Merged revisions 61596-61597 via svnmerge from

svn+ssh://pythondev@svn.python.org/python/trunk

........
  r61596 | martin.v.loewis | 2008-03-18 23:43:46 -0500 (Di, 18 Mär 2008) | 2 lines

  Import lib2to3.
........
  r61597 | martin.v.loewis | 2008-03-18 23:58:04 -0500 (Di, 18 Mär 2008) | 3 lines

  Initialized merge tracking via "svnmerge" with revisions "1-61595" from
  svn+ssh://pythondev@svn.python.org/sandbox/trunk/2to3/lib2to3
........

Lib/lib2to3/Grammar.txt

+# Grammar for Python
+
+# Note:  Changing the grammar specified in this file will most likely
+#        require corresponding changes in the parser module
+#        (../Modules/parsermodule.c).  If you can't make the changes to
+#        that module yourself, please co-ordinate the required changes
+#        with someone who can; ask around on python-dev for help.  Fred
+#        Drake <fdrake@acm.org> will probably be listening there.
+
+# NOTE WELL: You should also follow all the steps listed in PEP 306,
+# "How to Change Python's Grammar"
+
+# Commands for Kees Blom's railroad program
+#diagram:token NAME
+#diagram:token NUMBER
+#diagram:token STRING
+#diagram:token NEWLINE
+#diagram:token ENDMARKER
+#diagram:token INDENT
+#diagram:output\input python.bla
+#diagram:token DEDENT
+#diagram:output\textwidth 20.04cm\oddsidemargin  0.0cm\evensidemargin 0.0cm
+#diagram:rules
+
+# Start symbols for the grammar:
+#	file_input is a module or sequence of commands read from an input file;
+#	single_input is a single interactive statement;
+#	eval_input is the input for the eval() and input() functions.
+# NB: compound_stmt in single_input is followed by extra NEWLINE!
+file_input: (NEWLINE | stmt)* ENDMARKER
+single_input: NEWLINE | simple_stmt | compound_stmt NEWLINE
+eval_input: testlist NEWLINE* ENDMARKER
+
+decorator: '@' dotted_name [ '(' [arglist] ')' ] NEWLINE
+decorators: decorator+
+decorated: decorators (classdef | funcdef)
+funcdef: 'def' NAME parameters ['->' test] ':' suite
+parameters: '(' [typedargslist] ')'
+typedargslist: ((tfpdef ['=' test] ',')*
+                ('*' [tname] (',' tname ['=' test])* [',' '**' tname] | '**' tname)
+                | tfpdef ['=' test] (',' tfpdef ['=' test])* [','])
+tname: NAME [':' test]
+tfpdef: tname | '(' tfplist ')'
+tfplist: tfpdef (',' tfpdef)* [',']
+varargslist: ((vfpdef ['=' test] ',')*
+              ('*' [vname] (',' vname ['=' test])*  [',' '**' vname] | '**' vname)
+              | vfpdef ['=' test] (',' vfpdef ['=' test])* [','])
+vname: NAME
+vfpdef: vname | '(' vfplist ')'
+vfplist: vfpdef (',' vfpdef)* [',']
+
+stmt: simple_stmt | compound_stmt
+simple_stmt: small_stmt (';' small_stmt)* [';'] NEWLINE
+small_stmt: (expr_stmt | print_stmt  | del_stmt | pass_stmt | flow_stmt |
+             import_stmt | global_stmt | exec_stmt | assert_stmt)
+expr_stmt: testlist (augassign (yield_expr|testlist) |
+                     ('=' (yield_expr|testlist))*)
+augassign: ('+=' | '-=' | '*=' | '/=' | '%=' | '&=' | '|=' | '^=' |
+            '<<=' | '>>=' | '**=' | '//=')
+# For normal assignments, additional restrictions enforced by the interpreter
+print_stmt: 'print' ( [ test (',' test)* [','] ] |
+                      '>>' test [ (',' test)+ [','] ] )
+del_stmt: 'del' exprlist
+pass_stmt: 'pass'
+flow_stmt: break_stmt | continue_stmt | return_stmt | raise_stmt | yield_stmt
+break_stmt: 'break'
+continue_stmt: 'continue'
+return_stmt: 'return' [testlist]
+yield_stmt: yield_expr
+raise_stmt: 'raise' [test ['from' test | ',' test [',' test]]]
+import_stmt: import_name | import_from
+import_name: 'import' dotted_as_names
+import_from: ('from' ('.'* dotted_name | '.'+)
+              'import' ('*' | '(' import_as_names ')' | import_as_names))
+import_as_name: NAME ['as' NAME]
+dotted_as_name: dotted_name ['as' NAME]
+import_as_names: import_as_name (',' import_as_name)* [',']
+dotted_as_names: dotted_as_name (',' dotted_as_name)*
+dotted_name: NAME ('.' NAME)*
+global_stmt: ('global' | 'nonlocal') NAME (',' NAME)*
+exec_stmt: 'exec' expr ['in' test [',' test]]
+assert_stmt: 'assert' test [',' test]
+
+compound_stmt: if_stmt | while_stmt | for_stmt | try_stmt | with_stmt | funcdef | classdef | decorated
+if_stmt: 'if' test ':' suite ('elif' test ':' suite)* ['else' ':' suite]
+while_stmt: 'while' test ':' suite ['else' ':' suite]
+for_stmt: 'for' exprlist 'in' testlist ':' suite ['else' ':' suite]
+try_stmt: ('try' ':' suite
+           ((except_clause ':' suite)+
+	    ['else' ':' suite]
+	    ['finally' ':' suite] |
+	   'finally' ':' suite))
+with_stmt: 'with' test [ with_var ] ':' suite
+with_var: 'as' expr
+# NB compile.c makes sure that the default except clause is last
+except_clause: 'except' [test [(',' | 'as') test]]
+suite: simple_stmt | NEWLINE INDENT stmt+ DEDENT
+
+# Backward compatibility cruft to support:
+# [ x for x in lambda: True, lambda: False if x() ]
+# even while also allowing:
+# lambda x: 5 if x else 2
+# (But not a mix of the two)
+testlist_safe: old_test [(',' old_test)+ [',']]
+old_test: or_test | old_lambdef
+old_lambdef: 'lambda' [varargslist] ':' old_test
+
+test: or_test ['if' or_test 'else' test] | lambdef
+or_test: and_test ('or' and_test)*
+and_test: not_test ('and' not_test)*
+not_test: 'not' not_test | comparison
+comparison: expr (comp_op expr)*
+comp_op: '<'|'>'|'=='|'>='|'<='|'<>'|'!='|'in'|'not' 'in'|'is'|'is' 'not'
+expr: xor_expr ('|' xor_expr)*
+xor_expr: and_expr ('^' and_expr)*
+and_expr: shift_expr ('&' shift_expr)*
+shift_expr: arith_expr (('<<'|'>>') arith_expr)*
+arith_expr: term (('+'|'-') term)*
+term: factor (('*'|'/'|'%'|'//') factor)*
+factor: ('+'|'-'|'~') factor | power
+power: atom trailer* ['**' factor]
+atom: ('(' [yield_expr|testlist_gexp] ')' |
+       '[' [listmaker] ']' |
+       '{' [dictsetmaker] '}' |
+       '`' testlist1 '`' |
+       NAME | NUMBER | STRING+ | '.' '.' '.')
+listmaker: test ( comp_for | (',' test)* [','] )
+testlist_gexp: test ( comp_for | (',' test)* [','] )
+lambdef: 'lambda' [varargslist] ':' test
+trailer: '(' [arglist] ')' | '[' subscriptlist ']' | '.' NAME
+subscriptlist: subscript (',' subscript)* [',']
+subscript: test | [test] ':' [test] [sliceop]
+sliceop: ':' [test]
+exprlist: expr (',' expr)* [',']
+testlist: test (',' test)* [',']
+dictsetmaker: ( (test ':' test (comp_for | (',' test ':' test)* [','])) |
+                (test (comp_for | (',' test)* [','])) )
+
+classdef: 'class' NAME ['(' [arglist] ')'] ':' suite
+
+arglist: (argument ',')* (argument [',']| '*' test [',' '**' test] | '**' test)
+argument: test [comp_for] | test '=' test  # Really [keyword '='] test
+
+comp_iter: comp_for | comp_if
+comp_for: 'for' exprlist 'in' testlist_safe [comp_iter]
+comp_if: 'if' old_test [comp_iter]
+
+testlist1: test (',' test)*
+
+# not used in grammar, but may appear in "node" passed from Parser to Compiler
+encoding_decl: NAME
+
+yield_expr: 'yield' [testlist]

Lib/lib2to3/PatternGrammar.txt

+# Copyright 2006 Google, Inc. All Rights Reserved.
+# Licensed to PSF under a Contributor Agreement.
+
+# A grammar to describe tree matching patterns.
+# Not shown here:
+# - 'TOKEN' stands for any token (leaf node)
+# - 'any' stands for any node (leaf or interior)
+# With 'any' we can still specify the sub-structure.
+
+# The start symbol is 'Matcher'.
+
+Matcher: Alternatives ENDMARKER
+
+Alternatives: Alternative ('|' Alternative)*
+
+Alternative: (Unit | NegatedUnit)+
+
+Unit: [NAME '='] ( STRING [Repeater]
+                 | NAME [Details] [Repeater]
+                 | '(' Alternatives ')' [Repeater]
+                 | '[' Alternatives ']'
+		 )
+
+NegatedUnit: 'not' (STRING | NAME [Details] | '(' Alternatives ')')
+
+Repeater: '*' | '+' | '{' NUMBER [',' NUMBER] '}'
+
+Details: '<' Alternatives '>'

Lib/lib2to3/__init__.py

+#empty

Lib/lib2to3/fixes/__init__.py

+# Dummy file to make this directory a package.

Lib/lib2to3/fixes/basefix.py

+# Copyright 2006 Google, Inc. All Rights Reserved.
+# Licensed to PSF under a Contributor Agreement.
+
+"""Base class for fixers (optional, but recommended)."""
+
+# Python imports
+import logging
+import itertools
+
+# Get a usable 'set' constructor
+try:
+    set
+except NameError:
+    from sets import Set as set
+
+# Local imports
+from ..patcomp import PatternCompiler
+from .. import pygram
+
+class BaseFix(object):
+
+    """Optional base class for fixers.
+
+    The subclass name must be FixFooBar where FooBar is the result of
+    removing underscores and capitalizing the words of the fix name.
+    For example, the class name for a fixer named 'has_key' should be
+    FixHasKey.
+    """
+
+    PATTERN = None  # Most subclasses should override with a string literal
+    pattern = None  # Compiled pattern, set by compile_pattern()
+    options = None  # Options object passed to initializer
+    filename = None # The filename (set by set_filename)
+    logger = None   # A logger (set by set_filename)
+    numbers = itertools.count(1) # For new_name()
+    used_names = set() # A set of all used NAMEs
+    order = "post" # Does the fixer prefer pre- or post-order traversal
+    explicit = False # Is this ignored by refactor.py -f all?
+
+    # Shortcut for access to Python grammar symbols
+    syms = pygram.python_symbols
+
+    def __init__(self, options, log):
+        """Initializer.  Subclass may override.
+
+        Args:
+            options: an optparse.Values instance which can be used
+                to inspect the command line options.
+            log: a list to append warnings and other messages to.
+        """
+        self.options = options
+        self.log = log
+        self.compile_pattern()
+
+    def compile_pattern(self):
+        """Compiles self.PATTERN into self.pattern.
+
+        Subclass may override if it doesn't want to use
+        self.{pattern,PATTERN} in .match().
+        """
+        if self.PATTERN is not None:
+            self.pattern = PatternCompiler().compile_pattern(self.PATTERN)
+
+    def set_filename(self, filename):
+        """Set the filename, and a logger derived from it.
+
+        The main refactoring tool should call this.
+        """
+        self.filename = filename
+        self.logger = logging.getLogger(filename)
+
+    def match(self, node):
+        """Returns match for a given parse tree node.
+
+        Should return a true or false object (not necessarily a bool).
+        It may return a non-empty dict of matching sub-nodes as
+        returned by a matching pattern.
+
+        Subclass may override.
+        """
+        results = {"node": node}
+        return self.pattern.match(node, results) and results
+
+    def transform(self, node, results):
+        """Returns the transformation for a given parse tree node.
+
+        Args:
+          node: the root of the parse tree that matched the fixer.
+          results: a dict mapping symbolic names to part of the match.
+          
+        Returns:
+          None, or a node that is a modified copy of the
+          argument node.  The node argument may also be modified in-place to
+          effect the same change.
+
+        Subclass *must* override.
+        """
+        raise NotImplementedError()
+
+    def parenthesize(self, node):
+        """Wrapper around pygram.parenthesize()."""
+        return pygram.parenthesize(node)
+
+    def new_name(self, template="xxx_todo_changeme"):
+        """Return a string suitable for use as an identifier
+
+        The new name is guaranteed not to conflict with other identifiers.
+        """
+        name = template
+        while name in self.used_names:
+            name = template + str(self.numbers.next())
+        self.used_names.add(name)
+        return name
+
+    def log_message(self, message):
+        if self.first_log:
+            self.first_log = False
+            self.log.append("### In file %s ###" % self.filename)
+        self.log.append(message)
+
+    def cannot_convert(self, node, reason=None):
+        """Warn the user that a given chunk of code is not valid Python 3,
+        but that it cannot be converted automatically.
+
+        First argument is the top-level node for the code in question.
+        Optional second argument is why it can't be converted.
+        """
+        lineno = node.get_lineno()
+        for_output = node.clone()
+        for_output.set_prefix("")
+        msg = "Line %d: could not convert: %s"
+        self.log_message(msg % (lineno, for_output))
+        if reason:
+            self.log_message(reason)
+
+    def warning(self, node, reason):
+        """Used for warning the user about possible uncertainty in the
+        translation.
+
+        First argument is the top-level node for the code in question.
+        Optional second argument is why it can't be converted.
+        """
+        lineno = node.get_lineno()
+        self.log_message("Line %d: %s" % (lineno, reason))
+
+    def start_tree(self, tree, filename):
+        """Some fixers need to maintain tree-wide state.
+        This method is called once, at the start of tree fix-up.
+        
+        tree - the root node of the tree to be processed.
+        filename - the name of the file the tree came from.
+        """
+        self.used_names = tree.used_names
+        self.set_filename(filename)
+        self.numbers = itertools.count(1)
+        self.first_log = True
+
+    def finish_tree(self, tree, filename):
+        """Some fixers need to maintain tree-wide state.
+        This method is called once, at the conclusion of tree fix-up.
+        
+        tree - the root node of the tree to be processed.
+        filename - the name of the file the tree came from.
+        """
+        pass

Lib/lib2to3/fixes/fix_apply.py

+# Copyright 2006 Google, Inc. All Rights Reserved.
+# Licensed to PSF under a Contributor Agreement.
+
+"""Fixer for apply().
+
+This converts apply(func, v, k) into (func)(*v, **k)."""
+
+# Local imports
+from .. import pytree
+from ..pgen2 import token
+from . import basefix
+from .util import Call, Comma
+
+class FixApply(basefix.BaseFix):
+
+    PATTERN = """
+    power< 'apply'
+        trailer<
+            '('
+            arglist<
+                (not argument<NAME '=' any>) func=any ','
+                (not argument<NAME '=' any>) args=any [','
+                (not argument<NAME '=' any>) kwds=any] [',']
+            >
+            ')'
+        >
+    >
+    """
+
+    def transform(self, node, results):
+        syms = self.syms
+        assert results
+        func = results["func"]
+        args = results["args"]
+        kwds = results.get("kwds")
+        prefix = node.get_prefix()
+        func = func.clone()
+        if (func.type not in (token.NAME, syms.atom) and
+            (func.type != syms.power or
+             func.children[-2].type == token.DOUBLESTAR)):
+            # Need to parenthesize
+            func = self.parenthesize(func)
+        func.set_prefix("")
+        args = args.clone()
+        args.set_prefix("")
+        if kwds is not None:
+            kwds = kwds.clone()
+            kwds.set_prefix("")
+        l_newargs = [pytree.Leaf(token.STAR, "*"), args]
+        if kwds is not None:
+            l_newargs.extend([Comma(),
+                              pytree.Leaf(token.DOUBLESTAR, "**"),
+                              kwds])
+            l_newargs[-2].set_prefix(" ") # that's the ** token
+        # XXX Sometimes we could be cleverer, e.g. apply(f, (x, y) + t)
+        # can be translated into f(x, y, *t) instead of f(*(x, y) + t)
+        #new = pytree.Node(syms.power, (func, ArgList(l_newargs)))
+        return Call(func, l_newargs, prefix=prefix)

Lib/lib2to3/fixes/fix_basestring.py

+"""Fixer for basestring -> str."""
+# Author: Christian Heimes
+
+# Local imports
+from . import basefix
+from .util import Name
+
+class FixBasestring(basefix.BaseFix):
+
+    PATTERN = "'basestring'"
+
+    def transform(self, node, results):
+        return Name("str", prefix=node.get_prefix())

Lib/lib2to3/fixes/fix_buffer.py

+# Copyright 2007 Google, Inc. All Rights Reserved.
+# Licensed to PSF under a Contributor Agreement.
+
+"""Fixer that changes buffer(...) into memoryview(...)."""
+
+# Local imports
+from . import basefix
+from .util import Name
+
+
+class FixBuffer(basefix.BaseFix):
+
+    explicit = True # The user must ask for this fixer
+
+    PATTERN = """
+              power< name='buffer' trailer< '(' [any] ')' > >
+              """
+
+    def transform(self, node, results):
+        name = results["name"]
+        name.replace(Name("memoryview", prefix=name.get_prefix()))

Lib/lib2to3/fixes/fix_callable.py

+# Copyright 2007 Google, Inc. All Rights Reserved.
+# Licensed to PSF under a Contributor Agreement.
+
+"""Fixer for callable().
+
+This converts callable(obj) into hasattr(obj, '__call__')."""
+
+# Local imports
+from .. import pytree
+from . import basefix
+from .util import Call, Name, String
+
+class FixCallable(basefix.BaseFix):
+
+    # Ignore callable(*args) or use of keywords.
+    # Either could be a hint that the builtin callable() is not being used.
+    PATTERN = """
+    power< 'callable'
+           trailer< lpar='('
+                    ( not(arglist | argument<any '=' any>) func=any
+                      | func=arglist<(not argument<any '=' any>) any ','> )
+                    rpar=')' >
+           after=any*
+    >
+    """
+
+    def transform(self, node, results):
+        func = results["func"]
+
+        args = [func.clone(), String(', '), String("'__call__'")]
+        return Call(Name("hasattr"), args, prefix=node.get_prefix())

Lib/lib2to3/fixes/fix_dict.py

+# Copyright 2007 Google, Inc. All Rights Reserved.
+# Licensed to PSF under a Contributor Agreement.
+
+"""Fixer for dict methods.
+
+d.keys() -> list(d.keys())
+d.items() -> list(d.items())
+d.values() -> list(d.values())
+
+d.iterkeys() -> iter(d.keys())
+d.iteritems() -> iter(d.items())
+d.itervalues() -> iter(d.values())
+
+Except in certain very specific contexts: the iter() can be dropped
+when the context is list(), sorted(), iter() or for...in; the list()
+can be dropped when the context is list() or sorted() (but not iter()
+or for...in!). Special contexts that apply to both: list(), sorted(), tuple()
+set(), any(), all(), sum().
+
+Note: iter(d.keys()) could be written as iter(d) but since the
+original d.iterkeys() was also redundant we don't fix this.  And there
+are (rare) contexts where it makes a difference (e.g. when passing it
+as an argument to a function that introspects the argument).
+"""
+
+# Local imports
+from .. import pytree
+from .. import patcomp
+from ..pgen2 import token
+from . import basefix
+from .util import Name, Call, LParen, RParen, ArgList, Dot, set
+
+
+exempt = set(["sorted", "list", "set", "any", "all", "tuple", "sum"])
+iter_exempt = exempt | set(["iter"])
+
+
+class FixDict(basefix.BaseFix):
+    PATTERN = """
+    power< head=any+
+         trailer< '.' method=('keys'|'items'|'values'|
+                              'iterkeys'|'iteritems'|'itervalues') >
+         parens=trailer< '(' ')' >
+         tail=any*
+    >
+    """
+
+    def transform(self, node, results):
+        head = results["head"]
+        method = results["method"][0] # Extract node for method name
+        tail = results["tail"]
+        syms = self.syms
+        method_name = method.value
+        isiter = method_name.startswith("iter")
+        if isiter:
+            method_name = method_name[4:]
+        assert method_name in ("keys", "items", "values"), repr(method)
+        head = [n.clone() for n in head]
+        tail = [n.clone() for n in tail]
+        special = not tail and self.in_special_context(node, isiter)
+        args = head + [pytree.Node(syms.trailer,
+                                   [Dot(),
+                                    Name(method_name,
+                                         prefix=method.get_prefix())]),
+                       results["parens"].clone()]
+        new = pytree.Node(syms.power, args)
+        if not special:
+            new.set_prefix("")
+            new = Call(Name(isiter and "iter" or "list"), [new])
+        if tail:
+            new = pytree.Node(syms.power, [new] + tail)
+        new.set_prefix(node.get_prefix())
+        return new
+
+    P1 = "power< func=NAME trailer< '(' node=any ')' > any* >"
+    p1 = patcomp.compile_pattern(P1)
+
+    P2 = """for_stmt< 'for' any 'in' node=any ':' any* >
+            | comp_for< 'for' any 'in' node=any any* >
+         """
+    p2 = patcomp.compile_pattern(P2)
+
+    def in_special_context(self, node, isiter):
+        if node.parent is None:
+            return False
+        results = {}
+        if (node.parent.parent is not None and
+               self.p1.match(node.parent.parent, results) and
+               results["node"] is node):
+            if isiter:
+                # iter(d.iterkeys()) -> iter(d.keys()), etc.
+                return results["func"].value in iter_exempt
+            else:
+                # list(d.keys()) -> list(d.keys()), etc.
+                return results["func"].value in exempt
+        if not isiter:
+            return False
+        # for ... in d.iterkeys() -> for ... in d.keys(), etc.
+        return self.p2.match(node.parent, results) and results["node"] is node

Lib/lib2to3/fixes/fix_except.py

+"""Fixer for except statements with named exceptions.
+
+The following cases will be converted:
+
+- "except E, T:" where T is a name:
+    
+    except E as T:
+    
+- "except E, T:" where T is not a name, tuple or list:
+    
+        except E as t:
+            T = t
+        
+    This is done because the target of an "except" clause must be a
+    name.
+    
+- "except E, T:" where T is a tuple or list literal:
+    
+        except E as t:
+            T = t.args
+"""
+# Author: Collin Winter
+
+# Local imports
+from .. import pytree
+from ..pgen2 import token
+from . import basefix
+from .util import Assign, Attr, Name, is_tuple, is_list, reversed
+
+def find_excepts(nodes):
+    for i, n in enumerate(nodes):
+        if isinstance(n, pytree.Node):
+            if n.children[0].value == 'except':
+                yield (n, nodes[i+2])
+
+class FixExcept(basefix.BaseFix):
+
+    PATTERN = """
+    try_stmt< 'try' ':' suite
+                  cleanup=((except_clause ':' suite)+ ['else' ':' suite]
+                                                      ['finally' ':' suite]
+	                       | 'finally' ':' suite) >
+    """
+
+    def transform(self, node, results):
+        syms = self.syms
+
+        try_cleanup = [ch.clone() for ch in results['cleanup']]
+        for except_clause, e_suite in find_excepts(try_cleanup):
+            if len(except_clause.children) == 4:
+                (E, comma, N) = except_clause.children[1:4]
+                comma.replace(Name("as", prefix=" "))
+
+                if N.type != token.NAME:
+                    # Generate a new N for the except clause
+                    new_N = Name(self.new_name(), prefix=" ")
+                    target = N.clone()
+                    target.set_prefix("")
+                    N.replace(new_N)
+                    new_N = new_N.clone()
+
+                    # Insert "old_N = new_N" as the first statement in
+                    #  the except body. This loop skips leading whitespace
+                    #  and indents
+                    #TODO(cwinter) suite-cleanup
+                    suite_stmts = e_suite.children
+                    for i, stmt in enumerate(suite_stmts):
+                        if isinstance(stmt, pytree.Node):
+                            break
+
+                    # The assignment is different if old_N is a tuple or list
+                    # In that case, the assignment is old_N = new_N.args
+                    if is_tuple(N) or is_list(N):
+                        assign = Assign(target, Attr(new_N, Name('args')))
+                    else:
+                        assign = Assign(target, new_N)
+
+                    #TODO(cwinter) stopgap until children becomes a smart list
+                    for child in reversed(suite_stmts[:i]):
+                        e_suite.insert_child(0, child)
+                    e_suite.insert_child(i, assign)
+                elif N.get_prefix() == "":
+                    # No space after a comma is legal; no space after "as",
+                    # not so much.
+                    N.set_prefix(" ")
+
+        #TODO(cwinter) fix this when children becomes a smart list
+        children = [c.clone() for c in node.children[:3]] + try_cleanup
+        return pytree.Node(node.type, children)

Lib/lib2to3/fixes/fix_exec.py

+# Copyright 2006 Google, Inc. All Rights Reserved.
+# Licensed to PSF under a Contributor Agreement.
+
+"""Fixer for exec.
+
+This converts usages of the exec statement into calls to a built-in
+exec() function.
+
+exec code in ns1, ns2 -> exec(code, ns1, ns2)
+"""
+
+# Local imports
+from .. import pytree
+from . import basefix
+from .util import Comma, Name, Call
+
+
+class FixExec(basefix.BaseFix):
+
+    PATTERN = """
+    exec_stmt< 'exec' a=any 'in' b=any [',' c=any] >
+    |
+    exec_stmt< 'exec' (not atom<'(' [any] ')'>) a=any >
+    """
+
+    def transform(self, node, results):
+        assert results
+        syms = self.syms
+        a = results["a"]
+        b = results.get("b")
+        c = results.get("c")
+        args = [a.clone()]
+        args[0].set_prefix("")
+        if b is not None:
+            args.extend([Comma(), b.clone()])
+        if c is not None:
+            args.extend([Comma(), c.clone()])
+
+        return Call(Name("exec"), args, prefix=node.get_prefix())

Lib/lib2to3/fixes/fix_execfile.py

+# Copyright 2006 Google, Inc. All Rights Reserved.
+# Licensed to PSF under a Contributor Agreement.
+
+"""Fixer for execfile.
+
+This converts usages of the execfile function into calls to the built-in
+exec() function.
+"""
+
+from .. import pytree
+from . import basefix
+from .util import Comma, Name, Call, LParen, RParen, Dot
+
+
+class FixExecfile(basefix.BaseFix):
+
+    PATTERN = """
+    power< 'execfile' trailer< '(' arglist< filename=any [',' globals=any [',' locals=any ] ] > ')' > >
+    |
+    power< 'execfile' trailer< '(' filename=any ')' > >
+    """
+
+    def transform(self, node, results):
+        assert results
+        syms = self.syms
+        filename = results["filename"]
+        globals = results.get("globals")
+        locals = results.get("locals")
+        args = [Name('open'), LParen(), filename.clone(), RParen(), Dot(),
+                Name('read'), LParen(), RParen()]
+        args[0].set_prefix("")
+        if globals is not None:
+            args.extend([Comma(), globals.clone()])
+        if locals is not None:
+            args.extend([Comma(), locals.clone()])
+
+        return Call(Name("exec"), args, prefix=node.get_prefix())

Lib/lib2to3/fixes/fix_filter.py

+# Copyright 2007 Google, Inc. All Rights Reserved.
+# Licensed to PSF under a Contributor Agreement.
+
+"""Fixer that changes filter(F, X) into list(filter(F, X)).
+
+We avoid the transformation if the filter() call is directly contained
+in iter(<>), list(<>), tuple(<>), sorted(<>), ...join(<>), or
+for V in <>:.
+
+NOTE: This is still not correct if the original code was depending on
+filter(F, X) to return a string if X is a string and a tuple if X is a
+tuple.  That would require type inference, which we don't do.  Let
+Python 2.6 figure it out.
+"""
+
+# Local imports
+from .. import pytree
+from .. import patcomp
+from ..pgen2 import token
+from . import basefix
+from .util import Name, Call, ListComp, attr_chain, does_tree_import
+
+class FixFilter(basefix.BaseFix):
+
+    PATTERN = """
+    filter_lambda=power<
+        'filter'
+        trailer<
+            '('
+            arglist<
+                lambdef< 'lambda'
+                         (fp=NAME | vfpdef< '(' fp=NAME ')'> ) ':' xp=any
+                >
+                ','
+                it=any
+            >
+            ')'
+        >
+    >
+    |
+    power<
+        'filter'
+        trailer< '(' arglist< none='None' ',' seq=any > ')' >
+    >
+    |
+    power<
+        'filter'
+        args=trailer< '(' [any] ')' >
+    >
+    """
+
+    def start_tree(self, *args):
+        super(FixFilter, self).start_tree(*args)
+        self._new_filter = None
+
+    def has_new_filter(self, node):
+        if self._new_filter is not None:
+            return self._new_filter
+        self._new_filter = does_tree_import('future_builtins', 'filter', node)
+        return self._new_filter
+
+    def transform(self, node, results):
+        if self.has_new_filter(node):
+            # If filter is imported from future_builtins, we don't want to
+            # do anything here.
+            return
+
+        if "filter_lambda" in results:
+            new = ListComp(results.get("fp").clone(),
+                           results.get("fp").clone(),
+                           results.get("it").clone(),
+                           results.get("xp").clone())
+
+        elif "none" in results:
+            new = ListComp(Name("_f"),
+                           Name("_f"),
+                           results["seq"].clone(),
+                           Name("_f"))
+
+        else:
+            if in_special_context(node):
+                return None
+            new = node.clone()
+            new.set_prefix("")
+            new = Call(Name("list"), [new])
+        new.set_prefix(node.get_prefix())
+        return new
+
+P0 = """for_stmt< 'for' any 'in' node=any ':' any* >
+        | comp_for< 'for' any 'in' node=any any* >
+     """
+p0 = patcomp.compile_pattern(P0)
+
+P1 = """
+power<
+    ( 'iter' | 'list' | 'tuple' | 'sorted' | 'set' | 'sum' |
+      'any' | 'all' | (any* trailer< '.' 'join' >) )
+    trailer< '(' node=any ')' >
+    any*
+>
+"""
+p1 = patcomp.compile_pattern(P1)
+
+P2 = """
+power<
+    'sorted'
+    trailer< '(' arglist<node=any any*> ')' >
+    any*
+>
+"""
+p2 = patcomp.compile_pattern(P2)
+
+def in_special_context(node):
+    patterns = [p0, p1, p2]
+    for pattern, parent in zip(patterns, attr_chain(node, "parent")):
+        results = {}
+        if pattern.match(parent, results) and results["node"] is node:
+            return True
+    return False

Lib/lib2to3/fixes/fix_funcattrs.py

+"""Fix function attribute names (f.func_x -> f.__x__)."""
+# Author: Collin Winter
+
+# Local imports
+from . import basefix
+from .util import Name
+
+
+class FixFuncattrs(basefix.BaseFix):
+    PATTERN = """
+    power< any+ trailer< '.' attr=('func_closure' | 'func_doc' | 'func_globals'
+                                  | 'func_name' | 'func_defaults' | 'func_code'
+                                  | 'func_dict') > any* >
+    """
+
+    def transform(self, node, results):
+        attr = results["attr"][0]
+        attr.replace(Name(("__%s__" % attr.value[5:]),
+                          prefix=attr.get_prefix()))

Lib/lib2to3/fixes/fix_future.py

+"""Remove __future__ imports
+
+from __future__ import foo is replaced with an empty line.
+"""
+# Author: Christian Heimes 
+
+# Local imports
+from . import basefix
+from .util import BlankLine 
+
+class FixFuture(basefix.BaseFix):
+    PATTERN = """import_from< 'from' module_name="__future__" 'import' any >"""
+
+    def transform(self, node, results):
+        return BlankLine()
+

Lib/lib2to3/fixes/fix_has_key.py

+# Copyright 2006 Google, Inc. All Rights Reserved.
+# Licensed to PSF under a Contributor Agreement.
+
+"""Fixer for has_key().
+
+Calls to .has_key() methods are expressed in terms of the 'in'
+operator:
+
+    d.has_key(k) -> k in d
+
+CAVEATS:
+1) While the primary target of this fixer is dict.has_key(), the
+   fixer will change any has_key() method call, regardless of its
+   class.
+
+2) Cases like this will not be converted:
+
+    m = d.has_key
+    if m(k):
+        ...
+        
+   Only *calls* to has_key() are converted. While it is possible to
+   convert the above to something like
+   
+    m = d.__contains__
+    if m(k):
+        ...
+        
+   this is currently not done.
+"""
+
+# Local imports
+from .. import pytree
+from ..pgen2 import token
+from . import basefix
+from .util import Name
+
+
+class FixHasKey(basefix.BaseFix):
+
+    PATTERN = """
+    anchor=power<
+        before=any+
+        trailer< '.' 'has_key' >
+        trailer<
+            '('
+            ( not(arglist | argument<any '=' any>) arg=any
+            | arglist<(not argument<any '=' any>) arg=any ','>
+            )
+            ')'
+        >
+        after=any*
+    >
+    |
+    negation=not_test<
+        'not'
+        anchor=power<
+            before=any+
+            trailer< '.' 'has_key' >
+            trailer<
+                '('
+                ( not(arglist | argument<any '=' any>) arg=any
+                | arglist<(not argument<any '=' any>) arg=any ','>
+                )
+                ')'
+            >
+        >
+    >
+    """
+
+    def transform(self, node, results):
+        assert results
+        syms = self.syms
+        if (node.parent.type == syms.not_test and
+            self.pattern.match(node.parent)):
+            # Don't transform a node matching the first alternative of the
+            # pattern when its parent matches the second alternative
+            return None
+        negation = results.get("negation")
+        anchor = results["anchor"]
+        prefix = node.get_prefix()
+        before = [n.clone() for n in results["before"]]
+        arg = results["arg"].clone()
+        after = results.get("after")
+        if after:
+            after = [n.clone() for n in after]
+        if arg.type in (syms.comparison, syms.not_test, syms.and_test,
+                        syms.or_test, syms.test, syms.lambdef, syms.argument):
+            arg = self.parenthesize(arg)
+        if len(before) == 1:
+            before = before[0]
+        else:
+            before = pytree.Node(syms.power, before)
+        before.set_prefix(" ")
+        n_op = Name("in", prefix=" ")
+        if negation:
+            n_not = Name("not", prefix=" ")
+            n_op = pytree.Node(syms.comp_op, (n_not, n_op))
+        new = pytree.Node(syms.comparison, (arg, n_op, before))
+        if after:
+            new = self.parenthesize(new)
+            new = pytree.Node(syms.power, (new,) + tuple(after))
+        if node.parent.type in (syms.comparison, syms.expr, syms.xor_expr,
+                                syms.and_expr, syms.shift_expr,
+                                syms.arith_expr, syms.term,
+                                syms.factor, syms.power):
+            new = self.parenthesize(new)
+        new.set_prefix(prefix)
+        return new

Lib/lib2to3/fixes/fix_idioms.py

+"""Adjust some old Python 2 idioms to their modern counterparts.
+
+* Change some type comparisons to isinstance() calls:
+    type(x) == T -> isinstance(x, T)
+    type(x) is T -> isinstance(x, T)
+    type(x) != T -> not isinstance(x, T)
+    type(x) is not T -> not isinstance(x, T)
+    
+* Change "while 1:" into "while True:".
+
+* Change both
+
+    v = list(EXPR)
+    v.sort()
+    foo(v)
+
+and the more general
+
+    v = EXPR
+    v.sort()
+    foo(v)
+    
+into
+
+    v = sorted(EXPR)
+    foo(v)
+"""
+# Author: Jacques Frechet, Collin Winter
+
+# Local imports
+from . import basefix
+from .util import Call, Comma, Name, Node, syms
+
+CMP = "(n='!=' | '==' | 'is' | n=comp_op< 'is' 'not' >)"
+TYPE = "power< 'type' trailer< '(' x=any ')' > >"
+
+class FixIdioms(basefix.BaseFix):
+
+    explicit = True # The user must ask for this fixer
+
+    PATTERN = r"""
+        isinstance=comparison< %s %s T=any >
+        |
+        isinstance=comparison< T=any %s %s >
+        |
+        while_stmt< 'while' while='1' ':' any+ >
+        |
+        sorted=any<
+            any*
+            simple_stmt<
+              expr_stmt< id1=any '='
+                         power< list='list' trailer< '(' (not arglist<any+>) any ')' > >
+              >
+              '\n'
+            >
+            sort=
+            simple_stmt<
+              power< id2=any
+                     trailer< '.' 'sort' > trailer< '(' ')' >
+              >
+              '\n'
+            >
+            next=any*
+        >
+        |
+        sorted=any<
+            any*
+            simple_stmt< expr_stmt< id1=any '=' expr=any > '\n' >
+            sort=
+            simple_stmt<
+              power< id2=any
+                     trailer< '.' 'sort' > trailer< '(' ')' >
+              >
+              '\n'
+            >
+            next=any*
+        >
+    """ % (TYPE, CMP, CMP, TYPE)
+
+    def match(self, node):
+        r = super(FixIdioms, self).match(node)
+        # If we've matched one of the sort/sorted subpatterns above, we
+        # want to reject matches where the initial assignment and the
+        # subsequent .sort() call involve different identifiers.
+        if r and "sorted" in r:
+            if r["id1"] == r["id2"]:
+                return r
+            return None
+        return r
+
+    def transform(self, node, results):
+        if "isinstance" in results:
+            return self.transform_isinstance(node, results)
+        elif "while" in results:
+            return self.transform_while(node, results)
+        elif "sorted" in results:
+            return self.transform_sort(node, results)
+        else:
+            raise RuntimeError("Invalid match")
+
+    def transform_isinstance(self, node, results):
+        x = results["x"].clone() # The thing inside of type()
+        T = results["T"].clone() # The type being compared against
+        x.set_prefix("")
+        T.set_prefix(" ")
+        test = Call(Name("isinstance"), [x, Comma(), T])
+        if "n" in results:
+            test.set_prefix(" ")
+            test = Node(syms.not_test, [Name("not"), test])
+        test.set_prefix(node.get_prefix())
+        return test
+
+    def transform_while(self, node, results):
+        one = results["while"]
+        one.replace(Name("True", prefix=one.get_prefix()))
+
+    def transform_sort(self, node, results):
+        sort_stmt = results["sort"]
+        next_stmt = results["next"]
+        list_call = results.get("list")
+        simple_expr = results.get("expr")
+
+        if list_call:
+            list_call.replace(Name("sorted", prefix=list_call.get_prefix()))
+        elif simple_expr:
+            new = simple_expr.clone()
+            new.set_prefix("")
+            simple_expr.replace(Call(Name("sorted"), [new],
+                                     prefix=simple_expr.get_prefix()))
+        else:
+            raise RuntimeError("should not have reached here")
+        sort_stmt.remove()
+        if next_stmt:
+            next_stmt[0].set_prefix(sort_stmt.get_prefix())

Lib/lib2to3/fixes/fix_imports.py

+"""Fix incompatible imports and module references.
+
+Fixes:
+  * StringIO -> io
+  * cStringIO -> io
+  * md5 -> hashlib
+"""
+# Author: Collin Winter
+
+# Local imports
+from . import basefix
+from .util import Name, attr_chain, any, set
+import __builtin__
+builtin_names = [name for name in dir(__builtin__)
+                 if name not in ("__name__", "__doc__")]
+
+MAPPING = {"StringIO":  ("io", ["StringIO"]),
+           "cStringIO": ("io", ["StringIO"]),
+           "__builtin__" : ("builtins", builtin_names), 
+          }
+
+
+def alternates(members):
+    return "(" + "|".join(map(repr, members)) + ")"
+
+
+def build_pattern():
+    bare = set()
+    for old_module, (new_module, members) in MAPPING.items():
+        bare.add(old_module)
+        bare.update(members)
+        members = alternates(members)
+        yield """import_name< 'import' (module=%r
+                              | dotted_as_names< any* module=%r any* >) >
+              """ % (old_module, old_module)
+        yield """import_from< 'from' module_name=%r 'import'
+                   ( %s | import_as_name< %s 'as' any >) >
+              """ % (old_module, members, members)
+        yield """import_from< 'from' module_name=%r 'import' star='*' >
+              """ % old_module
+        yield """import_name< 'import'
+                              dotted_as_name< module_name=%r 'as' any > >
+              """ % old_module
+        yield """power< module_name=%r trailer< '.' %s > any* >
+              """ % (old_module, members)
+    yield """bare_name=%s""" % alternates(bare)
+
+
+class FixImports(basefix.BaseFix):
+    PATTERN = "|".join(build_pattern())
+
+    order = "pre" # Pre-order tree traversal
+
+    # Don't match the node if it's within another match
+    def match(self, node):
+        match = super(FixImports, self).match
+        results = match(node)
+        if results:
+            if any([match(obj) for obj in attr_chain(node, "parent")]):
+                return False
+            return results
+        return False
+
+    def start_tree(self, tree, filename):
+        super(FixImports, self).start_tree(tree, filename)
+        self.replace = {}
+
+    def transform(self, node, results):
+        import_mod = results.get("module")
+        mod_name = results.get("module_name")
+        bare_name = results.get("bare_name")
+        star = results.get("star")
+
+        if import_mod or mod_name:
+            new_name, members = MAPPING[(import_mod or mod_name).value]
+
+        if import_mod:
+            self.replace[import_mod.value] = new_name
+            import_mod.replace(Name(new_name, prefix=import_mod.get_prefix()))
+        elif mod_name:
+            if star:
+                self.cannot_convert(node, "Cannot handle star imports.")
+            else:
+                mod_name.replace(Name(new_name, prefix=mod_name.get_prefix()))
+        elif bare_name:
+            bare_name = bare_name[0]
+            new_name = self.replace.get(bare_name.value)
+            if new_name:
+              bare_name.replace(Name(new_name, prefix=bare_name.get_prefix()))

Lib/lib2to3/fixes/fix_input.py

+"""Fixer that changes input(...) into eval(input(...))."""
+# Author: Andre Roberge
+
+# Local imports
+from . import basefix
+from .util import Call, Name
+from .. import patcomp
+
+
+context = patcomp.compile_pattern("power< 'eval' trailer< '(' any ')' > >")
+
+
+class FixInput(basefix.BaseFix):
+
+    PATTERN = """
+              power< 'input' args=trailer< '(' [any] ')' > >
+              """
+
+    def transform(self, node, results):
+        # If we're already wrapped in a eval() call, we're done.
+        if context.match(node.parent.parent):
+            return
+
+        new = node.clone()
+        new.set_prefix("")
+        return Call(Name("eval"), [new], prefix=node.get_prefix())

Lib/lib2to3/fixes/fix_intern.py

+# Copyright 2006 Georg Brandl.
+# Licensed to PSF under a Contributor Agreement.
+
+"""Fixer for intern().
+
+intern(s) -> sys.intern(s)"""
+
+# Local imports
+from .. import pytree
+from . import basefix
+from .util import Name, Attr
+
+
+class FixIntern(basefix.BaseFix):
+
+    PATTERN = """
+    power< 'intern'
+           trailer< lpar='('
+                    ( not(arglist | argument<any '=' any>) obj=any
+                      | obj=arglist<(not argument<any '=' any>) any ','> )
+                    rpar=')' >
+           after=any*
+    >
+    """
+
+    def transform(self, node, results):
+        syms = self.syms
+        obj = results["obj"].clone()
+        if obj.type == syms.arglist:
+            newarglist = obj.clone()
+        else:
+            newarglist = pytree.Node(syms.arglist, [obj.clone()])
+        after = results["after"]
+        if after:
+            after = [n.clone() for n in after]
+        new = pytree.Node(syms.power,
+                          Attr(Name("sys"), Name("intern")) +
+                          [pytree.Node(syms.trailer,
+                                       [results["lpar"].clone(),
+                                        newarglist,
+                                        results["rpar"].clone()])] + after)
+        new.set_prefix(node.get_prefix())
+        return new

Lib/lib2to3/fixes/fix_itertools.py

+""" Fixer for itertools.(imap|ifilter|izip) --> (map|filter|zip) and
+    itertools.ifilterfalse --> itertools.filterfalse (bugs 2360-2363)
+
+    If itertools is imported as something else (ie: import itertools as it;
+    it.izip(spam, eggs)) method calls will not get fixed.
+    """
+
+# Local imports
+from . import basefix
+from .util import Name
+
+class FixItertools(basefix.BaseFix):
+    it_funcs = "('imap'|'ifilter'|'izip'|'ifilterfalse')"
+    PATTERN = """
+              power< it='itertools'
+                  trailer<
+                     dot='.' func=%(it_funcs)s > trailer< '(' [any] ')' > >
+              |
+              power< func=%(it_funcs)s trailer< '(' [any] ')' > >
+              """ %(locals())
+
+    def transform(self, node, results):
+        prefix = None
+        func = results['func'][0]
+        if 'it' in results and func.value != 'ifilterfalse':
+            dot, it = (results['dot'], results['it'])
+            # Remove the 'itertools'
+            prefix = it.get_prefix()
+            it.remove()
+            # Replace the node wich contains ('.', 'function') with the
+            # function (to be consistant with the second part of the pattern)
+            dot.remove()
+            func.parent.replace(func)
+
+        prefix = prefix or func.get_prefix()
+        func.replace(Name(func.value[1:], prefix=prefix))

Lib/lib2to3/fixes/fix_long.py

+# Copyright 2006 Google, Inc. All Rights Reserved.
+# Licensed to PSF under a Contributor Agreement.
+
+"""Fixer that turns 'long' into 'int' everywhere.
+
+This also strips the trailing 'L' or 'l' from long loterals.
+"""
+
+# Local imports
+from .. import pytree
+from . import basefix
+from .util import Name, Number
+
+
+class FixLong(basefix.BaseFix):
+
+    PATTERN = """
+    (long_type = 'long' | number = NUMBER)
+    """
+
+    static_long = Name("long")
+    static_int = Name("int")
+
+    def transform(self, node, results):
+        long_type = results.get("long_type")
+        number = results.get("number")
+        new = None
+        if long_type:
+            assert node == self.static_long, node
+            new = self.static_int.clone()
+        if number and node.value[-1] in ("l", "L"):
+            new = Number(node.value[:-1])
+        if new is not None:
+            new.set_prefix(node.get_prefix())
+            return new

