another ``ParseResults``
object, or with some entirely different object (common for parse actions that perform data
conversions). A convenient way to build a new parse result is to define the values
using a dict, and then create the return value using :class:`ParseResults.from_dict`.
If None is passed as the ``fn`` parse action, all previously added parse actions for this
expression are cleared.
Optional keyword arguments:
- ``call_during_try`` = (default= ``False``) indicate if parse action should be run during
lookaheads and alternate testing. For parse actions that have side effects, it is
important to only call the parse action once it is determined that it is being
called as part of a successful parse. For parse actions that perform additional
validation, then call_during_try should be passed as True, so that the validation
code is included in the preliminary "try" parses.
Note: the default parsing behavior is to expand tabs in the input string
before starting the parsing process. See :class:`parse_string` for more
information on parsing strings containing ``<TAB>`` s, and suggested
methods to maintain a consistent view of the parsed string, the parse
location, and line and column positions within the parsed string.
Example::
# parse dates in the form YYYY/MM/DD
# use parse action to convert toks from str to int at parse time
def convert_to_int(toks):
return int(toks[0])
# use a parse action to verify that the date is a valid date
def is_valid_date(instring, loc, toks):
from datetime import date
year, month, day = toks[::2]
try:
date(year, month, day)
except ValueError:
raise ParseException(instring, loc, "invalid date given")
integer = Word(nums)
date_str = integer + '/' + integer + '/' + integer
# add parse actions
integer.set_parse_action(convert_to_int)
date_str.set_parse_action(is_valid_date)
# note that integer fields are now ints, not strings
date_str.run_tests('''
# successful parse - note that integer fields were converted to ints
1999/12/31
# fail - invalid date
1999/13/31
''')
"""
if list(fns) == [None]:
self.parseAction = []
else:
if not all(callable(fn) for fn in fns):
raise TypeError("parse actions must be callable")
self.parseAction = [_trim_arity(fn) for fn in fns]
self.callDuringTry = kwargs.get(
"call_during_try", kwargs.get("callDuringTry", False)
)
return self
def add_parse_action(self, *fns: ParseAction, **kwargs) -> "ParserElement":
"""
Add one or more parse actions to expression's list of parse actions. See :class:`set_parse_action`.
See examples in :class:`copy`.
"""
self.parseAction += [_trim_arity(fn) for fn in fns]
self.callDuringTry = self.callDuringTry or kwargs.get(
"call_during_try", kwargs.get("callDuringTry", False)
)
return self
def add_condition(self, *fns: ParseCondition, **kwargs) -> "ParserElement":
"""Add a boolean predicate function to expression's list of parse actions. See
:class:`set_parse_action` for function call signatures. Unlike ``set_parse_action``,
functions passed to ``add_condition`` need to return boolean success/fail of the condition.
Optional keyword arguments:
- ``message`` = define a custom message to be used in the raised exception
- ``fatal`` = if True, will raise ParseFatalException to stop parsing immediately; otherwise will raise
ParseException
- ``call_during_try`` = boolean to indicate if this method should be called during internal tryParse calls,
default=False
Example::
integer = Word(nums).set_parse_action(lambda toks: int(toks[0]))
year_int = integer.copy()
year_int.add_condition(lambda toks: toks[0] >= 2000, message="Only support years 2000 and later")
date_str = year_int + '/' + integer + '/' + integer
result = date_str.parse_string("1999/12/31") # -> Exception: Only support years 2000 and later (at char 0),
(line:1, col:1)
"""
for fn in fns:
self.parseAction.append(
condition_as_parse_action(
fn,
message=str(kwargs.get("message")),
fatal=bool(kwargs.get("fatal", False)),
)
)
self.callDuringTry = self.callDuringTry or kwargs.get(
"call_during_try", kwargs.get("callDuringTry", False)
)
return self
def set_fail_action(self, fn: ParseFailAction) -> "ParserElement":
"""
Define action to perform if parsing fails at this expression.
Fail acton fn is a callable function that takes the arguments
``fn(s, loc, expr, err)`` where:
- ``s`` = string being parsed
- ``loc`` = location where expression match was attempted and failed
- ``expr`` = the parse expression that failed
- ``err`` = the exception thrown
The function returns no value. It may throw :class:`ParseFatalException`
if it is desired to stop parsing immediately."""
self.failAction = fn
return self
def _skipIgnorables(self, instring: str, loc: int) -> int:
if not self.ignoreExprs:
return loc
exprsFound = True
ignore_expr_fns = [e._parse for e in self.ignoreExprs]
while exprsFound:
exprsFound = False
for ignore_fn in ignore_expr_fns:
try:
while 1:
loc, dummy = ignore_fn(instring, loc)
exprsFound = True
except ParseException:
pass
return loc
def preParse(self, instring: str, loc: int) -> int:
if self.ignoreExprs:
loc = self._skipIgnorables(instring, loc)
if self.skipWhitespace:
instrlen = len(instring)
white_chars = self.whiteChars
while loc < instrlen and instring[loc] in white_chars:
loc += 1
return loc
def parseImpl(self, instring, loc, doActions=True):
return loc, []
def postParse(self, instring, loc, tokenlist):
return tokenlist
# @profile
def _parseNoCache(
self, instring, loc, doActions=True, callPreParse=True
) -> Tuple[int, ParseResults]:
TRY, MATCH, FAIL = 0, 1, 2
debugging = self.debug # and doActions)
len_instring = len(instring)
if debugging or self.failAction:
# print("Match {} at loc {}({}, {})".format(self, loc, lineno(loc, instring), col(loc, instring)))
try:
if callPreParse and self.callPreparse:
pre_loc = self.preParse(instring, loc)
else:
pre_loc = loc
tokens_start = pre_loc
if self.debugActions.debug_try:
self.debugActions.debug_try(instring, tokens_start, self, False)
if self.mayIndexError or pre_loc >= len_instring:
try:
loc, tokens = self.parseImpl(instring, pre_loc, doActions)
except IndexError:
raise ParseException(instring, len_instring, self.errmsg, self)
else:
loc, tokens = self.parseImpl(instring, pre_loc, doActions)
except Exception as err:
# print("Exception raised:", err)
if self.debugActions.debug_fail:
self.debugActions.debug_fail(
instring, tokens_start, self, err, False
)
if self.failAction:
self.failAction(instring, tokens_start, self, err)
raise
else:
if callPreParse and self.callPreparse:
pre_loc = self.preParse(instring, loc)
else:
pre_loc = loc
tokens_start = pre_loc
if self.mayIndexError or pre_loc >= len_instring:
try:
loc, tokens = self.parseImpl(instring, pre_loc, doActions)
except IndexError:
raise ParseException(instring, len_instring, self.errmsg, self)
else:
loc, tokens = self.parseImpl(instring, pre_loc, doActions)
tokens = self.postParse(instring, loc, tokens)
ret_tokens = ParseResults(
tokens, self.resultsName, asList=self.saveAsList, modal=self.modalResults
)
if self.parseAction and (doActions or self.callDuringTry):
if debugging:
try:
for fn in self.parseAction:
try:
tokens = fn(instring, tokens_start, ret_tokens) # type: ignore [call-arg, arg-type]
except IndexError as parse_action_exc:
exc = ParseException("exception raised in parse action")
raise exc from parse_action_exc
if tokens is not None and tokens is not ret_tokens:
ret_tokens = ParseResults(
tokens,
self.resultsName,
asList=self.saveAsList
and isinstance(tokens, (ParseResults, list)),
modal=self.modalResults,
)
except Exception as err:
# print "Exception raised in user parse action:", err
if self.debugActions.debug_fail:
self.debugActions.debug_fail(
instring, tokens_start, self, err, False
)
raise
else:
for fn in self.parseAction:
try:
tokens = fn(instring, tokens_start, ret_tokens) # type: ignore [call-arg, arg-type]
except IndexError as parse_action_exc:
exc = ParseException("exception raised in parse action")
raise exc from parse_action_exc
if tokens is not None and tokens is not ret_tokens:
ret_tokens = ParseResults(
tokens,
self.resultsName,
asList=self.saveAsList
and isinstance(tokens, (ParseResults, list)),
modal=self.modalResults,
)
if debugging:
# print("Matched", self, "->", ret_tokens.as_list())
if self.debugActions.debug_match:
self.debugActions.debug_match(
instring, tokens_start, loc, self, ret_tokens, False
)
return loc, ret_tokens
def try_parse(
self,
instring: str,
loc: int,
*,
raise_fatal: bool = False,
do_actions: bool = False,
) -> int:
try:
return self._parse(instring, loc, doActions=do_actions)[0]
except ParseFatalException:
if raise_fatal:
raise
raise ParseException(instring, loc, self.errmsg, self)
def can_parse_next(self, instring: str, loc: int, do_actions: bool = False) -> bool:
try:
self.try_parse(instring, loc, do_actions=do_actions)
except (ParseException, IndexError):
return False
else:
return True
# cache for left-recursion in Forward references
recursion_lock = RLock()
recursion_memos: typing.Dict[
Tuple[int, "Forward", bool], Tuple[int, Union[ParseResults, Exception]]
] = {}
class _CacheType(dict):
"""
class to help type checking
"""
not_in_cache: bool
def get(self, *args):
...
