Google Git
Sign in
foss-eda-tools / third_party / shuttle / sky130 / mpw-007 / slot-030 / 268a65bf77514ce05b2b1415de0dd66b6df5a35e / . / venv / lib / python3.9 / site-packages / pkg_resources / _vendor / pyparsing.py
blob: cf75e1e5fcbfe7eac41d2a9e446c5c980741087b [file] [log] [blame]
# module pyparsing.py
#
# Copyright (c) 2003-2018 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.
#
__doc__ = \
"""
pyparsing module - Classes and methods to define and execute parsing grammars
=============================================================================
The pyparsing module is an alternative approach to creating and executing simple grammars,
vs. the traditional lex/yacc approach, or the use of regular expressions. With pyparsing, you
don't need to learn a new syntax for defining grammars or matching expressions - the parsing module
provides a library of classes that you use to construct the grammar directly in Python.
Here is a program to parse "Hello, World!" (or any greeting of the form
C{"<salutation>, <addressee>!"}), built up using L{Word}, L{Literal}, and L{And} elements
(L{'+'<ParserElement.__add__>} operator gives L{And} expressions, strings are auto-converted to
L{Literal} expressions)::
from pyparsing import Word, alphas
# define grammar of a greeting
greet = Word(alphas) + "," + Word(alphas) + "!"
hello = "Hello, World!"
print (hello, "->", greet.parseString(hello))
The program outputs the following::
Hello, World! -> ['Hello', ',', 'World', '!']
The Python representation of the grammar is quite readable, owing to the self-explanatory
class names, and the use of '+', '|' and '^' operators.
The L{ParseResults} object returned from L{ParserElement.parseString<ParserElement.parseString>} can be accessed as a nested list, a dictionary, or an
object with named attributes.
The pyparsing module handles some of the problems that are typically vexing when writing text parsers:
- extra or missing whitespace (the above program will also handle "Hello,World!", "Hello , World !", etc.)
- quoted strings
- embedded comments
Getting Started -
-----------------
Visit the classes L{ParserElement} and L{ParseResults} to see the base classes that most other pyparsing
classes inherit from. Use the docstrings for examples of how to:
- construct literal match expressions from L{Literal} and L{CaselessLiteral} classes
- construct character word-group expressions using the L{Word} class
- see how to create repetitive expressions using L{ZeroOrMore} and L{OneOrMore} classes
- use L{'+'<And>}, L{'|'<MatchFirst>}, L{'^'<Or>}, and L{'&'<Each>} operators to combine simple expressions into more complex ones
- associate names with your parsed results using L{ParserElement.setResultsName}
- find some helpful expression short-cuts like L{delimitedList} and L{oneOf}
- find more useful common expressions in the L{pyparsing_common} namespace class
"""
__version__ = "2.2.1"
__versionTime__ = "18 Sep 2018 00:49 UTC"
__author__ = "Paul McGuire <ptmcg@users.sourceforge.net>"
import string
from weakref import ref as wkref
import copy
import sys
import warnings
import re
import sre_constants
import collections
import pprint
import traceback
import types
from datetime import datetime
try:
from _thread import RLock
except ImportError:
from threading import RLock
try:
# Python 3
from collections.abc import Iterable
from collections.abc import MutableMapping
except ImportError:
# Python 2.7
from collections import Iterable
from collections import MutableMapping
try:
from collections import OrderedDict as _OrderedDict
except ImportError:
try:
from ordereddict import OrderedDict as _OrderedDict
except ImportError:
_OrderedDict = None
#~ sys.stderr.write( "testing pyparsing module, version %s, %s\n" % (__version__,__versionTime__ ) )
__all__ = [
'And', 'CaselessKeyword', 'CaselessLiteral', 'CharsNotIn', 'Combine', 'Dict', 'Each', 'Empty',
'FollowedBy', 'Forward', 'GoToColumn', 'Group', 'Keyword', 'LineEnd', 'LineStart', 'Literal',
'MatchFirst', 'NoMatch', 'NotAny', 'OneOrMore', 'OnlyOnce', 'Optional', 'Or',
'ParseBaseException', 'ParseElementEnhance', 'ParseException', 'ParseExpression', 'ParseFatalException',
'ParseResults', 'ParseSyntaxException', 'ParserElement', 'QuotedString', 'RecursiveGrammarException',
'Regex', 'SkipTo', 'StringEnd', 'StringStart', 'Suppress', 'Token', 'TokenConverter',
'White', 'Word', 'WordEnd', 'WordStart', 'ZeroOrMore',
'alphanums', 'alphas', 'alphas8bit', 'anyCloseTag', 'anyOpenTag', 'cStyleComment', 'col',
'commaSeparatedList', 'commonHTMLEntity', 'countedArray', 'cppStyleComment', 'dblQuotedString',
'dblSlashComment', 'delimitedList', 'dictOf', 'downcaseTokens', 'empty', 'hexnums',
'htmlComment', 'javaStyleComment', 'line', 'lineEnd', 'lineStart', 'lineno',
'makeHTMLTags', 'makeXMLTags', 'matchOnlyAtCol', 'matchPreviousExpr', 'matchPreviousLiteral',
'nestedExpr', 'nullDebugAction', 'nums', 'oneOf', 'opAssoc', 'operatorPrecedence', 'printables',
'punc8bit', 'pythonStyleComment', 'quotedString', 'removeQuotes', 'replaceHTMLEntity',
'replaceWith', 'restOfLine', 'sglQuotedString', 'srange', 'stringEnd',
'stringStart', 'traceParseAction', 'unicodeString', 'upcaseTokens', 'withAttribute',
'indentedBlock', 'originalTextFor', 'ungroup', 'infixNotation','locatedExpr', 'withClass',
'CloseMatch', 'tokenMap', 'pyparsing_common',
]
system_version = tuple(sys.version_info)[:3]
PY_3 = system_version[0] == 3
if PY_3:
_MAX_INT = sys.maxsize
basestring = str
unichr = chr
_ustr = str
# build list of single arg builtins, that can be used as parse actions
singleArgBuiltins = [sum, len, sorted, reversed, list, tuple, set, any, all, min, max]
else:
_MAX_INT = sys.maxint
range = xrange
def _ustr(obj):
"""Drop-in replacement for str(obj) that tries to be Unicode friendly. It first tries
str(obj). If that fails with a UnicodeEncodeError, then it tries unicode(obj). It
then < returns the unicode object | encodes it with the default encoding | ... >.
"""
if isinstance(obj,unicode):
return obj
try:
# If this works, then _ustr(obj) has the same behaviour as str(obj), so
# it won't break any existing code.
return str(obj)
except UnicodeEncodeError:
# Else encode it
ret = unicode(obj).encode(sys.getdefaultencoding(), 'xmlcharrefreplace')
xmlcharref = Regex(r'&#\d+;')
xmlcharref.setParseAction(lambda t: '\\u' + hex(int(t[0][2:-1]))[2:])
return xmlcharref.transformString(ret)
# build list of single arg builtins, tolerant of Python version, that can be used as parse actions
singleArgBuiltins = []
import __builtin__
for fname in "sum len sorted reversed list tuple set any all min max".split():
try:
singleArgBuiltins.append(getattr(__builtin__,fname))
except AttributeError:
continue
_generatorType = type((y for y in range(1)))
def _xml_escape(data):
"""Escape &, <, >, ", ', etc. in a string of data."""
# ampersand must be replaced first
from_symbols = '&><"\''
to_symbols = ('&'+s+';' for s in "amp gt lt quot apos".split())
for from_,to_ in zip(from_symbols, to_symbols):
data = data.replace(from_, to_)
return data
class _Constants(object):
pass
alphas = string.ascii_uppercase + string.ascii_lowercase
nums = "0123456789"
hexnums = nums + "ABCDEFabcdef"
alphanums = alphas + nums
_bslash = chr(92)
printables = "".join(c for c in string.printable if c not in string.whitespace)
class ParseBaseException(Exception):
"""base exception class for all parsing runtime exceptions"""
# Performance tuning: we construct a *lot* of these, so keep this
# constructor as small and fast as possible
def __init__( self, pstr, loc=0, msg=None, elem=None ):
self.loc = loc
if msg is None:
self.msg = pstr
self.pstr = ""
else:
self.msg = msg
self.pstr = pstr
self.parserElement = elem
self.args = (pstr, loc, msg)
@classmethod
def _from_exception(cls, pe):
"""
internal factory method to simplify creating one type of ParseException
from another - avoids having __init__ signature conflicts among subclasses
"""
return cls(pe.pstr, pe.loc, pe.msg, pe.parserElement)
def __getattr__( self, aname ):
"""supported attributes by name are:
- lineno - returns the line number of the exception text
- col - returns the column number of the exception text
- line - returns the line containing the exception text
"""
if( aname == "lineno" ):
return lineno( self.loc, self.pstr )
elif( aname in ("col", "column") ):
return col( self.loc, self.pstr )
elif( aname == "line" ):
return line( self.loc, self.pstr )
else:
raise AttributeError(aname)
def __str__( self ):
return "%s (at char %d), (line:%d, col:%d)" % \
( self.msg, self.loc, self.lineno, self.column )
def __repr__( self ):
return _ustr(self)
def markInputline( self, markerString = ">!<" ):
"""Extracts the exception line from the input string, and marks
the location of the exception with a special symbol.
"""
line_str = self.line
line_column = self.column - 1
if markerString:
line_str = "".join((line_str[:line_column],
markerString, line_str[line_column:]))
return line_str.strip()
def __dir__(self):
return "lineno col line".split() + dir(type(self))
class ParseException(ParseBaseException):
"""
Exception thrown when parse expressions don't match class;
supported attributes by name are:
- lineno - returns the line number of the exception text
- col - returns the column number of the exception text
- line - returns the line containing the exception text
Example::
try:
Word(nums).setName("integer").parseString("ABC")
except ParseException as pe:
print(pe)
print("column: {}".format(pe.col))
prints::
Expected integer (at char 0), (line:1, col:1)
column: 1
"""
pass
class ParseFatalException(ParseBaseException):
"""user-throwable exception thrown when inconsistent parse content
is found; stops all parsing immediately"""
pass
class ParseSyntaxException(ParseFatalException):
"""just like L{ParseFatalException}, but thrown internally when an
L{ErrorStop<And._ErrorStop>} ('-' operator) indicates that parsing is to stop
immediately because an unbacktrackable syntax error has been found"""
pass
#~ class ReparseException(ParseBaseException):
#~ """Experimental class - parse actions can raise this exception to cause
#~ pyparsing to reparse the input string:
#~ - with a modified input string, and/or
#~ - with a modified start location
#~ Set the values of the ReparseException in the constructor, and raise the
#~ exception in a parse action to cause pyparsing to use the new string/location.
#~ Setting the values as None causes no change to be made.
#~ """
#~ def __init_( self, newstring, restartLoc ):
#~ self.newParseText = newstring
#~ self.reparseLoc = restartLoc
class RecursiveGrammarException(Exception):
"""exception thrown by L{ParserElement.validate} if the grammar could be improperly recursive"""
def __init__( self, parseElementList ):
self.parseElementTrace = parseElementList
def __str__( self ):
return "RecursiveGrammarException: %s" % self.parseElementTrace
class _ParseResultsWithOffset(object):
def __init__(self,p1,p2):
self.tup = (p1,p2)
def __getitem__(self,i):
return self.tup[i]
def __repr__(self):
return repr(self.tup[0])
def setOffset(self,i):
self.tup = (self.tup[0],i)
class ParseResults(object):
"""
Structured parse results, to provide multiple means of access to the parsed data:
- as a list (C{len(results)})
- by list index (C{results[0], results[1]}, etc.)
- by attribute (C{results.<resultsName>} - see L{ParserElement.setResultsName})
Example::
integer = Word(nums)
date_str = (integer.setResultsName("year") + '/'
+ integer.setResultsName("month") + '/'
+ integer.setResultsName("day"))
# equivalent form:
# date_str = integer("year") + '/' + integer("month") + '/' + integer("day")
# parseString returns a ParseResults object
result = date_str.parseString("1999/12/31")
def test(s, fn=repr):
print("%s -> %s" % (s, fn(eval(s))))
test("list(result)")
test("result[0]")
test("result['month']")
test("result.day")
test("'month' in result")
test("'minutes' in result")
test("result.dump()", str)
prints::
list(result) -> ['1999', '/', '12', '/', '31']
result[0] -> '1999'
result['month'] -> '12'
result.day -> '31'
'month' in result -> True
'minutes' in result -> False
result.dump() -> ['1999', '/', '12', '/', '31']
- day: 31
- month: 12
- year: 1999
"""
def __new__(cls, toklist=None, name=None, asList=True, modal=True ):
if isinstance(toklist, cls):
return toklist
retobj = object.__new__(cls)
retobj.__doinit = True
return retobj
# Performance tuning: we construct a *lot* of these, so keep this
# constructor as small and fast as possible
def __init__( self, toklist=None, name=None, asList=True, modal=True, isinstance=isinstance ):
if self.__doinit:
self.__doinit = False
self.__name = None
self.__parent = None
self.__accumNames = {}
self.__asList = asList
self.__modal = modal
if toklist is None:
toklist = []
if isinstance(toklist, list):
self.__toklist = toklist[:]
elif isinstance(toklist, _generatorType):
self.__toklist = list(toklist)
else:
self.__toklist = [toklist]
self.__tokdict = dict()
if name is not None and name:
if not modal:
self.__accumNames[name] = 0
if isinstance(name,int):
name = _ustr(name) # will always return a str, but use _ustr for consistency
self.__name = name
if not (isinstance(toklist, (type(None), basestring, list)) and toklist in (None,'',[])):
if isinstance(toklist,basestring):
toklist = [ toklist ]
if asList:
if isinstance(toklist,ParseResults):
self[name] = _ParseResultsWithOffset(toklist.copy(),0)
else:
self[name] = _ParseResultsWithOffset(ParseResults(toklist[0]),0)
self[name].__name = name
else:
try:
self[name] = toklist[0]
except (KeyError,TypeError,IndexError):
self[name] = toklist
def __getitem__( self, i ):
if isinstance( i, (int,slice) ):
return self.__toklist[i]
else:
if i not in self.__accumNames:
return self.__tokdict[i][-1][0]
else:
return ParseResults([ v[0] for v in self.__tokdict[i] ])
def __setitem__( self, k, v, isinstance=isinstance ):
if isinstance(v,_ParseResultsWithOffset):
self.__tokdict[k] = self.__tokdict.get(k,list()) + [v]
sub = v[0]
elif isinstance(k,(int,slice)):
self.__toklist[k] = v
sub = v
else:
self.__tokdict[k] = self.__tokdict.get(k,list()) + [_ParseResultsWithOffset(v,0)]
sub = v
if isinstance(sub,ParseResults):
sub.__parent = wkref(self)
def __delitem__( self, i ):
if isinstance(i,(int,slice)):
mylen = len( self.__toklist )
del self.__toklist[i]
# convert int to slice
if isinstance(i, int):
if i < 0:
i += mylen
i = slice(i, i+1)
# get removed indices
removed = list(range(*i.indices(mylen)))
removed.reverse()
# fixup indices in token dictionary
for name,occurrences in self.__tokdict.items():
for j in removed:
for k, (value, position) in enumerate(occurrences):
occurrences[k] = _ParseResultsWithOffset(value, position - (position > j))
else:
del self.__tokdict[i]
def __contains__( self, k ):
return k in self.__tokdict
def __len__( self ): return len( self.__toklist )
def __bool__(self): return ( not not self.__toklist )
__nonzero__ = __bool__
def __iter__( self ): return iter( self.__toklist )
def __reversed__( self ): return iter( self.__toklist[::-1] )
def _iterkeys( self ):
if hasattr(self.__tokdict, "iterkeys"):
return self.__tokdict.iterkeys()
else:
return iter(self.__tokdict)
def _itervalues( self ):
return (self[k] for k in self._iterkeys())
def _iteritems( self ):
return ((k, self[k]) for k in self._iterkeys())
if PY_3:
keys = _iterkeys
"""Returns an iterator of all named result keys (Python 3.x only)."""