Lib/lib2to3/fixes/fix_map.py

+# Copyright 2007 Google, Inc. All Rights Reserved.
+# Licensed to PSF under a Contributor Agreement.
+
+"""Fixer that changes map(F, ...) into list(map(F, ...)) unless there
+exists a 'from future_builtins import map' statement in the top-level
+namespace.
+
+As a special case, map(None, X) is changed into list(X).  (This is
+necessary because the semantics are changed in this case -- the new
+map(None, X) is equivalent to [(x,) for x in X].)
+
+We avoid the transformation (except for the special case mentioned
+above) if the map() call is directly contained in iter(<>), list(<>),
+tuple(<>), sorted(<>), ...join(<>), or for V in <>:.
+
+NOTE: This is still not correct if the original code was depending on
+map(F, X, Y, ...) to go on until the longest argument is exhausted,
+substituting None for missing values -- like zip(), it now stops as
+soon as the shortest argument is exhausted.
+"""
+
+# Local imports
+from .. import pytree
+from .. import patcomp
+from ..pgen2 import token
+from . import basefix
+from .util import Name, Call, ListComp, attr_chain, does_tree_import
+from ..pygram import python_symbols as syms
+
+class FixMap(basefix.BaseFix):
+
+    PATTERN = """
+    map_none=power<
+        'map'
+        trailer< '(' arglist< 'None' ',' arg=any [','] > ')' >
+    >
+    |
+    map_lambda=power<
+        'map'
+        trailer<
+            '('
+            arglist<
+                lambdef< 'lambda'
+                         (fp=NAME | vfpdef< '(' fp=NAME ')'> ) ':' xp=any
+                >
+                ','
+                it=any
+            >
+            ')'
+        >
+    >
+    |
+    power<
+        'map'
+        args=trailer< '(' [any] ')' >
+    >
+    """
+
+    def start_tree(self, *args):
+        super(FixMap, self).start_tree(*args)
+        self._future_map_found = None
+
+    def has_future_map(self, node):
+        if self._future_map_found is not None:
+            return self._future_map_found
+        self._future_map_found = does_tree_import('future_builtins', 'map', node)
+        return self._future_map_found
+
+    def transform(self, node, results):
+        if self.has_future_map(node):
+            # If a future map has been imported for this file, we won't
+            # be making any modifications
+            return
+            
+        if node.parent.type == syms.simple_stmt:
+            self.warning(node, "You should use a for loop here")
+            new = node.clone()
+            new.set_prefix("")
+            new = Call(Name("list"), [new])
+        elif "map_lambda" in results:
+            new = ListComp(results.get("xp").clone(),
+                           results.get("fp").clone(),
+                           results.get("it").clone())
+        else:
+            if "map_none" in results:
+                new = results["arg"].clone()
+            else:
+                if in_special_context(node):
+                    return None
+                new = node.clone()
+            new.set_prefix("")
+            new = Call(Name("list"), [new])
+        new.set_prefix(node.get_prefix())
+        return new
+
+P0 = """for_stmt< 'for' any 'in' node=any ':' any* >
+        | comp_for< 'for' any 'in' node=any any* >
+     """
+p0 = patcomp.compile_pattern(P0)
+
+P1 = """
+power<
+    ( 'iter' | 'list' | 'tuple' | 'sorted' | 'set' | 'sum' |
+      'any' | 'all' | (any* trailer< '.' 'join' >) )
+    trailer< '(' node=any ')' >
+    any*
+>
+"""
+p1 = patcomp.compile_pattern(P1)
+
+P2 = """
+power<
+    'sorted'
+    trailer< '(' arglist<node=any any*> ')' >
+    any*
+>
+"""
+p2 = patcomp.compile_pattern(P2)
+
+def in_special_context(node):
+    patterns = [p0, p1, p2]
+    for pattern, parent in zip(patterns, attr_chain(node, "parent")):
+        results = {}
+        if pattern.match(parent, results) and results["node"] is node:
+            return True
+    return False

Lib/lib2to3/fixes/fix_methodattrs.py

+"""Fix bound method attributes (method.im_? -> method.__?__).
+"""
+# Author: Christian Heimes
+
+# Local imports
+from . import basefix
+from .util import Name
+
+MAP = {
+    "im_func" : "__func__",
+    "im_self" : "__self__",
+    "im_class" : "__self__.__class__"
+    }
+
+class FixMethodattrs(basefix.BaseFix):
+    PATTERN = """
+    power< any+ trailer< '.' attr=('im_func' | 'im_self' | 'im_class') > any* >
+    """
+
+    def transform(self, node, results):
+        attr = results["attr"][0]
+        new = MAP[attr.value]
+        attr.replace(Name(new, prefix=attr.get_prefix()))

Lib/lib2to3/fixes/fix_ne.py

+# Copyright 2006 Google, Inc. All Rights Reserved.
+# Licensed to PSF under a Contributor Agreement.
+
+"""Fixer that turns <> into !=."""
+
+# Local imports
+from .. import pytree
+from ..pgen2 import token
+from . import basefix
+
+
+class FixNe(basefix.BaseFix):
+    # This is so simple that we don't need the pattern compiler.
+
+    def match(self, node):
+        # Override
+        return node.type == token.NOTEQUAL and node.value == "<>"
+
+    def transform(self, node, results):
+      new = pytree.Leaf(token.NOTEQUAL, "!=")
+      new.set_prefix(node.get_prefix())
+      return new

Lib/lib2to3/fixes/fix_next.py

+"""Fixer for it.next() -> next(it), per PEP 3114."""
+# Author: Collin Winter
+
+# Things that currently aren't covered:
+#   - listcomp "next" names aren't warned
+#   - "with" statement targets aren't checked
+
+# Local imports
+from ..pgen2 import token
+from ..pygram import python_symbols as syms
+from . import basefix
+from .util import Name, Call, find_binding, any
+
+bind_warning = "Calls to builtin next() possibly shadowed by global binding"
+
+
+class FixNext(basefix.BaseFix):
+    PATTERN = """
+    power< base=any+ trailer< '.' attr='next' > trailer< '(' ')' > >
+    |
+    power< head=any+ trailer< '.' attr='next' > not trailer< '(' ')' > >
+    |
+    classdef< 'class' any+ ':'
+              suite< any*
+                     funcdef< 'def'
+                              name='next'
+                              parameters< '(' NAME ')' > any+ >
+                     any* > >
+    |
+    global=global_stmt< 'global' any* 'next' any* >
+    |
+    mod=file_input< any+ >
+    """
+
+    order = "pre" # Pre-order tree traversal
+
+    def start_tree(self, tree, filename):
+        super(FixNext, self).start_tree(tree, filename)
+        self.shadowed_next = False
+
+    def transform(self, node, results):
+        assert results
+
+        base = results.get("base")
+        attr = results.get("attr")
+        name = results.get("name")
+        mod = results.get("mod")
+
+        if base:
+          if self.shadowed_next:
+              attr.replace(Name("__next__", prefix=attr.get_prefix()))
+          else:
+              base = [n.clone() for n in base]
+              base[0].set_prefix("")
+              node.replace(Call(Name("next", prefix=node.get_prefix()), base))
+        elif name:
+            n = Name("__next__", prefix=name.get_prefix())
+            name.replace(n)
+        elif attr:
+            # We don't do this transformation if we're assigning to "x.next".
+            # Unfortunately, it doesn't seem possible to do this in PATTERN,
+            #  so it's being done here.
+            if is_assign_target(node):
+                head = results["head"]
+                if "".join([str(n) for n in head]).strip() == '__builtin__':
+                    self.warning(node, bind_warning)
+                return
+            attr.replace(Name("__next__"))
+        elif "global" in results:
+            self.warning(node, bind_warning)
+            self.shadowed_next = True
+        elif mod:
+            n = find_binding('next', mod)
+            if n:
+                self.warning(n, bind_warning)
+                self.shadowed_next = True
+
+
+### The following functions help test if node is part of an assignment
+###  target.
+
+def is_assign_target(node):
+    assign = find_assign(node)
+    if assign is None:
+        return False
+
+    for child in assign.children:
+        if child.type == token.EQUAL:
+            return False
+        elif is_subtree(child, node):
+            return True
+    return False
+
+def find_assign(node):
+    if node.type == syms.expr_stmt:
+        return node
+    if node.type == syms.simple_stmt or node.parent is None:
+        return None
+    return find_assign(node.parent)
+
+def is_subtree(root, node):
+    if root == node:
+        return True
+    return any([is_subtree(c, node) for c in root.children])

Lib/lib2to3/fixes/fix_nonzero.py

+"""Fixer for __nonzero__ -> __bool__ methods."""
+# Author: Collin Winter
+
+# Local imports
+from .import basefix
+from .util import Name, syms
+
+class FixNonzero(basefix.BaseFix):
+    PATTERN = """
+    classdef< 'class' any+ ':'
+              suite< any*
+                     funcdef< 'def' name='__nonzero__'
+                              parameters< '(' NAME ')' > any+ >
+                     any* > >
+    """
+
+    def transform(self, node, results):
+        name = results["name"]
+        new = Name("__bool__", prefix=name.get_prefix())
+        name.replace(new)

Lib/lib2to3/fixes/fix_numliterals.py

+"""Fixer that turns 1L into 1, 0755 into 0o755.
+"""
+# Copyright 2007 Georg Brandl.
+# Licensed to PSF under a Contributor Agreement.
+
+# Local imports
+from ..pgen2 import token
+from .import basefix
+from .util import Number, set
+
+
+class FixNumliterals(basefix.BaseFix):
+    # This is so simple that we don't need the pattern compiler.
+
+    def match(self, node):
+        # Override
+        return (node.type == token.NUMBER and
+                (node.value.startswith("0") or node.value[-1] in "Ll"))
+
+    def transform(self, node, results):
+        val = node.value
+        if val[-1] in 'Ll':
+            val = val[:-1]
+        elif val.startswith('0') and val.isdigit() and len(set(val)) > 1:
+            val = "0o" + val[1:]
+
+        return Number(val, prefix=node.get_prefix())

Lib/lib2to3/fixes/fix_print.py

+# Copyright 2006 Google, Inc. All Rights Reserved.
+# Licensed to PSF under a Contributor Agreement.
+
+"""Fixer for print.
+
+Change:
+    'print'          into 'print()'
+    'print ...'	     into 'print(...)'
+    'print ... ,'    into 'print(..., end=" ")'
+    'print >>x, ...' into 'print(..., file=x)'
+"""
+
+# Local imports
+from .. import patcomp
+from .. import pytree
+from ..pgen2 import token
+from .import basefix
+from .util import Name, Call, Comma, String, is_tuple
+
+
+parend_expr = patcomp.compile_pattern(
+              """atom< '(' [atom|STRING|NAME] ')' >"""
+              )
+
+
+class FixPrint(basefix.BaseFix):
+
+    PATTERN = """
+              simple_stmt< bare='print' any > | print_stmt
+              """
+
+    def transform(self, node, results):
+        assert results
+        bare_print = results.get("bare")
+
+        if bare_print:
+            # Special-case print all by itself
+            bare_print.replace(Call(Name("print"), [],
+                               prefix=bare_print.get_prefix()))
+            return
+        assert node.children[0] == Name("print")
+        args = node.children[1:]
+        if len(args) == 1 and parend_expr.match(args[0]):
+            # We don't want to keep sticking parens around an
+            # already-parenthesised expression.
+            return
+
+        sep = end = file = None
+        if args and args[-1] == Comma():
+            args = args[:-1]
+            end = " "
+        if args and args[0] == pytree.Leaf(token.RIGHTSHIFT, ">>"):
+            assert len(args) >= 2
+            file = args[1].clone()
+            args = args[3:] # Strip a possible comma after the file expression
+        # Now synthesize a print(args, sep=..., end=..., file=...) node.
+        l_args = [arg.clone() for arg in args]
+        if l_args:
+            l_args[0].set_prefix("")
+        if sep is not None or end is not None or file is not None:
+            if sep is not None:
+                self.add_kwarg(l_args, "sep", String(repr(sep)))
+            if end is not None:
+                self.add_kwarg(l_args, "end", String(repr(end)))
+            if file is not None:
+                self.add_kwarg(l_args, "file", file)
+        n_stmt = Call(Name("print"), l_args)
+        n_stmt.set_prefix(node.get_prefix())
+        return n_stmt
+
+    def add_kwarg(self, l_nodes, s_kwd, n_expr):
+        # XXX All this prefix-setting may lose comments (though rarely)
+        n_expr.set_prefix("")
+        n_argument = pytree.Node(self.syms.argument,
+                                 (Name(s_kwd),
+                                  pytree.Leaf(token.EQUAL, "="),
+                                  n_expr))
+        if l_nodes:
+            l_nodes.append(Comma())
+            n_argument.set_prefix(" ")
+        l_nodes.append(n_argument)

Lib/lib2to3/fixes/fix_raise.py

+"""Fixer for 'raise E, V, T'
+
+raise         -> raise
+raise E       -> raise E
+raise E, V    -> raise E(V)
+raise E, V, T -> raise E(V).with_traceback(T)
+
+raise (((E, E'), E''), E'''), V -> raise E(V)
+raise "foo", V, T               -> warns about string exceptions
+
+
+CAVEATS:
+1) "raise E, V" will be incorrectly translated if V is an exception
+   instance. The correct Python 3 idiom is
+   
+        raise E from V
+        
+   but since we can't detect instance-hood by syntax alone and since
+   any client code would have to be changed as well, we don't automate
+   this.
+"""
+# Author: Collin Winter
+
+# Local imports
+from .. import pytree
+from ..pgen2 import token
+from .import basefix
+from .util import Name, Call, Attr, ArgList, is_tuple
+
+class FixRaise(basefix.BaseFix):
+
+    PATTERN = """
+    raise_stmt< 'raise' exc=any [',' val=any [',' tb=any]] >
+    """
+
+    def transform(self, node, results):
+        syms = self.syms
+
+        exc = results["exc"].clone()
+        if exc.type is token.STRING:
+            self.cannot_convert(node, "Python 3 does not support string exceptions")
+            return
+
+        # Python 2 supports
+        #  raise ((((E1, E2), E3), E4), E5), V
+        # as a synonym for
+        #  raise E1, V
+        # Since Python 3 will not support this, we recurse down any tuple
+        # literals, always taking the first element.
+        if is_tuple(exc):
+          while is_tuple(exc):
+              # exc.children[1:-1] is the unparenthesized tuple
+              # exc.children[1].children[0] is the first element of the tuple
+              exc = exc.children[1].children[0].clone()
+          exc.set_prefix(" ")
+
+        if "val" not in results:
+            # One-argument raise
+            new = pytree.Node(syms.raise_stmt, [Name("raise"), exc])
+            new.set_prefix(node.get_prefix())
+            return new
+
+        val = results["val"].clone()
+        if is_tuple(val):
+            args = [c.clone() for c in val.children[1:-1]]
+        else:
+            val.set_prefix("")
+            args = [val]
+
+        if "tb" in results:
+            tb = results["tb"].clone()
+            tb.set_prefix("")
+
+            e = Call(exc, args)
+            with_tb = Attr(e, Name('with_traceback')) + [ArgList([tb])]
+            new = pytree.Node(syms.simple_stmt, [Name("raise")] + with_tb)
+            new.set_prefix(node.get_prefix())
+            return new
+        else:
+            return pytree.Node(syms.raise_stmt,
+                               [Name("raise"), Call(exc, args)],
+                               prefix=node.get_prefix())

Lib/lib2to3/fixes/fix_raw_input.py

+"""Fixer that changes raw_input(...) into input(...)."""
+# Author: Andre Roberge
+
+# Local imports
+from .import basefix
+from .util import Name
+
+class FixRawInput(basefix.BaseFix):
+
+    PATTERN = """
+              power< name='raw_input' trailer< '(' [any] ')' > >
+              """
+
+    def transform(self, node, results):
+        name = results["name"]
+        name.replace(Name("input", prefix=name.get_prefix()))

Lib/lib2to3/fixes/fix_renames.py

+"""Fix incompatible renames
+
+Fixes:
+  * sys.maxint -> sys.maxsize
+"""
+# Author: Christian Heimes
+# based on Collin Winter's fix_import
+
+# Local imports
+from .import basefix
+from .util import Name, attr_chain, any, set
+
+MAPPING = {"sys":  {"maxint" : "maxsize"},
+          }
+LOOKUP = {}
+
+def alternates(members):
+    return "(" + "|".join(map(repr, members)) + ")"
+
+
+def build_pattern():
+    #bare = set()
+    for module, replace in MAPPING.items():
+        for old_attr, new_attr in replace.items():
+            LOOKUP[(module, old_attr)] = new_attr
+            #bare.add(module)
+            #bare.add(old_attr)
+            #yield """
+            #      import_name< 'import' (module=%r
+            #          | dotted_as_names< any* module=%r any* >) >
+            #      """ % (module, module)
+            yield """
+                  import_from< 'from' module_name=%r 'import'
+                      ( attr_name=%r | import_as_name< attr_name=%r 'as' any >) >
+                  """ % (module, old_attr, old_attr)
+            yield """
+                  power< module_name=%r trailer< '.' attr_name=%r > any* >
+                  """ % (module, old_attr)
+    #yield """bare_name=%s""" % alternates(bare)
+
+
+class FixRenames(basefix.BaseFix):
+    PATTERN = "|".join(build_pattern())
+
+    order = "pre" # Pre-order tree traversal
+
+    # Don't match the node if it's within another match
+    def match(self, node):
+        match = super(FixRenames, self).match
+        results = match(node)
+        if results:
+            if any([match(obj) for obj in attr_chain(node, "parent")]):
+                return False
+            return results
+        return False
+
+    #def start_tree(self, tree, filename):
+    #    super(FixRenames, self).start_tree(tree, filename)
+    #    self.replace = {}
+
+    def transform(self, node, results):
+        mod_name = results.get("module_name")
+        attr_name = results.get("attr_name")
+        #bare_name = results.get("bare_name")
+        #import_mod = results.get("module")
+      
+        if mod_name and attr_name:
+            new_attr = LOOKUP[(mod_name.value, attr_name.value)]
+            attr_name.replace(Name(new_attr, prefix=attr_name.get_prefix()))
+

Lib/lib2to3/fixes/fix_repr.py

+# Copyright 2006 Google, Inc. All Rights Reserved.
+# Licensed to PSF under a Contributor Agreement.
+
+"""Fixer that transforms `xyzzy` into repr(xyzzy)."""
+
+# Local imports
+from .import basefix
+from .util import Call, Name
+
+
+class FixRepr(basefix.BaseFix):
+
+    PATTERN = """
+              atom < '`' expr=any '`' >
+              """
+
+    def transform(self, node, results):
+      expr = results["expr"].clone()
+
+      if expr.type == self.syms.testlist1:
+          expr = self.parenthesize(expr)
+      return Call(Name("repr"), [expr], prefix=node.get_prefix())

Lib/lib2to3/fixes/fix_standarderror.py

+# Copyright 2007 Google, Inc. All Rights Reserved.
+# Licensed to PSF under a Contributor Agreement.
+
+"""Fixer for StandardError -> Exception."""
+
+# Local imports
+from .import basefix
+from .util import Name
+
+
+class FixStandarderror(basefix.BaseFix):
+
+    PATTERN = """
+              'StandardError'
+              """
+
+    def transform(self, node, results):
+        return Name("Exception", prefix=node.get_prefix())

Lib/lib2to3/fixes/fix_throw.py

+"""Fixer for generator.throw(E, V, T).
+
+g.throw(E)       -> g.throw(E)
+g.throw(E, V)    -> g.throw(E(V))
+g.throw(E, V, T) -> g.throw(E(V).with_traceback(T))
+
+g.throw("foo"[, V[, T]]) will warn about string exceptions."""
+# Author: Collin Winter
+
+# Local imports
+from .. import pytree
+from ..pgen2 import token
+from .import basefix
+from .util import Name, Call, ArgList, Attr, is_tuple
+
+class FixThrow(basefix.BaseFix):
+
+    PATTERN = """
+    power< any trailer< '.' 'throw' >
+           trailer< '(' args=arglist< exc=any ',' val=any [',' tb=any] > ')' >
+    >
+    |
+    power< any trailer< '.' 'throw' > trailer< '(' exc=any ')' > >
+    """
+
+    def transform(self, node, results):
+        syms = self.syms
+
+        exc = results["exc"].clone()
+        if exc.type is token.STRING:
+            self.cannot_convert(node, "Python 3 does not support string exceptions")
+            return
+
+        # Leave "g.throw(E)" alone
+        val = results.get("val")
+        if val is None:
+            return
+
+        val = val.clone()
+        if is_tuple(val):
+            args = [c.clone() for c in val.children[1:-1]]
+        else:
+            val.set_prefix("")
+            args = [val]
+
+        throw_args = results["args"]
+
+        if "tb" in results:
+            tb = results["tb"].clone()
+            tb.set_prefix("")
+
+            e = Call(exc, args)
+            with_tb = Attr(e, Name('with_traceback')) + [ArgList([tb])]
+            throw_args.replace(pytree.Node(syms.power, with_tb))
+        else:
+            throw_args.replace(Call(exc, args))

Lib/lib2to3/fixes/fix_tuple_params.py

+"""Fixer for function definitions with tuple parameters.
+
+def func(((a, b), c), d):
+    ...
+    
+    ->
+
+def func(x, d):
+    ((a, b), c) = x
+    ...
+
+It will also support lambdas:
+  
+    lambda (x, y): x + y -> lambda t: t[0] + t[1]
+
+    # The parens are a syntax error in Python 3
+    lambda (x): x + y -> lambda x: x + y
+"""
+# Author: Collin Winter
+
+# Local imports
+from .. import pytree
+from ..pgen2 import token
+from .import basefix
+from .util import Assign, Name, Newline, Number, Subscript, syms
+
+def is_docstring(stmt):
+    return isinstance(stmt, pytree.Node) and \
+           stmt.children[0].type == token.STRING
+
+class FixTupleParams(basefix.BaseFix):
+    PATTERN = """
+              funcdef< 'def' any parameters< '(' args=any ')' >
+                       ['->' any] ':' suite=any+ >
+              |
+              lambda=
+              lambdef< 'lambda' args=vfpdef< '(' inner=any ')' >
+                       ':' body=any
+              >
+              """
+
+    def transform(self, node, results):
+        if "lambda" in results:
+            return self.transform_lambda(node, results)
+
+        new_lines = []
+        suite = results["suite"]
+        args = results["args"]
+        # This crap is so "def foo(...): x = 5; y = 7" is handled correctly.
+        # TODO(cwinter): suite-cleanup
+        if suite[0].children[1].type == token.INDENT:
+            start = 2
+            indent = suite[0].children[1].value
+            end = Newline()
+        else:
+            start = 0
+            indent = "; "
+            end = pytree.Leaf(token.INDENT, "")
+
+        # We need access to self for new_name(), and making this a method
+        #  doesn't feel right. Closing over self and new_lines makes the
+        #  code below cleaner.
+        def handle_tuple(tuple_arg, add_prefix=False):
+            n = Name(self.new_name())
+            arg = tuple_arg.clone()
+            arg.set_prefix("")
+            stmt = Assign(arg, n.clone())
+            if add_prefix:
+                n.set_prefix(" ")
+            tuple_arg.replace(n)
+            new_lines.append(pytree.Node(syms.simple_stmt,
+                                         [stmt, end.clone()]))
+
+        if args.type == syms.tfpdef:
+            handle_tuple(args)
+        elif args.type == syms.typedargslist:
+            for i, arg in enumerate(args.children):
+                if arg.type == syms.tfpdef:
+                    # Without add_prefix, the emitted code is correct,
+                    #  just ugly.
+                    handle_tuple(arg, add_prefix=(i > 0))
+
+        if not new_lines:
+            return node
+
+        # This isn't strictly necessary, but it plays nicely with other fixers.
+        # TODO(cwinter) get rid of this when children becomes a smart list
+        for line in new_lines:
+            line.parent = suite[0]
+
+        # TODO(cwinter) suite-cleanup
+        after = start
+        if start == 0:
+            new_lines[0].set_prefix(" ")
+        elif is_docstring(suite[0].children[start]):
+            new_lines[0].set_prefix(indent)
+            after = start + 1
+
+        suite[0].children[after:after] = new_lines
+        for i in range(after+1, after+len(new_lines)+1):
+            suite[0].children[i].set_prefix(indent)
+        suite[0].changed()
+
+    def transform_lambda(self, node, results):
+        args = results["args"]
+        body = results["body"]
+        inner = simplify_args(results["inner"])
+
+        # Replace lambda ((((x)))): x  with lambda x: x
+        if inner.type == token.NAME:
+            inner = inner.clone()
+            inner.set_prefix(" ")
+            args.replace(inner)
+            return
+
+        params = find_params(args)
+        to_index = map_to_index(params)
+        tup_name = self.new_name(tuple_name(params))
+
+        new_param = Name(tup_name, prefix=" ")
+        args.replace(new_param.clone())
+        for n in body.post_order():
+            if n.type == token.NAME and n.value in to_index:
+                subscripts = [c.clone() for c in to_index[n.value]]
+                new = pytree.Node(syms.power,
+                                  [new_param.clone()] + subscripts)
+                new.set_prefix(n.get_prefix())
+                n.replace(new)
+
+
+### Helper functions for transform_lambda()
+
+def simplify_args(node):
+    if node.type in (syms.vfplist, token.NAME):
+        return node
+    elif node.type == syms.vfpdef:
+        # These look like vfpdef< '(' x ')' > where x is NAME
+        # or another vfpdef instance (leading to recursion).
+        while node.type == syms.vfpdef:
+            node = node.children[1]
+        return node
+    raise RuntimeError("Received unexpected node %s" % node)
+
+def find_params(node):
+    if node.type == syms.vfpdef:
+        return find_params(node.children[1])
+    elif node.type == token.NAME:
+        return node.value
+    return [find_params(c) for c in node.children if c.type != token.COMMA]
+
+def map_to_index(param_list, prefix=[], d=None):
+    if d is None:
+        d = {}
+    for i, obj in enumerate(param_list):
+        trailer = [Subscript(Number(i))]
+        if isinstance(obj, list):
+            map_to_index(obj, trailer, d=d)
+        else:
+            d[obj] = prefix + trailer
+    return d
+
+def tuple_name(param_list):
+    l = []
+    for obj in param_list:
+        if isinstance(obj, list):
+            l.append(tuple_name(obj))
+        else:
+            l.append(obj)
+    return "_".join(l)

Lib/lib2to3/fixes/fix_types.py

+# Copyright 2007 Google, Inc. All Rights Reserved.
+# Licensed to PSF under a Contributor Agreement.
+
+"""Fixer for removing uses of the types module.
+
+These work for only the known names in the types module.  The forms above
+can include types. or not.  ie, It is assumed the module is imported either as:
+
+    import types
+    from types import ... # either * or specific types
+
+The import statements are not modified.
+
+There should be another fixer that handles at least the following constants:
+
+   type([]) -> list
+   type(()) -> tuple
+   type('') -> str
+
+"""
+
+# Local imports
+from ..pgen2 import token
+from .import basefix
+from .util import Name
+
+_TYPE_MAPPING = {
+        'BooleanType' : 'bool',
+        'BufferType' : 'memoryview',
+        'ClassType' : 'type',
+        'ComplexType' : 'complex',
+        'DictType': 'dict',
+        'DictionaryType' : 'dict',
+        'EllipsisType' : 'type(Ellipsis)',
+        #'FileType' : 'io.IOBase',
+        'FloatType': 'float',
+        'IntType': 'int',
+        'ListType': 'list',
+        'LongType': 'int',
+        'ObjectType' : 'object',
+        'NoneType': 'type(None)',
+        'NotImplementedType' : 'type(NotImplemented)',
+        'SliceType' : 'slice',
+        'StringType': 'bytes', # XXX ?
+        'StringTypes' : 'str', # XXX ?
+        'TupleType': 'tuple',
+        'TypeType' : 'type',
+        'UnicodeType': 'str',
+        'XRangeType' : 'range',
+    }
+
+_pats = ["power< 'types' trailer< '.' name='%s' > >" % t for t in _TYPE_MAPPING]
+
+class FixTypes(basefix.BaseFix):
+
+    PATTERN = '|'.join(_pats)
+
+    def transform(self, node, results):
+        new_value = _TYPE_MAPPING.get(results["name"].value)
+        if new_value:
+            return Name(new_value, prefix=node.get_prefix())
+        return None

Lib/lib2to3/fixes/fix_unicode.py

+"""Fixer that changes unicode to str, unichr to chr, and u"..." into "...".
+
+"""
+
+import re
+from ..pgen2 import token
+from .import basefix
+
+class FixUnicode(basefix.BaseFix):
+
+  PATTERN = "STRING | NAME<'unicode' | 'unichr'>"
+
+  def transform(self, node, results):
+    if node.type == token.NAME:
+      if node.value == "unicode":
+        new = node.clone()
+        new.value = "str"
+        return new
+      if node.value == "unichr":
+        new = node.clone()
+        new.value = "chr"
+        return new
+      # XXX Warn when __unicode__ found?
+    elif node.type == token.STRING:
+      if re.match(r"[uU][rR]?[\'\"]", node.value):
+        new = node.clone()
+        new.value = new.value[1:]
+        return new

Lib/lib2to3/fixes/fix_ws_comma.py

+"""Fixer that changes 'a ,b' into 'a, b'.
+
+This also changes '{a :b}' into '{a: b}', but does not touch other
+uses of colons.  It does not touch other uses of whitespace.
+
+"""
+
+from .. import pytree
+from ..pgen2 import token
+from .import basefix
+
+class FixWsComma(basefix.BaseFix):
+
+  explicit = True # The user must ask for this fixers
+
+  PATTERN = """
+  any<(not(',') any)+ ',' ((not(',') any)+ ',')* [not(',') any]>
+  """
+
+  COMMA = pytree.Leaf(token.COMMA, ",")
+  COLON = pytree.Leaf(token.COLON, ":")
+  SEPS = (COMMA, COLON)
+
+  def transform(self, node, results):
+    new = node.clone()
+    comma = False
+    for child in new.children:
+      if child in self.SEPS:
+        prefix = child.get_prefix()
+        if prefix.isspace() and "\n" not in prefix:
+          child.set_prefix("")
+        comma = True
+      else:
+        if comma:
+          prefix = child.get_prefix()
+          if not prefix:
+            child.set_prefix(" ")
+        comma = False
+    return new