def set(self, *args):
...
# argument cache for optimizing repeated calls when backtracking through recursive expressions
packrat_cache = (
_CacheType()
) # set later by enable_packrat(); this is here so that reset_cache() doesn't fail
packrat_cache_lock = RLock()
packrat_cache_stats = [0, 0]
# this method gets repeatedly called during backtracking with the same arguments -
# we can cache these arguments and save ourselves the trouble of re-parsing the contained expression
def _parseCache(
self, instring, loc, doActions=True, callPreParse=True
) -> Tuple[int, ParseResults]:
HIT, MISS = 0, 1
TRY, MATCH, FAIL = 0, 1, 2
lookup = (self, instring, loc, callPreParse, doActions)
with ParserElement.packrat_cache_lock:
cache = ParserElement.packrat_cache
value = cache.get(lookup)
if value is cache.not_in_cache:
ParserElement.packrat_cache_stats[MISS] += 1
try:
value = self._parseNoCache(instring, loc, doActions, callPreParse)
except ParseBaseException as pe:
# cache a copy of the exception, without the traceback
cache.set(lookup, pe.__class__(*pe.args))
raise
else:
cache.set(lookup, (value[0], value[1].copy(), loc))
return value
else:
ParserElement.packrat_cache_stats[HIT] += 1
if self.debug and self.debugActions.debug_try:
try:
self.debugActions.debug_try(instring, loc, self, cache_hit=True) # type: ignore [call-arg]
except TypeError:
pass
if isinstance(value, Exception):
if self.debug and self.debugActions.debug_fail:
try:
self.debugActions.debug_fail(
instring, loc, self, value, cache_hit=True # type: ignore [call-arg]
)
except TypeError:
pass
raise value
value = cast(Tuple[int, ParseResults, int], value)
loc_, result, endloc = value[0], value[1].copy(), value[2]
if self.debug and self.debugActions.debug_match:
try:
self.debugActions.debug_match(
instring, loc_, endloc, self, result, cache_hit=True # type: ignore [call-arg]
)
except TypeError:
pass
return loc_, result
_parse = _parseNoCache
@staticmethod
def reset_cache() -> None:
ParserElement.packrat_cache.clear()
ParserElement.packrat_cache_stats[:] = [0] * len(
ParserElement.packrat_cache_stats
)
ParserElement.recursion_memos.clear()
_packratEnabled = False
_left_recursion_enabled = False
@staticmethod
def disable_memoization() -> None:
"""
Disables active Packrat or Left Recursion parsing and their memoization
This method also works if neither Packrat nor Left Recursion are enabled.
This makes it safe to call before activating Packrat nor Left Recursion
to clear any previous settings.
"""
ParserElement.reset_cache()
ParserElement._left_recursion_enabled = False
ParserElement._packratEnabled = False
ParserElement._parse = ParserElement._parseNoCache
@staticmethod
def enable_left_recursion(
cache_size_limit: typing.Optional[int] = None, *, force=False
) -> None:
"""
Enables "bounded recursion" parsing, which allows for both direct and indirect
left-recursion. During parsing, left-recursive :class:`Forward` elements are
repeatedly matched with a fixed recursion depth that is gradually increased
until finding the longest match.
Example::
from pip._vendor import pyparsing as pp
pp.ParserElement.enable_left_recursion()
E = pp.Forward("E")
num = pp.Word(pp.nums)
# match `num`, or `num '+' num`, or `num '+' num '+' num`, ...
E <<= E + '+' - num | num
print(E.parse_string("1+2+3"))
Recursion search naturally memoizes matches of ``Forward`` elements and may
thus skip reevaluation of parse actions during backtracking. This may break
programs with parse actions which rely on strict ordering of side-effects.
Parameters:
- ``cache_size_limit`` - (default=``None``) - memoize at most this many
``Forward`` elements during matching; if ``None`` (the default),
memoize all ``Forward`` elements.
Bounded Recursion parsing works similar but not identical to Packrat parsing,
thus the two cannot be used together. Use ``force=True`` to disable any
previous, conflicting settings.
"""
if force:
ParserElement.disable_memoization()
elif ParserElement._packratEnabled:
raise RuntimeError("Packrat and Bounded Recursion are not compatible")
if cache_size_limit is None:
ParserElement.recursion_memos = _UnboundedMemo() # type: ignore[assignment]
elif cache_size_limit > 0:
ParserElement.recursion_memos = _LRUMemo(capacity=cache_size_limit) # type: ignore[assignment]
else:
raise NotImplementedError("Memo size of %s" % cache_size_limit)
ParserElement._left_recursion_enabled = True
@staticmethod
def enable_packrat(cache_size_limit: int = 128, *, force: bool = False) -> None:
"""
Enables "packrat" parsing, which adds memoizing to the parsing logic.
Repeated parse attempts at the same string location (which happens
often in many complex grammars) can immediately return a cached value,
instead of re-executing parsing/validating code. Memoizing is done of
both valid results and parsing exceptions.
Parameters:
- ``cache_size_limit`` - (default= ``128``) - if an integer value is provided
will limit the size of the packrat cache; if None is passed, then
the cache size will be unbounded; if 0 is passed, the cache will
be effectively disabled.
This speedup may break existing programs that use parse actions that
have side-effects. For this reason, packrat parsing is disabled when
you first import pyparsing. To activate the packrat feature, your
program must call the class method :class:`ParserElement.enable_packrat`.
For best results, call ``enable_packrat()`` immediately after
importing pyparsing.
Example::
from pip._vendor import pyparsing
pyparsing.ParserElement.enable_packrat()
Packrat parsing works similar but not identical to Bounded Recursion parsing,
thus the two cannot be used together. Use ``force=True`` to disable any
previous, conflicting settings.
"""
if force:
ParserElement.disable_memoization()
elif ParserElement._left_recursion_enabled:
raise RuntimeError("Packrat and Bounded Recursion are not compatible")
if not ParserElement._packratEnabled:
ParserElement._packratEnabled = True
if cache_size_limit is None:
ParserElement.packrat_cache = _UnboundedCache()
else:
ParserElement.packrat_cache = _FifoCache(cache_size_limit) # type: ignore[assignment]
ParserElement._parse = ParserElement._parseCache
def parse_string(
self, instring: str, parse_all: bool = False, *, parseAll: bool = False
) -> ParseResults:
"""
Parse a string with respect to the parser definition. This function is intended as the primary interface to the
client code.
:param instring: The input string to be parsed.
