Package pyparsing :: Module pyparsing
[frames] | no frames]

Module pyparsing

source code

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 "<salutation>, <addressee>!"):

   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 parsed results returned from 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:


Version: 2.0.2

Author: Paul McGuire <ptmcg@users.sourceforge.net>

Classes
  ParseBaseException
base exception class for all parsing runtime exceptions
  ParseException
exception thrown when parse expressions don't match class; supported attributes by name are:
  ParseFatalException
user-throwable exception thrown when inconsistent parse content is found; stops all parsing immediately
  ParseSyntaxException
just like ParseFatalException, but thrown internally when an ErrorStop ('-' operator) indicates that parsing is to stop immediately because an unbacktrackable syntax error has been found
  RecursiveGrammarException
exception thrown by validate() if the grammar could be improperly recursive
  ParseResults
Structured parse results, to provide multiple means of access to the parsed data:
  ParserElement
Abstract base level parser element class.
  Token
Abstract ParserElement subclass, for defining atomic matching patterns.
  Empty
An empty token, will always match.
  NoMatch
A token that will never match.
  Literal
Token to exactly match a specified string.
  Keyword
Token to exactly match a specified string as a keyword, that is, it must be immediately followed by a non-keyword character.
  CaselessLiteral
Token to match a specified string, ignoring case of letters.
  CaselessKeyword
  Word
Token for matching words composed of allowed character sets.
  Regex
Token for matching strings that match a given regular expression.
  QuotedString
Token for matching strings that are delimited by quoting characters.
  CharsNotIn
Token for matching words composed of characters *not* in a given set.
  White
Special matching class for matching whitespace.
  GoToColumn
Token to advance to a specific column of input text; useful for tabular report scraping.
  LineStart
Matches if current position is at the beginning of a line within the parse string
  LineEnd
Matches if current position is at the end of a line within the parse string
  StringStart
Matches if current position is at the beginning of the parse string
  StringEnd
Matches if current position is at the end of the parse string
  WordStart
Matches if the current position is at the beginning of a Word, and is not preceded by any character in a given set of wordChars (default=printables).
  WordEnd
Matches if the current position is at the end of a Word, and is not followed by any character in a given set of wordChars (default=printables).
  ParseExpression
Abstract subclass of ParserElement, for combining and post-processing parsed tokens.
  And
Requires all given ParseExpressions to be found in the given order.
  Or
Requires that at least one ParseExpression is found.
  MatchFirst
Requires that at least one ParseExpression is found.
  Each
Requires all given ParseExpressions to be found, but in any order.
  ParseElementEnhance
Abstract subclass of ParserElement, for combining and post-processing parsed tokens.
  FollowedBy
Lookahead matching of the given parse expression.
  NotAny
Lookahead to disallow matching with the given parse expression.
  ZeroOrMore
Optional repetition of zero or more of the given expression.
  OneOrMore
Repetition of one or more of the given expression.
  Optional
Optional matching of the given expression.
  SkipTo
Token for skipping over all undefined text until the matched expression is found.
  Forward
Forward declaration of an expression to be defined later - used for recursive grammars, such as algebraic infix notation.
  TokenConverter
Abstract subclass of ParseExpression, for converting parsed results.
  Upcase
Converter to upper case all matching tokens.
  Combine
Converter to concatenate all matching tokens to a single string.
  Group
Converter to return the matched tokens as a list - useful for returning tokens of ZeroOrMore and OneOrMore expressions.
  Dict
Converter to return a repetitive expression as a list, but also as a dictionary.
  Suppress
Converter for ignoring the results of a parsed expression.
  OnlyOnce
Wrapper for parse actions, to ensure they are only called once.
Functions
 
col(loc, strg)
Returns current column within a string, counting newlines as line separators.
source code
 
lineno(loc, strg)
Returns current line number within a string, counting newlines as line separators.
source code
 
line(loc, strg)
Returns the line of text containing loc within a string, counting newlines as line separators.
source code
 
nullDebugAction(*args)
'Do-nothing' debug action, to suppress debugging output during parsing.
source code
 
traceParseAction(f)
Decorator for debugging parse actions.
source code
 
delimitedList(expr, delim=',', combine=False)
Helper to define a delimited list of expressions - the delimiter defaults to ','.
source code
 
countedArray(expr, intExpr=None)
Helper to define a counted list of expressions.
source code
 
matchPreviousLiteral(expr)
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.
source code
 
matchPreviousExpr(expr)
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.
source code
 
oneOf(strs, caseless=False, useRegex=True)
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.
source code
 
dictOf(key, value)
Helper to easily and clearly define a dictionary by specifying the respective patterns for the key and value.
source code
 
originalTextFor(expr, asString=True)
Helper to return the original, untokenized text for a given expression.
source code
 
ungroup(expr)
Helper to undo pyparsing's default grouping of And expressions, even if all but one are non-empty.
source code
 
locatedExpr(expr)
Helper to decorate a returned token with its starting and ending locations in the input string.
source code
 
srange(s)
Helper to easily define string ranges for use in Word construction.
source code
 
matchOnlyAtCol(n)
Helper method for defining parse actions that require matching at a specific column in the input text.
source code
 
replaceWith(replStr)
Helper method for common parse actions that simply return a literal value.
source code
 
