Fixed bug in SkipTo when ignore argument is found at the beginning of the range of text to be skipped.

Fixed docstring in White class.
Fixed obscure bug when calling __getitem__ of an object that replaces ParseResults during a parse action.

git-svn-id: svn://svn.code.sf.net/p/pyparsing/code/src@176 9bf210a0-9d2d-494c-87cf-cfb32e7dff7b

3 files changed, 40 insertions, 17 deletions

diff --git a/CHANGES b/CHANGES
index a81fcc0..291fbd8 100644
--- a/CHANGES
+++ b/CHANGES
@@ -7,6 +7,12 @@ Version 1.5.2 - ???
 - Fixed bug in SkipTo/failOn handling - caught by eagle eye 

   cpennington on the pyparsing wiki!

 

+- Fixed second bug in SkipTo when using the ignore constructor

+  argument, reported by Catherine Devlin, thanks!

+  

+- Fixed obscure bug reported by Eike Welk when using a class

+  as a ParseAction with an errant __getitem__ method.

+

 - Simplified exception stack traces when reporting parse 

   exceptions back to caller of parseString or parseFile - thanks

   to a tip from Peter Otten on comp.lang.python.

diff --git a/pyparsing.py b/pyparsing.py
index 17e3b9e..bec5c33 100644
--- a/pyparsing.py
+++ b/pyparsing.py
@@ -59,7 +59,7 @@ The pyparsing module handles some of the problems that are typically vexing when
 """

 

 __version__ = "1.5.2"

-__versionTime__ = "5 December 2008 10:27"

+__versionTime__ = "20 December 2008 05:55"

 __author__ = "Paul McGuire <ptmcg@users.sourceforge.net>"

 

 import string

@@ -110,6 +110,9 @@ if not _PY3K:
            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.

@@ -309,7 +312,7 @@ class ParseResults(object):
                 else:

                     try:

                         self[name] = toklist[0]

-                    except (KeyError,TypeError):

+                    except (KeyError,TypeError,IndexError):

                         self[name] = toklist

 

     def __getitem__( self, i ):

@@ -1974,7 +1977,7 @@ 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 " \\t\\n".  Also takes optional min, max, and exact arguments,

+       matched; default is " \\t\\r\\n".  Also takes optional min, max, and exact arguments,

        as defined for the Word class."""

     whiteStrs = {

         " " : "<SPC>",

@@ -2534,7 +2537,7 @@ class Each(ParseExpression):
             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 ]

+        matchOrder += list(e for e in self.exprs if isinstance(e,Optional) and e.expr in tmpOpt)

 

         resultlist = []

         for e in matchOrder:

@@ -2815,9 +2818,7 @@ class SkipTo(ParseElementEnhance):
     """

     def __init__( self, other, include=False, ignore=None, failOn=None ):

         super( SkipTo, self ).__init__( other )

-        if ignore is not None:

-            self.expr = self.expr.copy()

-            self.expr.ignore(ignore)

+        self.ignoreExpr = ignore

         self.mayReturnEmpty = True

         self.mayIndexError = False

         self.includeMatch = include

@@ -2845,7 +2846,13 @@ class SkipTo(ParseElementEnhance):
                         failParse = True

                         raise ParseException(instring, loc, "Found expression " + str(self.failOn))

                     failParse = False

-                loc = expr._skipIgnorables( instring, loc )

+                if self.ignoreExpr is not None:

+                    while 1:

+                        try:

+                            loc = self.ignoreExpr.tryParse(instring,loc)

+                            print "found ignoreExpr, advance to", loc

+                        except ParseBaseException:

+                            break

                 expr._parse( instring, loc, doActions=False, callPreParse=False )

                 skipText = instring[startLoc:loc]

                 if self.includeMatch:

diff --git a/pyparsing_py3.py b/pyparsing_py3.py
index 2a9da0d..a9d8412 100644
--- a/pyparsing_py3.py
+++ b/pyparsing_py3.py
@@ -59,7 +59,7 @@ The pyparsing module handles some of the problems that are typically vexing when
 """

 

 __version__ = "1.5.2Py3"

-__versionTime__ = "8 November 2008 05:36"

+__versionTime__ = "20 December 2008 06:00"

 __author__ = "Paul McGuire <ptmcg@users.sourceforge.net>"

 

 import string

@@ -110,6 +110,9 @@ if not _PY3K:
            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.

@@ -309,7 +312,7 @@ class ParseResults(object):
                 else:

                     try:

                         self[name] = toklist[0]

-                    except (KeyError,TypeError):

+                    except (KeyError,TypeError,IndexError):

                         self[name] = toklist

 

     def __getitem__( self, i ):

@@ -1974,7 +1977,7 @@ 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 " \\t\\n".  Also takes optional min, max, and exact arguments,

+       matched; default is " \\t\\r\\n".  Also takes optional min, max, and exact arguments,

        as defined for the Word class."""

     whiteStrs = {

         " " : "<SPC>",

@@ -2208,7 +2211,10 @@ class ParseExpression(ParserElement):
         elif isinstance( exprs, basestring ):

             self.exprs = [ Literal( exprs ) ]

         else:

-            self.exprs = list( exprs )

+            try:

+                self.exprs = list( exprs )

+            except TypeError:

+                self.exprs = [ exprs ]

         self.callPreparse = False

 

     def __getitem__( self, i ):

@@ -2812,9 +2818,7 @@ class SkipTo(ParseElementEnhance):
     """

     def __init__( self, other, include=False, ignore=None, failOn=None ):

         super( SkipTo, self ).__init__( other )

-        if ignore is not None:

-            self.expr = self.expr.copy()

-            self.expr.ignore(ignore)

+        self.ignoreExpr = ignore

         self.mayReturnEmpty = True

         self.mayIndexError = False

         self.includeMatch = include

@@ -2842,7 +2846,13 @@ class SkipTo(ParseElementEnhance):
                         failParse = True

                         raise ParseException(instring, loc, "Found expression " + str(self.failOn))

                     failParse = False

-                loc = expr._skipIgnorables( instring, loc )

+                if self.ignoreExpr is not None:

+                    while 1:

+                        try:

+                            loc = self.ignoreExpr.tryParse(instring,loc)

+                            print "found ignoreExpr, advance to", loc

+                        except ParseBaseException:

+                            break

                 expr._parse( instring, loc, doActions=False, callPreParse=False )

                 skipText = instring[startLoc:loc]

                 if self.includeMatch:

@@ -3195,7 +3205,7 @@ def oneOf( strs, caseless=False, useRegex=True ):
         parseElementClass = Literal

 

     if isinstance(strs,(list,tuple)):

-        symbols = strs[:]

+        symbols = list(strs[:])

     elif isinstance(strs,basestring):

         symbols = strs.split()

     else: