Module pyparsing

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 ParserElement.parseString for more information on parsing strings containing <TAB>s, and suggested methods to maintain a consistent view of the parsed string, the parse location, and line and column positions within the parsed string.

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 ParserElement.parseString for more information on parsing strings containing <TAB>s, and suggested methods to maintain a consistent view of the parsed string, the parse location, and line and column positions within the parsed string.

Decorator for debugging parse actions.

When the parse action is called, this decorator will print ">> entering method-name(line:current_source_line, parse_location, matched_tokens)". When the parse action completes, the decorator will print "<<" followed by the returned value, or any exception that the parse action raised.

Example:

   wd = Word(alphas)

   @traceParseAction
   def remove_duplicate_chars(tokens):
       return ''.join(sorted(set(''.join(tokens)))

   wds = OneOrMore(wd).setParseAction(remove_duplicate_chars)
   print(wds.parseString("slkdjs sld sldd sdlf sdljf"))

prints:

   >>entering remove_duplicate_chars(line: 'slkdjs sld sldd sdlf sdljf', 0, (['slkdjs', 'sld', 'sldd', 'sdlf', 'sdljf'], {}))
   <<leaving remove_duplicate_chars (ret: 'dfjkls')
   ['dfjkls']

Helper to define a delimited list of expressions - the delimiter defaults to ','. By default, the list elements and delimiters can have intervening whitespace, and comments, but this can be overridden by passing combine=True in the constructor. If combine is set to True, the matching tokens are returned as a single token string, with the delimiters included; otherwise, the matching tokens are returned as a list of tokens, with the delimiters suppressed.

Example:

   delimitedList(Word(alphas)).parseString("aa,bb,cc") # -> ['aa', 'bb', 'cc']
   delimitedList(Word(hexnums), delim=':', combine=True).parseString("AA:BB:CC:DD:EE") # -> ['AA:BB:CC:DD:EE']

Helper to define a counted list of expressions. This helper defines a pattern of the form:

   integer expr expr expr...

where the leading integer tells how many expr expressions follow. The matched tokens returns the array of expr tokens as a list - the leading count token is suppressed.

If intExpr is specified, it should be a pyparsing expression that produces an integer value.

Example:

   countedArray(Word(alphas)).parseString('2 ab cd ef')  # -> ['ab', 'cd']

   # in this parser, the leading integer value is given in binary,
   # '10' indicating that 2 values are in the array
   binaryConstant = Word('01').setParseAction(lambda t: int(t[0], 2))
   countedArray(Word(alphas), intExpr=binaryConstant).parseString('10 ab cd ef')  # -> ['ab', 'cd']

Helper to define an expression that is indirectly defined from the tokens matched in a previous expression, that is, it looks for a 'repeat' of a previous expression. For example:

   first = Word(nums)
   second = matchPreviousLiteral(first)
   matchExpr = first + ":" + second

will match "1:1", but not "1:2". Because this matches a previous literal, will also match the leading "1:1" in "1:10". If this is not desired, use matchPreviousExpr. Do not use with packrat parsing enabled.

Helper to define an expression that is indirectly defined from the tokens matched in a previous expression, that is, it looks for a 'repeat' of a previous expression. For example:

   first = Word(nums)
   second = matchPreviousExpr(first)
   matchExpr = first + ":" + second

will match "1:1", but not "1:2". Because this matches by expressions, will not match the leading "1:1" in "1:10"; the expressions are evaluated first, and then compared, so "1" is compared with "10". Do not use with packrat parsing enabled.

Helper to quickly define a set of alternative Literals, and makes sure to do longest-first testing when there is a conflict, regardless of the input order, but returns a MatchFirst for best performance.

Parameters:

• strs - a string of space-delimited literals, or a collection of string literals
• caseless - (default=False) - treat all literals as caseless
• useRegex - (default=True) - as an optimization, will generate a Regex object; otherwise, will generate a MatchFirst object (if caseless=True, or if creating a Regex raises an exception)

Example:

   comp_oper = oneOf("< = > <= >= !=")
   var = Word(alphas)
   number = Word(nums)
   term = var | number
   comparison_expr = term + comp_oper + term
   print(comparison_expr.searchString("B = 12  AA=23 B<=AA AA>12"))

prints:

   [['B', '=', '12'], ['AA', '=', '23'], ['B', '<=', 'AA'], ['AA', '>', '12']]

Helper to easily and clearly define a dictionary by specifying the respective patterns for the key and value. Takes care of defining the Dict, ZeroOrMore, and Group tokens in the proper order. The key pattern can include delimiting markers or punctuation, as long as they are suppressed, thereby leaving the significant key text. The value pattern can include named results, so that the Dict results can include named token fields.