values = _itervalues
"""Returns an iterator of all named result values (Python 3.x only)."""
items = _iteritems
"""Returns an iterator of all named result key-value tuples (Python 3.x only)."""
else:
iterkeys = _iterkeys
"""Returns an iterator of all named result keys (Python 2.x only)."""
itervalues = _itervalues
"""Returns an iterator of all named result values (Python 2.x only)."""
iteritems = _iteritems
"""Returns an iterator of all named result key-value tuples (Python 2.x only)."""
def keys( self ):
"""Returns all named result keys (as a list in Python 2.x, as an iterator in Python 3.x)."""
return list(self.iterkeys())
def values( self ):
"""Returns all named result values (as a list in Python 2.x, as an iterator in Python 3.x)."""
return list(self.itervalues())
def items( self ):
"""Returns all named result key-values (as a list of tuples in Python 2.x, as an iterator in Python 3.x)."""
return list(self.iteritems())
def haskeys( self ):
"""Since keys() returns an iterator, this method is helpful in bypassing
code that looks for the existence of any defined results names."""
return bool(self.__tokdict)
def pop( self, *args, **kwargs):
"""
Removes and returns item at specified index (default=C{last}).
Supports both C{list} and C{dict} semantics for C{pop()}. If passed no
argument or an integer argument, it will use C{list} semantics
and pop tokens from the list of parsed tokens. If passed a
non-integer argument (most likely a string), it will use C{dict}
semantics and pop the corresponding value from any defined
results names. A second default return value argument is
supported, just as in C{dict.pop()}.
Example::
def remove_first(tokens):
tokens.pop(0)
print(OneOrMore(Word(nums)).parseString("0 123 321")) # -> ['0', '123', '321']
print(OneOrMore(Word(nums)).addParseAction(remove_first).parseString("0 123 321")) # -> ['123', '321']
label = Word(alphas)
patt = label("LABEL") + OneOrMore(Word(nums))
print(patt.parseString("AAB 123 321").dump())
# Use pop() in a parse action to remove named result (note that corresponding value is not
# removed from list form of results)
def remove_LABEL(tokens):
tokens.pop("LABEL")
return tokens
patt.addParseAction(remove_LABEL)
print(patt.parseString("AAB 123 321").dump())
prints::
['AAB', '123', '321']
- LABEL: AAB
['AAB', '123', '321']
"""
if not args:
args = [-1]
for k,v in kwargs.items():
if k == 'default':
args = (args[0], v)
else:
raise TypeError("pop() got an unexpected keyword argument '%s'" % k)
if (isinstance(args[0], int) or
len(args) == 1 or
args[0] in self):
index = args[0]
ret = self[index]
del self[index]
return ret
else:
defaultvalue = args[1]
return defaultvalue
def get(self, key, defaultValue=None):
"""
Returns named result matching the given key, or if there is no
such name, then returns the given C{defaultValue} or C{None} if no
C{defaultValue} is specified.
Similar to C{dict.get()}.
Example::
integer = Word(nums)
date_str = integer("year") + '/' + integer("month") + '/' + integer("day")
result = date_str.parseString("1999/12/31")
print(result.get("year")) # -> '1999'
print(result.get("hour", "not specified")) # -> 'not specified'
print(result.get("hour")) # -> None
"""
if key in self:
return self[key]
else:
return defaultValue
def insert( self, index, insStr ):
"""
Inserts new element at location index in the list of parsed tokens.
Similar to C{list.insert()}.
Example::
print(OneOrMore(Word(nums)).parseString("0 123 321")) # -> ['0', '123', '321']
# use a parse action to insert the parse location in the front of the parsed results
def insert_locn(locn, tokens):
tokens.insert(0, locn)
print(OneOrMore(Word(nums)).addParseAction(insert_locn).parseString("0 123 321")) # -> [0, '0', '123', '321']
"""
self.__toklist.insert(index, insStr)
# fixup indices in token dictionary
for name,occurrences in self.__tokdict.items():
for k, (value, position) in enumerate(occurrences):
occurrences[k] = _ParseResultsWithOffset(value, position + (position > index))
def append( self, item ):
"""
Add single element to end of ParseResults list of elements.
Example::
print(OneOrMore(Word(nums)).parseString("0 123 321")) # -> ['0', '123', '321']
# use a parse action to compute the sum of the parsed integers, and add it to the end
def append_sum(tokens):
tokens.append(sum(map(int, tokens)))
print(OneOrMore(Word(nums)).addParseAction(append_sum).parseString("0 123 321")) # -> ['0', '123', '321', 444]
"""
self.__toklist.append(item)
def extend( self, itemseq ):
"""
Add sequence of elements to end of ParseResults list of elements.
Example::
patt = OneOrMore(Word(alphas))
# use a parse action to append the reverse of the matched strings, to make a palindrome
def make_palindrome(tokens):
tokens.extend(reversed([t[::-1] for t in tokens]))
return ''.join(tokens)
print(patt.addParseAction(make_palindrome).parseString("lskdj sdlkjf lksd")) # -> 'lskdjsdlkjflksddsklfjkldsjdksl'
"""
if isinstance(itemseq, ParseResults):
self += itemseq
else:
self.__toklist.extend(itemseq)
def clear( self ):
"""
Clear all elements and results names.
"""
del self.__toklist[:]
self.__tokdict.clear()
def __getattr__( self, name ):
try:
return self[name]
except KeyError:
return ""
if name in self.__tokdict:
if name not in self.__accumNames:
return self.__tokdict[name][-1][0]
else:
return ParseResults([ v[0] for v in self.__tokdict[name] ])
else:
return ""
def __add__( self, other ):
ret = self.copy()
ret += other
return ret
def __iadd__( self, other ):
if other.__tokdict:
offset = len(self.__toklist)
addoffset = lambda a: offset if a<0 else a+offset
otheritems = other.__tokdict.items()
otherdictitems = [(k, _ParseResultsWithOffset(v[0],addoffset(v[1])) )
for (k,vlist) in otheritems for v in vlist]
for k,v in otherdictitems:
self[k] = v
if isinstance(v[0],ParseResults):
v[0].__parent = wkref(self)
self.__toklist += other.__toklist
self.__accumNames.update( other.__accumNames )
return self
def __radd__(self, other):
if isinstance(other,int) and other == 0:
# useful for merging many ParseResults using sum() builtin
return self.copy()
else:
# this may raise a TypeError - so be it
return other + self
def __repr__( self ):
return "(%s, %s)" % ( repr( self.__toklist ), repr( self.__tokdict ) )
def __str__( self ):
return '[' + ', '.join(_ustr(i) if isinstance(i, ParseResults) else repr(i) for i in self.__toklist) + ']'
def _asStringList( self, sep='' ):
out = []
for item in self.__toklist:
if out and sep:
out.append(sep)
if isinstance( item, ParseResults ):
out += item._asStringList()
else:
out.append( _ustr(item) )
return out
def asList( self ):
"""
Returns the parse results as a nested list of matching tokens, all converted to strings.
Example::
patt = OneOrMore(Word(alphas))
result = patt.parseString("sldkj lsdkj sldkj")
# even though the result prints in string-like form, it is actually a pyparsing ParseResults
print(type(result), result) # -> <class 'pyparsing.ParseResults'> ['sldkj', 'lsdkj', 'sldkj']
# Use asList() to create an actual list
result_list = result.asList()
print(type(result_list), result_list) # -> <class 'list'> ['sldkj', 'lsdkj', 'sldkj']
"""
return [res.asList() if isinstance(res,ParseResults) else res for res in self.__toklist]
def asDict( self ):
"""
Returns the named parse results as a nested dictionary.
Example::
integer = Word(nums)
date_str = integer("year") + '/' + integer("month") + '/' + integer("day")
result = date_str.parseString('12/31/1999')
print(type(result), repr(result)) # -> <class 'pyparsing.ParseResults'> (['12', '/', '31', '/', '1999'], {'day': [('1999', 4)], 'year': [('12', 0)], 'month': [('31', 2)]})
result_dict = result.asDict()
print(type(result_dict), repr(result_dict)) # -> <class 'dict'> {'day': '1999', 'year': '12', 'month': '31'}
# even though a ParseResults supports dict-like access, sometime you just need to have a dict
import json
print(json.dumps(result)) # -> Exception: TypeError: ... is not JSON serializable
print(json.dumps(result.asDict())) # -> {"month": "31", "day": "1999", "year": "12"}
"""
if PY_3:
item_fn = self.items
else:
item_fn = self.iteritems
def toItem(obj):
if isinstance(obj, ParseResults):
if obj.haskeys():
return obj.asDict()
else:
return [toItem(v) for v in obj]
else:
return obj
return dict((k,toItem(v)) for k,v in item_fn())
def copy( self ):
"""
Returns a new copy of a C{ParseResults} object.
"""
ret = ParseResults( self.__toklist )
ret.__tokdict = self.__tokdict.copy()
ret.__parent = self.__parent
ret.__accumNames.update( self.__accumNames )
ret.__name = self.__name
return ret
def asXML( self, doctag=None, namedItemsOnly=False, indent="", formatted=True ):
"""
(Deprecated) Returns the parse results as XML. Tags are created for tokens and lists that have defined results names.
"""
nl = "\n"
out = []
namedItems = dict((v[1],k) for (k,vlist) in self.__tokdict.items()
for v in vlist)
nextLevelIndent = indent + " "
# collapse out indents if formatting is not desired
if not formatted:
indent = ""
nextLevelIndent = ""
nl = ""
selfTag = None
if doctag is not None:
selfTag = doctag
else:
if self.__name:
selfTag = self.__name
if not selfTag:
if namedItemsOnly:
return ""
else:
selfTag = "ITEM"
out += [ nl, indent, "<", selfTag, ">" ]
for i,res in enumerate(self.__toklist):
if isinstance(res,ParseResults):
if i in namedItems:
out += [ res.asXML(namedItems[i],
namedItemsOnly and doctag is None,
nextLevelIndent,
formatted)]
else:
out += [ res.asXML(None,
namedItemsOnly and doctag is None,
nextLevelIndent,
formatted)]
else:
# individual token, see if there is a name for it
resTag = None
if i in namedItems:
resTag = namedItems[i]
if not resTag:
if namedItemsOnly:
continue
else:
resTag = "ITEM"
xmlBodyText = _xml_escape(_ustr(res))
out += [ nl, nextLevelIndent, "<", resTag, ">",
xmlBodyText,
"</", resTag, ">" ]
out += [ nl, indent, "</", selfTag, ">" ]
return "".join(out)
def __lookup(self,sub):
for k,vlist in self.__tokdict.items():
for v,loc in vlist:
if sub is v:
return k
return None
def getName(self):
r"""
Returns the results name for this token expression. Useful when several
different expressions might match at a particular location.
Example::
integer = Word(nums)
ssn_expr = Regex(r"\d\d\d-\d\d-\d\d\d\d")
house_number_expr = Suppress('#') + Word(nums, alphanums)
user_data = (Group(house_number_expr)("house_number")
| Group(ssn_expr)("ssn")
| Group(integer)("age"))
user_info = OneOrMore(user_data)
result = user_info.parseString("22 111-22-3333 #221B")
for item in result:
print(item.getName(), ':', item[0])
prints::
age : 22
ssn : 111-22-3333
house_number : 221B
"""
if self.__name:
return self.__name
elif self.__parent:
par = self.__parent()
if par:
return par.__lookup(self)
else:
return None
elif (len(self) == 1 and
len(self.__tokdict) == 1 and
next(iter(self.__tokdict.values()))[0][1] in (0,-1)):
return next(iter(self.__tokdict.keys()))
else:
return None
def dump(self, indent='', depth=0, full=True):
"""
Diagnostic method for listing out the contents of a C{ParseResults}.
Accepts an optional C{indent} argument so that this string can be embedded
in a nested display of other data.
Example::
integer = Word(nums)
date_str = integer("year") + '/' + integer("month") + '/' + integer("day")
result = date_str.parseString('12/31/1999')
print(result.dump())
prints::
['12', '/', '31', '/', '1999']
- day: 1999
- month: 31
- year: 12
"""
out = []
NL = '\n'
out.append( indent+_ustr(self.asList()) )
if full:
if self.haskeys():
items = sorted((str(k), v) for k,v in self.items())
for k,v in items:
if out:
out.append(NL)
out.append( "%s%s- %s: " % (indent,(' '*depth), k) )
if isinstance(v,ParseResults):
if v:
out.append( v.dump(indent,depth+1) )
else:
out.append(_ustr(v))
else:
out.append(repr(v))
elif any(isinstance(vv,ParseResults) for vv in self):
v = self
for i,vv in enumerate(v):
if isinstance(vv,ParseResults):
out.append("\n%s%s[%d]:\n%s%s%s" % (indent,(' '*(depth)),i,indent,(' '*(depth+1)),vv.dump(indent,depth+1) ))
else:
out.append("\n%s%s[%d]:\n%s%s%s" % (indent,(' '*(depth)),i,indent,(' '*(depth+1)),_ustr(vv)))
return "".join(out)
def pprint(self, *args, **kwargs):
"""
Pretty-printer for parsed results as a list, using the C{pprint} module.
