ERROR:root:code for hash md5 was not found

Discussion in 'Python' started by mike, Jan 11, 2012.

  1. mike

    mike Guest

    Hi,

    We are running are running Python program on Redhat 5.5.

    When executing our program we get the following error ( see below).

    Any ideas what this is due to?

    br,

    //mike

    /pysibelius/lib/common/
    DataTypes.py
    Overwriten ...
    ERROR:root:code for hash md5 was not found.
    Traceback (most recent call last):
    File "/python/lib/python2.7/hashlib.py", line 139, in <module>
    globals()[__func_name] = __get_hash(__func_name)
    File "/python/lib/python2.7/hashlib.py", line 91, in
    __get_builtin_constructor
    raise ValueError('unsupported hash type %s' % name)
    ValueError: unsupported hash type md5
    ERROR:root:code for hash sha1 was not found
    mike, Jan 11, 2012
    #1
    1. Advertising

  2. Le 11/01/2012 12:19, mike a écrit :
    > Hi,
    >
    > We are running are running Python program on Redhat 5.5.
    >
    > When executing our program we get the following error ( see below).
    >
    > Any ideas what this is due to?


    On my computer hashlib has "md5" :



    Python 2.6.6 (r266:84292, Sep 15 2010, 16:22:56)
    [GCC 4.4.5] on linux2
    Type "help", "copyright", "credits" or "license" for more information.
    >>> import hashlib
    >>> h=hashlib.new("md5")
    >>> h.update("Hello")
    >>> h.hexdigest()

    'd992641f1b2b9c08b569c0a17c4e7cb8'

    While if I ask for a hash method that hashlib does not know, I get the
    same error as you :

    >>> g=hashlib.new("bla")

    Traceback (most recent call last):
    File "<stdin>", line 1, in <module>
    File "/usr/lib/python2.6/hashlib.py", line 101, in __hash_new
    return __get_builtin_constructor(name)(string)
    File "/usr/lib/python2.6/hashlib.py", line 80, in
    __get_builtin_constructor
    raise ValueError, "unsupported hash type"
    ValueError: unsupported hash type


    You should give us a more comprehensive example of your problem.

    Laurent
    Laurent Claessens, Jan 11, 2012
    #2
    1. Advertising

  3. mike

    mike Guest

    On Jan 11, 12:28 pm, Laurent Claessens <> wrote:
    > Le 11/01/2012 12:19, mike a crit :
    >
    > > Hi,

    >
    > > We are running are running Python program on Redhat 5.5.

    >
    > > When executing our program we get the following error ( see below).

    >
    > > Any ideas what this is due to?

    >
    > On my computer hashlib has "md5" :
    >
    > Python 2.6.6 (r266:84292, Sep 15 2010, 16:22:56)
    > [GCC 4.4.5] on linux2
    > Type "help", "copyright", "credits" or "license" for more information.>>>import hashlib
    > >>> h=hashlib.new("md5")
    > >>> h.update("Hello")
    > >>> h.hexdigest()

    >
    > 'd992641f1b2b9c08b569c0a17c4e7cb8'
    >
    > While if I ask for a hash method that hashlib does not know, I get the
    > same error as you :
    >
    >  >>> g=hashlib.new("bla")
    > Traceback (most recent call last):
    >    File "<stdin>", line 1, in <module>
    >    File "/usr/lib/python2.6/hashlib.py", line 101, in __hash_new
    >      return __get_builtin_constructor(name)(string)
    >    File "/usr/lib/python2.6/hashlib.py", line 80, in
    > __get_builtin_constructor
    >      raise ValueError, "unsupported hash type"
    > ValueError: unsupported hash type
    >
    > You should give us a more comprehensive example of your problem.
    >
    > Laurent


    Hi,

    Thanks for reply.

    I did some more digging and found that our class imports a "yacc.py"
    that uses

    import re, types, sys, cStringIO, hashlib, os.path

    so it has hashlib.

    yacc.py seems to be an old version 1.3 ( I found 2.3 -->).