Lib/lib2to3/fixes/fix_xrange.py

+# Copyright 2007 Google, Inc. All Rights Reserved.
+# Licensed to PSF under a Contributor Agreement.
+
+"""Fixer that changes xrange(...) into range(...)."""
+
+# Local imports
+from .import basefix
+from .util import Name
+
+class FixXrange(basefix.BaseFix):
+
+    PATTERN = """
+              power< name='xrange' trailer< '(' [any] ')' > >
+              """
+
+    def transform(self, node, results):
+        name = results["name"]
+        name.replace(Name("range", prefix=name.get_prefix()))

Lib/lib2to3/fixes/fix_xreadlines.py

+"""Fix "for x in f.xreadlines()" -> "for x in f".
+
+This fixer will also convert g(f.xreadlines) into g(f.__iter__)."""
+# Author: Collin Winter
+
+# Local imports
+from .import basefix
+from .util import Name
+
+
+class FixXreadlines(basefix.BaseFix):
+    PATTERN = """
+    power< call=any+ trailer< '.' 'xreadlines' > trailer< '(' ')' > >
+    |
+    power< any+ trailer< '.' no_call='xreadlines' > >
+    """
+
+    def transform(self, node, results):
+        no_call = results.get("no_call")
+
+        if no_call:
+            no_call.replace(Name("__iter__", prefix=no_call.get_prefix()))
+        else:
+            node.replace([x.clone() for x in results["call"]])

Lib/lib2to3/patcomp.py

+# Copyright 2006 Google, Inc. All Rights Reserved.
+# Licensed to PSF under a Contributor Agreement.
+
+"""Pattern compiler.
+
+The grammer is taken from PatternGrammar.txt.
+
+The compiler compiles a pattern to a pytree.*Pattern instance.
+"""
+
+__author__ = "Guido van Rossum <guido@python.org>"
+
+# Python imports
+import os
+
+# Fairly local imports
+from .pgen2 import driver
+from .pgen2 import literals
+from .pgen2 import token
+from .pgen2 import tokenize
+
+# Really local imports
+from . import pytree
+from . import pygram
+
+# The pattern grammar file
+_PATTERN_GRAMMAR_FILE = os.path.join(os.path.dirname(__file__),
+                                     "PatternGrammar.txt")
+
+
+def tokenize_wrapper(input):
+    """Tokenizes a string suppressing significant whitespace."""
+    skip = (token.NEWLINE, token.INDENT, token.DEDENT)
+    tokens = tokenize.generate_tokens(driver.generate_lines(input).next)
+    for quintuple in tokens:
+        type, value, start, end, line_text = quintuple
+        if type not in skip:
+            yield quintuple
+
+
+class PatternCompiler(object):
+
+    def __init__(self, grammar_file=_PATTERN_GRAMMAR_FILE):
+        """Initializer.
+
+        Takes an optional alternative filename for the pattern grammar.
+        """
+        self.grammar = driver.load_grammar(grammar_file)
+        self.syms = pygram.Symbols(self.grammar)
+        self.pygrammar = pygram.python_grammar
+        self.pysyms = pygram.python_symbols
+        self.driver = driver.Driver(self.grammar, convert=pattern_convert)
+
+    def compile_pattern(self, input, debug=False):
+        """Compiles a pattern string to a nested pytree.*Pattern object."""
+        tokens = tokenize_wrapper(input)
+        root = self.driver.parse_tokens(tokens, debug=debug)
+        return self.compile_node(root)
+
+    def compile_node(self, node):
+        """Compiles a node, recursively.
+
+        This is one big switch on the node type.
+        """
+        # XXX Optimize certain Wildcard-containing-Wildcard patterns
+        # that can be merged
+        if node.type == self.syms.Matcher:
+            node = node.children[0] # Avoid unneeded recursion
+
+        if node.type == self.syms.Alternatives:
+            # Skip the odd children since they are just '|' tokens
+            alts = [self.compile_node(ch) for ch in node.children[::2]]
+            if len(alts) == 1:
+                return alts[0]
+            p = pytree.WildcardPattern([[a] for a in alts], min=1, max=1)
+            return p.optimize()
+
+        if node.type == self.syms.Alternative:
+            units = [self.compile_node(ch) for ch in node.children]
+            if len(units) == 1:
+                return units[0]
+            p = pytree.WildcardPattern([units], min=1, max=1)
+            return p.optimize()
+
+        if node.type == self.syms.NegatedUnit:
+            pattern = self.compile_basic(node.children[1:])
+            p = pytree.NegatedPattern(pattern)
+            return p.optimize()
+
+        assert node.type == self.syms.Unit
+
+        name = None
+        nodes = node.children
+        if len(nodes) >= 3 and nodes[1].type == token.EQUAL:
+            name = nodes[0].value
+            nodes = nodes[2:]
+        repeat = None
+        if len(nodes) >= 2 and nodes[-1].type == self.syms.Repeater:
+            repeat = nodes[-1]
+            nodes = nodes[:-1]
+
+        # Now we've reduced it to: STRING | NAME [Details] | (...) | [...]
+        pattern = self.compile_basic(nodes, repeat)
+
+        if repeat is not None:
+            assert repeat.type == self.syms.Repeater
+            children = repeat.children
+            child = children[0]
+            if child.type == token.STAR:
+                min = 0
+                max = pytree.HUGE
+            elif child.type == token.PLUS:
+                min = 1
+                max = pytree.HUGE
+            elif child.type == token.LBRACE:
+                assert children[-1].type == token.RBRACE
+                assert  len(children) in (3, 5)
+                min = max = self.get_int(children[1])
+                if len(children) == 5:
+                    max = self.get_int(children[3])
+            else:
+                assert False
+            if min != 1 or max != 1:
+                pattern = pattern.optimize()
+                pattern = pytree.WildcardPattern([[pattern]], min=min, max=max)
+
+        if name is not None:
+            pattern.name = name
+        return pattern.optimize()
+
+    def compile_basic(self, nodes, repeat=None):
+        # Compile STRING | NAME [Details] | (...) | [...]
+        assert len(nodes) >= 1
+        node = nodes[0]
+        if node.type == token.STRING:
+            value = literals.evalString(node.value)
+            return pytree.LeafPattern(content=value)
+        elif node.type == token.NAME:
+            value = node.value
+            if value.isupper():
+                if value not in TOKEN_MAP:
+                    raise SyntaxError("Invalid token: %r" % value)
+                return pytree.LeafPattern(TOKEN_MAP[value])
+            else:
+                if value == "any":
+                    type = None
+                elif not value.startswith("_"):
+                    type = getattr(self.pysyms, value, None)
+                    if type is None:
+                        raise SyntaxError("Invalid symbol: %r" % value)
+                if nodes[1:]: # Details present
+                    content = [self.compile_node(nodes[1].children[1])]
+                else:
+                    content = None
+                return pytree.NodePattern(type, content)
+        elif node.value == "(":
+            return self.compile_node(nodes[1])
+        elif node.value == "[":
+            assert repeat is None
+            subpattern = self.compile_node(nodes[1])
+            return pytree.WildcardPattern([[subpattern]], min=0, max=1)
+        assert False, node
+
+    def get_int(self, node):
+        assert node.type == token.NUMBER
+        return int(node.value)
+
+
+# Map named tokens to the type value for a LeafPattern
+TOKEN_MAP = {"NAME": token.NAME,
+             "STRING": token.STRING,
+             "NUMBER": token.NUMBER,
+             "TOKEN": None}
+
+
+def pattern_convert(grammar, raw_node_info):
+    """Converts raw node information to a Node or Leaf instance."""
+    type, value, context, children = raw_node_info
+    if children or type in grammar.number2symbol:
+        return pytree.Node(type, children, context=context)
+    else:
+        return pytree.Leaf(type, value, context=context)
+
+
+def compile_pattern(pattern):
+    return PatternCompiler().compile_pattern(pattern)

Lib/lib2to3/pgen2/__init__.py

+# Copyright 2004-2005 Elemental Security, Inc. All Rights Reserved.
+# Licensed to PSF under a Contributor Agreement.
+
+"""The pgen2 package."""

Lib/lib2to3/pgen2/conv.py

+# Copyright 2004-2005 Elemental Security, Inc. All Rights Reserved.
+# Licensed to PSF under a Contributor Agreement.
+
+"""Convert graminit.[ch] spit out by pgen to Python code.
+
+Pgen is the Python parser generator.  It is useful to quickly create a
+parser from a grammar file in Python's grammar notation.  But I don't
+want my parsers to be written in C (yet), so I'm translating the
+parsing tables to Python data structures and writing a Python parse
+engine.
+
+Note that the token numbers are constants determined by the standard
+Python tokenizer.  The standard token module defines these numbers and
+their names (the names are not used much).  The token numbers are
+hardcoded into the Python tokenizer and into pgen.  A Python
+implementation of the Python tokenizer is also available, in the
+standard tokenize module.
+
+On the other hand, symbol numbers (representing the grammar's
+non-terminals) are assigned by pgen based on the actual grammar
+input.
+
+Note: this module is pretty much obsolete; the pgen module generates
+equivalent grammar tables directly from the Grammar.txt input file
+without having to invoke the Python pgen C program.
+
+"""
+
+# Python imports
+import re
+
+# Local imports
+from pgen2 import grammar, token
+
+
+class Converter(grammar.Grammar):
+    """Grammar subclass that reads classic pgen output files.
+
+    The run() method reads the tables as produced by the pgen parser
+    generator, typically contained in two C files, graminit.h and
+    graminit.c.  The other methods are for internal use only.
+
+    See the base class for more documentation.
+
+    """
+
+    def run(self, graminit_h, graminit_c):
+        """Load the grammar tables from the text files written by pgen."""
+        self.parse_graminit_h(graminit_h)
+        self.parse_graminit_c(graminit_c)
+        self.finish_off()
+
+    def parse_graminit_h(self, filename):
+        """Parse the .h file writen by pgen.  (Internal)
+
+        This file is a sequence of #define statements defining the
+        nonterminals of the grammar as numbers.  We build two tables
+        mapping the numbers to names and back.
+
+        """
+        try:
+            f = open(filename)
+        except IOError, err:
+            print "Can't open %s: %s" % (filename, err)
+            return False
+        self.symbol2number = {}
+        self.number2symbol = {}
+        lineno = 0
+        for line in f:
+            lineno += 1
+            mo = re.match(r"^#define\s+(\w+)\s+(\d+)$", line)
+            if not mo and line.strip():
+                print "%s(%s): can't parse %s" % (filename, lineno,
+                                                  line.strip())
+            else:
+                symbol, number = mo.groups()
+                number = int(number)
+                assert symbol not in self.symbol2number
+                assert number not in self.number2symbol
+                self.symbol2number[symbol] = number
+                self.number2symbol[number] = symbol
+        return True
+
+    def parse_graminit_c(self, filename):
+        """Parse the .c file writen by pgen.  (Internal)
+
+        The file looks as follows.  The first two lines are always this:
+
+        #include "pgenheaders.h"
+        #include "grammar.h"
+
+        After that come four blocks:
+
+        1) one or more state definitions
+        2) a table defining dfas
+        3) a table defining labels
+        4) a struct defining the grammar
+
+        A state definition has the following form:
+        - one or more arc arrays, each of the form:
+          static arc arcs_<n>_<m>[<k>] = {
+                  {<i>, <j>},
+                  ...
+          };
+        - followed by a state array, of the form:
+          static state states_<s>[<t>] = {
+                  {<k>, arcs_<n>_<m>},
+                  ...
+          };
+
+        """
+        try:
+            f = open(filename)
+        except IOError, err:
+            print "Can't open %s: %s" % (filename, err)
+            return False
+        # The code below essentially uses f's iterator-ness!
+        lineno = 0
+
+        # Expect the two #include lines
+        lineno, line = lineno+1, f.next()
+        assert line == '#include "pgenheaders.h"\n', (lineno, line)
+        lineno, line = lineno+1, f.next()
+        assert line == '#include "grammar.h"\n', (lineno, line)
+
+        # Parse the state definitions
+        lineno, line = lineno+1, f.next()
+        allarcs = {}
+        states = []
+        while line.startswith("static arc "):
+            while line.startswith("static arc "):
+                mo = re.match(r"static arc arcs_(\d+)_(\d+)\[(\d+)\] = {$",
+                              line)
+                assert mo, (lineno, line)
+                n, m, k = map(int, mo.groups())
+                arcs = []
+                for _ in range(k):
+                    lineno, line = lineno+1, f.next()
+                    mo = re.match(r"\s+{(\d+), (\d+)},$", line)
+                    assert mo, (lineno, line)
+                    i, j = map(int, mo.groups())
+                    arcs.append((i, j))
+                lineno, line = lineno+1, f.next()
+                assert line == "};\n", (lineno, line)
+                allarcs[(n, m)] = arcs
+                lineno, line = lineno+1, f.next()
+            mo = re.match(r"static state states_(\d+)\[(\d+)\] = {$", line)
+            assert mo, (lineno, line)
+            s, t = map(int, mo.groups())
+            assert s == len(states), (lineno, line)
+            state = []
+            for _ in range(t):
+                lineno, line = lineno+1, f.next()
+                mo = re.match(r"\s+{(\d+), arcs_(\d+)_(\d+)},$", line)
+                assert mo, (lineno, line)
+                k, n, m = map(int, mo.groups())
+                arcs = allarcs[n, m]
+                assert k == len(arcs), (lineno, line)
+                state.append(arcs)
+            states.append(state)
+            lineno, line = lineno+1, f.next()
+            assert line == "};\n", (lineno, line)
+            lineno, line = lineno+1, f.next()
+        self.states = states
+
+        # Parse the dfas
+        dfas = {}
+        mo = re.match(r"static dfa dfas\[(\d+)\] = {$", line)
+        assert mo, (lineno, line)
+        ndfas = int(mo.group(1))
+        for i in range(ndfas):
+            lineno, line = lineno+1, f.next()
+            mo = re.match(r'\s+{(\d+), "(\w+)", (\d+), (\d+), states_(\d+),$',
+                          line)
+            assert mo, (lineno, line)
+            symbol = mo.group(2)
+            number, x, y, z = map(int, mo.group(1, 3, 4, 5))
+            assert self.symbol2number[symbol] == number, (lineno, line)
+            assert self.number2symbol[number] == symbol, (lineno, line)
+            assert x == 0, (lineno, line)
+            state = states[z]
+            assert y == len(state), (lineno, line)
+            lineno, line = lineno+1, f.next()
+            mo = re.match(r'\s+("(?:\\\d\d\d)*")},$', line)
+            assert mo, (lineno, line)
+            first = {}
+            rawbitset = eval(mo.group(1))
+            for i, c in enumerate(rawbitset):
+                byte = ord(c)
+                for j in range(8):
+                    if byte & (1<<j):
+                        first[i*8 + j] = 1
+            dfas[number] = (state, first)
+        lineno, line = lineno+1, f.next()
+        assert line == "};\n", (lineno, line)
+        self.dfas = dfas
+
+        # Parse the labels
+        labels = []
+        lineno, line = lineno+1, f.next()
+        mo = re.match(r"static label labels\[(\d+)\] = {$", line)
+        assert mo, (lineno, line)
+        nlabels = int(mo.group(1))
+        for i in range(nlabels):
+            lineno, line = lineno+1, f.next()
+            mo = re.match(r'\s+{(\d+), (0|"\w+")},$', line)
+            assert mo, (lineno, line)
+            x, y = mo.groups()
+            x = int(x)
+            if y == "0":
+                y = None
+            else:
+                y = eval(y)
+            labels.append((x, y))
+        lineno, line = lineno+1, f.next()
+        assert line == "};\n", (lineno, line)
+        self.labels = labels
+
+        # Parse the grammar struct
+        lineno, line = lineno+1, f.next()
+        assert line == "grammar _PyParser_Grammar = {\n", (lineno, line)
+        lineno, line = lineno+1, f.next()
+        mo = re.match(r"\s+(\d+),$", line)
+        assert mo, (lineno, line)
+        ndfas = int(mo.group(1))
+        assert ndfas == len(self.dfas)
+        lineno, line = lineno+1, f.next()
+        assert line == "\tdfas,\n", (lineno, line)
+        lineno, line = lineno+1, f.next()
+        mo = re.match(r"\s+{(\d+), labels},$", line)
+        assert mo, (lineno, line)
+        nlabels = int(mo.group(1))
+        assert nlabels == len(self.labels), (lineno, line)
+        lineno, line = lineno+1, f.next()
+        mo = re.match(r"\s+(\d+)$", line)
+        assert mo, (lineno, line)
+        start = int(mo.group(1))
+        assert start in self.number2symbol, (lineno, line)
+        self.start = start
+        lineno, line = lineno+1, f.next()
+        assert line == "};\n", (lineno, line)
+        try:
+            lineno, line = lineno+1, f.next()
+        except StopIteration:
+            pass
+        else:
+            assert 0, (lineno, line)
+
+    def finish_off(self):
+        """Create additional useful structures.  (Internal)."""
+        self.keywords = {} # map from keyword strings to arc labels
+        self.tokens = {}   # map from numeric token values to arc labels
+        for ilabel, (type, value) in enumerate(self.labels):
+            if type == token.NAME and value is not None:
+                self.keywords[value] = ilabel
+            elif value is None:
+                self.tokens[type] = ilabel