Accepts additional positional or keyword args as defined for the
C{pprint.pprint} method. (U{http://docs.python.org/3/library/pprint.html#pprint.pprint})
Example::
ident = Word(alphas, alphanums)
num = Word(nums)
func = Forward()
term = ident | num | Group('(' + func + ')')
func <<= ident + Group(Optional(delimitedList(term)))
result = func.parseString("fna a,b,(fnb c,d,200),100")
result.pprint(width=40)
prints::
['fna',
['a',
'b',
['(', 'fnb', ['c', 'd', '200'], ')'],
'100']]
"""
pprint.pprint(self.asList(), *args, **kwargs)
# add support for pickle protocol
def __getstate__(self):
return ( self.__toklist,
( self.__tokdict.copy(),
self.__parent is not None and self.__parent() or None,
self.__accumNames,
self.__name ) )
def __setstate__(self,state):
self.__toklist = state[0]
(self.__tokdict,
par,
inAccumNames,
self.__name) = state[1]
self.__accumNames = {}
self.__accumNames.update(inAccumNames)
if par is not None:
self.__parent = wkref(par)
else:
self.__parent = None
def __getnewargs__(self):
return self.__toklist, self.__name, self.__asList, self.__modal
def __dir__(self):
return (dir(type(self)) + list(self.keys()))
MutableMapping.register(ParseResults)
def col (loc,strg):
"""Returns current column within a string, counting newlines as line separators.
The first column is number 1.
Note: the default parsing behavior is to expand tabs in the input string
before starting the parsing process. See L{I{ParserElement.parseString}<ParserElement.parseString>} for more information
on parsing strings containing C{<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.
"""
s = strg
return 1 if 0<loc<len(s) and s[loc-1] == '\n' else loc - s.rfind("\n", 0, loc)
def lineno(loc,strg):
"""Returns current line number within a string, counting newlines as line separators.
The first line is number 1.
Note: the default parsing behavior is to expand tabs in the input string
before starting the parsing process. See L{I{ParserElement.parseString}<ParserElement.parseString>} for more information
on parsing strings containing C{<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.
"""
return strg.count("\n",0,loc) + 1
def line( loc, strg ):
"""Returns the line of text containing loc within a string, counting newlines as line separators.
"""
lastCR = strg.rfind("\n", 0, loc)
nextCR = strg.find("\n", loc)
if nextCR >= 0:
return strg[lastCR+1:nextCR]
else:
return strg[lastCR+1:]
def _defaultStartDebugAction( instring, loc, expr ):
print (("Match " + _ustr(expr) + " at loc " + _ustr(loc) + "(%d,%d)" % ( lineno(loc,instring), col(loc,instring) )))
def _defaultSuccessDebugAction( instring, startloc, endloc, expr, toks ):
print ("Matched " + _ustr(expr) + " -> " + str(toks.asList()))
def _defaultExceptionDebugAction( instring, loc, expr, exc ):
print ("Exception raised:" + _ustr(exc))
def nullDebugAction(*args):
"""'Do-nothing' debug action, to suppress debugging output during parsing."""
pass
# Only works on Python 3.x - nonlocal is toxic to Python 2 installs
#~ 'decorator to trim function calls to match the arity of the target'
#~ def _trim_arity(func, maxargs=3):
#~ if func in singleArgBuiltins:
#~ return lambda s,l,t: func(t)
#~ limit = 0
#~ foundArity = False
#~ def wrapper(*args):
#~ nonlocal limit,foundArity
#~ while 1:
#~ try:
#~ ret = func(*args[limit:])
#~ foundArity = True
#~ return ret
#~ except TypeError:
#~ if limit == maxargs or foundArity:
#~ raise
#~ limit += 1
#~ continue
#~ return wrapper
# this version is Python 2.x-3.x cross-compatible
'decorator to trim function calls to match the arity of the target'
def _trim_arity(func, maxargs=2):
if func in singleArgBuiltins:
return lambda s,l,t: func(t)
limit = [0]
foundArity = [False]
# traceback return data structure changed in Py3.5 - normalize back to plain tuples
if system_version[:2] >= (3,5):
def extract_stack(limit=0):
# special handling for Python 3.5.0 - extra deep call stack by 1
offset = -3 if system_version == (3,5,0) else -2
frame_summary = traceback.extract_stack(limit=-offset+limit-1)[offset]
return [frame_summary[:2]]
def extract_tb(tb, limit=0):
frames = traceback.extract_tb(tb, limit=limit)
frame_summary = frames[-1]
return [frame_summary[:2]]
else:
extract_stack = traceback.extract_stack
extract_tb = traceback.extract_tb
# 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
LINE_DIFF = 6
# 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!!!!
this_line = extract_stack(limit=2)[-1]
pa_call_line_synth = (this_line[0], this_line[1]+LINE_DIFF)
def wrapper(*args):
while 1:
try:
ret = func(*args[limit[0]:])
foundArity[0] = True
return ret
except TypeError:
# re-raise TypeErrors if they did not come from our arity testing
if foundArity[0]:
raise
else:
try:
tb = sys.exc_info()[-1]
if not extract_tb(tb, limit=2)[-1][:2] == pa_call_line_synth:
raise
finally:
del tb
if limit[0] <= maxargs:
limit[0] += 1
continue
raise
# copy func name to wrapper for sensible debug output
func_name = "<parse action>"
try:
func_name = getattr(func, '__name__',
getattr(func, '__class__').__name__)
except Exception:
func_name = str(func)
wrapper.__name__ = func_name
return wrapper
class ParserElement(object):
"""Abstract base level parser element class."""
DEFAULT_WHITE_CHARS = " \n\t\r"
verbose_stacktrace = False
@staticmethod
def setDefaultWhitespaceChars( chars ):
r"""
Overrides the default whitespace chars
Example::
# default whitespace chars are space, <TAB> and newline
OneOrMore(Word(alphas)).parseString("abc def\nghi jkl") # -> ['abc', 'def', 'ghi', 'jkl']
# change to just treat newline as significant
ParserElement.setDefaultWhitespaceChars(" \t")
OneOrMore(Word(alphas)).parseString("abc def\nghi jkl") # -> ['abc', 'def']
"""
ParserElement.DEFAULT_WHITE_CHARS = chars
@staticmethod
def inlineLiteralsUsing(cls):
"""
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.parseString("1999/12/31") # -> ['1999', '/', '12', '/', '31']
# change to Suppress
ParserElement.inlineLiteralsUsing(Suppress)
date_str = integer("year") + '/' + integer("month") + '/' + integer("day")
date_str.parseString("1999/12/31") # -> ['1999', '12', '31']
"""
ParserElement._literalStringClass = cls
def __init__( self, savelist=False ):
self.parseAction = list()
self.failAction = None
#~ self.name = "<unknown>" # don't define self.name, let subclasses try/except upcall
self.strRepr = None
self.resultsName = None
self.saveAsList = savelist
self.skipWhitespace = True
self.whiteChars = ParserElement.DEFAULT_WHITE_CHARS
self.copyDefaultWhiteChars = True
self.mayReturnEmpty = False # used when checking for left-recursion
self.keepTabs = False
self.ignoreExprs = list()
self.debug = False
self.streamlined = False
self.mayIndexError = True # used to optimize exception handling for subclasses that don't advance parse index
self.errmsg = ""
self.modalResults = True # used to mark results names as modal (report only last) or cumulative (list all)
self.debugActions = ( None, None, None ) #custom debug actions
self.re = None
self.callPreparse = True # used to avoid redundant calls to preParse
self.callDuringTry = False
def copy( self ):
"""
Make a copy of this C{ParserElement}. Useful for defining different parse actions
for the same parsing pattern, using copies of the original parse element.
Example::
integer = Word(nums).setParseAction(lambda toks: int(toks[0]))
integerK = integer.copy().addParseAction(lambda toks: toks[0]*1024) + Suppress("K")
integerM = integer.copy().addParseAction(lambda toks: toks[0]*1024*1024) + Suppress("M")
print(OneOrMore(integerK | integerM | integer).parseString("5K 100 640K 256M"))
prints::
[5120, 100, 655360, 268435456]
Equivalent form of C{expr.copy()} is just C{expr()}::
integerM = integer().addParseAction(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 = ParserElement.DEFAULT_WHITE_CHARS
return cpy
def setName( self, name ):
"""
Define name for this expression, makes debugging and exception messages clearer.
Example::
Word(nums).parseString("ABC") # -> Exception: Expected W:(0123...) (at char 0), (line:1, col:1)
Word(nums).setName("integer").parseString("ABC") # -> Exception: Expected integer (at char 0), (line:1, col:1)
"""
self.name = name
self.errmsg = "Expected " + self.name
if hasattr(self,"exception"):
self.exception.msg = self.errmsg
return self
def setResultsName( self, name, listAllMatches=False ):
"""
Define name for referencing matching tokens as a nested attribute
of the returned parse results.
NOTE: this returns a *copy* of the original C{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,
C{expr("name")} in place of C{expr.setResultsName("name")} -
see L{I{__call__}<__call__>}.
Example::
date_str = (integer.setResultsName("year") + '/'
+ integer.setResultsName("month") + '/'
+ integer.setResultsName("day"))
# equivalent form:
date_str = integer("year") + '/' + integer("month") + '/' + integer("day")
"""
newself = self.copy()
if name.endswith("*"):
name = name[:-1]
listAllMatches=True
newself.resultsName = name
newself.modalResults = not listAllMatches
return newself
def setBreak(self,breakFlag = True):
"""Method to invoke the Python pdb debugger when this element is
about to be parsed. Set C{breakFlag} to True to enable, False to
disable.
"""
if breakFlag:
_parseMethod = self._parse
def breaker(instring, loc, doActions=True, callPreParse=True):
import pdb
pdb.set_trace()
return _parseMethod( instring, loc, doActions, callPreParse )
breaker._originalParseMethod = _parseMethod
self._parse = breaker
else:
if hasattr(self._parse,"_originalParseMethod"):
self._parse = self._parse._originalParseMethod
return self
def setParseAction( self, *fns, **kwargs ):
"""
Define one or more actions to perform when successfully matching parse element definition.
Parse action fn is a callable method with 0-3 arguments, called as C{fn(s,loc,toks)},
C{fn(loc,toks)}, C{fn(toks)}, or just C{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 C{L{ParseResults}} object
If the functions in fns modify the tokens, they can return them as the return
value from fn, and the modified list of tokens will replace the original.
Otherwise, fn does not need to return any value.
Optional keyword arguments:
- callDuringTry = (default=C{False}) indicate if parse action should be run during lookaheads and alternate testing
Note: the default parsing behavior is to expand tabs in the input string
before starting the parsing process. See L{I{parseString}<parseString>} for more information
on parsing strings containing C{<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::
integer = Word(nums)
date_str = integer + '/' + integer + '/' + integer
date_str.parseString("1999/12/31") # -> ['1999', '/', '12', '/', '31']
# use parse action to convert to ints at parse time
integer = Word(nums).setParseAction(lambda toks: int(toks[0]))
date_str = integer + '/' + integer + '/' + integer
# note that integer fields are now ints, not strings
date_str.parseString("1999/12/31") # -> [1999, '/', 12, '/', 31]
"""
self.parseAction = list(map(_trim_arity, list(fns)))
self.callDuringTry = kwargs.get("callDuringTry", False)
return self
def addParseAction( self, *fns, **kwargs ):
"""
Add one or more parse actions to expression's list of parse actions. See L{I{setParseAction}<setParseAction>}.
See examples in L{I{copy}<copy>}.
"""
self.parseAction += list(map(_trim_arity, list(fns)))
self.callDuringTry = self.callDuringTry or kwargs.get("callDuringTry", False)
return self
def addCondition(self, *fns, **kwargs):
"""Add a boolean predicate function to expression's list of parse actions. See
L{I{setParseAction}<setParseAction>} for function call signatures. Unlike C{setParseAction},
functions passed to C{addCondition} 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
Example::
integer = Word(nums).setParseAction(lambda toks: int(toks[0]))
year_int = integer.copy()
year_int.addCondition(lambda toks: toks[0] >= 2000, message="Only support years 2000 and later")
date_str = year_int + '/' + integer + '/' + integer
result = date_str.parseString("1999/12/31") # -> Exception: Only support years 2000 and later (at char 0), (line:1, col:1)
"""
msg = kwargs.get("message", "failed user-defined condition")
exc_type = ParseFatalException if kwargs.get("fatal", False) else ParseException
for fn in fns:
def pa(s,l,t):
if not bool(_trim_arity(fn)(s,l,t)):
raise exc_type(s,l,msg)
self.parseAction.append(pa)
self.callDuringTry = self.callDuringTry or kwargs.get("callDuringTry", False)
return self
def setFailAction( self, fn ):
"""Define action to perform if parsing fails at this expression.
Fail acton fn is a callable function that takes the arguments
C{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 C{L{ParseFatalException}}
if it is desired to stop parsing immediately."""
self.failAction = fn
return self
def _skipIgnorables( self, instring, loc ):
exprsFound = True
while exprsFound:
exprsFound = False
for e in self.ignoreExprs:
try:
while 1:
loc,dummy = e._parse( instring, loc )
exprsFound = True
except ParseException:
pass
return loc
def preParse( self, instring, loc ):
if self.ignoreExprs:
loc = self._skipIgnorables( instring, loc )
if self.skipWhitespace:
wt = self.whiteChars
instrlen = len(instring)
while loc < instrlen and instring[loc] in wt:
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 ):
debugging = ( self.debug ) #and doActions )
if debugging or self.failAction:
#~ print ("Match",self,"at loc",loc,"(%d,%d)" % ( lineno(loc,instring), col(loc,instring) ))
if (self.debugActions[0] ):
self.debugActions[0]( instring, loc, self )
if callPreParse and self.callPreparse:
preloc = self.preParse( instring, loc )
else:
preloc = loc
tokensStart = preloc
try:
try:
loc,tokens = self.parseImpl( instring, preloc, doActions )
except IndexError:
raise ParseException( instring, len(instring), self.errmsg, self )
except ParseBaseException as err:
#~ print ("Exception raised:", err)
if self.debugActions[2]:
self.debugActions[2]( instring, tokensStart, self, err )
if self.failAction:
self.failAction( instring, tokensStart, self, err )
raise
else:
if callPreParse and self.callPreparse:
preloc = self.preParse( instring, loc )
else:
preloc = loc
tokensStart = preloc
if self.mayIndexError or preloc >= len(instring):
try:
loc,tokens = self.parseImpl( instring, preloc, doActions )
except IndexError:
raise ParseException( instring, len(instring), self.errmsg, self )
else:
loc,tokens = self.parseImpl( instring, preloc, doActions )
tokens = self.postParse( instring, loc, tokens )
retTokens = 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:
tokens = fn( instring, tokensStart, retTokens )
if tokens is not None:
retTokens = ParseResults( tokens,
self.resultsName,
asList=self.saveAsList and isinstance(tokens,(ParseResults,list)),
modal=self.modalResults )
except ParseBaseException as err:
#~ print "Exception raised in user parse action:", err
if (self.debugActions[2] ):
self.debugActions[2]( instring, tokensStart, self, err )
raise
else:
for fn in self.parseAction:
tokens = fn( instring, tokensStart, retTokens )
if tokens is not None:
retTokens = ParseResults( tokens,
self.resultsName,
asList=self.saveAsList and isinstance(tokens,(ParseResults,list)),
modal=self.modalResults )
if debugging:
#~ print ("Matched",self,"->",retTokens.asList())
if (self.debugActions[1] ):
self.debugActions[1]( instring, tokensStart, loc, self, retTokens )
return loc, retTokens
def tryParse( self, instring, loc ):
try:
return self._parse( instring, loc, doActions=False )[0]
except ParseFatalException:
raise ParseException( instring, loc, self.errmsg, self)
def canParseNext(self, instring, loc):
try:
self.tryParse(instring, loc)
except (ParseException, IndexError):
return False
else:
return True
class _UnboundedCache(object):
def __init__(self):
cache = {}
self.not_in_cache = not_in_cache = object()
def get(self, key):
return cache.get(key, not_in_cache)
def set(self, key, value):
cache[key] = value
def clear(self):
cache.clear()
def cache_len(self):
return len(cache)
self.get = types.MethodType(get, self)
self.set = types.MethodType(set, self)
self.clear = types.MethodType(clear, self)
self.__len__ = types.MethodType(cache_len, self)
if _OrderedDict is not None:
class _FifoCache(object):
def __init__(self, size):
self.not_in_cache = not_in_cache = object()
cache = _OrderedDict()
def get(self, key):
return cache.get(key, not_in_cache)
def set(self, key, value):
cache[key] = value
while len(cache) > size:
try:
cache.popitem(False)
except KeyError:
pass
def clear(self):
cache.clear()
def cache_len(self):
return len(cache)
self.get = types.MethodType(get, self)
self.set = types.MethodType(set, self)
self.clear = types.MethodType(clear, self)
self.__len__ = types.MethodType(cache_len, self)
else:
class _FifoCache(object):
def __init__(self, size):
self.not_in_cache = not_in_cache = object()
cache = {}
key_fifo = collections.deque([], size)
def get(self, key):
return cache.get(key, not_in_cache)
def set(self, key, value):
cache[key] = value
while len(key_fifo) > size:
cache.pop(key_fifo.popleft(), None)
key_fifo.append(key)
def clear(self):
cache.clear()
key_fifo.clear()
def cache_len(self):
return len(cache)
self.get = types.MethodType(get, self)
self.set = types.MethodType(set, self)
self.clear = types.MethodType(clear, self)
self.__len__ = types.MethodType(cache_len, self)
# argument cache for optimizing repeated calls when backtracking through recursive expressions
packrat_cache = {} # this is set later by enabledPackrat(); this is here so that resetCache() 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 ):
HIT, MISS = 0, 1
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()))
return value
else:
ParserElement.packrat_cache_stats[HIT] += 1
if isinstance(value, Exception):
raise value
return (value[0], value[1].copy())
_parse = _parseNoCache
@staticmethod
def resetCache():
ParserElement.packrat_cache.clear()
ParserElement.packrat_cache_stats[:] = [0] * len(ParserElement.packrat_cache_stats)
_packratEnabled = False
@staticmethod
def enablePackrat(cache_size_limit=128):
"""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=C{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 C{ParserElement.enablePackrat()}. If
your program uses C{psyco} to "compile as you go", you must call
C{enablePackrat} before calling C{psyco.full()}. If you do not do this,
Python will crash. For best results, call C{enablePackrat()} immediately
after importing pyparsing.
Example::
import pyparsing
pyparsing.ParserElement.enablePackrat()
"""
if not ParserElement._packratEnabled:
ParserElement._packratEnabled = True
if cache_size_limit is None:
ParserElement.packrat_cache = ParserElement._UnboundedCache()
else:
ParserElement.packrat_cache = ParserElement._FifoCache(cache_size_limit)
ParserElement._parse = ParserElement._parseCache
def parseString( self, instring, parseAll=False ):
"""
Execute the parse expression with the given string.
This is the main interface to the client code, once the complete
expression has been built.
If you want the grammar to require that the entire input string be
successfully parsed, then set C{parseAll} to True (equivalent to ending
the grammar with C{L{StringEnd()}}).
Note: C{parseString} implicitly calls C{expandtabs()} on the input string,
in order to report proper column numbers in parse actions.
If the input string contains tabs and
the grammar uses parse actions that use the C{loc} argument to index into the
string being parsed, you can ensure you have a consistent view of the input
string by:
- calling C{parseWithTabs} on your grammar before calling C{parseString}
(see L{I{parseWithTabs}<parseWithTabs>})
- define your parse action using the full C{(s,loc,toks)} signature, and
reference the input string using the parse action's C{s} argument
- explictly expand the tabs in your input string before calling
C{parseString}
Example::
Word('a').parseString('aaaaabaaa') # -> ['aaaaa']
Word('a').parseString('aaaaabaaa', parseAll=True) # -> Exception: Expected end of text
"""
ParserElement.resetCache()
if not self.streamlined:
self.streamline()
#~ self.saveAsList = True
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, clears out pyparsing internal stack trace
raise exc
else:
return tokens
def scanString( self, instring, maxMatches=_MAX_INT, overlap=False ):
"""
Scan the input string for expression matches. Each match will return the
matching tokens, start location, and end location. May be called with optional
C{maxMatches} argument, to clip scanning after 'n' matches are found. If
C{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 L{I{parseString}<parseString>} for more information on parsing
strings with embedded tabs.
Example::
source = "sldjf123lsdjjkf345sldkjf879lkjsfd987"
print(source)
for tokens,start,end in Word(alphas).scanString(source):
print(' '*start + '^'*(end-start))
print(' '*start + tokens[0])
prints::
sldjf123lsdjjkf345sldkjf879lkjsfd987
^^^^^
sldjf
^^^^^^^
lsdjjkf
^^^^^^
sldkjf
^^^^^^
lkjsfd
"""
if not self.streamlined:
self.streamline()
for e in self.ignoreExprs:
e.streamline()
if not self.keepTabs:
instring = _ustr(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 = preparseFn( instring, loc )
nextLoc,tokens = parseFn( instring, preloc, callPreParse=False )
except ParseException:
loc = preloc+1
else:
if nextLoc > loc:
matches += 1
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
def transformString( self, instring ):
"""
Extension to C{L{scanString}}, to modify matching text with modified tokens that may
be returned from a parse action. To use C{transformString}, define a grammar and
attach a parse action to it that modifies the returned token list.
Invoking C{transformString()} on a target string will then scan for matches,
and replace the matched text patterns according to the logic in the parse
action. C{transformString()} returns the resulting transformed string.
Example::
wd = Word(alphas)
wd.setParseAction(lambda toks: toks[0].title())
print(wd.transformString("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 = []
lastE = 0
# force preservation of <TAB>s, to minimize unwanted transformation of string, and to
# keep string locs straight between transformString and scanString
self.keepTabs = True
try:
for t,s,e in self.scanString( instring ):
out.append( instring[lastE:s] )
if t:
if isinstance(t,ParseResults):
out += t.asList()
elif isinstance(t,list):
out += t
else:
out.append(t)
lastE = e
out.append(instring[lastE:])
out = [o for o in out if o]
return "".join(map(_ustr,_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
def searchString( self, instring, maxMatches=_MAX_INT ):
"""
Another extension to C{L{scanString}}, simplifying the access to the tokens found
to match the given parse expression. May be called with optional
C{maxMatches} 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.searchString("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.searchString("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']
"""
try:
return ParseResults([ t for t,s,e in self.scanString( instring, maxMatches ) ])
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
def split(self, instring, maxsplit=_MAX_INT, includeSeparators=False):
"""
Generator method to split a string using the given expression as a separator.
May be called with optional C{maxsplit} argument, to limit the number of splits;
and the optional C{includeSeparators} argument (default=C{False}), if the separating
matching text should be included in the split results.
Example::
punc = oneOf(list(".,;:/-!?"))
print(list(punc.split("This, this?, this sentence, is badly punctuated!")))
prints::
['This', ' this', '', ' this sentence', ' is badly punctuated', '']
"""
splits = 0
last = 0
for t,s,e in self.scanString(instring, maxMatches=maxsplit):
yield instring[last:s]
if includeSeparators:
yield t[0]
last = e
yield instring[last:]
def __add__(self, other ):
"""
Implementation of + operator - returns C{L{And}}. Adding strings to a ParserElement
converts them to L{Literal}s by default.
Example::
greet = Word(alphas) + "," + Word(alphas) + "!"
hello = "Hello, World!"
print (hello, "->", greet.parseString(hello))
Prints::
Hello, World! -> ['Hello', ',', 'World', '!']
"""
if isinstance( other, basestring ):
other = ParserElement._literalStringClass( other )
if not isinstance( other, ParserElement ):
warnings.warn("Cannot combine element of type %s with ParserElement" % type(other),
SyntaxWarning, stacklevel=2)
return None
return And( [ self, other ] )
def __radd__(self, other ):
"""
Implementation of + operator when left operand is not a C{L{ParserElement}}
"""
if isinstance( other, basestring ):
other = ParserElement._literalStringClass( other )
if not isinstance( other, ParserElement ):
warnings.warn("Cannot combine element of type %s with ParserElement" % type(other),
SyntaxWarning, stacklevel=2)
return None
return other + self
def __sub__(self, other):
"""
Implementation of - operator, returns C{L{And}} with error stop
"""
if isinstance( other, basestring ):
other = ParserElement._literalStringClass( other )
if not isinstance( other, ParserElement ):
warnings.warn("Cannot combine element of type %s with ParserElement" % type(other),
SyntaxWarning, stacklevel=2)
return None
return self + And._ErrorStop() + other
def __rsub__(self, other ):
"""
Implementation of - operator when left operand is not a C{L{ParserElement}}
"""
if isinstance( other, basestring ):
other = ParserElement._literalStringClass( other )
if not isinstance( other, ParserElement ):
warnings.warn("Cannot combine element of type %s with ParserElement" % type(other),
SyntaxWarning, stacklevel=2)
return None
return other - self
def __mul__(self,other):
"""
Implementation of * operator, allows use of C{expr * 3} in place of
C{expr + expr + expr}. Expressions may also me multiplied by a 2-integer
tuple, similar to C{{min,max}} multipliers in regular expressions. Tuples
may also include C{None} as in:
- C{expr*(n,None)} or C{expr*(n,)} is equivalent
to C{expr*n + L{ZeroOrMore}(expr)}
(read as "at least n instances of C{expr}")
- C{expr*(None,n)} is equivalent to C{expr*(0,n)}
(read as "0 to n instances of C{expr}")
- C{expr*(None,None)} is equivalent to C{L{ZeroOrMore}(expr)}
- C{expr*(1,None)} is equivalent to C{L{OneOrMore}(expr)}
Note that C{expr*(None,n)} does not raise an exception if
more than n exprs exist in the input stream; that is,
C{expr*(None,n)} does not enforce a maximum number of expr
occurrences. If this behavior is desired, then write
C{expr*(None,n) + ~expr}
"""
if isinstance(other,int):
minElements, optElements = other,0
elif isinstance(other,tuple):
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:
raise TypeError("cannot multiply 'ParserElement' and ('%s','%s') objects", type(other[0]),type(other[1]))
else:
raise TypeError("cannot multiply 'ParserElement' and '%s' objects", type(other))
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:
raise ValueError("cannot multiply ParserElement by 0 or (0,0)")
if (optElements):
def makeOptionalList(n):
if n>1:
return Optional(self + makeOptionalList(n-1))
else:
return Optional(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):
return self.__mul__(other)
def __or__(self, other ):
"""
Implementation of | operator - returns C{L{MatchFirst}}
"""
if isinstance( other, basestring ):
other = ParserElement._literalStringClass( other )
if not isinstance( other, ParserElement ):
warnings.warn("Cannot combine element of type %s with ParserElement" % type(other),
SyntaxWarning, stacklevel=2)
return None
return MatchFirst( [ self, other ] )
def __ror__(self, other ):
"""
Implementation of | operator when left operand is not a C{L{ParserElement}}
"""
if isinstance( other, basestring ):
other = ParserElement._literalStringClass( other )
if not isinstance( other, ParserElement ):
warnings.warn("Cannot combine element of type %s with ParserElement" % type(other),
SyntaxWarning, stacklevel=2)
return None
return other | self
def __xor__(self, other ):
"""
Implementation of ^ operator - returns C{L{Or}}
"""
if isinstance( other, basestring ):
other = ParserElement._literalStringClass( other )
if not isinstance( other, ParserElement ):
warnings.warn("Cannot combine element of type %s with ParserElement" % type(other),
SyntaxWarning, stacklevel=2)
return None
return Or( [ self, other ] )
def __rxor__(self, other ):
"""
Implementation of ^ operator when left operand is not a C{L{ParserElement}}
"""
if isinstance( other, basestring ):
other = ParserElement._literalStringClass( other )
if not isinstance( other, ParserElement ):
warnings.warn("Cannot combine element of type %s with ParserElement" % type(other),
SyntaxWarning, stacklevel=2)
return None
return other ^ self
def __and__(self, other ):
"""
Implementation of & operator - returns C{L{Each}}
"""
if isinstance( other, basestring ):
other = ParserElement._literalStringClass( other )
if not isinstance( other, ParserElement ):
warnings.warn("Cannot combine element of type %s with ParserElement" % type(other),
SyntaxWarning, stacklevel=2)
return None
return Each( [ self, other ] )
def __rand__(self, other ):
"""
Implementation of & operator when left operand is not a C{L{ParserElement}}
"""
if isinstance( other, basestring ):
other = ParserElement._literalStringClass( other )
if not isinstance( other, ParserElement ):
warnings.warn("Cannot combine element of type %s with ParserElement" % type(other),
SyntaxWarning, stacklevel=2)
return None
return other & self
def __invert__( self ):
"""
Implementation of ~ operator - returns C{L{NotAny}}
"""
return NotAny( self )
def __call__(self, name=None):
"""
Shortcut for C{L{setResultsName}}, with C{listAllMatches=False}.
If C{name} is given with a trailing C{'*'} character, then C{listAllMatches} will be
passed as C{True}.
If C{name} is omitted, same as calling C{L{copy}}.
Example::
# these are equivalent
userdata = Word(alphas).setResultsName("name") + Word(nums+"-").setResultsName("socsecno")
userdata = Word(alphas)("name") + Word(nums+"-")("socsecno")
"""
if name is not None:
return self.setResultsName(name)
else:
return self.copy()
def suppress( self ):
"""
Suppresses the output of this C{ParserElement}; useful to keep punctuation from
cluttering up returned output.
"""
return Suppress( self )
def leaveWhitespace( self ):
"""
Disables the skipping of whitespace before matching the characters in the
C{ParserElement}'s defined pattern. This is normally only used internally by
the pyparsing module, but may be needed in some whitespace-sensitive grammars.
"""
self.skipWhitespace = False
return self
def setWhitespaceChars( self, chars ):
"""
Overrides the default whitespace chars
"""
self.skipWhitespace = True
self.whiteChars = chars
self.copyDefaultWhiteChars = False
return self
def parseWithTabs( self ):
"""
Overrides default behavior to expand C{<TAB>}s to spaces before parsing the input string.
Must be called before C{parseString} when the input grammar contains elements that
match C{<TAB>} characters.