:param parse_all: If set, the entire input string must match the grammar.
:param parseAll: retained for pre-PEP8 compatibility, will be removed in a future release.
:raises ParseException: Raised if ``parse_all`` is set and the input string does not match the whole grammar.
:returns: the parsed data as a :class:`ParseResults` object, which may be accessed as a `list`, a `dict`, or
an object with attributes if the given parser includes results names.
If the input string is required to match the entire grammar, ``parse_all`` flag must be set to ``True``. This
is also equivalent to ending the grammar with :class:`StringEnd`\\ ().
To report proper column numbers, ``parse_string`` operates on a copy of the input string where all tabs are
converted to spaces (8 spaces per tab, as per the default in ``string.expandtabs``). If the input string
contains tabs and the grammar uses parse actions that use the ``loc`` argument to index into the string
being parsed, one can ensure a consistent view of the input string by doing one of the following:
- calling ``parse_with_tabs`` on your grammar before calling ``parse_string`` (see :class:`parse_with_tabs`),
- define your parse action using the full ``(s,loc,toks)`` signature, and reference the input string using the
parse action's ``s`` argument, or
- explicitly expand the tabs in your input string before calling ``parse_string``.
Examples:
By default, partial matches are OK.
>>> res = Word('a').parse_string('aaaaabaaa')
>>> print(res)
['aaaaa']
The parsing behavior varies by the inheriting class of this abstract class. Please refer to the children
directly to see more examples.
It raises an exception if parse_all flag is set and instring does not match the whole grammar.
>>> res = Word('a').parse_string('aaaaabaaa', parse_all=True)
Traceback (most recent call last):
...
pyparsing.ParseException: Expected end of text, found 'b' (at char 5), (line:1, col:6)
"""
parseAll = parse_all or parseAll
ParserElement.reset_cache()
if not self.streamlined:
self.streamline()
for e in self.ignoreExprs:
e.streamline()
if not self.keepTabs:
instring = instring.expandtabs()
try:
loc, tokens = self._parse(instring, 0)
if parseAll:
loc = self.preParse(instring, loc)
se = Empty() + StringEnd()
se._parse(instring, loc)
except ParseBaseException as exc:
if ParserElement.verbose_stacktrace:
raise
else:
# catch and re-raise exception from here, clearing out pyparsing internal stack trace
raise exc.with_traceback(None)
else:
return tokens
def scan_string(
self,
instring: str,
max_matches: int = _MAX_INT,
overlap: bool = False,
*,
debug: bool = False,
maxMatches: int = _MAX_INT,
) -> Generator[Tuple[ParseResults, int, int], None, None]:
"""
Scan the input string for expression matches. Each match will return the
matching tokens, start location, and end location. May be called with optional
``max_matches`` argument, to clip scanning after 'n' matches are found. If
``overlap`` is specified, then overlapping matches will be reported.
Note that the start and end locations are reported relative to the string
being parsed. See :class:`parse_string` for more information on parsing
strings with embedded tabs.
Example::
source = "sldjf123lsdjjkf345sldkjf879lkjsfd987"
print(source)
for tokens, start, end in Word(alphas).scan_string(source):
print(' '*start + '^'*(end-start))
print(' '*start + tokens[0])
prints::
sldjf123lsdjjkf345sldkjf879lkjsfd987
^^^^^
sldjf
^^^^^^^
lsdjjkf
^^^^^^
sldkjf
^^^^^^
lkjsfd
"""
maxMatches = min(maxMatches, max_matches)
if not self.streamlined:
self.streamline()
for e in self.ignoreExprs:
e.streamline()
if not self.keepTabs:
instring = str(instring).expandtabs()
instrlen = len(instring)
loc = 0
preparseFn = self.preParse
parseFn = self._parse
ParserElement.resetCache()
matches = 0
try:
while loc <= instrlen and matches < maxMatches:
try:
preloc: int = preparseFn(instring, loc)
nextLoc: int
tokens: ParseResults
nextLoc, tokens = parseFn(instring, preloc, callPreParse=False)
except ParseException:
loc = preloc + 1
else:
if nextLoc > loc:
matches += 1
if debug:
print(
{
"tokens": tokens.asList(),
"start": preloc,
"end": nextLoc,
}
)
yield tokens, preloc, nextLoc
if overlap:
nextloc = preparseFn(instring, loc)
if nextloc > loc:
loc = nextLoc
else:
loc += 1
else:
loc = nextLoc
else:
loc = preloc + 1
except ParseBaseException as exc:
if ParserElement.verbose_stacktrace:
raise
else:
# catch and re-raise exception from here, clears out pyparsing internal stack trace
raise exc.with_traceback(None)
def transform_string(self, instring: str, *, debug: bool = False) -> str:
"""
Extension to :class:`scan_string`, to modify matching text with modified tokens that may
be returned from a parse action. To use ``transform_string``, define a grammar and
attach a parse action to it that modifies the returned token list.
Invoking ``transform_string()`` on a target string will then scan for matches,
and replace the matched text patterns according to the logic in the parse
action. ``transform_string()`` returns the resulting transformed string.
Example::
wd = Word(alphas)
wd.set_parse_action(lambda toks: toks[0].title())
print(wd.transform_string("now is the winter of our discontent made glorious summer by this sun of york."))
prints::
Now Is The Winter Of Our Discontent Made Glorious Summer By This Sun Of York.
"""
out: List[str] = []
lastE = 0
# force preservation of <TAB>s, to minimize unwanted transformation of string, and to
# keep string locs straight between transform_string and scan_string
self.keepTabs = True
try:
for t, s, e in self.scan_string(instring, debug=debug):
out.append(instring[lastE:s])
if t:
if isinstance(t, ParseResults):
out += t.as_list()
elif isinstance(t, Iterable) and not isinstance(t, str_type):
out.extend(t)
else:
out.append(t)
lastE = e
out.append(instring[lastE:])
out = [o for o in out if o]
return "".join([str(s) for s in _flatten(out)])
except ParseBaseException as exc:
if ParserElement.verbose_stacktrace:
raise
else:
# catch and re-raise exception from here, clears out pyparsing internal stack trace
raise exc.with_traceback(None)
def search_string(
self,
instring: str,
max_matches: int = _MAX_INT,
*,
debug: bool = False,
maxMatches: int = _MAX_INT,
) -> ParseResults:
"""
Another extension to :class:`scan_string`, simplifying the access to the tokens found
to match the given parse expression. May be called with optional
``max_matches`` argument, to clip searching after 'n' matches are found.
Example::
# a capitalized word starts with an uppercase letter, followed by zero or more lowercase letters
cap_word = Word(alphas.upper(), alphas.lower())
print(cap_word.search_string("More than Iron, more than Lead, more than Gold I need Electricity"))
# the sum() builtin can be used to merge results into a single ParseResults object
print(sum(cap_word.search_string("More than Iron, more than Lead, more than Gold I need Electricity")))
prints::
[['More'], ['Iron'], ['Lead'], ['Gold'], ['I'], ['Electricity']]
['More', 'Iron', 'Lead', 'Gold', 'I', 'Electricity']
"""
maxMatches = min(maxMatches, max_matches)
try:
return ParseResults(
[t for t, s, e in self.scan_string(instring, maxMatches, debug=debug)]
)
except ParseBaseException as exc:
if ParserElement.verbose_stacktrace:
raise
else:
# catch and re-raise exception from here, clears out pyparsing internal stack trace
raise exc.with_traceback(None)
def split(
self,
instring: str,
maxsplit: int = _MAX_INT,
include_separators: bool = False,
*,
includeSeparators=False,
) -> Generator[str, None, None]:
"""
Generator method to split a string using the given expression as a separator.
May be called with optional ``maxsplit`` argument, to limit the number of splits;
and the optional ``include_separators`` argument (default= ``False``), if the separating
matching text should be included in the split results.