Lib/lib2to3/pgen2/driver.py

+# Copyright 2004-2005 Elemental Security, Inc. All Rights Reserved.
+# Licensed to PSF under a Contributor Agreement.
+
+# Modifications:
+# Copyright 2006 Google, Inc. All Rights Reserved.
+# Licensed to PSF under a Contributor Agreement.
+
+"""Parser driver.
+
+This provides a high-level interface to parse a file into a syntax tree.
+
+"""
+
+__author__ = "Guido van Rossum <guido@python.org>"
+
+__all__ = ["Driver", "load_grammar"]
+
+# Python imports
+import os
+import logging
+import sys
+
+# Pgen imports
+from . import grammar, parse, token, tokenize
+
+
+class Driver(object):
+
+    def __init__(self, grammar, convert=None, logger=None):
+        self.grammar = grammar
+        if logger is None:
+            logger = logging.getLogger()
+        self.logger = logger
+        self.convert = convert
+
+    def parse_tokens(self, tokens, debug=False):
+        """Parse a series of tokens and return the syntax tree."""
+        # XXX Move the prefix computation into a wrapper around tokenize.
+        p = parse.Parser(self.grammar, self.convert)
+        p.setup()
+        lineno = 1
+        column = 0
+        type = value = start = end = line_text = None
+        prefix = ""
+        for quintuple in tokens:
+            type, value, start, end, line_text = quintuple
+            if start != (lineno, column):
+                assert (lineno, column) <= start, ((lineno, column), start)
+                s_lineno, s_column = start
+                if lineno < s_lineno:
+                    prefix += "\n" * (s_lineno - lineno)
+                    lineno = s_lineno
+                    column = 0
+                if column < s_column:
+                    prefix += line_text[column:s_column]
+                    column = s_column
+            if type in (tokenize.COMMENT, tokenize.NL):
+                prefix += value
+                lineno, column = end
+                if value.endswith("\n"):
+                    lineno += 1
+                    column = 0
+                continue
+            if type == token.OP:
+                type = grammar.opmap[value]
+            if debug:
+                self.logger.debug("%s %r (prefix=%r)",
+                                  token.tok_name[type], value, prefix)
+            if p.addtoken(type, value, (prefix, start)):
+                if debug:
+                    self.logger.debug("Stop.")
+                break
+            prefix = ""
+            lineno, column = end
+            if value.endswith("\n"):
+                lineno += 1
+                column = 0
+        else:
+            # We never broke out -- EOF is too soon (how can this happen???)
+            raise parse.ParseError("incomplete input", t, v, x)
+        return p.rootnode
+
+    def parse_stream_raw(self, stream, debug=False):
+        """Parse a stream and return the syntax tree."""
+        tokens = tokenize.generate_tokens(stream.readline)
+        return self.parse_tokens(tokens, debug)
+
+    def parse_stream(self, stream, debug=False):
+        """Parse a stream and return the syntax tree."""
+        return self.parse_stream_raw(stream, debug)
+
+    def parse_file(self, filename, debug=False):
+        """Parse a file and return the syntax tree."""
+        stream = open(filename)
+        try:
+            return self.parse_stream(stream, debug)
+        finally:
+            stream.close()
+
+    def parse_string(self, text, debug=False):
+        """Parse a string and return the syntax tree."""
+        tokens = tokenize.generate_tokens(generate_lines(text).next)
+        return self.parse_tokens(tokens, debug)
+
+
+def generate_lines(text):
+    """Generator that behaves like readline without using StringIO."""
+    for line in text.splitlines(True):
+        yield line
+    while True:
+        yield ""
+
+
+def load_grammar(gt="Grammar.txt", gp=None,
+                 save=True, force=False, logger=None):
+    """Load the grammar (maybe from a pickle)."""
+    if logger is None:
+        logger = logging.getLogger()
+    if gp is None:
+        head, tail = os.path.splitext(gt)
+        if tail == ".txt":
+            tail = ""
+        gp = head + tail + ".".join(map(str, sys.version_info)) + ".pickle"
+    if force or not _newer(gp, gt):
+        logger.info("Generating grammar tables from %s", gt)
+        from pgen2 import pgen
+        g = pgen.generate_grammar(gt)
+        if save:
+            logger.info("Writing grammar tables to %s", gp)
+            g.dump(gp)
+    else:
+        g = grammar.Grammar()
+        g.load(gp)
+    return g
+
+
+def _newer(a, b):
+    """Inquire whether file a was written since file b."""
+    if not os.path.exists(a):
+        return False
+    if not os.path.exists(b):
+        return True
+    return os.path.getmtime(a) >= os.path.getmtime(b)

Lib/lib2to3/pgen2/grammar.py

+# Copyright 2004-2005 Elemental Security, Inc. All Rights Reserved.
+# Licensed to PSF under a Contributor Agreement.
+
+"""This module defines the data structures used to represent a grammar.
+
+These are a bit arcane because they are derived from the data
+structures used by Python's 'pgen' parser generator.
+
+There's also a table here mapping operators to their names in the
+token module; the Python tokenize module reports all operators as the
+fallback token code OP, but the parser needs the actual token code.
+
+"""
+
+# Python imports
+import pickle
+
+# Local imports
+from . import token, tokenize
+
+
+class Grammar(object):
+    """Pgen parsing tables tables conversion class.
+
+    Once initialized, this class supplies the grammar tables for the
+    parsing engine implemented by parse.py.  The parsing engine
+    accesses the instance variables directly.  The class here does not
+    provide initialization of the tables; several subclasses exist to
+    do this (see the conv and pgen modules).
+
+    The load() method reads the tables from a pickle file, which is
+    much faster than the other ways offered by subclasses.  The pickle
+    file is written by calling dump() (after loading the grammar
+    tables using a subclass).  The report() method prints a readable
+    representation of the tables to stdout, for debugging.
+
+    The instance variables are as follows:
+
+    symbol2number -- a dict mapping symbol names to numbers.  Symbol
+                     numbers are always 256 or higher, to distinguish
+                     them from token numbers, which are between 0 and
+                     255 (inclusive).
+
+    number2symbol -- a dict mapping numbers to symbol names;
+                     these two are each other's inverse.
+
+    states        -- a list of DFAs, where each DFA is a list of
+                     states, each state is is a list of arcs, and each
+                     arc is a (i, j) pair where i is a label and j is
+                     a state number.  The DFA number is the index into
+                     this list.  (This name is slightly confusing.)
+                     Final states are represented by a special arc of
+                     the form (0, j) where j is its own state number.
+
+    dfas          -- a dict mapping symbol numbers to (DFA, first)
+                     pairs, where DFA is an item from the states list
+                     above, and first is a set of tokens that can
+                     begin this grammar rule (represented by a dict
+                     whose values are always 1).
+
+    labels        -- a list of (x, y) pairs where x is either a token
+                     number or a symbol number, and y is either None
+                     or a string; the strings are keywords.  The label
+                     number is the index in this list; label numbers
+                     are used to mark state transitions (arcs) in the
+                     DFAs.
+
+    start         -- the number of the grammar's start symbol.
+
+    keywords      -- a dict mapping keyword strings to arc labels.
+
+    tokens        -- a dict mapping token numbers to arc labels.
+
+    """
+
+    def __init__(self):
+        self.symbol2number = {}
+        self.number2symbol = {}
+        self.states = []
+        self.dfas = {}
+        self.labels = [(0, "EMPTY")]
+        self.keywords = {}
+        self.tokens = {}
+        self.symbol2label = {}
+        self.start = 256
+
+    def dump(self, filename):
+        """Dump the grammar tables to a pickle file."""
+        f = open(filename, "wb")
+        pickle.dump(self.__dict__, f, 2)
+        f.close()
+
+    def load(self, filename):
+        """Load the grammar tables from a pickle file."""
+        f = open(filename, "rb")
+        d = pickle.load(f)
+        f.close()
+        self.__dict__.update(d)
+
+    def report(self):
+        """Dump the grammar tables to standard output, for debugging."""
+        from pprint import pprint
+        print "s2n"
+        pprint(self.symbol2number)
+        print "n2s"
+        pprint(self.number2symbol)
+        print "states"
+        pprint(self.states)
+        print "dfas"
+        pprint(self.dfas)
+        print "labels"
+        pprint(self.labels)
+        print "start", self.start
+
+
+# Map from operator to number (since tokenize doesn't do this)
+
+opmap_raw = """
+( LPAR
+) RPAR
+[ LSQB
+] RSQB
+: COLON
+, COMMA
+; SEMI
++ PLUS
+- MINUS
+* STAR
+/ SLASH
+| VBAR
+& AMPER
+< LESS
+> GREATER
+= EQUAL
+. DOT
+% PERCENT
+` BACKQUOTE
+{ LBRACE
+} RBRACE
+@ AT
+== EQEQUAL
+!= NOTEQUAL
+<> NOTEQUAL
+<= LESSEQUAL
+>= GREATEREQUAL
+~ TILDE
+^ CIRCUMFLEX
+<< LEFTSHIFT
+>> RIGHTSHIFT
+** DOUBLESTAR
++= PLUSEQUAL
+-= MINEQUAL
+*= STAREQUAL
+/= SLASHEQUAL
+%= PERCENTEQUAL
+&= AMPEREQUAL
+|= VBAREQUAL
+^= CIRCUMFLEXEQUAL
+<<= LEFTSHIFTEQUAL
+>>= RIGHTSHIFTEQUAL
+**= DOUBLESTAREQUAL
+// DOUBLESLASH
+//= DOUBLESLASHEQUAL
+-> RARROW
+"""
+
+opmap = {}
+for line in opmap_raw.splitlines():
+    if line:
+        op, name = line.split()
+        opmap[op] = getattr(token, name)

Lib/lib2to3/pgen2/literals.py

+# Copyright 2004-2005 Elemental Security, Inc. All Rights Reserved.
+# Licensed to PSF under a Contributor Agreement.
+
+"""Safely evaluate Python string literals without using eval()."""
+
+import re
+
+simple_escapes = {"a": "\a",
+                  "b": "\b",
+                  "f": "\f",
+                  "n": "\n",
+                  "r": "\r",
+                  "t": "\t",
+                  "v": "\v",
+                  "'": "'",
+                  '"': '"',
+                  "\\": "\\"}
+
+def escape(m):
+    all, tail = m.group(0, 1)
+    assert all.startswith("\\")
+    esc = simple_escapes.get(tail)
+    if esc is not None:
+        return esc
+    if tail.startswith("x"):
+        hexes = tail[1:]
+        if len(hexes) < 2:
+            raise ValueError("invalid hex string escape ('\\%s')" % tail)
+        try:
+            i = int(hexes, 16)
+        except ValueError:
+            raise ValueError("invalid hex string escape ('\\%s')" % tail)
+    else:
+        try:
+            i = int(tail, 8)
+        except ValueError:
+            raise ValueError("invalid octal string escape ('\\%s')" % tail)
+    return chr(i)
+
+def evalString(s):
+    assert s.startswith("'") or s.startswith('"'), repr(s[:1])
+    q = s[0]
+    if s[:3] == q*3:
+        q = q*3
+    assert s.endswith(q), repr(s[-len(q):])
+    assert len(s) >= 2*len(q)
+    s = s[len(q):-len(q)]
+    return re.sub(r"\\(\'|\"|\\|[abfnrtv]|x.{0,2}|[0-7]{1,3})", escape, s)
+
+def test():
+    for i in range(256):
+        c = chr(i)
+        s = repr(c)
+        e = evalString(s)
+        if e != c:
+            print i, c, s, e
+
+
+if __name__ == "__main__":
+    test()