"""
self.keepTabs = True
return self
def ignore( self, other ):
"""
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 = OneOrMore(Word(alphas))
patt.parseString('ablaj /* comment */ lskjd') # -> ['ablaj']
patt.ignore(cStyleComment)
patt.parseString('ablaj /* comment */ lskjd') # -> ['ablaj', 'lskjd']
"""
if isinstance(other, basestring):
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 setDebugActions( self, startAction, successAction, exceptionAction ):
"""
Enable display of debugging messages while doing pattern matching.
"""
self.debugActions = (startAction or _defaultStartDebugAction,
successAction or _defaultSuccessDebugAction,
exceptionAction or _defaultExceptionDebugAction)
self.debug = True
return self
def setDebug( self, flag=True ):
"""
Enable display of debugging messages while doing pattern matching.
Set C{flag} to True to enable, False to disable.
Example::
wd = Word(alphas).setName("alphaword")
integer = Word(nums).setName("numword")
term = wd | integer
# turn on debugging for wd
wd.setDebug()
OneOrMore(term).parseString("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 L{setDebugActions}. Prior to attempting
to match the C{wd} expression, the debugging message C{"Match <exprname> at loc <n>(<line>,<col>)"}
is shown. Then if the parse succeeds, a C{"Matched"} message is shown, or an C{"Exception raised"}
message is shown. Also note the use of L{setName} 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 C{Word} expression without calling C{setName} is C{"W:(ABCD...)"}.
"""
if flag:
self.setDebugActions( _defaultStartDebugAction, _defaultSuccessDebugAction, _defaultExceptionDebugAction )
else:
self.debug = False
return self
def __str__( self ):
return self.name
def __repr__( self ):
return _ustr(self)
def streamline( self ):
self.streamlined = True
self.strRepr = None
return self
def checkRecursion( self, parseElementList ):
pass
def validate( self, validateTrace=[] ):
"""
Check defined expressions for valid structure, check for infinite recursive definitions.
"""
self.checkRecursion( [] )
def parseFile( self, file_or_filename, parseAll=False ):
"""
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.
"""
try:
file_contents = file_or_filename.read()
except AttributeError:
with open(file_or_filename, "r") as f:
file_contents = f.read()
try:
return self.parseString(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
def __eq__(self,other):
if isinstance(other, ParserElement):
return self is other or vars(self) == vars(other)
elif isinstance(other, basestring):
return self.matches(other)
else:
return super(ParserElement,self)==other
def __ne__(self,other):
return not (self == other)
def __hash__(self):
return hash(id(self))
def __req__(self,other):
return self == other
def __rne__(self,other):
return not (self == other)
def matches(self, testString, parseAll=True):
"""
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:
- testString - to test against this expression for a match
- parseAll - (default=C{True}) - flag to pass to C{L{parseString}} when running tests
Example::
expr = Word(nums)
assert expr.matches("100")
"""
try:
self.parseString(_ustr(testString), parseAll=parseAll)
return True
except ParseBaseException:
return False
def runTests(self, tests, parseAll=True, comment='#', fullDump=True, printResults=True, failureTests=False):
"""
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
- parseAll - (default=C{True}) - flag to pass to C{L{parseString}} when running tests
- comment - (default=C{'#'}) - expression for indicating embedded comments in the test
string; pass None to disable comment filtering
- fullDump - (default=C{True}) - dump results as list followed by results names in nested outline;
if False, only dump nested list
- printResults - (default=C{True}) prints test output to stdout
- failureTests - (default=C{False}) indicates if these tests are expected to fail parsing
Returns: a (success, results) tuple, where success indicates that all tests succeeded
(or failed if C{failureTests} 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.runTests('''
# 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.runTests('''
# stray character
100Z
# missing leading digit before '.'
-.100
# too many '.'
3.14.159
''', failureTests=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.runTest(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'.)
"""
if isinstance(tests, basestring):
tests = list(map(str.strip, tests.rstrip().splitlines()))
if isinstance(comment, basestring):
comment = Literal(comment)
allResults = []
comments = []
success = True
for t in tests:
if comment is not None and comment.matches(t, False) or comments and not t:
comments.append(t)
continue
if not t:
continue
out = ['\n'.join(comments), t]
comments = []
try:
t = t.replace(r'\n','\n')
result = self.parseString(t, parseAll=parseAll)
out.append(result.dump(full=fullDump))
success = success and not failureTests
except ParseBaseException as pe:
fatal = "(FATAL)" if isinstance(pe, ParseFatalException) else ""
if '\n' in t:
out.append(line(pe.loc, t))
out.append(' '*(col(pe.loc,t)-1) + '^' + fatal)
else:
out.append(' '*pe.loc + '^' + fatal)
out.append("FAIL: " + str(pe))
success = success and failureTests
result = pe
except Exception as exc:
out.append("FAIL-EXCEPTION: " + str(exc))
success = success and failureTests
result = exc
if printResults:
if fullDump:
out.append('')
print('\n'.join(out))
allResults.append((t, result))
return success, allResults
class Token(ParserElement):
"""
Abstract C{ParserElement} subclass, for defining atomic matching patterns.
"""
def __init__( self ):
super(Token,self).__init__( savelist=False )
class Empty(Token):
"""
An empty token, will always match.
"""
def __init__( self ):
super(Empty,self).__init__()
self.name = "Empty"
self.mayReturnEmpty = True
self.mayIndexError = False
class NoMatch(Token):
"""
A token that will never match.
"""
def __init__( self ):
super(NoMatch,self).__init__()
self.name = "NoMatch"
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').parseString('blah') # -> ['blah']
Literal('blah').parseString('blahfooblah') # -> ['blah']
Literal('blah').parseString('bla') # -> Exception: Expected "blah"
For case-insensitive matching, use L{CaselessLiteral}.
For keyword matching (force word break before and after the matched string),
use L{Keyword} or L{CaselessKeyword}.
"""
def __init__( self, matchString ):
super(Literal,self).__init__()
self.match = matchString
self.matchLen = len(matchString)
try:
self.firstMatchChar = matchString[0]
except IndexError:
warnings.warn("null string passed to Literal; use Empty() instead",
SyntaxWarning, stacklevel=2)
self.__class__ = Empty
self.name = '"%s"' % _ustr(self.match)
self.errmsg = "Expected " + self.name
self.mayReturnEmpty = False
self.mayIndexError = False
# Performance tuning: this routine gets called a *lot*
# if this is a single character match string and the first character matches,
# short-circuit as quickly as possible, and avoid calling startswith
#~ @profile
def parseImpl( self, instring, loc, doActions=True ):
if (instring[loc] == self.firstMatchChar and
(self.matchLen==1 or instring.startswith(self.match,loc)) ):
return loc+self.matchLen, self.match
raise ParseException(instring, loc, self.errmsg, self)
_L = Literal
ParserElement._literalStringClass = Literal
class Keyword(Token):
"""
Token to exactly match a specified string as a keyword, that is, it must be
immediately followed by a non-keyword character. Compare with C{L{Literal}}:
- C{Literal("if")} will match the leading C{'if'} in C{'ifAndOnlyIf'}.
- C{Keyword("if")} will not; it will only match the leading C{'if'} in C{'if x=1'}, or C{'if(y==2)'}
Accepts two optional constructor arguments in addition to the keyword string:
- C{identChars} is a string of characters that would be valid identifier characters,
defaulting to all alphanumerics + "_" and "$"
- C{caseless} allows case-insensitive matching, default is C{False}.
Example::
Keyword("start").parseString("start") # -> ['start']
Keyword("start").parseString("starting") # -> Exception
For case-insensitive matching, use L{CaselessKeyword}.
"""
DEFAULT_KEYWORD_CHARS = alphanums+"_$"
def __init__( self, matchString, identChars=None, caseless=False ):
super(Keyword,self).__init__()
if identChars is None:
identChars = Keyword.DEFAULT_KEYWORD_CHARS
self.match = matchString
self.matchLen = len(matchString)
try:
self.firstMatchChar = matchString[0]
except IndexError:
warnings.warn("null string passed to Keyword; use Empty() instead",
SyntaxWarning, stacklevel=2)
self.name = '"%s"' % self.match
self.errmsg = "Expected " + self.name
self.mayReturnEmpty = False
self.mayIndexError = False
self.caseless = caseless
if caseless:
self.caselessmatch = matchString.upper()
identChars = identChars.upper()
self.identChars = set(identChars)
def parseImpl( self, instring, loc, doActions=True ):
if self.caseless:
if ( (instring[ loc:loc+self.matchLen ].upper() == self.caselessmatch) and
(loc >= len(instring)-self.matchLen or instring[loc+self.matchLen].upper() not in self.identChars) and
(loc == 0 or instring[loc-1].upper() not in self.identChars) ):
return loc+self.matchLen, self.match
else:
if (instring[loc] == self.firstMatchChar and
(self.matchLen==1 or instring.startswith(self.match,loc)) and
(loc >= len(instring)-self.matchLen or instring[loc+self.matchLen] not in self.identChars) and
(loc == 0 or instring[loc-1] not in self.identChars) ):
return loc+self.matchLen, self.match
raise ParseException(instring, loc, self.errmsg, self)
def copy(self):
c = super(Keyword,self).copy()
c.identChars = Keyword.DEFAULT_KEYWORD_CHARS
return c
@staticmethod
def setDefaultKeywordChars( chars ):
"""Overrides the default Keyword chars
"""
Keyword.DEFAULT_KEYWORD_CHARS = 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::
OneOrMore(CaselessLiteral("CMD")).parseString("cmd CMD Cmd10") # -> ['CMD', 'CMD', 'CMD']
(Contrast with example for L{CaselessKeyword}.)
"""
def __init__( self, matchString ):
super(CaselessLiteral,self).__init__( matchString.upper() )
# Preserve the defining literal.
self.returnString = matchString
self.name = "'%s'" % self.returnString
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 L{Keyword}.
Example::
OneOrMore(CaselessKeyword("CMD")).parseString("cmd CMD Cmd10") # -> ['CMD', 'CMD']
(Contrast with example for L{CaselessLiteral}.)
"""
def __init__( self, matchString, identChars=None ):
super(CaselessKeyword,self).__init__( matchString, identChars, caseless=True )
def parseImpl( self, instring, loc, doActions=True ):
if ( (instring[ loc:loc+self.matchLen ].upper() == self.caselessmatch) and
(loc >= len(instring)-self.matchLen or instring[loc+self.matchLen].upper() not in self.identChars) ):
return loc+self.matchLen, self.match
raise ParseException(instring, loc, self.errmsg, self)
class CloseMatch(Token):
"""
A variation on L{Literal} which matches "close" matches, that is,
strings with at most 'n' mismatching characters. C{CloseMatch} takes parameters:
- C{match_string} - string to be matched
- C{maxMismatches} - (C{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:
- C{mismatches} - a list of the positions within the match_string where mismatches were found
- C{original} - the original match_string used to compare against the input string
If C{mismatches} is an empty list, then the match was an exact match.
Example::
patt = CloseMatch("ATCATCGAATGGA")
patt.parseString("ATCATCGAAXGGA") # -> (['ATCATCGAAXGGA'], {'mismatches': [[9]], 'original': ['ATCATCGAATGGA']})
patt.parseString("ATCAXCGAAXGGA") # -> Exception: Expected 'ATCATCGAATGGA' (with up to 1 mismatches) (at char 0), (line:1, col:1)
# exact match
patt.parseString("ATCATCGAATGGA") # -> (['ATCATCGAATGGA'], {'mismatches': [[]], 'original': ['ATCATCGAATGGA']})
# close match allowing up to 2 mismatches
patt = CloseMatch("ATCATCGAATGGA", maxMismatches=2)
patt.parseString("ATCAXCGAAXGGA") # -> (['ATCAXCGAAXGGA'], {'mismatches': [[4, 9]], 'original': ['ATCATCGAATGGA']})
"""
def __init__(self, match_string, maxMismatches=1):
super(CloseMatch,self).__init__()
self.name = match_string
self.match_string = match_string
self.maxMismatches = maxMismatches
self.errmsg = "Expected %r (with up to %d mismatches)" % (self.match_string, self.maxMismatches)
self.mayIndexError = False
self.mayReturnEmpty = False
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 enumerate(zip(instring[loc:maxloc], self.match_string)):
src,mat = s_m
if src != mat:
mismatches.append(match_stringloc)
if len(mismatches) > maxMismatches:
break
else:
loc = match_stringloc + 1
results = ParseResults([instring[start:loc]])
results['original'] = self.match_string
results['mismatches'] = mismatches
return loc, results
raise ParseException(instring, loc, self.errmsg, self)
class Word(Token):
"""
Token for matching words composed of allowed character sets.
Defined with string containing all allowed initial characters,
an optional string containing allowed body characters (if omitted,
defaults to the initial character set), and an optional minimum,
maximum, and/or exact length. The default value for C{min} is 1 (a
minimum value < 1 is not valid); the default values for C{max} and C{exact}
are 0, meaning no maximum or exact length restriction. An optional
C{excludeChars} parameter can list characters that might be found in
the input C{bodyChars} string; useful to define a word of all printables
except for one or two characters, for instance.
L{srange} is useful for defining custom character set strings for defining
C{Word} expressions, using range notation from regular expression character sets.
A common mistake is to use C{Word} to match a specific literal string, as in
C{Word("Address")}. Remember that C{Word} uses the string argument to define
I{sets} of matchable characters. This expression would match "Add", "AAA",
"dAred", or any other word made up of the characters 'A', 'd', 'r', 'e', and 's'.
To match an exact literal string, use L{Literal} or L{Keyword}.
pyparsing includes helper strings for building Words:
- L{alphas}
- L{nums}
- L{alphanums}
- L{hexnums}
- L{alphas8bit} (alphabetic characters in ASCII range 128-255 - accented, tilded, umlauted, etc.)
- L{punc8bit} (non-alphabetic characters in ASCII range 128-255 - currency, symbols, superscripts, diacriticals, etc.)
- L{printables} (any non-whitespace character)
Example::
# a word composed of digits
integer = Word(nums) # equivalent to Word("0123456789") or Word(srange("0-9"))
# a word with a leading capital, and zero or more lowercase
capital_word = Word(alphas.upper(), alphas.lower())
# hostnames are alphanumeric, with leading alpha, and '-'
hostname = Word(alphas, alphanums+'-')
# roman numeral (not a strict parser, accepts invalid mix of characters)
roman = Word("IVXLCDM")
# any string of non-whitespace characters, except for ','
csv_value = Word(printables, excludeChars=",")
"""
def __init__( self, initChars, bodyChars=None, min=1, max=0, exact=0, asKeyword=False, excludeChars=None ):
super(Word,self).__init__()
if excludeChars:
initChars = ''.join(c for c in initChars if c not in excludeChars)
if bodyChars:
bodyChars = ''.join(c for c in bodyChars if c not in excludeChars)
self.initCharsOrig = initChars
self.initChars = set(initChars)
if bodyChars :
self.bodyCharsOrig = bodyChars
self.bodyChars = set(bodyChars)
else:
self.bodyCharsOrig = initChars
self.bodyChars = set(initChars)
self.maxSpecified = max > 0
if min < 1:
raise ValueError("cannot specify a minimum length < 1; use Optional(Word()) if zero-length word is permitted")
self.minLen = min
if max > 0:
self.maxLen = max
else:
self.maxLen = _MAX_INT
if exact > 0:
self.maxLen = exact
self.minLen = exact
self.name = _ustr(self)
self.errmsg = "Expected " + self.name
self.mayIndexError = False
self.asKeyword = asKeyword
if ' ' not in self.initCharsOrig+self.bodyCharsOrig and (min==1 and max==0 and exact==0):
if self.bodyCharsOrig == self.initCharsOrig:
self.reString = "[%s]+" % _escapeRegexRangeChars(self.initCharsOrig)
elif len(self.initCharsOrig) == 1:
self.reString = "%s[%s]*" % \
(re.escape(self.initCharsOrig),
_escapeRegexRangeChars(self.bodyCharsOrig),)
else:
self.reString = "[%s][%s]*" % \
(_escapeRegexRangeChars(self.initCharsOrig),
_escapeRegexRangeChars(self.bodyCharsOrig),)
if self.asKeyword:
self.reString = r"\b"+self.reString+r"\b"
try:
self.re = re.compile( self.reString )
except Exception:
self.re = None
def parseImpl( self, instring, loc, doActions=True ):
if self.re:
result = self.re.match(instring,loc)
if not result:
raise ParseException(instring, loc, self.errmsg, self)
loc = result.end()
return loc, result.group()
if not(instring[ loc ] in self.initChars):
raise ParseException(instring, loc, self.errmsg, self)
start = loc
loc += 1
instrlen = len(instring)
bodychars = self.bodyChars
maxloc = start + self.maxLen
maxloc = min( maxloc, instrlen )
while loc < maxloc and instring[loc] in bodychars:
loc += 1
throwException = False
if loc - start < self.minLen:
throwException = True
if self.maxSpecified and loc < instrlen and instring[loc] in bodychars:
throwException = True
if self.asKeyword:
if (start>0 and instring[start-1] in bodychars) or (loc<instrlen and instring[loc] in bodychars):
throwException = True
if throwException:
raise ParseException(instring, loc, self.errmsg, self)
return loc, instring[start:loc]
def __str__( self ):
try:
return super(Word,self).__str__()
except Exception:
pass
if self.strRepr is None:
def charsAsStr(s):
if len(s)>4:
return s[:4]+"..."
else:
return s
if ( self.initCharsOrig != self.bodyCharsOrig ):
self.strRepr = "W:(%s,%s)" % ( charsAsStr(self.initCharsOrig), charsAsStr(self.bodyCharsOrig) )
else:
self.strRepr = "W:(%s)" % charsAsStr(self.initCharsOrig)
return self.strRepr
class Regex(Token):
r"""
Token for matching strings that match a given regular expression.
Defined with string specifying the regular expression in a form recognized by the inbuilt Python re module.
If the given regex contains named groups (defined using C{(?P<name>...)}), these will be preserved as
named parse results.
Example::
realnum = Regex(r"[+-]?\d+\.\d*")
date = Regex(r'(?P<year>\d{4})-(?P<month>\d\d?)-(?P<day>\d\d?)')
# ref: http://stackoverflow.com/questions/267399/how-do-you-match-only-valid-roman-numerals-with-a-regular-expression
roman = Regex(r"M{0,4}(CM|CD|D?C{0,3})(XC|XL|L?X{0,3})(IX|IV|V?I{0,3})")
"""
compiledREtype = type(re.compile("[A-Z]"))
def __init__( self, pattern, flags=0):
"""The parameters C{pattern} and C{flags} are passed to the C{re.compile()} function as-is. See the Python C{re} module for an explanation of the acceptable patterns and flags."""
super(Regex,self).__init__()
if isinstance(pattern, basestring):
if not pattern:
warnings.warn("null string passed to Regex; use Empty() instead",
SyntaxWarning, stacklevel=2)
self.pattern = pattern
self.flags = flags
try:
self.re = re.compile(self.pattern, self.flags)
self.reString = self.pattern
except sre_constants.error:
warnings.warn("invalid pattern (%s) passed to Regex" % pattern,
SyntaxWarning, stacklevel=2)
raise
elif isinstance(pattern, Regex.compiledREtype):
self.re = pattern
self.pattern = \
self.reString = str(pattern)
self.flags = flags
else:
raise ValueError("Regex may only be constructed with a string or a compiled RE object")
self.name = _ustr(self)
self.errmsg = "Expected " + self.name
self.mayIndexError = False
self.mayReturnEmpty = True
def parseImpl( self, instring, loc, doActions=True ):
result = self.re.match(instring,loc)
if not result:
raise ParseException(instring, loc, self.errmsg, self)
loc = result.end()
d = result.groupdict()
ret = ParseResults(result.group())
if d:
for k in d:
ret[k] = d[k]
return loc,ret
def __str__( self ):
try:
return super(Regex,self).__str__()
except Exception:
pass
if self.strRepr is None:
self.strRepr = "Re:(%s)" % repr(self.pattern)
return self.strRepr
class QuotedString(Token):
r"""
Token for matching strings that are delimited by quoting characters.
Defined with the following parameters:
- quoteChar - string of one or more characters defining the quote delimiting string
- escChar - character to escape quotes, typically backslash (default=C{None})
- escQuote - special quote sequence to escape an embedded quote string (such as SQL's "" to escape an embedded ") (default=C{None})
- multiline - boolean indicating whether quotes can span multiple lines (default=C{False})
- unquoteResults - boolean indicating whether the matched text should be unquoted (default=C{True})
- endQuoteChar - string of one or more characters defining the end of the quote delimited string (default=C{None} => same as quoteChar)
- convertWhitespaceEscapes - convert escaped whitespace (C{'\t'}, C{'\n'}, etc.) to actual whitespace (default=C{True})
Example::
qs = QuotedString('"')
print(qs.searchString('lsjdf "This is the quote" sldjf'))
complex_qs = QuotedString('{{', endQuoteChar='}}')
print(complex_qs.searchString('lsjdf {{This is the "quote"}} sldjf'))
sql_qs = QuotedString('"', escQuote='""')
print(sql_qs.searchString('lsjdf "This is the quote with ""embedded"" quotes" sldjf'))
prints::
[['This is the quote']]
[['This is the "quote"']]
[['This is the quote with "embedded" quotes']]
"""
def __init__( self, quoteChar, escChar=None, escQuote=None, multiline=False, unquoteResults=True, endQuoteChar=None, convertWhitespaceEscapes=True):
super(QuotedString,self).__init__()
# remove white space from quote chars - wont work anyway
quoteChar = quoteChar.strip()
if not quoteChar:
warnings.warn("quoteChar cannot be the empty string",SyntaxWarning,stacklevel=2)
raise SyntaxError()
if endQuoteChar is None:
endQuoteChar = quoteChar
else:
endQuoteChar = endQuoteChar.strip()
if not endQuoteChar:
warnings.warn("endQuoteChar cannot be the empty string",SyntaxWarning,stacklevel=2)
raise SyntaxError()
self.quoteChar = quoteChar
self.quoteCharLen = len(quoteChar)
self.firstQuoteChar = quoteChar[0]
self.endQuoteChar = endQuoteChar
self.endQuoteCharLen = len(endQuoteChar)
self.escChar = escChar
self.escQuote = escQuote
self.unquoteResults = unquoteResults
self.convertWhitespaceEscapes = convertWhitespaceEscapes
if multiline:
self.flags = re.MULTILINE | re.DOTALL
self.pattern = r'%s(?:[^%s%s]' % \
( re.escape(self.quoteChar),
_escapeRegexRangeChars(self.endQuoteChar[0]),
(escChar is not None and _escapeRegexRangeChars(escChar) or '') )
else:
self.flags = 0
self.pattern = r'%s(?:[^%s\n\r%s]' % \
( re.escape(self.quoteChar),
_escapeRegexRangeChars(self.endQuoteChar[0]),
(escChar is not None and _escapeRegexRangeChars(escChar) or '') )
if len(self.endQuoteChar) > 1:
self.pattern += (
'|(?:' + ')|(?:'.join("%s[^%s]" % (re.escape(self.endQuoteChar[:i]),
_escapeRegexRangeChars(self.endQuoteChar[i]))
for i in range(len(self.endQuoteChar)-1,0,-1)) + ')'
)
if escQuote:
self.pattern += (r'|(?:%s)' % re.escape(escQuote))
if escChar:
self.pattern += (r'|(?:%s.)' % re.escape(escChar))
self.escCharReplacePattern = re.escape(self.escChar)+"(.)"
self.pattern += (r')*%s' % re.escape(self.endQuoteChar))
try:
self.re = re.compile(self.pattern, self.flags)
self.reString = self.pattern
except sre_constants.error:
warnings.warn("invalid pattern (%s) passed to Regex" % self.pattern,
SyntaxWarning, stacklevel=2)
raise
self.name = _ustr(self)
self.errmsg = "Expected " + self.name
self.mayIndexError = False
self.mayReturnEmpty = True
def parseImpl( self, instring, loc, doActions=True ):
result = instring[loc] == self.firstQuoteChar and self.re.match(instring,loc) or None
if not result:
raise ParseException(instring, loc, self.errmsg, self)
loc = result.end()
ret = result.group()
if self.unquoteResults:
# strip off quotes
ret = ret[self.quoteCharLen:-self.endQuoteCharLen]
if isinstance(ret,basestring):
# replace escaped whitespace
if '\\' in ret and self.convertWhitespaceEscapes:
ws_map = {
r'\t' : '\t',
r'\n' : '\n',
r'\f' : '\f',
r'\r' : '\r',
}
for wslit,wschar in ws_map.items():
ret = ret.replace(wslit, wschar)
# replace escaped characters
if self.escChar:
ret = re.sub(self.escCharReplacePattern, r"\g<1>", ret)
# replace escaped quotes
if self.escQuote:
ret = ret.replace(self.escQuote, self.endQuoteChar)
return loc, ret
def __str__( self ):
try:
return super(QuotedString,self).__str__()
except Exception:
pass
if self.strRepr is None:
self.strRepr = "quoted string, starting with %s ending with %s" % (self.quoteChar, self.endQuoteChar)
return self.strRepr
class CharsNotIn(Token):
"""
Token for matching words composed of characters I{not} in a given set (will
include whitespace in matched characters if not listed in the provided exclusion set - see example).
Defined with string containing all disallowed characters, and an optional
minimum, maximum, and/or exact length. The default value for C{min} is 1 (a
minimum value < 1 is not valid); the default values for C{max} and C{exact}
are 0, meaning no maximum or exact length restriction.
Example::
# define a comma-separated-value as anything that is not a ','
csv_value = CharsNotIn(',')
print(delimitedList(csv_value).parseString("dkls,lsdkjf,s12 34,@!#,213"))
prints::
['dkls', 'lsdkjf', 's12 34', '@!#', '213']
"""
def __init__( self, notChars, min=1, max=0, exact=0 ):
super(CharsNotIn,self).__init__()
self.skipWhitespace = False
self.notChars = notChars
if min < 1:
raise ValueError("cannot specify a minimum length < 1; use Optional(CharsNotIn()) if zero-length char group is permitted")
self.minLen = min
if max > 0:
self.maxLen = max
else:
self.maxLen = _MAX_INT
if exact > 0:
self.maxLen = exact
self.minLen = exact
self.name = _ustr(self)
self.errmsg = "Expected " + self.name
self.mayReturnEmpty = ( self.minLen == 0 )
self.mayIndexError = False
def parseImpl( self, instring, loc, doActions=True ):
if instring[loc] in self.notChars:
raise ParseException(instring, loc, self.errmsg, self)
start = loc
loc += 1
notchars = self.notChars
maxlen = min( start+self.maxLen, len(instring) )
while loc < maxlen and \
(instring[loc] not in notchars):
loc += 1
if loc - start < self.minLen:
raise ParseException(instring, loc, self.errmsg, self)
return loc, instring[start:loc]
def __str__( self ):
try:
return super(CharsNotIn, self).__str__()
except Exception:
pass
if self.strRepr is None:
if len(self.notChars) > 4:
self.strRepr = "!W:(%s...)" % self.notChars[:4]
else:
self.strRepr = "!W:(%s)" % self.notChars
return self.strRepr
class White(Token):
"""
Special matching class for matching whitespace. Normally, whitespace is ignored
by pyparsing grammars. This class is included when some whitespace structures
are significant. Define with a string containing the whitespace characters to be
matched; default is C{" \\t\\r\\n"}. Also takes optional C{min}, C{max}, and C{exact} arguments,
as defined for the C{L{Word}} class.
"""
whiteStrs = {
" " : "<SPC>",
"\t": "<TAB>",
"\n": "<LF>",
"\r": "<CR>",
"\f": "<FF>",
}
def __init__(self, ws=" \t\r\n", min=1, max=0, exact=0):
super(White,self).__init__()
self.matchWhite = ws
self.setWhitespaceChars( "".join(c for c in self.whiteChars if c not in self.matchWhite) )
#~ self.leaveWhitespace()
self.name = ("".join(White.whiteStrs[c] for c in self.matchWhite))
self.mayReturnEmpty = True
self.errmsg = "Expected " + self.name
self.minLen = min
if max > 0:
self.maxLen = max
else:
self.maxLen = _MAX_INT
if exact > 0:
self.maxLen = exact
self.minLen = exact
def parseImpl( self, instring, loc, doActions=True ):
if not(instring[ loc ] in self.matchWhite):
raise ParseException(instring, loc, self.errmsg, self)
start = loc
loc += 1
maxloc = start + self.maxLen
maxloc = min( maxloc, len(instring) )
while loc < maxloc and instring[loc] in self.matchWhite:
loc += 1
if loc - start < self.minLen:
raise ParseException(instring, loc, self.errmsg, self)
return loc, instring[start:loc]
class _PositionToken(Token):
def __init__( self ):
super(_PositionToken,self).__init__()
self.name=self.__class__.__name__
self.mayReturnEmpty = True
self.mayIndexError = False
class GoToColumn(_PositionToken):
"""
Token to advance to a specific column of input text; useful for tabular report scraping.