Example::
punc = one_of(list(".,;:/-!?"))
print(list(punc.split("This, this?, this sentence, is badly punctuated!")))
prints::
['This', ' this', '', ' this sentence', ' is badly punctuated', '']
"""
includeSeparators = includeSeparators or include_separators
last = 0
for t, s, e in self.scan_string(instring, max_matches=maxsplit):
yield instring[last:s]
if includeSeparators:
yield t[0]
last = e
yield instring[last:]
def __add__(self, other) -> "ParserElement":
"""
Implementation of ``+`` operator - returns :class:`And`. Adding strings to a :class:`ParserElement`
converts them to :class:`Literal`\\ s by default.
Example::
greet = Word(alphas) + "," + Word(alphas) + "!"
hello = "Hello, World!"
print(hello, "->", greet.parse_string(hello))
prints::
Hello, World! -> ['Hello', ',', 'World', '!']
``...`` may be used as a parse expression as a short form of :class:`SkipTo`::
Literal('start') + ... + Literal('end')
is equivalent to::
Literal('start') + SkipTo('end')("_skipped*") + Literal('end')
Note that the skipped text is returned with '_skipped' as a results name,
and to support having multiple skips in the same parser, the value returned is
a list of all skipped text.
"""
if other is Ellipsis:
return _PendingSkip(self)
if isinstance(other, str_type):
other = self._literalStringClass(other)
if not isinstance(other, ParserElement):
return NotImplemented
return And([self, other])
def __radd__(self, other) -> "ParserElement":
"""
Implementation of ``+`` operator when left operand is not a :class:`ParserElement`
"""
if other is Ellipsis:
return SkipTo(self)("_skipped*") + self
if isinstance(other, str_type):
other = self._literalStringClass(other)
if not isinstance(other, ParserElement):
return NotImplemented
return other + self
def __sub__(self, other) -> "ParserElement":
"""
Implementation of ``-`` operator, returns :class:`And` with error stop
"""
if isinstance(other, str_type):
other = self._literalStringClass(other)
if not isinstance(other, ParserElement):
return NotImplemented
return self + And._ErrorStop() + other
def __rsub__(self, other) -> "ParserElement":
"""
Implementation of ``-`` operator when left operand is not a :class:`ParserElement`
"""
if isinstance(other, str_type):
other = self._literalStringClass(other)
if not isinstance(other, ParserElement):
return NotImplemented
return other - self
def __mul__(self, other) -> "ParserElement":
"""
Implementation of ``*`` operator, allows use of ``expr * 3`` in place of
``expr + expr + expr``. Expressions may also be multiplied by a 2-integer
tuple, similar to ``{min, max}`` multipliers in regular expressions. Tuples
may also include ``None`` as in:
- ``expr*(n, None)`` or ``expr*(n, )`` is equivalent
to ``expr*n + ZeroOrMore(expr)``
(read as "at least n instances of ``expr``")
- ``expr*(None, n)`` is equivalent to ``expr*(0, n)``
(read as "0 to n instances of ``expr``")
- ``expr*(None, None)`` is equivalent to ``ZeroOrMore(expr)``
- ``expr*(1, None)`` is equivalent to ``OneOrMore(expr)``
Note that ``expr*(None, n)`` does not raise an exception if
more than n exprs exist in the input stream; that is,
``expr*(None, n)`` does not enforce a maximum number of expr
occurrences. If this behavior is desired, then write
``expr*(None, n) + ~expr``
"""
if other is Ellipsis:
other = (0, None)
elif isinstance(other, tuple) and other[:1] == (Ellipsis,):
other = ((0,) + other[1:] + (None,))[:2]
if isinstance(other, int):
minElements, optElements = other, 0
elif isinstance(other, tuple):
other = tuple(o if o is not Ellipsis else None for o in other)
other = (other + (None, None))[:2]
if other[0] is None:
other = (0, other[1])
if isinstance(other[0], int) and other[1] is None:
if other[0] == 0:
return ZeroOrMore(self)
if other[0] == 1:
return OneOrMore(self)
else:
return self * other[0] + ZeroOrMore(self)
elif isinstance(other[0], int) and isinstance(other[1], int):
minElements, optElements = other
optElements -= minElements
else:
return NotImplemented
else:
return NotImplemented
if minElements < 0:
raise ValueError("cannot multiply ParserElement by negative value")
if optElements < 0:
raise ValueError(
"second tuple value must be greater or equal to first tuple value"
)
if minElements == optElements == 0:
return And([])
if optElements:
def makeOptionalList(n):
if n > 1:
return Opt(self + makeOptionalList(n - 1))
else:
return Opt(self)
if minElements:
if minElements == 1:
ret = self + makeOptionalList(optElements)
else:
ret = And([self] * minElements) + makeOptionalList(optElements)
else:
ret = makeOptionalList(optElements)
else:
if minElements == 1:
ret = self
else:
ret = And([self] * minElements)
return ret
def __rmul__(self, other) -> "ParserElement":
return self.__mul__(other)
def __or__(self, other) -> "ParserElement":
"""
Implementation of ``|`` operator - returns :class:`MatchFirst`
"""
if other is Ellipsis:
return _PendingSkip(self, must_skip=True)
if isinstance(other, str_type):
# `expr | ""` is equivalent to `Opt(expr)`
if other == "":
return Opt(self)
other = self._literalStringClass(other)
if not isinstance(other, ParserElement):
return NotImplemented
return MatchFirst([self, other])
def __ror__(self, other) -> "ParserElement":
"""
Implementation of ``|`` operator when left operand is not a :class:`ParserElement`
"""
if isinstance(other, str_type):
other = self._literalStringClass(other)
if not isinstance(other, ParserElement):
return NotImplemented
return other | self
def __xor__(self, other) -> "ParserElement":
"""
Implementation of ``^`` operator - returns :class:`Or`
"""
if isinstance(other, str_type):
other = self._literalStringClass(other)
if not isinstance(other, ParserElement):
return NotImplemented
return Or([self, other])
def __rxor__(self, other) -> "ParserElement":
"""
Implementation of ``^`` operator when left operand is not a :class:`ParserElement`
"""
if isinstance(other, str_type):
other = self._literalStringClass(other)
if not isinstance(other, ParserElement):
return NotImplemented
return other ^ self
def __and__(self, other) -> "ParserElement":
"""
Implementation of ``&`` operator - returns :class:`Each`
"""
if isinstance(other, str_type):
other = self._literalStringClass(other)
if not isinstance(other, ParserElement):
return NotImplemented
return Each([self, other])
def __rand__(self, other) -> "ParserElement":
"""
Implementation of ``&`` operator when left operand is not a :class:`ParserElement`
"""
if isinstance(other, str_type):
other = self._literalStringClass(other)
if not isinstance(other, ParserElement):
return NotImplemented
return other & self
def __invert__(self) -> "ParserElement":
"""
Implementation of ``~`` operator - returns :class:`NotAny`
"""
return NotAny(self)
# disable __iter__ to override legacy use of sequential access to __getitem__ to
# iterate over a sequence
__iter__ = None
def __getitem__(self, key):
"""
use ``[]`` indexing notation as a short form for expression repetition:
- ``expr[n]`` is equivalent to ``expr*n``
- ``expr[m, n]`` is equivalent to ``expr*(m, n)``
- ``expr[n, ...]`` or ``expr[n,]`` is equivalent
to ``expr*n + ZeroOrMore(expr)``
(read as "at least n instances of ``expr``")
- ``expr[..., n]`` is equivalent to ``expr*(0, n)``
(read as "0 to n instances of ``expr``")
- ``expr[...]`` and ``expr[0, ...]`` are equivalent to ``ZeroOrMore(expr)``
- ``expr[1, ...]`` is equivalent to ``OneOrMore(expr)``
``None`` may be used in place of ``...``.