Example:

   text = "shape: SQUARE posn: upper left color: light blue texture: burlap"
   attr_expr = (label + Suppress(':') + OneOrMore(data_word, stopOn=label).setParseAction(' '.join))
   print(OneOrMore(attr_expr).parseString(text).dump())
   
   attr_label = label
   attr_value = Suppress(':') + OneOrMore(data_word, stopOn=label).setParseAction(' '.join)

   # similar to Dict, but simpler call format
   result = dictOf(attr_label, attr_value).parseString(text)
   print(result.dump())
   print(result['shape'])
   print(result.shape)  # object attribute access works too
   print(result.asDict())

prints:

   [['shape', 'SQUARE'], ['posn', 'upper left'], ['color', 'light blue'], ['texture', 'burlap']]
   - color: light blue
   - posn: upper left
   - shape: SQUARE
   - texture: burlap
   SQUARE
   SQUARE
   {'color': 'light blue', 'shape': 'SQUARE', 'posn': 'upper left', 'texture': 'burlap'}

Helper to return the original, untokenized text for a given expression. Useful to restore the parsed fields of an HTML start tag into the raw tag text itself, or to revert separate tokens with intervening whitespace back to the original matching input text. By default, returns astring containing the original parsed text.

If the optional asString argument is passed as False, then the return value is a ParseResults containing any results names that were originally matched, and a single token containing the original matched text from the input string. So if the expression passed to originalTextFor contains expressions with defined results names, you must set asString to False if you want to preserve those results name values.

Example:

   src = "this is test <b> bold <i>text</i> </b> normal text "
   for tag in ("b","i"):
       opener,closer = makeHTMLTags(tag)
       patt = originalTextFor(opener + SkipTo(closer) + closer)
       print(patt.searchString(src)[0])

prints:

   ['<b> bold <i>text</i> </b>']
   ['<i>text</i>']

Helper to decorate a returned token with its starting and ending locations in the input string. This helper adds the following results names:

• locn_start = location where matched expression begins
• locn_end = location where matched expression ends
• value = the actual parsed results

Be careful if the input text contains <TAB> characters, you may want to call ParserElement.parseWithTabs

Example:

   wd = Word(alphas)
   for match in locatedExpr(wd).searchString("ljsdf123lksdjjf123lkkjj1222"):
       print(match)

prints:

   [[0, 'ljsdf', 5]]
   [[8, 'lksdjjf', 15]]
   [[18, 'lkkjj', 23]]

Helper to easily define string ranges for use in Word construction. Borrows syntax from regexp '[]' string range definitions:

   srange("[0-9]")   -> "0123456789"
   srange("[a-z]")   -> "abcdefghijklmnopqrstuvwxyz"
   srange("[a-z$_]") -> "abcdefghijklmnopqrstuvwxyz$_"

The input string must be enclosed in []'s, and the returned string is the expanded character set joined into a single string. The values enclosed in the []'s may be:

• a single character
• an escaped character with a leading backslash (such as \- or \])
• an escaped hex character with a leading '\x' (\x21, which is a '!' character) (\0x## is also supported for backwards compatibility)
• an escaped octal character with a leading '\0' (\041, which is a '!' character)
• a range of any of the above, separated by a dash ('a-z', etc.)
• any combination of the above ('aeiouy', 'a-zA-Z0-9_$', etc.)

Helper method for common parse actions that simply return a literal value. Especially useful when used with transformString().

Example:

   num = Word(nums).setParseAction(lambda toks: int(toks[0]))
   na = oneOf("N/A NA").setParseAction(replaceWith(math.nan))
   term = na | num
   
   OneOrMore(term).parseString("324 234 N/A 234") # -> [324, 234, nan, 234]

Helper parse action for removing quotation marks from parsed quoted strings.