removeQuotes(s, l, t)
Helper parse action for removing quotation marks from parsed quoted strings.
source code
 
upcaseTokens(s, l, t)
Helper parse action to convert tokens to upper case.
source code
 
downcaseTokens(s, l, t)
Helper parse action to convert tokens to lower case.
source code
 
keepOriginalText(s, startLoc, t)
DEPRECATED - use new helper method originalTextFor.
source code
 
makeHTMLTags(tagStr)
Helper to construct opening and closing tag expressions for HTML, given a tag name
source code
 
makeXMLTags(tagStr)
Helper to construct opening and closing tag expressions for XML, given a tag name
source code
 
withAttribute(*args, **attrDict)
Helper to create a validating parse action to be used with start tags created with makeXMLTags or makeHTMLTags.
source code
 
infixNotation(baseExpr, opList, lpar=Suppress:("("), rpar=Suppress:(")"))
Helper method for constructing grammars of expressions made up of operators working in a precedence hierarchy.
source code
 
operatorPrecedence(baseExpr, opList, lpar=Suppress:("("), rpar=Suppress:(")"))
Helper method for constructing grammars of expressions made up of operators working in a precedence hierarchy.
source code
 
nestedExpr(opener='(', closer=')', content=None, ignoreExpr=quotedString using single or double quotes)
Helper method for defining nested lists enclosed in opening and closing delimiters ("(" and ")" are the default).
source code
 
indentedBlock(blockStatementExpr, indentStack, indent=True)
Helper method for defining space-delimited indentation blocks, such as those used to define block statements in Python source code.
source code
 
replaceHTMLEntity(t) source code
Variables
  alphas = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ'
  nums = '0123456789'
  hexnums = '0123456789ABCDEFabcdef'
  alphanums = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVW...
  printables = '0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKL...
  empty = empty
  lineStart = lineStart
  lineEnd = lineEnd
  stringStart = stringStart
  stringEnd = stringEnd
  opAssoc = _Constants()
  dblQuotedString = string enclosed in double quotes
  sglQuotedString = string enclosed in single quotes
  quotedString = quotedString using single or double quotes
  unicodeString = Combine:({"u" quotedString using single or dou...
  alphas8bit = u'ÀÁÂÃÄÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖØÙÚÛÜÝÞßàáâãäåæçèéêëìíîï...
  punc8bit = u'¡¢£¤¥¦§¨©ª«¬­®¯°±²³´µ¶·¸¹º»¼½¾¿×÷'
  commonHTMLEntity = Combine:({"&" Re:('gt|lt|amp|nbsp|quot') ";"})
  cStyleComment = C style comment
  htmlComment = Re:('<!--[\\s\\S]*?-->')
  restOfLine = Re:('.*')
  dblSlashComment = // comment
  cppStyleComment = C++ style comment
  javaStyleComment = C++ style comment
  pythonStyleComment = Python style comment
  commaSeparatedList = commaSeparatedList
  anyCloseTag = </W:(abcd...,abcd...)>
  anyOpenTag = <W:(abcd...,abcd...)>
Function Details

col(loc, strg)

source code 

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.

lineno(loc, strg)

source code 

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.

delimitedList(expr, delim=',', combine=False)

source code 

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.

countedArray(expr, intExpr=None)

source code 

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.

matchPreviousLiteral(expr)

source code 

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.

matchPreviousExpr(expr)

source code 

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.

oneOf(strs, caseless=False, useRegex=True)

source code 

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 list 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)

dictOf(key, value)

source code 

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.

originalTextFor(expr, asString=True)

source code 

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. Simpler to use than the parse action keepOriginalText, and does not require the inspect module to chase up the call stack. By default, returns a string 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.

locatedExpr(expr)

source code 

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

srange(s)

source code 

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.)

replaceWith(replStr)

source code 

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

removeQuotes(s, l, t)

source code 

Helper parse action for removing quotation marks from parsed quoted strings. To use, add this parse action to quoted string using:

 quotedString.setParseAction( removeQuotes )

keepOriginalText(s, startLoc, t)

source code 

DEPRECATED - use new helper method originalTextFor. Helper parse action to preserve original parsed text, overriding any nested parse actions.

withAttribute(*args, **attrDict)

source code 

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.

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

infixNotation(baseExpr, opList, lpar=Suppress:("("), rpar=Suppress:(")"))

source code 

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.

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)
  • lpar - expression for matching left-parentheses (default=Suppress('('))
  • rpar - expression for matching right-parentheses (default=Suppress(')'))

operatorPrecedence(baseExpr, opList, lpar=Suppress:("("), rpar=Suppress:(")"))

source code 

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.

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)
  • lpar - expression for matching left-parentheses (default=Suppress('('))
  • rpar - expression for matching right-parentheses (default=Suppress(')'))

nestedExpr(opener='(', closer=')', content=None, ignoreExpr=quotedString using single or double quotes)

source code 

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.

indentedBlock(blockStatementExpr, indentStack, indent=True)

source code 

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.


Variables Details

alphanums

Value:
'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789'

printables

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

unicodeString

Value:
Combine:({"u" quotedString using single or double quotes})

alphas8bit

Value:
u'ÀÁÂÃÄÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖØÙÚÛÜÝÞßàáâãäåæçèéêëìíîïðñòóôõöøùúûüýþÿ'