Note that ``expr[..., n]`` and ``expr[m, n]`` do not raise an exception
if more than ``n`` ``expr``\\ s exist in the input stream. If this behavior is
desired, then write ``expr[..., n] + ~expr``.
For repetition with a stop_on expression, use slice notation:
- ``expr[...: end_expr]`` and ``expr[0, ...: end_expr]`` are equivalent to ``ZeroOrMore(expr, stop_on=end_expr)``
- ``expr[1, ...: end_expr]`` is equivalent to ``OneOrMore(expr, stop_on=end_expr)``
"""
stop_on_defined = False
stop_on = NoMatch()
if isinstance(key, slice):
key, stop_on = key.start, key.stop
if key is None:
key = ...
stop_on_defined = True
elif isinstance(key, tuple) and isinstance(key[-1], slice):
key, stop_on = (key[0], key[1].start), key[1].stop
stop_on_defined = True
# convert single arg keys to tuples
if isinstance(key, str_type):
key = (key,)
try:
iter(key)
except TypeError:
key = (key, key)
if len(key) > 2:
raise TypeError(
f"only 1 or 2 index arguments supported ({key[:5]}{f'... [{len(key)}]' if len(key) > 5 else ''})"
)
# clip to 2 elements
ret = self * tuple(key[:2])
ret = typing.cast(_MultipleMatch, ret)
if stop_on_defined:
ret.stopOn(stop_on)
return ret
def __call__(self, name: typing.Optional[str] = None) -> "ParserElement":
"""
Shortcut for :class:`set_results_name`, with ``list_all_matches=False``.
If ``name`` is given with a trailing ``'*'`` character, then ``list_all_matches`` will be
passed as ``True``.
If ``name`` is omitted, same as calling :class:`copy`.
Example::
# these are equivalent
userdata = Word(alphas).set_results_name("name") + Word(nums + "-").set_results_name("socsecno")
userdata = Word(alphas)("name") + Word(nums + "-")("socsecno")
"""
if name is not None:
return self._setResultsName(name)
else:
return self.copy()
def suppress(self) -> "ParserElement":
"""
Suppresses the output of this :class:`ParserElement`; useful to keep punctuation from
cluttering up returned output.
"""
return Suppress(self)
def ignore_whitespace(self, recursive: bool = True) -> "ParserElement":
"""
Enables the skipping of whitespace before matching the characters in the
:class:`ParserElement`'s defined pattern.
:param recursive: If ``True`` (the default), also enable whitespace skipping in child elements (if any)
"""
self.skipWhitespace = True
return self
def leave_whitespace(self, recursive: bool = True) -> "ParserElement":
"""
Disables the skipping of whitespace before matching the characters in the
:class:`ParserElement`'s defined pattern. This is normally only used internally by
the pyparsing module, but may be needed in some whitespace-sensitive grammars.
:param recursive: If true (the default), also disable whitespace skipping in child elements (if any)
"""
self.skipWhitespace = False
return self
def set_whitespace_chars(
self, chars: Union[Set[str], str], copy_defaults: bool = False
) -> "ParserElement":
"""
Overrides the default whitespace chars
"""
self.skipWhitespace = True
self.whiteChars = set(chars)
self.copyDefaultWhiteChars = copy_defaults
return self
def parse_with_tabs(self) -> "ParserElement":
"""
Overrides default behavior to expand ``<TAB>`` s to spaces before parsing the input string.
Must be called before ``parse_string`` when the input grammar contains elements that
match ``<TAB>`` characters.
"""
self.keepTabs = True
return self
def ignore(self, other: "ParserElement") -> "ParserElement":
"""
Define expression to be ignored (e.g., comments) while doing pattern
matching; may be called repeatedly, to define multiple comment or other
ignorable patterns.
Example::
patt = Word(alphas)[1, ...]
patt.parse_string('ablaj /* comment */ lskjd')
# -> ['ablaj']
patt.ignore(c_style_comment)
patt.parse_string('ablaj /* comment */ lskjd')
# -> ['ablaj', 'lskjd']
"""
import typing
if isinstance(other, str_type):
other = Suppress(other)
if isinstance(other, Suppress):
if other not in self.ignoreExprs:
self.ignoreExprs.append(other)
else:
self.ignoreExprs.append(Suppress(other.copy()))
return self
def set_debug_actions(
self,
start_action: DebugStartAction,
success_action: DebugSuccessAction,
exception_action: DebugExceptionAction,
) -> "ParserElement":
"""
Customize display of debugging messages while doing pattern matching:
- ``start_action`` - method to be called when an expression is about to be parsed;
should have the signature ``fn(input_string: str, location: int, expression: ParserElement, cache_hit: bool)``
- ``success_action`` - method to be called when an expression has successfully parsed;
should have the signature ``fn(input_string: str, start_location: int, end_location: int, expression: ParserELement, parsed_tokens: ParseResults, cache_hit: bool)``
- ``exception_action`` - method to be called when expression fails to parse;
should have the signature ``fn(input_string: str, location: int, expression: ParserElement, exception: Exception, cache_hit: bool)``
"""
self.debugActions = self.DebugActions(
start_action or _default_start_debug_action, # type: ignore[truthy-function]
success_action or _default_success_debug_action, # type: ignore[truthy-function]
exception_action or _default_exception_debug_action, # type: ignore[truthy-function]
)
self.debug = True
return self
def set_debug(self, flag: bool = True, recurse: bool = False) -> "ParserElement":
"""
Enable display of debugging messages while doing pattern matching.
Set ``flag`` to ``True`` to enable, ``False`` to disable.
Set ``recurse`` to ``True`` to set the debug flag on this expression and all sub-expressions.
Example::
wd = Word(alphas).set_name("alphaword")
integer = Word(nums).set_name("numword")
term = wd | integer
# turn on debugging for wd
wd.set_debug()
term[1, ...].parse_string("abc 123 xyz 890")
prints::
Match alphaword at loc 0(1,1)
Matched alphaword -> ['abc']
Match alphaword at loc 3(1,4)
Exception raised:Expected alphaword (at char 4), (line:1, col:5)
Match alphaword at loc 7(1,8)
Matched alphaword -> ['xyz']
Match alphaword at loc 11(1,12)
Exception raised:Expected alphaword (at char 12), (line:1, col:13)
Match alphaword at loc 15(1,16)
Exception raised:Expected alphaword (at char 15), (line:1, col:16)
The output shown is that produced by the default debug actions - custom debug actions can be
specified using :class:`set_debug_actions`. Prior to attempting
to match the ``wd`` expression, the debugging message ``"Match <exprname> at loc <n>(<line>,<col>)"``
is shown. Then if the parse succeeds, a ``"Matched"`` message is shown, or an ``"Exception raised"``
message is shown. Also note the use of :class:`set_name` to assign a human-readable name to the expression,
which makes debugging and exception messages easier to understand - for instance, the default
name created for the :class:`Word` expression without calling ``set_name`` is ``"W:(A-Za-z)"``.
"""
if recurse:
for expr in self.visit_all():
expr.set_debug(flag, recurse=False)
return self
if flag:
self.set_debug_actions(
_default_start_debug_action,
_default_success_debug_action,
_default_exception_debug_action,
)
else:
self.debug = False
return self
@property
def default_name(self) -> str:
if self._defaultName is None:
self._defaultName = self._generateDefaultName()
return self._defaultName
@abstractmethod
def _generateDefaultName(self) -> str:
"""
Child classes must define this method, which defines how the ``default_name`` is set.
"""
def set_name(self, name: str) -> "ParserElement":
"""
Define name for this expression, makes debugging and exception messages clearer.