    Reading about hashlib it seems to be dependent on os installation of
    OpenSSL but I cannot find out how.

    br,

    //mike

    yacc.py
    ===============
    #-----------------------------------------------------------------------------
    # ply: yacc.py
    #
    # Author: David M. Beazley ()
    # Department of Computer Science
    # University of Chicago
    # Chicago, IL 60637
    #
    # Copyright (C) 2001, David M. Beazley
    #
    # $Header: /cvsroot/ply/ply/yacc.py,v 1.15 2002/12/12 21:16:18 beazley
    Exp $
    #
    # This library is free software; you can redistribute it and/or
    # modify it under the terms of the GNU Lesser General Public
    # License as published by the Free Software Foundation; either
    # version 2.1 of the License, or (at your option) any later version.
    #
    # This library is distributed in the hope that it will be useful,
    # but WITHOUT ANY WARRANTY; without even the implied warranty of
    # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
    # Lesser General Public License for more details.
    #
    # You should have received a copy of the GNU Lesser General Public
    # License along with this library; if not, write to the Free Software
    # Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
    02111-1307 USA
    #
    # See the file COPYING for a complete copy of the LGPL.
    #
    #
    # This implements an LR parser that is constructed from grammar rules
    defined
    # as Python functions. Roughly speaking, this module is a cross
    between
    # John Aycock's Spark system and the GNU bison utility.
    #
    # Disclaimer: This is a work in progress. SLR parsing seems to work
    fairly
    # well and there is extensive error checking. LALR(1) is in progress.
    The
    # rest of this file is a bit of a mess. Please pardon the dust.
    #
    # The current implementation is only somewhat object-oriented. The
    # LR parser itself is defined in terms of an object (which allows
    multiple
    # parsers to co-exist). However, most of the variables used during
    table
    # construction are defined in terms of global variables. Users
    shouldn't
    # notice unless they are trying to define multiple parsers at the same
    # time using threads (in which case they should have their head
    examined).
    #-----------------------------------------------------------------------------

    __version__ = "1.3"

    #-----------------------------------------------------------------------------
    # === User configurable parameters ===
    #
    # Change these to modify the default behavior of yacc (if you wish)
    #-----------------------------------------------------------------------------

    yaccdebug = 1 # Debugging mode. If set, yacc
    generates a
    # a 'parser.out' file in the current
    directory

    debug_file = 'parser.out' # Default name of the debugging file
    tab_module = 'parsetab' # Default name of the table module
    default_lr = 'SLR' # Default LR table generation method

    error_count = 3 # Number of symbols that must be
    shifted to leave recovery mode

    import re, types, sys, cStringIO, hashlib, os.path

    # Exception raised for yacc-related errors
    class YaccError(Exception): pass

    #-----------------------------------------------------------------------------
    # === LR Parsing Engine ===
    #
    # The following classes are used for the LR parser itself. These are
    not
    # used during table construction and are independent of the actual LR
    # table generation algorithm
    #-----------------------------------------------------------------------------

    # This class is used to hold non-terminal grammar symbols during
    parsing.
    # It normally has the following attributes set:
    # .type = Grammar symbol type
    # .value = Symbol value
    # .lineno = Starting line number
    # .endlineno = Ending line number (optional, set
    automatically)

    class YaccSymbol:
    def __str__(self): return self.type
    def __repr__(self): return str(self)

    # This class is a wrapper around the objects actually passed to each
    # grammar rule. Index lookup and assignment actually assign the
    # .value attribute of the underlying YaccSymbol object.
    # The lineno() method returns the line number of a given
    # item (or 0 if not defined). The linespan() method returns
    # a tuple of (startline,endline) representing the range of lines
    # for a symbol.

    class YaccSlice:
    def __init__(self,s):
    self.slice = s
    self.pbstack = []

    def __getitem__(self,n):
    return self.slice[n].value

    def __setitem__(self,n,v):
    self.slice[n].value = v

    def lineno(self,n):
    return getattr(self.slice[n],"lineno",0)

    def linespan(self,n):
    startline = getattr(self.slice[n],"lineno",0)
    endline = getattr(self.slice[n],"endlineno",startline)
    return startline,endline

    def pushback(self,n):
    if n <= 0:
    raise ValueError, "Expected a positive value"
    if n > (len(self.slice)-1):
    raise ValueError, "Can't push %d tokens. Only %d are
    available." % (n,len(self.slice)-1)
    for i in range(0,n):
    self.pbstack.append(self.slice[-i-1])

    # The LR Parsing engine. This is defined as a class so that multiple
    parsers
    # can exist in the same process. A user never instantiates this
    directly.
    # Instead, the global yacc() function should be used to create a
    suitable Parser
    # object.

    class Parser:
    def __init__(self,magic=None):

    # This is a hack to keep users from trying to instantiate a
    Parser
    # object directly.

    if magic != "xyzzy":
    raise YaccError, "Can't instantiate Parser. Use yacc()
    instead."

    # Reset internal state
    self.productions = None # List of productions
    self.errorfunc = None # Error handling function
    self.action = { } # LR Action table
    self.goto = { } # LR goto table
    self.require = { } # Attribute require table
    self.method = "Unknown LR" # Table construction method
    used

    def errok(self):
    self.errorcount = 0

    def restart(self):
    del self.statestack[:]
    del self.symstack[:]
    sym = YaccSymbol()
    sym.type = '$'
    self.symstack.append(sym)
    self.statestack.append(0)

    def parse(self,input=None,lexer=None,debug=0):
    lookahead = None # Current lookahead symbol
    lookaheadstack = [ ] # Stack of lookahead symbols
    actions = self.action # Local reference to action
    table
    goto = self.goto # Local reference to goto
    table
    prod = self.productions # Local reference to
    production list
    pslice = YaccSlice(None) # Slice object passed to
    grammar rules
    pslice.parser = self # Parser object
    self.errorcount = 0 # Used during error recovery

    # If no lexer was given, we will try to use the lex module
    if not lexer:
    import lex as lexer

    pslice.lexer = lexer

    # If input was supplied, pass to lexer
    if input:
    lexer.input(input)

    # Tokenize function
    get_token = lexer.token

    statestack = [ ] # Stack of parsing states
    self.statestack = statestack
    symstack = [ ] # Stack of grammar symbols
    self.symstack = symstack

    errtoken = None # Err token

    # The start state is assumed to be (0,$)
    statestack.append(0)
    sym = YaccSymbol()
    sym.type = '$'
    symstack.append(sym)

    while 1:
    # Get the next symbol on the input. If a lookahead symbol
    # is already set, we just use that. Otherwise, we'll pull
    # the next token off of the lookaheadstack or from the
    lexer
    if not lookahead:
    if not lookaheadstack:
    lookahead = get_token() # Get the next token
    else:
    lookahead = lookaheadstack.pop()
    if not lookahead:
    lookahead = YaccSymbol()
    lookahead.type = '$'
    if debug:
    print "#%-20s : %s" % (lookahead, [xx.type for xx in
    symstack])

    # Check the action table
    s = statestack[-1]
    ltype = lookahead.type
    t = actions.get((s,ltype),None)

    if t is not None:
    if t > 0:
    # shift a symbol on the stack
    if ltype == '$':
    # Error, end of input
    print "#yacc: Parse error. EOF"
    return
    statestack.append(t)
    symstack.append(lookahead)
    lookahead = None

    # Decrease error count on successful shift
    if self.errorcount > 0:
    self.errorcount -= 1

    continue

    if t < 0:
    # reduce a symbol on the stack, emit a production
    p = prod[-t]
    pname = p.name
    plen = p.len

    # Get production function
    sym = YaccSymbol()
    sym.type = pname # Production name
    sym.value = None

    if plen:
    targ = symstack[-plen-1:]
    targ[0] = sym
    try:
    sym.lineno = targ[1].lineno
    sym.endlineno =
    getattr(targ[-1],"endlineno",targ[-1].lineno)
    except AttributeError:
    sym.lineno = 0
    del symstack[-plen:]
    del statestack[-plen:]
    else:
    sym.lineno = 0
    targ = [ sym ]
    pslice.slice = targ
    pslice.pbstack = []
    # Call the grammar rule with our special slice
    object
    p.func(pslice)