"""
def __init__( self, colno ):
super(GoToColumn,self).__init__()
self.col = colno
def preParse( self, instring, loc ):
if col(loc,instring) != self.col:
instrlen = len(instring)
if self.ignoreExprs:
loc = self._skipIgnorables( instring, loc )
while loc < instrlen and instring[loc].isspace() and col( loc, instring ) != self.col :
loc += 1
return loc
def parseImpl( self, instring, loc, doActions=True ):
thiscol = col( loc, instring )
if thiscol > self.col:
raise ParseException( instring, loc, "Text not in expected column", self )
newloc = loc + self.col - thiscol
ret = instring[ loc: newloc ]
return newloc, ret
class LineStart(_PositionToken):
"""
Matches if current position is at the beginning of a line within the parse string
Example::
test = '''\
AAA this line
AAA and this line
AAA but not this one
B AAA and definitely not this one
'''
for t in (LineStart() + 'AAA' + restOfLine).searchString(test):
print(t)
Prints::
['AAA', ' this line']
['AAA', ' and this line']
"""
def __init__( self ):
super(LineStart,self).__init__()
self.errmsg = "Expected start of line"
def parseImpl( self, instring, loc, doActions=True ):
if col(loc, instring) == 1:
return loc, []
raise ParseException(instring, loc, self.errmsg, self)
class LineEnd(_PositionToken):
"""
Matches if current position is at the end of a line within the parse string
"""
def __init__( self ):
super(LineEnd,self).__init__()
self.setWhitespaceChars( ParserElement.DEFAULT_WHITE_CHARS.replace("\n","") )
self.errmsg = "Expected end of line"
def parseImpl( self, instring, loc, doActions=True ):
if loc<len(instring):
if instring[loc] == "\n":
return loc+1, "\n"
else:
raise ParseException(instring, loc, self.errmsg, self)
elif loc == len(instring):
return loc+1, []
else:
raise ParseException(instring, loc, self.errmsg, self)
class StringStart(_PositionToken):
"""
Matches if current position is at the beginning of the parse string
"""
def __init__( self ):
super(StringStart,self).__init__()
self.errmsg = "Expected start of text"
def parseImpl( self, instring, loc, doActions=True ):
if loc != 0:
# see if entire string up to here is just whitespace and ignoreables
if loc != self.preParse( instring, 0 ):
raise ParseException(instring, loc, self.errmsg, self)
return loc, []
class StringEnd(_PositionToken):
"""
Matches if current position is at the end of the parse string
"""
def __init__( self ):
super(StringEnd,self).__init__()
self.errmsg = "Expected end of text"
def parseImpl( self, instring, loc, doActions=True ):
if loc < len(instring):
raise ParseException(instring, loc, self.errmsg, self)
elif loc == len(instring):
return loc+1, []
elif loc > len(instring):
return loc, []
else:
raise ParseException(instring, loc, self.errmsg, self)
class WordStart(_PositionToken):
"""
Matches if the current position is at the beginning of a Word, and
is not preceded by any character in a given set of C{wordChars}
(default=C{printables}). To emulate the C{\b} behavior of regular expressions,
use C{WordStart(alphanums)}. C{WordStart} will also match at the beginning of
the string being parsed, or at the beginning of a line.
"""
def __init__(self, wordChars = printables):
super(WordStart,self).__init__()
self.wordChars = set(wordChars)
self.errmsg = "Not at the start of a word"
def parseImpl(self, instring, loc, doActions=True ):
if loc != 0:
if (instring[loc-1] in self.wordChars or
instring[loc] not in self.wordChars):
raise ParseException(instring, loc, self.errmsg, self)
return loc, []
class WordEnd(_PositionToken):
"""
Matches if the current position is at the end of a Word, and
is not followed by any character in a given set of C{wordChars}
(default=C{printables}). To emulate the C{\b} behavior of regular expressions,
use C{WordEnd(alphanums)}. C{WordEnd} will also match at the end of
the string being parsed, or at the end of a line.
"""
def __init__(self, wordChars = printables):
super(WordEnd,self).__init__()
self.wordChars = set(wordChars)
self.skipWhitespace = False
self.errmsg = "Not at the end of a word"
def parseImpl(self, instring, loc, doActions=True ):
instrlen = len(instring)
if instrlen>0 and loc<instrlen:
if (instring[loc] in self.wordChars or
instring[loc-1] not in self.wordChars):
raise ParseException(instring, loc, self.errmsg, self)
return loc, []
class ParseExpression(ParserElement):
"""
Abstract subclass of ParserElement, for combining and post-processing parsed tokens.
"""
def __init__( self, exprs, savelist = False ):
super(ParseExpression,self).__init__(savelist)
if isinstance( exprs, _generatorType ):
exprs = list(exprs)
if isinstance( exprs, basestring ):
self.exprs = [ ParserElement._literalStringClass( exprs ) ]
elif isinstance( exprs, Iterable ):
exprs = list(exprs)
# if sequence of strings provided, wrap with Literal
if all(isinstance(expr, basestring) for expr in exprs):
exprs = map(ParserElement._literalStringClass, exprs)
self.exprs = list(exprs)
else:
try:
self.exprs = list( exprs )
except TypeError:
self.exprs = [ exprs ]
self.callPreparse = False
def __getitem__( self, i ):
return self.exprs[i]
def append( self, other ):
self.exprs.append( other )
self.strRepr = None
return self
def leaveWhitespace( self ):
"""Extends C{leaveWhitespace} defined in base class, and also invokes C{leaveWhitespace} on
all contained expressions."""
self.skipWhitespace = False
self.exprs = [ e.copy() for e in self.exprs ]
for e in self.exprs:
e.leaveWhitespace()
return self
def ignore( self, other ):
if isinstance( other, Suppress ):
if other not in self.ignoreExprs:
super( ParseExpression, self).ignore( other )
for e in self.exprs:
e.ignore( self.ignoreExprs[-1] )
else:
super( ParseExpression, self).ignore( other )
for e in self.exprs:
e.ignore( self.ignoreExprs[-1] )
return self
def __str__( self ):
try:
return super(ParseExpression,self).__str__()
except Exception:
pass
if self.strRepr is None:
self.strRepr = "%s:(%s)" % ( self.__class__.__name__, _ustr(self.exprs) )
return self.strRepr
def streamline( self ):
super(ParseExpression,self).streamline()
for e in self.exprs:
e.streamline()
# collapse nested And's of the form And( And( And( a,b), c), d) to And( a,b,c,d )
# but only if there are no parse actions or resultsNames on the nested And's
# (likewise for Or's and MatchFirst's)
if ( len(self.exprs) == 2 ):
other = self.exprs[0]
if ( isinstance( other, self.__class__ ) and
not(other.parseAction) and
other.resultsName is None and
not other.debug ):
self.exprs = other.exprs[:] + [ self.exprs[1] ]
self.strRepr = None
self.mayReturnEmpty |= other.mayReturnEmpty
self.mayIndexError |= other.mayIndexError
other = self.exprs[-1]
if ( isinstance( other, self.__class__ ) and
not(other.parseAction) and
other.resultsName is None and
not other.debug ):
self.exprs = self.exprs[:-1] + other.exprs[:]
self.strRepr = None
self.mayReturnEmpty |= other.mayReturnEmpty
self.mayIndexError |= other.mayIndexError
self.errmsg = "Expected " + _ustr(self)
return self
def setResultsName( self, name, listAllMatches=False ):
ret = super(ParseExpression,self).setResultsName(name,listAllMatches)
return ret
def validate( self, validateTrace=[] ):
tmp = validateTrace[:]+[self]
for e in self.exprs:
e.validate(tmp)
self.checkRecursion( [] )
def copy(self):
ret = super(ParseExpression,self).copy()
ret.exprs = [e.copy() for e in self.exprs]
return ret
class And(ParseExpression):
"""
Requires all given C{ParseExpression}s to be found in the given order.
Expressions may be separated by whitespace.
May be constructed using the C{'+'} operator.
May also be constructed using the C{'-'} operator, which will suppress backtracking.
Example::
integer = Word(nums)
name_expr = OneOrMore(Word(alphas))
expr = And([integer("id"),name_expr("name"),integer("age")])
# more easily written as:
expr = integer("id") + name_expr("name") + integer("age")
"""
class _ErrorStop(Empty):
def __init__(self, *args, **kwargs):
super(And._ErrorStop,self).__init__(*args, **kwargs)
self.name = '-'
self.leaveWhitespace()
def __init__( self, exprs, savelist = True ):
super(And,self).__init__(exprs, savelist)
self.mayReturnEmpty = all(e.mayReturnEmpty for e in self.exprs)
self.setWhitespaceChars( self.exprs[0].whiteChars )
self.skipWhitespace = self.exprs[0].skipWhitespace
self.callPreparse = True
def parseImpl( self, instring, loc, doActions=True ):
# pass False as last arg to _parse for first element, since we already
# pre-parsed the string as part of our And pre-parsing
loc, resultlist = self.exprs[0]._parse( instring, loc, doActions, callPreParse=False )
errorStop = False
for e in self.exprs[1:]:
if isinstance(e, And._ErrorStop):
errorStop = True
continue
if errorStop:
try:
loc, exprtokens = e._parse( instring, loc, doActions )
except ParseSyntaxException:
raise
except ParseBaseException as pe:
pe.__traceback__ = None
raise ParseSyntaxException._from_exception(pe)
except IndexError:
raise ParseSyntaxException(instring, len(instring), self.errmsg, self)
else:
loc, exprtokens = e._parse( instring, loc, doActions )
if exprtokens or exprtokens.haskeys():
resultlist += exprtokens
return loc, resultlist
def __iadd__(self, other ):
if isinstance( other, basestring ):
other = ParserElement._literalStringClass( other )
return self.append( other ) #And( [ self, other ] )
def checkRecursion( self, parseElementList ):
subRecCheckList = parseElementList[:] + [ self ]
for e in self.exprs:
e.checkRecursion( subRecCheckList )
if not e.mayReturnEmpty:
break
def __str__( self ):
if hasattr(self,"name"):
return self.name
if self.strRepr is None:
self.strRepr = "{" + " ".join(_ustr(e) for e in self.exprs) + "}"
return self.strRepr
class Or(ParseExpression):
"""
Requires that at least one C{ParseExpression} is found.
If two expressions match, the expression that matches the longest string will be used.
May be constructed using the C{'^'} operator.
Example::
# construct Or using '^' operator
number = Word(nums) ^ Combine(Word(nums) + '.' + Word(nums))
print(number.searchString("123 3.1416 789"))
prints::
[['123'], ['3.1416'], ['789']]
"""
def __init__( self, exprs, savelist = False ):
super(Or,self).__init__(exprs, savelist)
if self.exprs:
self.mayReturnEmpty = any(e.mayReturnEmpty for e in self.exprs)
else:
self.mayReturnEmpty = True
def parseImpl( self, instring, loc, doActions=True ):
maxExcLoc = -1
maxException = None
matches = []
for e in self.exprs:
try:
loc2 = e.tryParse( instring, loc )
except ParseException as err:
err.__traceback__ = None
if err.loc > maxExcLoc:
maxException = err
maxExcLoc = err.loc
except IndexError:
if len(instring) > maxExcLoc:
maxException = ParseException(instring,len(instring),e.errmsg,self)
maxExcLoc = len(instring)
else:
# save match among all matches, to retry longest to shortest
matches.append((loc2, e))
if matches:
matches.sort(key=lambda x: -x[0])
for _,e in matches:
try:
return e._parse( instring, loc, doActions )
except ParseException as err:
err.__traceback__ = None
if err.loc > maxExcLoc:
maxException = err
maxExcLoc = err.loc
if maxException is not None:
maxException.msg = self.errmsg
raise maxException
else:
raise ParseException(instring, loc, "no defined alternatives to match", self)
def __ixor__(self, other ):
if isinstance( other, basestring ):
other = ParserElement._literalStringClass( other )
return self.append( other ) #Or( [ self, other ] )
def __str__( self ):
if hasattr(self,"name"):
return self.name
if self.strRepr is None:
self.strRepr = "{" + " ^ ".join(_ustr(e) for e in self.exprs) + "}"
return self.strRepr
def checkRecursion( self, parseElementList ):
subRecCheckList = parseElementList[:] + [ self ]
for e in self.exprs:
e.checkRecursion( subRecCheckList )
class MatchFirst(ParseExpression):
"""
Requires that at least one C{ParseExpression} is found.
If two expressions match, the first one listed is the one that will match.
May be constructed using the C{'|'} operator.
Example::
# construct MatchFirst using '|' operator
# watch the order of expressions to match
number = Word(nums) | Combine(Word(nums) + '.' + Word(nums))
print(number.searchString("123 3.1416 789")) # Fail! -> [['123'], ['3'], ['1416'], ['789']]
# put more selective expression first
number = Combine(Word(nums) + '.' + Word(nums)) | Word(nums)
print(number.searchString("123 3.1416 789")) # Better -> [['123'], ['3.1416'], ['789']]
"""
def __init__( self, exprs, savelist = False ):
super(MatchFirst,self).__init__(exprs, savelist)
if self.exprs:
self.mayReturnEmpty = any(e.mayReturnEmpty for e in self.exprs)
else:
self.mayReturnEmpty = True
def parseImpl( self, instring, loc, doActions=True ):
maxExcLoc = -1
maxException = None
for e in self.exprs:
try:
ret = e._parse( instring, loc, doActions )
return ret
except ParseException as err:
if err.loc > maxExcLoc:
maxException = err
maxExcLoc = err.loc
except IndexError:
if len(instring) > maxExcLoc:
maxException = ParseException(instring,len(instring),e.errmsg,self)
maxExcLoc = len(instring)
# only got here if no expression matched, raise exception for match that made it the furthest
else:
if maxException is not None:
maxException.msg = self.errmsg
raise maxException
else:
raise ParseException(instring, loc, "no defined alternatives to match", self)
def __ior__(self, other ):
if isinstance( other, basestring ):
other = ParserElement._literalStringClass( other )
return self.append( other ) #MatchFirst( [ self, other ] )
def __str__( self ):
if hasattr(self,"name"):
return self.name
if self.strRepr is None:
self.strRepr = "{" + " | ".join(_ustr(e) for e in self.exprs) + "}"
return self.strRepr
def checkRecursion( self, parseElementList ):
subRecCheckList = parseElementList[:] + [ self ]
for e in self.exprs:
e.checkRecursion( subRecCheckList )
class Each(ParseExpression):
"""
Requires all given C{ParseExpression}s to be found, but in any order.
Expressions may be separated by whitespace.
May be constructed using the C{'&'} operator.