Example::
Word(nums).parse_string("ABC") # -> Exception: Expected W:(0-9) (at char 0), (line:1, col:1)
Word(nums).set_name("integer").parse_string("ABC") # -> Exception: Expected integer (at char 0), (line:1, col:1)
"""
self.customName = name
self.errmsg = "Expected " + self.name
if __diag__.enable_debug_on_named_expressions:
self.set_debug()
return self
@property
def name(self) -> str:
# This will use a user-defined name if available, but otherwise defaults back to the auto-generated name
return self.customName if self.customName is not None else self.default_name
def __str__(self) -> str:
return self.name
def __repr__(self) -> str:
return str(self)
def streamline(self) -> "ParserElement":
self.streamlined = True
self._defaultName = None
return self
def recurse(self) -> List["ParserElement"]:
return []
def _checkRecursion(self, parseElementList):
subRecCheckList = parseElementList[:] + [self]
for e in self.recurse():
e._checkRecursion(subRecCheckList)
def validate(self, validateTrace=None) -> None:
"""
Check defined expressions for valid structure, check for infinite recursive definitions.
"""
warnings.warn(
"ParserElement.validate() is deprecated, and should not be used to check for left recursion",
DeprecationWarning,
stacklevel=2,
)
self._checkRecursion([])
def parse_file(
self,
file_or_filename: Union[str, Path, TextIO],
encoding: str = "utf-8",
parse_all: bool = False,
*,
parseAll: bool = False,
) -> ParseResults:
"""
Execute the parse expression on the given file or filename.
If a filename is specified (instead of a file object),
the entire file is opened, read, and closed before parsing.
"""
parseAll = parseAll or parse_all
try:
file_or_filename = typing.cast(TextIO, file_or_filename)
file_contents = file_or_filename.read()
except AttributeError:
file_or_filename = typing.cast(str, file_or_filename)
with open(file_or_filename, "r", encoding=encoding) as f:
file_contents = f.read()
try:
return self.parse_string(file_contents, parseAll)
except ParseBaseException as exc:
if ParserElement.verbose_stacktrace:
raise
else:
# catch and re-raise exception from here, clears out pyparsing internal stack trace
raise exc.with_traceback(None)
def __eq__(self, other):
if self is other:
return True
elif isinstance(other, str_type):
return self.matches(other, parse_all=True)
elif isinstance(other, ParserElement):
return vars(self) == vars(other)
return False
def __hash__(self):
return id(self)
def matches(
self, test_string: str, parse_all: bool = True, *, parseAll: bool = True
) -> bool:
"""
Method for quick testing of a parser against a test string. Good for simple
inline microtests of sub expressions while building up larger parser.
Parameters:
- ``test_string`` - to test against this expression for a match
- ``parse_all`` - (default= ``True``) - flag to pass to :class:`parse_string` when running tests
Example::
expr = Word(nums)
assert expr.matches("100")
"""
parseAll = parseAll and parse_all
try:
self.parse_string(str(test_string), parse_all=parseAll)
return True
except ParseBaseException:
return False
def run_tests(
self,
tests: Union[str, List[str]],
parse_all: bool = True,
comment: typing.Optional[Union["ParserElement", str]] = "#",
full_dump: bool = True,
print_results: bool = True,
failure_tests: bool = False,
post_parse: typing.Optional[Callable[[str, ParseResults], str]] = None,
file: typing.Optional[TextIO] = None,
with_line_numbers: bool = False,
*,
parseAll: bool = True,
fullDump: bool = True,
printResults: bool = True,
failureTests: bool = False,
postParse: typing.Optional[Callable[[str, ParseResults], str]] = None,
) -> Tuple[bool, List[Tuple[str, Union[ParseResults, Exception]]]]:
"""
Execute the parse expression on a series of test strings, showing each
test, the parsed results or where the parse failed. Quick and easy way to
run a parse expression against a list of sample strings.
Parameters:
- ``tests`` - a list of separate test strings, or a multiline string of test strings
- ``parse_all`` - (default= ``True``) - flag to pass to :class:`parse_string` when running tests
- ``comment`` - (default= ``'#'``) - expression for indicating embedded comments in the test
string; pass None to disable comment filtering
- ``full_dump`` - (default= ``True``) - dump results as list followed by results names in nested outline;
if False, only dump nested list
- ``print_results`` - (default= ``True``) prints test output to stdout
- ``failure_tests`` - (default= ``False``) indicates if these tests are expected to fail parsing
- ``post_parse`` - (default= ``None``) optional callback for successful parse results; called as
`fn(test_string, parse_results)` and returns a string to be added to the test output
- ``file`` - (default= ``None``) optional file-like object to which test output will be written;
if None, will default to ``sys.stdout``
- ``with_line_numbers`` - default= ``False``) show test strings with line and column numbers
Returns: a (success, results) tuple, where success indicates that all tests succeeded
(or failed if ``failure_tests`` is True), and the results contain a list of lines of each
test's output
Example::
number_expr = pyparsing_common.number.copy()
result = number_expr.run_tests('''
# unsigned integer
100
# negative integer
-100
# float with scientific notation
6.02e23
# integer with scientific notation
1e-12
''')
print("Success" if result[0] else "Failed!")
result = number_expr.run_tests('''
# stray character
100Z
# missing leading digit before '.'
-.100
# too many '.'
3.14.159
''', failure_tests=True)
print("Success" if result[0] else "Failed!")
prints::
# unsigned integer
100
[100]
# negative integer
-100
[-100]
# float with scientific notation
6.02e23
[6.02e+23]
# integer with scientific notation
1e-12
[1e-12]
Success
# stray character
100Z
^
FAIL: Expected end of text (at char 3), (line:1, col:4)
# missing leading digit before '.'
-.100
^
FAIL: Expected {real number with scientific notation | real number | signed integer} (at char 0), (line:1, col:1)
# too many '.'
3.14.159
^
FAIL: Expected end of text (at char 4), (line:1, col:5)
Success
Each test string must be on a single line. If you want to test a string that spans multiple
lines, create a test like this::
expr.run_tests(r"this is a test\\n of strings that spans \\n 3 lines")
(Note that this is a raw string literal, you must include the leading ``'r'``.)
"""
from .testing import pyparsing_test
parseAll = parseAll and parse_all
fullDump = fullDump and full_dump
printResults = printResults and print_results
failureTests = failureTests or failure_tests
postParse = postParse or post_parse
if isinstance(tests, str_type):
tests = typing.cast(str, tests)
line_strip = type(tests).strip
tests = [line_strip(test_line) for test_line in tests.rstrip().splitlines()]
comment_specified = comment is not None
if comment_specified:
if isinstance(comment, str_type):
comment = typing.cast(str, comment)
comment = Literal(comment)
comment = typing.cast(ParserElement, comment)
if file is None:
file = sys.stdout
print_ = file.write
result: Union[ParseResults, Exception]
allResults: List[Tuple[str, Union[ParseResults, Exception]]] = []
comments: List[str] = []
success = True
NL = Literal(r"\n").add_parse_action(replace_with("\n")).ignore(quoted_string)
BOM = "\ufeff"
for t in tests:
if comment_specified and comment.matches(t, False) or comments and not t:
comments.append(
pyparsing_test.with_line_numbers(t) if with_line_numbers else t
)
continue
if not t:
continue
out = [
"\n" + "\n".join(comments) if comments else "",
pyparsing_test.with_line_numbers(t) if with_line_numbers else t,
]
comments = []
try:
# convert newline marks to actual newlines, and strip leading BOM if present
t = NL.transform_string(t.lstrip(BOM))
result = self.parse_string(t, parse_all=parseAll)
except ParseBaseException as pe:
fatal = "(FATAL)" if isinstance(pe, ParseFatalException) else ""
out.append(pe.explain())
out.append("FAIL: " + str(pe))
if ParserElement.verbose_stacktrace:
out.extend(traceback.format_tb(pe.__traceback__))
success = success and failureTests
result = pe
except Exception as exc:
out.append(f"FAIL-EXCEPTION: {type(exc).__name__}: {exc}")
if ParserElement.verbose_stacktrace:
out.extend(traceback.format_tb(exc.__traceback__))
success = success and failureTests
result = exc
else:
success = success and not failureTests
if postParse is not None:
try:
pp_value = postParse(t, result)
if pp_value is not None:
if isinstance(pp_value, ParseResults):
out.append(pp_value.dump())
else:
out.append(str(pp_value))
else:
out.append(result.dump())
except Exception as e:
out.append(result.dump(full=fullDump))
out.append(
f"{postParse.__name__} failed: {type(e).__name__}: {e}"
)
else:
out.append(result.dump(full=fullDump))
out.append("")
if printResults:
print_("\n".join(out))
allResults.append((t, result))
return success, allResults
def create_diagram(
self,
output_html: Union[TextIO, Path, str],
vertical: int = 3,
show_results_names: bool = False,
show_groups: bool = False,
embed: bool = False,
**kwargs,
) -> None:
"""
Create a railroad diagram for the parser.