    # Validate attributes of the resulting value
    attribute
    # if require:
    # try:
    # t0 = targ[0]
    # r = Requires.get(t0.type,None)
    # t0d = t0.__dict__
    # if r:
    # for field in r:
    # tn = t0
    # for fname in field:
    # try:
    # tf = tn.__dict__
    # tn = tf.get(fname)
    # except StandardError:
    # tn = None
    # if not tn:
    # print "#%s:%d: Rule %s
    doesn't set required attribute '%s'" % \
    #
    (p.file,p.line,p.name,".".join(field))
    # except TypeError,LookupError:
    # print "#Bad requires directive " % r
    # pass


    # If there was a pushback, put that on the stack
    if pslice.pbstack:
    lookaheadstack.append(lookahead)
    for _t in pslice.pbstack:
    lookaheadstack.append(_t)
    lookahead = None

    symstack.append(sym)
    statestack.append(goto[statestack[-1],pname])
    continue

    if t == 0:
    n = symstack[-1]
    return getattr(n,"value",None)

    if t == None:
    # We have some kind of parsing error here. To handle
    this,
    # we are going to push the current token onto the
    tokenstack
    # and replace it with an 'error' token. If there are
    any synchronization
    # rules, they may catch it.
    #
    # In addition to pushing the error token, we call call
    the user defined p_error()
    # function if this is the first syntax error. This
    function is only called
    # if errorcount == 0.

    if not self.errorcount:
    self.errorcount = error_count
    errtoken = lookahead
    if errtoken.type == '$':
    errtoken = None # End of file!
    if self.errorfunc:
    global errok,token,restart
    errok = self.errok # Set some special
    functions available in error recovery
    token = get_token
    restart = self.restart
    tok = self.errorfunc(errtoken)
    del errok, token, restart # Delete special
    functions

    if not self.errorcount:
    # User must have done some kind of panic
    mode recovery on their own. The returned token
    # is the next lookahead
    lookahead = tok
    errtoken = None
    continue
    else:
    if errtoken:
    if hasattr(errtoken,"lineno"): lineno =
    lookahead.lineno
    else: lineno = 0
    if lineno:
    print "#yacc: Syntax error at line %d,
    token=%s" % (lineno, errtoken.type)
    else:
    print "#yacc: Syntax error, token=%s"
    % errtoken.type
    else:
    print "#yacc: Parse error in input. EOF"
    return

    else:
    self.errorcount = error_count

    # case 1: the statestack only has 1 entry on it. If
    we're in this state, the
    # entire parse has been rolled back and we're
    completely hosed. The token is
    # discarded and we just keep going.

    if len(statestack) <= 1 and lookahead.type != '$':
    lookahead = None
    errtoken = None
    # Nuke the pushback stack
    del lookaheadstack[:]
    continue

    # case 2: the statestack has a couple of entries on
    it, but we're
    # at the end of the file. nuke the top entry and
    generate an error token

    # Start nuking entries on the stack
    if lookahead.type == '$':
    # Whoa. We're really hosed here. Bail out
    return

    if lookahead.type != 'error':
    sym = symstack[-1]
    if sym.type == 'error':
    # Hmmm. Error is on top of stack, we'll just
    nuke input
    # symbol and continue
    lookahead = None
    continue
    t = YaccSymbol()
    t.type = 'error'
    if hasattr(lookahead,"lineno"):
    t.lineno = lookahead.lineno
    t.value = lookahead
    lookaheadstack.append(lookahead)
    lookahead = t
    else:
    symstack.pop()
    statestack.pop()

    continue

    # Call an error function here
    raise RuntimeError, "yacc: internal parser error!!!\n"

    #
    -----------------------------------------------------------------------------
    # === Parser Construction ===
    #
    # The following functions and variables are used to implement the
    yacc() function
    # itself. This is pretty hairy stuff involving lots of error
    checking,
    # construction of LR items, kernels, and so forth. Although a lot of
    # this work is done using global variables, the resulting Parser
    object
    # is completely self contained--meaning that it is safe to repeatedly
    # call yacc() with different grammars in the same application.
    #
    -----------------------------------------------------------------------------