Example::
color = oneOf("RED ORANGE YELLOW GREEN BLUE PURPLE BLACK WHITE BROWN")
shape_type = oneOf("SQUARE CIRCLE TRIANGLE STAR HEXAGON OCTAGON")
integer = Word(nums)
shape_attr = "shape:" + shape_type("shape")
posn_attr = "posn:" + Group(integer("x") + ',' + integer("y"))("posn")
color_attr = "color:" + color("color")
size_attr = "size:" + integer("size")
# use Each (using operator '&') to accept attributes in any order
# (shape and posn are required, color and size are optional)
shape_spec = shape_attr & posn_attr & Optional(color_attr) & Optional(size_attr)
shape_spec.runTests('''
shape: SQUARE color: BLACK posn: 100, 120
shape: CIRCLE size: 50 color: BLUE posn: 50,80
color:GREEN size:20 shape:TRIANGLE posn:20,40
'''
)
prints::
shape: SQUARE color: BLACK posn: 100, 120
['shape:', 'SQUARE', 'color:', 'BLACK', 'posn:', ['100', ',', '120']]
- color: BLACK
- posn: ['100', ',', '120']
- x: 100
- y: 120
- shape: SQUARE
shape: CIRCLE size: 50 color: BLUE posn: 50,80
['shape:', 'CIRCLE', 'size:', '50', 'color:', 'BLUE', 'posn:', ['50', ',', '80']]
- color: BLUE
- posn: ['50', ',', '80']
- x: 50
- y: 80
- shape: CIRCLE
- size: 50
color: GREEN size: 20 shape: TRIANGLE posn: 20,40
['color:', 'GREEN', 'size:', '20', 'shape:', 'TRIANGLE', 'posn:', ['20', ',', '40']]
- color: GREEN
- posn: ['20', ',', '40']
- x: 20
- y: 40
- shape: TRIANGLE
- size: 20
"""
def __init__( self, exprs, savelist = True ):
super(Each,self).__init__(exprs, savelist)
self.mayReturnEmpty = all(e.mayReturnEmpty for e in self.exprs)
self.skipWhitespace = True
self.initExprGroups = True
def parseImpl( self, instring, loc, doActions=True ):
if self.initExprGroups:
self.opt1map = dict((id(e.expr),e) for e in self.exprs if isinstance(e,Optional))
opt1 = [ e.expr for e in self.exprs if isinstance(e,Optional) ]
opt2 = [ e for e in self.exprs if e.mayReturnEmpty and not isinstance(e,Optional)]
self.optionals = opt1 + opt2
self.multioptionals = [ e.expr for e in self.exprs if isinstance(e,ZeroOrMore) ]
self.multirequired = [ e.expr for e in self.exprs if isinstance(e,OneOrMore) ]
self.required = [ e for e in self.exprs if not isinstance(e,(Optional,ZeroOrMore,OneOrMore)) ]
self.required += self.multirequired
self.initExprGroups = False
tmpLoc = loc
tmpReqd = self.required[:]
tmpOpt = self.optionals[:]
matchOrder = []
keepMatching = True
while keepMatching:
tmpExprs = tmpReqd + tmpOpt + self.multioptionals + self.multirequired
failed = []
for e in tmpExprs:
try:
tmpLoc = e.tryParse( instring, tmpLoc )
except ParseException:
failed.append(e)
else:
matchOrder.append(self.opt1map.get(id(e),e))
if e in tmpReqd:
tmpReqd.remove(e)
elif e in tmpOpt:
tmpOpt.remove(e)
if len(failed) == len(tmpExprs):
keepMatching = False
if tmpReqd:
missing = ", ".join(_ustr(e) for e in tmpReqd)
raise ParseException(instring,loc,"Missing one or more required elements (%s)" % missing )
# add any unmatched Optionals, in case they have default values defined
matchOrder += [e for e in self.exprs if isinstance(e,Optional) and e.expr in tmpOpt]
resultlist = []
for e in matchOrder:
loc,results = e._parse(instring,loc,doActions)
resultlist.append(results)
finalResults = sum(resultlist, ParseResults([]))
return loc, finalResults
def __str__( self ):
if hasattr(self,"name"):
return self.name
if self.strRepr is None:
self.strRepr = "{" + " & ".join(_ustr(e) for e in self.exprs) + "}"
return self.strRepr
def checkRecursion( self, parseElementList ):
subRecCheckList = parseElementList[:] + [ self ]
for e in self.exprs:
e.checkRecursion( subRecCheckList )
class ParseElementEnhance(ParserElement):
"""
Abstract subclass of C{ParserElement}, for combining and post-processing parsed tokens.
"""
def __init__( self, expr, savelist=False ):
super(ParseElementEnhance,self).__init__(savelist)
if isinstance( expr, basestring ):
if issubclass(ParserElement._literalStringClass, Token):
expr = ParserElement._literalStringClass(expr)
else:
expr = ParserElement._literalStringClass(Literal(expr))
self.expr = expr
self.strRepr = None
if expr is not None:
self.mayIndexError = expr.mayIndexError
self.mayReturnEmpty = expr.mayReturnEmpty
self.setWhitespaceChars( expr.whiteChars )
self.skipWhitespace = expr.skipWhitespace
self.saveAsList = expr.saveAsList
self.callPreparse = expr.callPreparse
self.ignoreExprs.extend(expr.ignoreExprs)
def parseImpl( self, instring, loc, doActions=True ):
if self.expr is not None:
return self.expr._parse( instring, loc, doActions, callPreParse=False )
else:
raise ParseException("",loc,self.errmsg,self)
def leaveWhitespace( self ):
self.skipWhitespace = False
self.expr = self.expr.copy()
if self.expr is not None:
self.expr.leaveWhitespace()
return self
def ignore( self, other ):
if isinstance( other, Suppress ):
if other not in self.ignoreExprs:
super( ParseElementEnhance, self).ignore( other )
if self.expr is not None:
self.expr.ignore( self.ignoreExprs[-1] )
else:
super( ParseElementEnhance, self).ignore( other )
if self.expr is not None:
self.expr.ignore( self.ignoreExprs[-1] )
return self
def streamline( self ):
super(ParseElementEnhance,self).streamline()
if self.expr is not None:
self.expr.streamline()
return self
def checkRecursion( self, parseElementList ):
if self in parseElementList:
raise RecursiveGrammarException( parseElementList+[self] )
subRecCheckList = parseElementList[:] + [ self ]
if self.expr is not None:
self.expr.checkRecursion( subRecCheckList )
def validate( self, validateTrace=[] ):
tmp = validateTrace[:]+[self]
if self.expr is not None:
self.expr.validate(tmp)
self.checkRecursion( [] )
def __str__( self ):
try:
return super(ParseElementEnhance,self).__str__()
except Exception:
pass
if self.strRepr is None and self.expr is not None:
self.strRepr = "%s:(%s)" % ( self.__class__.__name__, _ustr(self.expr) )
return self.strRepr
class FollowedBy(ParseElementEnhance):
"""
Lookahead matching of the given parse expression. C{FollowedBy}
does I{not} advance the parsing position within the input string, it only
verifies that the specified parse expression matches at the current
position. C{FollowedBy} always returns a null token list.
Example::
# use FollowedBy to match a label only if it is followed by a ':'
data_word = Word(alphas)
label = data_word + FollowedBy(':')
attr_expr = Group(label + Suppress(':') + OneOrMore(data_word, stopOn=label).setParseAction(' '.join))
OneOrMore(attr_expr).parseString("shape: SQUARE color: BLACK posn: upper left").pprint()
prints::
[['shape', 'SQUARE'], ['color', 'BLACK'], ['posn', 'upper left']]
"""
def __init__( self, expr ):
super(FollowedBy,self).__init__(expr)
self.mayReturnEmpty = True
def parseImpl( self, instring, loc, doActions=True ):
self.expr.tryParse( instring, loc )
return loc, []
class NotAny(ParseElementEnhance):
"""
Lookahead to disallow matching with the given parse expression. C{NotAny}
does I{not} advance the parsing position within the input string, it only
verifies that the specified parse expression does I{not} match at the current
position. Also, C{NotAny} does I{not} skip over leading whitespace. C{NotAny}
always returns a null token list. May be constructed using the '~' operator.
Example::
"""
def __init__( self, expr ):
super(NotAny,self).__init__(expr)
#~ self.leaveWhitespace()
self.skipWhitespace = False # do NOT use self.leaveWhitespace(), don't want to propagate to exprs
self.mayReturnEmpty = True
self.errmsg = "Found unwanted token, "+_ustr(self.expr)
def parseImpl( self, instring, loc, doActions=True ):
if self.expr.canParseNext(instring, loc):
raise ParseException(instring, loc, self.errmsg, self)
return loc, []
def __str__( self ):
if hasattr(self,"name"):
return self.name
if self.strRepr is None:
self.strRepr = "~{" + _ustr(self.expr) + "}"
return self.strRepr
class _MultipleMatch(ParseElementEnhance):
def __init__( self, expr, stopOn=None):
super(_MultipleMatch, self).__init__(expr)
self.saveAsList = True
ender = stopOn
if isinstance(ender, basestring):
ender = ParserElement._literalStringClass(ender)
self.not_ender = ~ender if ender is not None else None
def parseImpl( self, instring, loc, doActions=True ):
self_expr_parse = self.expr._parse
self_skip_ignorables = self._skipIgnorables
check_ender = self.not_ender is not None
if check_ender:
try_not_ender = self.not_ender.tryParse
# must be at least one (but first see if we are the stopOn sentinel;
# if so, fail)
if check_ender:
try_not_ender(instring, loc)
loc, tokens = self_expr_parse( instring, loc, doActions, callPreParse=False )
try:
hasIgnoreExprs = (not not self.ignoreExprs)
while 1:
if check_ender:
try_not_ender(instring, loc)
if hasIgnoreExprs:
preloc = self_skip_ignorables( instring, loc )
else:
preloc = loc
loc, tmptokens = self_expr_parse( instring, preloc, doActions )
if tmptokens or tmptokens.haskeys():
tokens += tmptokens
except (ParseException,IndexError):
pass
return loc, tokens
class OneOrMore(_MultipleMatch):
"""
Repetition of one or more of the given expression.
Parameters:
- expr - expression that must match one or more times
- stopOn - (default=C{None}) - expression for a terminating sentinel
(only required if the sentinel would ordinarily match the repetition
expression)
Example::
data_word = Word(alphas)
label = data_word + FollowedBy(':')
attr_expr = Group(label + Suppress(':') + OneOrMore(data_word).setParseAction(' '.join))
text = "shape: SQUARE posn: upper left color: BLACK"
OneOrMore(attr_expr).parseString(text).pprint() # Fail! read 'color' as data instead of next label -> [['shape', 'SQUARE color']]
# use stopOn attribute for OneOrMore to avoid reading label string as part of the data
attr_expr = Group(label + Suppress(':') + OneOrMore(data_word, stopOn=label).setParseAction(' '.join))
OneOrMore(attr_expr).parseString(text).pprint() # Better -> [['shape', 'SQUARE'], ['posn', 'upper left'], ['color', 'BLACK']]
# could also be written as
(attr_expr * (1,)).parseString(text).pprint()
"""
def __str__( self ):
if hasattr(self,"name"):
return self.name
if self.strRepr is None:
self.strRepr = "{" + _ustr(self.expr) + "}..."
return self.strRepr
class ZeroOrMore(_MultipleMatch):
"""
Optional repetition of zero or more of the given expression.
Parameters:
- expr - expression that must match zero or more times
- stopOn - (default=C{None}) - expression for a terminating sentinel
(only required if the sentinel would ordinarily match the repetition
expression)
Example: similar to L{OneOrMore}
"""
def __init__( self, expr, stopOn=None):
super(ZeroOrMore,self).__init__(expr, stopOn=stopOn)
self.mayReturnEmpty = True
def parseImpl( self, instring, loc, doActions=True ):
try:
return super(ZeroOrMore, self).parseImpl(instring, loc, doActions)
except (ParseException,IndexError):
return loc, []
def __str__( self ):
if hasattr(self,"name"):
return self.name
if self.strRepr is None:
self.strRepr = "[" + _ustr(self.expr) + "]..."
return self.strRepr
class _NullToken(object):
def __bool__(self):
return False
__nonzero__ = __bool__
def __str__(self):
return ""
_optionalNotMatched = _NullToken()
class Optional(ParseElementEnhance):
"""
Optional matching of the given expression.
Parameters:
- expr - expression that must match zero or more times
- default (optional) - value to be returned if the optional expression is not found.
Example::
# US postal code can be a 5-digit zip, plus optional 4-digit qualifier
zip = Combine(Word(nums, exact=5) + Optional('-' + Word(nums, exact=4)))
zip.runTests('''
# traditional ZIP code
12345
# ZIP+4 form
12101-0001
# invalid ZIP
98765-
''')
prints::
# traditional ZIP code
12345
['12345']
# ZIP+4 form
12101-0001
['12101-0001']
# invalid ZIP
98765-
^
FAIL: Expected end of text (at char 5), (line:1, col:6)
"""
def __init__( self, expr, default=_optionalNotMatched ):
super(Optional,self).__init__( expr, savelist=False )
self.saveAsList = self.expr.saveAsList
self.defaultValue = default
self.mayReturnEmpty = True
def parseImpl( self, instring, loc, doActions=True ):
try:
loc, tokens = self.expr._parse( instring, loc, doActions, callPreParse=False )
except (ParseException,IndexError):
if self.defaultValue is not _optionalNotMatched:
if self.expr.resultsName:
tokens = ParseResults([ self.defaultValue ])
tokens[self.expr.resultsName] = self.defaultValue
else:
tokens = [ self.defaultValue ]
else:
tokens = []
return loc, tokens
def __str__( self ):
if hasattr(self,"name"):
return self.name
if self.strRepr is None:
self.strRepr = "[" + _ustr(self.expr) + "]"
return self.strRepr
class SkipTo(ParseElementEnhance):
"""
Token for skipping over all undefined text until the matched expression is found.
Parameters:
- expr - target expression marking the end of the data to be skipped
- include - (default=C{False}) if True, the target expression is also parsed
(the skipped text and target expression are returned as a 2-element list).
- ignore - (default=C{None}) used to define grammars (typically quoted strings and
comments) that might contain false matches to the target expression
- failOn - (default=C{None}) define expressions that are not allowed to be
included in the skipped test; if found before the target expression is found,
the SkipTo is not a match
Example::
report = '''
Outstanding Issues Report - 1 Jan 2000
# | Severity | Description | Days Open
-----+----------+-------------------------------------------+-----------
101 | Critical | Intermittent system crash | 6
94 | Cosmetic | Spelling error on Login ('log|n') | 14
79 | Minor | System slow when running too many reports | 47
'''
integer = Word(nums)
SEP = Suppress('|')
# use SkipTo to simply match everything up until the next SEP
# - ignore quoted strings, so that a '|' character inside a quoted string does not match
# - parse action will call token.strip() for each matched token, i.e., the description body
string_data = SkipTo(SEP, ignore=quotedString)
string_data.setParseAction(tokenMap(str.strip))
ticket_expr = (integer("issue_num") + SEP
+ string_data("sev") + SEP
+ string_data("desc") + SEP
+ integer("days_open"))
for tkt in ticket_expr.searchString(report):
print tkt.dump()
prints::
['101', 'Critical', 'Intermittent system crash', '6']
- days_open: 6
- desc: Intermittent system crash
- issue_num: 101
- sev: Critical
['94', 'Cosmetic', "Spelling error on Login ('log|n')", '14']
- days_open: 14
- desc: Spelling error on Login ('log|n')
- issue_num: 94
- sev: Cosmetic
['79', 'Minor', 'System slow when running too many reports', '47']
- days_open: 47
- desc: System slow when running too many reports
- issue_num: 79
- sev: Minor
"""
def __init__( self, other, include=False, ignore=None, failOn=None ):
super( SkipTo, self ).__init__( other )
self.ignoreExpr = ignore
self.mayReturnEmpty = True </