Parameters:
- ``output_html`` (str or file-like object) - output target for generated
diagram HTML
- ``vertical`` (int) - threshold for formatting multiple alternatives vertically
instead of horizontally (default=3)
- ``show_results_names`` - bool flag whether diagram should show annotations for
defined results names
- ``show_groups`` - bool flag whether groups should be highlighted with an unlabeled surrounding box
- ``embed`` - bool flag whether generated HTML should omit <HEAD>, <BODY>, and <DOCTYPE> tags to embed
the resulting HTML in an enclosing HTML source
- ``head`` - str containing additional HTML to insert into the <HEAD> section of the generated code;
can be used to insert custom CSS styling
- ``body`` - str containing additional HTML to insert at the beginning of the <BODY> section of the
generated code
Additional diagram-formatting keyword arguments can also be included;
see railroad.Diagram class.
"""
try:
from .diagram import to_railroad, railroad_to_html
except ImportError as ie:
raise Exception(
"must ``pip install pyparsing[diagrams]`` to generate parser railroad diagrams"
) from ie
self.streamline()
railroad = to_railroad(
self,
vertical=vertical,
show_results_names=show_results_names,
show_groups=show_groups,
diagram_kwargs=kwargs,
)
if isinstance(output_html, (str, Path)):
with open(output_html, "w", encoding="utf-8") as diag_file:
diag_file.write(railroad_to_html(railroad, embed=embed, **kwargs))
else:
# we were passed a file-like object, just write to it
output_html.write(railroad_to_html(railroad, embed=embed, **kwargs))
# Compatibility synonyms
# fmt: off
@staticmethod
@replaced_by_pep8(inline_literals_using)
def inlineLiteralsUsing(): ...
@staticmethod
@replaced_by_pep8(set_default_whitespace_chars)
def setDefaultWhitespaceChars(): ...
@replaced_by_pep8(set_results_name)
def setResultsName(self): ...
@replaced_by_pep8(set_break)
def setBreak(self): ...
@replaced_by_pep8(set_parse_action)
def setParseAction(self): ...
@replaced_by_pep8(add_parse_action)
def addParseAction(self): ...
@replaced_by_pep8(add_condition)
def addCondition(self): ...
@replaced_by_pep8(set_fail_action)
def setFailAction(self): ...
@replaced_by_pep8(try_parse)
def tryParse(self): ...
@staticmethod
@replaced_by_pep8(enable_left_recursion)
def enableLeftRecursion(): ...
@staticmethod
@replaced_by_pep8(enable_packrat)
def enablePackrat(): ...
@replaced_by_pep8(parse_string)
def parseString(self): ...
@replaced_by_pep8(scan_string)
def scanString(self): ...
@replaced_by_pep8(transform_string)
def transformString(self): ...
@replaced_by_pep8(search_string)
def searchString(self): ...
@replaced_by_pep8(ignore_whitespace)
def ignoreWhitespace(self): ...
@replaced_by_pep8(leave_whitespace)
def leaveWhitespace(self): ...
@replaced_by_pep8(set_whitespace_chars)
def setWhitespaceChars(self): ...
@replaced_by_pep8(parse_with_tabs)
def parseWithTabs(self): ...
@replaced_by_pep8(set_debug_actions)
def setDebugActions(self): ...
@replaced_by_pep8(set_debug)
def setDebug(self): ...
@replaced_by_pep8(set_name)
def setName(self): ...
@replaced_by_pep8(parse_file)
def parseFile(self): ...
@replaced_by_pep8(run_tests)
def runTests(self): ...
canParseNext = can_parse_next
resetCache = reset_cache
defaultName = default_name
# fmt: on
class _PendingSkip(ParserElement):
# internal placeholder class to hold a place were '...' is added to a parser element,
# once another ParserElement is added, this placeholder will be replaced with a SkipTo
def __init__(self, expr: ParserElement, must_skip: bool = False):
super().__init__()
self.anchor = expr
self.must_skip = must_skip
def _generateDefaultName(self) -> str:
return str(self.anchor + Empty()).replace("Empty", "...")
def __add__(self, other) -> "ParserElement":
skipper = SkipTo(other).set_name("...")("_skipped*")
if self.must_skip:
def must_skip(t):
if not t._skipped or t._skipped.as_list() == [""]:
del t[0]
t.pop("_skipped", None)
def show_skip(t):
if t._skipped.as_list()[-1:] == [""]:
t.pop("_skipped")
t["_skipped"] = "missing <" + repr(self.anchor) + ">"
return (
self.anchor + skipper().add_parse_action(must_skip)
| skipper().add_parse_action(show_skip)
) + other
return self.anchor + skipper + other
def __repr__(self):
return self.defaultName
def parseImpl(self, *args):
raise Exception(
"use of `...` expression without following SkipTo target expression"
)
class Token(ParserElement):
"""Abstract :class:`ParserElement` subclass, for defining atomic
matching patterns.
"""
def __init__(self):
super().__init__(savelist=False)
def _generateDefaultName(self) -> str:
return type(self).__name__
class NoMatch(Token):
"""
A token that will never match.
"""
def __init__(self):
super().__init__()
self.mayReturnEmpty = True
self.mayIndexError = False
self.errmsg = "Unmatchable token"
def parseImpl(self, instring, loc, doActions=True):
raise ParseException(instring, loc, self.errmsg, self)
class Literal(Token):
"""
Token to exactly match a specified string.
Example::
Literal('blah').parse_string('blah') # -> ['blah']
Literal('blah').parse_string('blahfooblah') # -> ['blah']
Literal('blah').parse_string('bla') # -> Exception: Expected "blah"
For case-insensitive matching, use :class:`CaselessLiteral`.
For keyword matching (force word break before and after the matched string),
use :class:`Keyword` or :class:`CaselessKeyword`.
"""
def __new__(cls, match_string: str = "", *, matchString: str = ""):
# Performance tuning: select a subclass with optimized parseImpl
if cls is Literal:
match_string = matchString or match_string
if not match_string:
return super().__new__(Empty)
if len(match_string) == 1:
return super().__new__(_SingleCharLiteral)
# Default behavior
return super().__new__(cls)
# Needed to make copy.copy() work correctly if we customize __new__
def __getnewargs__(self):
return (self.match,)
def __init__(self, match_string: str = "", *, matchString: str = ""):
super().__init__()
match_string = matchString or match_string
self.match = match_string
self.matchLen = len(match_string)
self.firstMatchChar = match_string[:1]
self.errmsg = "Expected " + self.name
self.mayReturnEmpty = False
self.mayIndexError = False
def _generateDefaultName(self) -> str:
return repr(self.match)
def parseImpl(self, instring, loc, doActions=True):
if instring[loc] == self.firstMatchChar and instring.startswith(
self.match, loc
):
return loc + self.matchLen, self.match
raise ParseException(instring, loc, self.errmsg, self)
class Empty(Literal):
"""
An empty token, will always match.