Example:

   # by default, quotation marks are included in parsed results
   quotedString.parseString("'Now is the Winter of our Discontent'") # -> ["'Now is the Winter of our Discontent'"]

   # use removeQuotes to strip quotation marks from parsed results
   quotedString.setParseAction(removeQuotes)
   quotedString.parseString("'Now is the Winter of our Discontent'") # -> ["Now is the Winter of our Discontent"]

Helper to define a parse action by mapping a function to all elements of a ParseResults list.If any additional args are passed, they are forwarded to the given function as additional arguments after the token, as in hex_integer = Word(hexnums).setParseAction(tokenMap(int, 16)), which will convert the parsed data to an integer using base 16.

Example (compare the last to example in ParserElement.transformString:

   hex_ints = OneOrMore(Word(hexnums)).setParseAction(tokenMap(int, 16))
   hex_ints.runTests('''
       00 11 22 aa FF 0a 0d 1a
       ''')
   
   upperword = Word(alphas).setParseAction(tokenMap(str.upper))
   OneOrMore(upperword).runTests('''
       my kingdom for a horse
       ''')

   wd = Word(alphas).setParseAction(tokenMap(str.title))
   OneOrMore(wd).setParseAction(' '.join).runTests('''
       now is the winter of our discontent made glorious summer by this sun of york
       ''')

prints:

   00 11 22 aa FF 0a 0d 1a
   [0, 17, 34, 170, 255, 10, 13, 26]

   my kingdom for a horse
   ['MY', 'KINGDOM', 'FOR', 'A', 'HORSE']

   now is the winter of our discontent made glorious summer by this sun of york
   ['Now Is The Winter Of Our Discontent Made Glorious Summer By This Sun Of York']

(Deprecated) Helper parse action to convert tokens to upper case. Deprecated in favor of pyparsing_common.upcaseTokens

(Deprecated) Helper parse action to convert tokens to lower case. Deprecated in favor of pyparsing_common.downcaseTokens

Helper to construct opening and closing tag expressions for HTML, given a tag name. Matches tags in either upper or lower case, attributes with namespaces and with quoted or unquoted values.

Example:

   text = '<td>More info at the <a href="http://pyparsing.wikispaces.com">pyparsing</a> wiki page</td>'
   # makeHTMLTags returns pyparsing expressions for the opening and closing tags as a 2-tuple
   a,a_end = makeHTMLTags("A")
   link_expr = a + SkipTo(a_end)("link_text") + a_end
   
   for link in link_expr.searchString(text):
       # attributes in the <A> tag (like "href" shown here) are also accessible as named results
       print(link.link_text, '->', link.href)

prints:

   pyparsing -> http://pyparsing.wikispaces.com

Helper to construct opening and closing tag expressions for XML, given a tag name. Matches tags only in the given upper/lower case.

Example: similar to makeHTMLTags

Helper to create a validating parse action to be used with start tags created with makeXMLTags or makeHTMLTags. Use withAttribute to qualify a starting tag with a required attribute value, to avoid false matches on common tags such as <TD> or <DIV>.

Call withAttribute with a series of attribute names and values. Specify the list of filter attributes names and values as:

• keyword arguments, as in (align="right"), or
• as an explicit dict with ** operator, when an attribute name is also a Python reserved word, as in **{"class":"Customer", "align":"right"}
• a list of name-value tuples, as in ( ("ns1:class", "Customer"), ("ns2:align","right") )

For attribute names with a namespace prefix, you must use the second form. Attribute names are matched insensitive to upper/lower case.

If just testing for class (with or without a namespace), use withClass.

To verify that the attribute exists, but without specifying a value, pass withAttribute.ANY_VALUE as the value.