Lib/lib2to3/pgen2/parse.py

+# Copyright 2004-2005 Elemental Security, Inc. All Rights Reserved.
+# Licensed to PSF under a Contributor Agreement.
+
+"""Parser engine for the grammar tables generated by pgen.
+
+The grammar table must be loaded first.
+
+See Parser/parser.c in the Python distribution for additional info on
+how this parsing engine works.
+
+"""
+
+# Get a usable 'set' constructor
+try:
+    set
+except NameError:
+    from sets import Set as set
+
+# Local imports
+from . import token
+
+class ParseError(Exception):
+    """Exception to signal the parser is stuck."""
+
+    def __init__(self, msg, type, value, context):
+        Exception.__init__(self, "%s: type=%r, value=%r, context=%r" %
+                           (msg, type, value, context))
+        self.msg = msg
+        self.type = type
+        self.value = value
+        self.context = context
+
+class Parser(object):
+    """Parser engine.
+
+    The proper usage sequence is:
+
+    p = Parser(grammar, [converter])  # create instance
+    p.setup([start])                  # prepare for parsing
+    <for each input token>:
+        if p.addtoken(...):           # parse a token; may raise ParseError
+            break
+    root = p.rootnode                 # root of abstract syntax tree
+
+    A Parser instance may be reused by calling setup() repeatedly.
+
+    A Parser instance contains state pertaining to the current token
+    sequence, and should not be used concurrently by different threads
+    to parse separate token sequences.
+
+    See driver.py for how to get input tokens by tokenizing a file or
+    string.
+
+    Parsing is complete when addtoken() returns True; the root of the
+    abstract syntax tree can then be retrieved from the rootnode
+    instance variable.  When a syntax error occurs, addtoken() raises
+    the ParseError exception.  There is no error recovery; the parser
+    cannot be used after a syntax error was reported (but it can be
+    reinitialized by calling setup()).
+
+    """
+
+    def __init__(self, grammar, convert=None):
+        """Constructor.
+
+        The grammar argument is a grammar.Grammar instance; see the
+        grammar module for more information.
+
+        The parser is not ready yet for parsing; you must call the
+        setup() method to get it started.
+
+        The optional convert argument is a function mapping concrete
+        syntax tree nodes to abstract syntax tree nodes.  If not
+        given, no conversion is done and the syntax tree produced is
+        the concrete syntax tree.  If given, it must be a function of
+        two arguments, the first being the grammar (a grammar.Grammar
+        instance), and the second being the concrete syntax tree node
+        to be converted.  The syntax tree is converted from the bottom
+        up.
+
+        A concrete syntax tree node is a (type, value, context, nodes)
+        tuple, where type is the node type (a token or symbol number),
+        value is None for symbols and a string for tokens, context is
+        None or an opaque value used for error reporting (typically a
+        (lineno, offset) pair), and nodes is a list of children for
+        symbols, and None for tokens.
+
+        An abstract syntax tree node may be anything; this is entirely
+        up to the converter function.
+
+        """
+        self.grammar = grammar
+        self.convert = convert or (lambda grammar, node: node)
+
+    def setup(self, start=None):
+        """Prepare for parsing.
+
+        This *must* be called before starting to parse.
+
+        The optional argument is an alternative start symbol; it
+        defaults to the grammar's start symbol.
+
+        You can use a Parser instance to parse any number of programs;
+        each time you call setup() the parser is reset to an initial
+        state determined by the (implicit or explicit) start symbol.
+
+        """
+        if start is None:
+            start = self.grammar.start
+        # Each stack entry is a tuple: (dfa, state, node).
+        # A node is a tuple: (type, value, context, children),
+        # where children is a list of nodes or None, and context may be None.
+        newnode = (start, None, None, [])
+        stackentry = (self.grammar.dfas[start], 0, newnode)
+        self.stack = [stackentry]
+        self.rootnode = None
+        self.used_names = set() # Aliased to self.rootnode.used_names in pop()
+
+    def addtoken(self, type, value, context):
+        """Add a token; return True iff this is the end of the program."""
+        # Map from token to label
+        ilabel = self.classify(type, value, context)
+        # Loop until the token is shifted; may raise exceptions
+        while True:
+            dfa, state, node = self.stack[-1]
+            states, first = dfa
+            arcs = states[state]
+            # Look for a state with this label
+            for i, newstate in arcs:
+                t, v = self.grammar.labels[i]
+                if ilabel == i:
+                    # Look it up in the list of labels
+                    assert t < 256
+                    # Shift a token; we're done with it
+                    self.shift(type, value, newstate, context)
+                    # Pop while we are in an accept-only state
+                    state = newstate
+                    while states[state] == [(0, state)]:
+                        self.pop()
+                        if not self.stack:
+                            # Done parsing!
+                            return True
+                        dfa, state, node = self.stack[-1]
+                        states, first = dfa
+                    # Done with this token
+                    return False
+                elif t >= 256:
+                    # See if it's a symbol and if we're in its first set
+                    itsdfa = self.grammar.dfas[t]
+                    itsstates, itsfirst = itsdfa
+                    if ilabel in itsfirst:
+                        # Push a symbol
+                        self.push(t, self.grammar.dfas[t], newstate, context)
+                        break # To continue the outer while loop
+            else:
+                if (0, state) in arcs:
+                    # An accepting state, pop it and try something else
+                    self.pop()
+                    if not self.stack:
+                        # Done parsing, but another token is input
+                        raise ParseError("too much input",
+                                         type, value, context)
+                else:
+                    # No success finding a transition
+                    raise ParseError("bad input", type, value, context)
+
+    def classify(self, type, value, context):
+        """Turn a token into a label.  (Internal)"""
+        if type == token.NAME:
+            # Keep a listing of all used names
+            self.used_names.add(value)
+            # Check for reserved words
+            ilabel = self.grammar.keywords.get(value)
+            if ilabel is not None:
+                return ilabel
+        ilabel = self.grammar.tokens.get(type)
+        if ilabel is None:
+            raise ParseError("bad token", type, value, context)
+        return ilabel
+
+    def shift(self, type, value, newstate, context):
+        """Shift a token.  (Internal)"""
+        dfa, state, node = self.stack[-1]
+        newnode = (type, value, context, None)
+        newnode = self.convert(self.grammar, newnode)
+        if newnode is not None:
+            node[-1].append(newnode)
+        self.stack[-1] = (dfa, newstate, node)
+
+    def push(self, type, newdfa, newstate, context):
+        """Push a nonterminal.  (Internal)"""
+        dfa, state, node = self.stack[-1]
+        newnode = (type, None, context, [])
+        self.stack[-1] = (dfa, newstate, node)
+        self.stack.append((newdfa, 0, newnode))
+
+    def pop(self):
+        """Pop a nonterminal.  (Internal)"""
+        popdfa, popstate, popnode = self.stack.pop()
+        newnode = self.convert(self.grammar, popnode)
+        if newnode is not None:
+            if self.stack:
+                dfa, state, node = self.stack[-1]
+                node[-1].append(newnode)
+            else:
+                self.rootnode = newnode
+                self.rootnode.used_names = self.used_names

Lib/lib2to3/pgen2/token.py

+#! /usr/bin/env python
+
+"""Token constants (from "token.h")."""
+
+#  Taken from Python (r53757) and modified to include some tokens
+#   originally monkeypatched in by pgen2.tokenize
+
+#--start constants--
+ENDMARKER = 0
+NAME = 1
+NUMBER = 2
+STRING = 3
+NEWLINE = 4
+INDENT = 5
+DEDENT = 6
+LPAR = 7
+RPAR = 8
+LSQB = 9
+RSQB = 10
+COLON = 11
+COMMA = 12
+SEMI = 13
+PLUS = 14
+MINUS = 15
+STAR = 16
+SLASH = 17
+VBAR = 18
+AMPER = 19
+LESS = 20
+GREATER = 21
+EQUAL = 22
+DOT = 23
+PERCENT = 24
+BACKQUOTE = 25
+LBRACE = 26
+RBRACE = 27
+EQEQUAL = 28
+NOTEQUAL = 29
+LESSEQUAL = 30
+GREATEREQUAL = 31
+TILDE = 32
+CIRCUMFLEX = 33
+LEFTSHIFT = 34
+RIGHTSHIFT = 35
+DOUBLESTAR = 36
+PLUSEQUAL = 37
+MINEQUAL = 38
+STAREQUAL = 39
+SLASHEQUAL = 40
+PERCENTEQUAL = 41
+AMPEREQUAL = 42
+VBAREQUAL = 43
+CIRCUMFLEXEQUAL = 44
+LEFTSHIFTEQUAL = 45
+RIGHTSHIFTEQUAL = 46
+DOUBLESTAREQUAL = 47
+DOUBLESLASH = 48
+DOUBLESLASHEQUAL = 49
+AT = 50
+OP = 51
+COMMENT = 52
+NL = 53
+RARROW = 54
+ERRORTOKEN = 55
+N_TOKENS = 56
+NT_OFFSET = 256
+#--end constants--
+
+tok_name = {}
+for _name, _value in globals().items():
+    if type(_value) is type(0):
+        tok_name[_value] = _name
+
+
+def ISTERMINAL(x):
+    return x < NT_OFFSET
+
+def ISNONTERMINAL(x):
+    return x >= NT_OFFSET
+
+def ISEOF(x):
+    return x == ENDMARKER

Lib/lib2to3/pygram.py

+# Copyright 2006 Google, Inc. All Rights Reserved.
+# Licensed to PSF under a Contributor Agreement.
+
+"""Export the Python grammar and symbols."""
+
+# Python imports
+import os
+
+# Local imports
+from .pgen2 import token
+from .pgen2 import driver
+from . import pytree
+
+# The grammar file
+_GRAMMAR_FILE = os.path.join(os.path.dirname(__file__), "Grammar.txt")
+
+
+class Symbols(object):
+
+    def __init__(self, grammar):
+        """Initializer.
+
+        Creates an attribute for each grammar symbol (nonterminal),
+        whose value is the symbol's type (an int >= 256).
+        """
+        for name, symbol in grammar.symbol2number.iteritems():
+            setattr(self, name, symbol)
+
+
+python_grammar = driver.load_grammar(_GRAMMAR_FILE)
+python_symbols = Symbols(python_grammar)
+
+
+def parenthesize(node):
+    return pytree.Node(python_symbols.atom,
+                       (pytree.Leaf(token.LPAR, "("),
+                        node,
+                        pytree.Leaf(token.RPAR, ")")))

Lib/lib2to3/tests/__init__.py

+"""Make tests/ into a package. This allows us to "import tests" and
+have tests.all_tests be a TestSuite representing all test cases
+from all test_*.py files in tests/."""
+# Author: Collin Winter
+
+import os
+import os.path
+import unittest
+import types
+
+from . import support
+
+all_tests = unittest.TestSuite()
+
+tests_dir = os.path.join(os.path.dirname(__file__), '..', 'tests')
+tests = [t[0:-3] for t in os.listdir(tests_dir)
+                        if t.startswith('test_') and t.endswith('.py')]
+
+loader = unittest.TestLoader()
+
+for t in tests:
+    __import__("",globals(),locals(),[t],level=1)
+    mod = globals()[t]
+    all_tests.addTests(loader.loadTestsFromModule(mod))

Lib/lib2to3/tests/benchmark.py

+#!/usr/bin/env python2.5
+"""
+This is a benchmarking script to test the speed of 2to3's pattern matching
+system. It's equivalent to "refactor.py -f all" for every Python module
+in sys.modules, but without engaging the actual transformations.
+"""
+
+__author__ = "Collin Winter <collinw at gmail.com>"
+
+# Python imports
+import os.path
+import sys
+from time import time
+
+# Test imports
+from support import adjust_path
+adjust_path()
+
+# Local imports
+from .. import refactor
+
+### Mock code for refactor.py and the fixers
+###############################################################################
+class Options:
+    def __init__(self, **kwargs):
+        for k, v in kwargs.items():
+            setattr(self, k, v)
+
+        self.verbose = False
+        
+def dummy_transform(*args, **kwargs):
+    pass
+    
+### Collect list of modules to match against
+###############################################################################
+files = []
+for mod in sys.modules.values():
+    if mod is None or not hasattr(mod, '__file__'):
+        continue
+    f = mod.__file__
+    if f.endswith('.pyc'):
+        f = f[:-1]
+    if f.endswith('.py'):
+        files.append(f)
+
+### Set up refactor and run the benchmark
+###############################################################################
+options = Options(fix=["all"], print_function=False, doctests_only=False)
+refactor = refactor.RefactoringTool(options)
+for fixer in refactor.fixers:
+    # We don't want them to actually fix the tree, just match against it.
+    fixer.transform = dummy_transform
+
+t = time()
+for f in files:
+    print "Matching", f
+    refactor.refactor_file(f)
+print "%d seconds to match %d files" % (time() - t, len(sys.modules))

Lib/lib2to3/tests/pytree_idempotency.py

+#!/usr/bin/env python2.5
+# Copyright 2006 Google, Inc. All Rights Reserved.
+# Licensed to PSF under a Contributor Agreement.
+
+"""Main program for testing the infrastructure."""
+
+__author__ = "Guido van Rossum <guido@python.org>"
+
+# Support imports (need to be imported first)
+import support
+
+# Python imports
+import os
+import sys
+import logging
+
+# Local imports
+from .. import pytree
+import pgen2
+from pgen2 import driver
+
+logging.basicConfig()
+
+def main():
+    gr = driver.load_grammar("Grammar.txt")
+    dr = driver.Driver(gr, convert=pytree.convert)
+
+    fn = "example.py"
+    tree = dr.parse_file(fn, debug=True)
+    if not diff(fn, tree):
+        print "No diffs."
+    if not sys.argv[1:]:
+        return # Pass a dummy argument to run the complete test suite below
+
+    problems = []
+
+    # Process every imported module
+    for name in sys.modules:
+        mod = sys.modules[name]
+        if mod is None or not hasattr(mod, "__file__"):
+            continue
+        fn = mod.__file__
+        if fn.endswith(".pyc"):
+            fn = fn[:-1]
+        if not fn.endswith(".py"):
+            continue
+        print >>sys.stderr, "Parsing", fn
+        tree = dr.parse_file(fn, debug=True)
+        if diff(fn, tree):
+            problems.append(fn)
+
+    # Process every single module on sys.path (but not in packages)
+    for dir in sys.path:
+        try:
+            names = os.listdir(dir)
+        except os.error:
+            continue
+        print >>sys.stderr, "Scanning", dir, "..."
+        for name in names:
+            if not name.endswith(".py"):
+                continue
+            print >>sys.stderr, "Parsing", name
+            fn = os.path.join(dir, name)
+            try:
+                tree = dr.parse_file(fn, debug=True)
+            except pgen2.parse.ParseError, err:
+                print "ParseError:", err
+            else:
+                if diff(fn, tree):
+                    problems.append(fn)
+
+    # Show summary of problem files
+    if not problems:
+        print "No problems.  Congratulations!"
+    else:
+        print "Problems in following files:"
+        for fn in problems:
+            print "***", fn
+
+def diff(fn, tree):
+    f = open("@", "w")
+    try:
+        f.write(str(tree))
+    finally:
+        f.close()
+    try:
+        return os.system("diff -u %s @" % fn)
+    finally:
+        os.remove("@")
+
+if __name__ == "__main__":
+    main()

Lib/lib2to3/tests/support.py

+"""Support code for test_*.py files"""
+# Author: Collin Winter
+
+# Python imports
+import unittest
+import sys
+import os
+import os.path
+import re
+from textwrap import dedent
+
+#sys.path.insert(0, os.path.join(os.path.dirname(__file__), ".."))
+
+# Local imports
+from .. import pytree
+from ..pgen2 import driver
+
+test_dir = os.path.dirname(__file__)
+proj_dir = os.path.normpath(os.path.join(test_dir, ".."))
+grammar_path = os.path.join(test_dir, "..", "Grammar.txt")
+grammar = driver.load_grammar(grammar_path)
+driver = driver.Driver(grammar, convert=pytree.convert)
+
+def parse_string(string):
+    return driver.parse_string(reformat(string), debug=True)
+
+# Python 2.3's TestSuite is not iter()-able
+if sys.version_info < (2, 4):
+    def TestSuite_iter(self):
+        return iter(self._tests)
+    unittest.TestSuite.__iter__ = TestSuite_iter
+
+def run_all_tests(test_mod=None, tests=None):
+    if tests is None:
+        tests = unittest.TestLoader().loadTestsFromModule(test_mod)
+    unittest.TextTestRunner(verbosity=2).run(tests)
+
+def reformat(string):
+    return dedent(string) + "\n\n"
+
+def all_project_files():
+    for dirpath, dirnames, filenames in os.walk(proj_dir):
+        for filename in filenames:
+            if filename.endswith(".py"):
+                yield os.path.join(dirpath, filename)
+
+TestCase = unittest.TestCase

Lib/lib2to3/tests/test_all_fixers.py

+#!/usr/bin/env python2.5
+"""Tests that run all fixer modules over an input stream.
+
+This has been broken out into its own test module because of its
+running time.
+"""
+# Author: Collin Winter
+
+# Testing imports
+try:
+    from . import support
+except ImportError:
+    import support
+
+# Python imports
+import unittest
+
+# Local imports
+from .. import pytree
+from .. import refactor
+
+class Options:
+    def __init__(self, **kwargs):
+        for k, v in kwargs.items():
+            setattr(self, k, v)
+        self.verbose = False
+
+class Test_all(support.TestCase):
+    def setUp(self):
+        options = Options(fix=["all", "idioms", "ws_comma"],
+                          print_function=False)
+        self.refactor = refactor.RefactoringTool(options)
+
+    def test_all_project_files(self):
+        for filepath in support.all_project_files():
+            print "Fixing %s..." % filepath
+            self.refactor.refactor_string(open(filepath).read(), filepath)
+
+
+if __name__ == "__main__":
+    import __main__
+    support.run_all_tests(__main__)