"""
def __init__(self, match_string="", *, matchString=""):
super().__init__("")
self.mayReturnEmpty = True
self.mayIndexError = False
def _generateDefaultName(self) -> str:
return "Empty"
def parseImpl(self, instring, loc, doActions=True):
return loc, []
class _SingleCharLiteral(Literal):
def parseImpl(self, instring, loc, doActions=True):
if instring[loc] == self.firstMatchChar:
return loc + 1, self.match
raise ParseException(instring, loc, self.errmsg, self)
ParserElement._literalStringClass = Literal
class Keyword(Token):
"""
Token to exactly match a specified string as a keyword, that is,
it must be immediately preceded and followed by whitespace or
non-keyword characters. Compare with :class:`Literal`:
- ``Literal("if")`` will match the leading ``'if'`` in
``'ifAndOnlyIf'``.
- ``Keyword("if")`` will not; it will only match the leading
``'if'`` in ``'if x=1'``, or ``'if(y==2)'``
Accepts two optional constructor arguments in addition to the
keyword string:
- ``ident_chars`` is a string of characters that would be valid
identifier characters, defaulting to all alphanumerics + "_" and
"$"
- ``caseless`` allows case-insensitive matching, default is ``False``.
Example::
Keyword("start").parse_string("start") # -> ['start']
Keyword("start").parse_string("starting") # -> Exception
For case-insensitive matching, use :class:`CaselessKeyword`.
"""
DEFAULT_KEYWORD_CHARS = alphanums + "_$"
def __init__(
self,
match_string: str = "",
ident_chars: typing.Optional[str] = None,
caseless: bool = False,
*,
matchString: str = "",
identChars: typing.Optional[str] = None,
):
super().__init__()
identChars = identChars or ident_chars
if identChars is None:
identChars = Keyword.DEFAULT_KEYWORD_CHARS
match_string = matchString or match_string
self.match = match_string
self.matchLen = len(match_string)
try:
self.firstMatchChar = match_string[0]
except IndexError:
raise ValueError("null string passed to Keyword; use Empty() instead")
self.errmsg = f"Expected {type(self).__name__} {self.name}"
self.mayReturnEmpty = False
self.mayIndexError = False
self.caseless = caseless
if caseless:
self.caselessmatch = match_string.upper()
identChars = identChars.upper()
self.identChars = set(identChars)
def _generateDefaultName(self) -> str:
return repr(self.match)
def parseImpl(self, instring, loc, doActions=True):
errmsg = self.errmsg
errloc = loc
if self.caseless:
if instring[loc : loc + self.matchLen].upper() == self.caselessmatch:
if loc == 0 or instring[loc - 1].upper() not in self.identChars:
if (
loc >= len(instring) - self.matchLen
or instring[loc + self.matchLen].upper() not in self.identChars
):
return loc + self.matchLen, self.match
else:
# followed by keyword char
errmsg += ", was immediately followed by keyword character"
errloc = loc + self.matchLen
else:
# preceded by keyword char
errmsg += ", keyword was immediately preceded by keyword character"
errloc = loc - 1
# else no match just raise plain exception
else:
if (
instring[loc] == self.firstMatchChar
and self.matchLen == 1
or instring.startswith(self.match, loc)
):
if loc == 0 or instring[loc - 1] not in self.identChars:
if (
loc >= len(instring) - self.matchLen
or instring[loc + self.matchLen] not in self.identChars
):
return loc + self.matchLen, self.match
else:
# followed by keyword char
errmsg += (
", keyword was immediately followed by keyword character"
)
errloc = loc + self.matchLen
else:
# preceded by keyword char
errmsg += ", keyword was immediately preceded by keyword character"
errloc = loc - 1
# else no match just raise plain exception
raise ParseException(instring, errloc, errmsg, self)
@staticmethod
def set_default_keyword_chars(chars) -> None:
"""
Overrides the default characters used by :class:`Keyword` expressions.
"""
Keyword.DEFAULT_KEYWORD_CHARS = chars
setDefaultKeywordChars = set_default_keyword_chars
class CaselessLiteral(Literal):
"""
Token to match a specified string, ignoring case of letters.
Note: the matched results will always be in the case of the given
match string, NOT the case of the input text.
Example::
CaselessLiteral("CMD")[1, ...].parse_string("cmd CMD Cmd10")
# -> ['CMD', 'CMD', 'CMD']
(Contrast with example for :class:`CaselessKeyword`.)
"""
def __init__(self, match_string: str = "", *, matchString: str = ""):
match_string = matchString or match_string
super().__init__(match_string.upper())
# Preserve the defining literal.
self.returnString = match_string
self.errmsg = "Expected " + self.name
def parseImpl(self, instring, loc, doActions=True):
if instring[loc : loc + self.matchLen].upper() == self.match:
return loc + self.matchLen, self.returnString
raise ParseException(instring, loc, self.errmsg, self)
class CaselessKeyword(Keyword):
"""
Caseless version of :class:`Keyword`.
Example::
CaselessKeyword("CMD")[1, ...].parse_string("cmd CMD Cmd10")
# -> ['CMD', 'CMD']
(Contrast with example for :class:`CaselessLiteral`.)
"""
def __init__(
self,
match_string: str = "",
ident_chars: typing.Optional[str] = None,
*,
matchString: str = "",
identChars: typing.Optional[str] = None,
):
identChars = identChars or ident_chars
match_string = matchString or match_string
super().__init__(match_string, identChars, caseless=True)
class CloseMatch(Token):
"""A variation on :class:`Literal` which matches "close" matches,
that is, strings with at most 'n' mismatching characters.
:class:`CloseMatch` takes parameters:
- ``match_string`` - string to be matched
- ``caseless`` - a boolean indicating whether to ignore casing when comparing characters
- ``max_mismatches`` - (``default=1``) maximum number of
mismatches allowed to count as a match
The results from a successful parse will contain the matched text
from the input string and the following named results:
- ``mismatches`` - a list of the positions within the
match_string where mismatches were found
- ``original`` - the original match_string used to compare
against the input string
If ``mismatches`` is an empty list, then the match was an exact
match.
Example::
patt = CloseMatch("ATCATCGAATGGA")
patt.parse_string("ATCATCGAAXGGA") # -> (['ATCATCGAAXGGA'], {'mismatches': [[9]], 'original': ['ATCATCGAATGGA']})
patt.parse_string("ATCAXCGAAXGGA") # -> Exception: Expected 'ATCATCGAATGGA' (with up to 1 mismatches) (at char 0), (line:1, col:1)
# exact match
patt.parse_string("ATCATCGAATGGA") # -> (['ATCATCGAATGGA'], {'mismatches': [[]], 'original': ['ATCATCGAATGGA']})
# close match allowing up to 2 mismatches
patt = CloseMatch("ATCATCGAATGGA", max_mismatches=2)
patt.parse_string("ATCAXCGAAXGGA") # -> (['ATCAXCGAAXGGA'], {'mismatches': [[4, 9]], 'original': ['ATCATCGAATGGA']})
"""
def __init__(
self,
match_string: str,
max_mismatches: typing.Optional[int] = None,
*,
maxMismatches: int = 1,
caseless=False,
):
maxMismatches = max_mismatches if max_mismatches is not None else maxMismatches
super().__init__()
self.match_string = match_string
self.maxMismatches = maxMismatches
self.errmsg = f"Expected {self.match_string!r} (with up to {self.maxMismatches} mismatches)"
self.caseless = caseless
self.mayIndexError = False
self.mayReturnEmpty = False
def _generateDefaultName(self) -> str:
return f"{type(self).__name__}:{self.match_string!r}"
def parseImpl(self, instring, loc, doActions=True):
start = loc
instrlen = len(instring)
maxloc = start + len(self.match_string)
if maxloc <= instrlen:
match_string = self.match_string
match_stringloc = 0
mismatches = []
maxMismatches = self.maxMismatches
for match_stringloc, s_m in enume
Соседние файлы в папке pyparsing