Example:

   html = '''
       <div>
       Some text
       <div type="grid">1 4 0 1 0</div>
       <div type="graph">1,3 2,3 1,1</div>
       <div>this has no type</div>
       </div>
           
   '''
   div,div_end = makeHTMLTags("div")

   # only match div tag having a type attribute with value "grid"
   div_grid = div().setParseAction(withAttribute(type="grid"))
   grid_expr = div_grid + SkipTo(div | div_end)("body")
   for grid_header in grid_expr.searchString(html):
       print(grid_header.body)
   
   # construct a match with any div tag having a type attribute, regardless of the value
   div_any_type = div().setParseAction(withAttribute(type=withAttribute.ANY_VALUE))
   div_expr = div_any_type + SkipTo(div | div_end)("body")
   for div_header in div_expr.searchString(html):
       print(div_header.body)

prints:

   1 4 0 1 0

   1 4 0 1 0
   1,3 2,3 1,1

Simplified version of withAttribute when matching on a div class - made difficult because class is a reserved word in Python.

Example:

   html = '''
       <div>
       Some text
       <div class="grid">1 4 0 1 0</div>
       <div class="graph">1,3 2,3 1,1</div>
       <div>this &lt;div&gt; has no class</div>
       </div>
           
   '''
   div,div_end = makeHTMLTags("div")
   div_grid = div().setParseAction(withClass("grid"))
   
   grid_expr = div_grid + SkipTo(div | div_end)("body")
   for grid_header in grid_expr.searchString(html):
       print(grid_header.body)
   
   div_any_type = div().setParseAction(withClass(withAttribute.ANY_VALUE))
   div_expr = div_any_type + SkipTo(div | div_end)("body")
   for div_header in div_expr.searchString(html):
       print(div_header.body)

prints:

   1 4 0 1 0

   1 4 0 1 0
   1,3 2,3 1,1

Helper method for constructing grammars of expressions made up of operators working in a precedence hierarchy. Operators may be unary or binary, left- or right-associative. Parse actions can also be attached to operator expressions. The generated parser will also recognize the use of parentheses to override operator precedences (see example below).

Note: if you define a deep operator list, you may see performance issues when using infixNotation. See ParserElement.enablePackrat for a mechanism to potentially improve your parser performance.

Parameters:

• baseExpr - expression representing the most basic element for the nested
• opList - list of tuples, one for each operator precedence level in the expression grammar; each tuple is of the form (opExpr, numTerms, rightLeftAssoc, parseAction), where:
  • opExpr is the pyparsing expression for the operator; may also be a string, which will be converted to a Literal; if numTerms is 3, opExpr is a tuple of two expressions, for the two operators separating the 3 terms
  • numTerms is the number of terms for this operator (must be 1, 2, or 3)
  • rightLeftAssoc is the indicator whether the operator is right or left associative, using the pyparsing-defined constants opAssoc.RIGHT and opAssoc.LEFT.
  • parseAction is the parse action to be associated with expressions matching this operator expression (the parse action tuple member may be omitted); if the parse action is passed a tuple or list of functions, this is equivalent to calling setParseAction(*fn) (ParserElement.setParseAction)
• lpar - expression for matching left-parentheses (default=Suppress('('))
• rpar - expression for matching right-parentheses (default=Suppress(')'))

Example:

   # simple example of four-function arithmetic with ints and variable names
   integer = pyparsing_common.signed_integer
   varname = pyparsing_common.identifier 
   
   arith_expr = infixNotation(integer | varname,
       [
       ('-', 1, opAssoc.RIGHT),
       (oneOf('* /'), 2, opAssoc.LEFT),
       (oneOf('+ -'), 2, opAssoc.LEFT),
       ])
   
   arith_expr.runTests('''
       5+3*6
       (5+3)*6
       -2--11
       ''', fullDump=False)

prints:

   5+3*6
   [[5, '+', [3, '*', 6]]]

   (5+3)*6
   [[[5, '+', 3], '*', 6]]

   -2--11
   [[['-', 2], '-', ['-', 11]]]

(Deprecated) Former name of infixNotation, will be dropped in a future release.

Helper method for defining nested lists enclosed in opening and closing delimiters ("(" and ")" are the default).

Parameters:

• opener - opening character for a nested list (default="("); can also be a pyparsing expression
• closer - closing character for a nested list (default=")"); can also be a pyparsing expression
• content - expression for items within the nested lists (default=None)
• ignoreExpr - expression for ignoring opening and closing delimiters (default=quotedString)

If an expression is not provided for the content argument, the nested expression will capture all whitespace-delimited content between delimiters as a list of separate values.

Use the ignoreExpr argument to define expressions that may contain opening or closing characters that should not be treated as opening or closing characters for nesting, such as quotedString or a comment expression. Specify multiple expressions using an Or or MatchFirst. The default is quotedString, but if no expressions are to be ignored, then pass None for this argument.

Example:

   data_type = oneOf("void int short long char float double")
   decl_data_type = Combine(data_type + Optional(Word('*')))
   ident = Word(alphas+'_', alphanums+'_')
   number = pyparsing_common.number
   arg = Group(decl_data_type + ident)
   LPAR,RPAR = map(Suppress, "()")

   code_body = nestedExpr('{', '}', ignoreExpr=(quotedString | cStyleComment))

   c_function = (decl_data_type("type") 
                 + ident("name")
                 + LPAR + Optional(delimitedList(arg), [])("args") + RPAR 
                 + code_body("body"))
   c_function.ignore(cStyleComment)
   
   source_code = '''
       int is_odd(int x) { 
           return (x%2); 
       }
           
       int dec_to_hex(char hchar) { 
           if (hchar >= '0' && hchar <= '9') { 
               return (ord(hchar)-ord('0')); 
           } else { 
               return (10+ord(hchar)-ord('A'));
           } 
       }
   '''
   for func in c_function.searchString(source_code):
       print("%(name)s (%(type)s) args: %(args)s" % func)

prints:

   is_odd (int) args: [['int', 'x']]
   dec_to_hex (int) args: [['char', 'hchar']]

Helper method for defining space-delimited indentation blocks, such as those used to define block statements in Python source code.

Parameters:

• blockStatementExpr - expression defining syntax of statement that is repeated within the indented block
• indentStack - list created by caller to manage indentation stack (multiple statementWithIndentedBlock expressions within a single grammar should share a common indentStack)
• indent - boolean indicating whether block must be indented beyond the the current level; set to False for block of left-most statements (default=True)

A valid block must contain at least one blockStatement.

Example:

   data = '''
   def A(z):
     A1
     B = 100
     G = A2
     A2
     A3
   B
   def BB(a,b,c):
     BB1
     def BBA():
       bba1
       bba2
       bba3
   C
   D
   def spam(x,y):
        def eggs(z):
            pass
   '''


   indentStack = [1]
   stmt = Forward()

   identifier = Word(alphas, alphanums)
   funcDecl = ("def" + identifier + Group( "(" + Optional( delimitedList(identifier) ) + ")" ) + ":")
   func_body = indentedBlock(stmt, indentStack)
   funcDef = Group( funcDecl + func_body )

   rvalue = Forward()
   funcCall = Group(identifier + "(" + Optional(delimitedList(rvalue)) + ")")
   rvalue << (funcCall | identifier | Word(nums))
   assignment = Group(identifier + "=" + rvalue)
   stmt << ( funcDef | assignment | identifier )

   module_body = OneOrMore(stmt)

   parseTree = module_body.parseString(data)
   parseTree.pprint()

prints:

   [['def',
     'A',
     ['(', 'z', ')'],
     ':',
     [['A1'], [['B', '=', '100']], [['G', '=', 'A2']], ['A2'], ['A3']]],
    'B',
    ['def',
     'BB',
     ['(', 'a', 'b', 'c', ')'],
     ':',
     [['BB1'], [['def', 'BBA', ['(', ')'], ':', [['bba1'], ['bba2'], ['bba3']]]]]],
    'C',
    'D',
    ['def',
     'spam',
     ['(', 'x', 'y', ')'],
     ':',
     [[['def', 'eggs', ['(', 'z', ')'], ':', [['pass']]]]]]] 

alphanums

Value:

'ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789'

printables

Value:

'0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ!"#$%&\ '()*+,-./:;<=>?@[\\]^_`{|}~'

alphas8bit

Value:

u'ÀÁÂÃÄÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖØÙÚÛÜÝÞßàáâãäåæçèéêëìíîïðñòóôõöøùúûüýþÿ'

commaSeparatedList

(Deprecated) Predefined expression of 1 or more printable words or quoted strings, separated by commas. This expression is deprecated in favor of pyparsing_common.comma_separated_list.

Value:

commaSeparatedList