    #
    -----------------------------------------------------------------------------
    # validate_file()
    #
    # This function checks to see if there are duplicated p_rulename()
    functions
    # in the parser module file. Without this function, it is really easy
    for
    # users to make mistakes by cutting and pasting code fragments (and
    it's a real
    # bugger to try and figure out why the resulting parser doesn't
    work). Therefore,
    # we just do a little regular expression pattern matching of def
    statements
    # to try and detect duplicates.
    #
    -----------------------------------------------------------------------------

    def validate_file(filename):
    base,ext = os.path.splitext(filename)
    if ext != '.py': return 1 # No idea. Assume it's okay.

    try:
    f = open(filename)
    lines = f.readlines()
    f.close()
    except IOError:
    return 1 # Oh well

    # Match def p_funcname(
    fre = re.compile(r'\s*def\s+(p_[a-zA-Z_0-9]*)\(')
    counthash = { }
    linen = 1
    noerror = 1
    for l in lines:
    m = fre.match(l)
    if m:
    name = m.group(1)
    prev = counthash.get(name)
    if not prev:
    counthash[name] = linen
    else:
    print "#%s:%d: Function %s redefined. Previously
    defined on line %d" % (filename,linen,name,prev)
    noerror = 0
    linen += 1
    return noerror

    # This function looks for functions that might be grammar rules, but
    which don't have the proper p_suffix.
    def validate_dict(d):
    for n,v in d.items():
    if n[0:2] == 'p_' and isinstance(v,types.FunctionType):
    continue
    if n[0:2] == 't_': continue

    if n[0:2] == 'p_':
    print "#yacc: Warning. '%s' not defined as a function" % n
    if isinstance(v,types.FunctionType) and
    v.func_code.co_argcount == 1:
    try:
    doc = v.__doc__.split(" ")
    if doc[1] == ':':
    print "#%s:%d: Warning. Possible grammar rule '%s'
    defined without p_ prefix." % (v.func_code.co_filename,
    v.func_code.co_firstlineno,n)
    except StandardError:
    pass

    #
    -----------------------------------------------------------------------------
    # === GRAMMAR FUNCTIONS ===
    #
    # The following global variables and functions are used to store,
    manipulate,
    # and verify the grammar rules specified by the user.
    #
    -----------------------------------------------------------------------------

    # Initialize all of the global variables used during grammar
    construction
    def initialize_vars():
    global Productions, Prodnames, Prodmap, Terminals
    global Nonterminals, First, Follow, Precedence, LRitems
    global Errorfunc, Signature, Requires

    Productions = [None] # A list of all of the productions. The
    first
    # entry is always reserved for the purpose
    of
    # building an augmented grammar

    Prodnames = { } # A dictionary mapping the names of
    nonterminals to a list of all
    # productions of that nonterminal.

    Prodmap = { } # A dictionary that is only used to detect
    duplicate
    # productions.

    Terminals = { } # A dictionary mapping the names of
    terminal symbols to a
    # list of the rules where they are used.

    Nonterminals = { } # A dictionary mapping names of
    nonterminals to a list
    # of rule numbers where they are used.

    First = { } # A dictionary of precomputed FIRST(x)
    symbols

    Follow = { } # A dictionary of precomputed FOLLOW(x)
    symbols

    Precedence = { } # Precedence rules for each terminal.
    Contains tuples of the
    # form ('right',level) or ('nonassoc',
    level) or ('left',level)

    LRitems = [ ] # A list of all LR items for the grammar.
    These are the
    # productions with the "dot" like E -> E .
    PLUS E

    Errorfunc = None # User defined error handler

    Signature = hashlib.md5() # Digital signature of the grammar
    rules, precedence
    # and other information. Used to
    determined when a
    # parsing table needs to be
    regenerated.

    Requires = { } # Requires list

    # File objects used when creating the parser.out debugging file
    global _vf, _vfc
    _vf = cStringIO.StringIO()
    _vfc = cStringIO.StringIO()

    #
    -----------------------------------------------------------------------------
    # class Production:
    #
    # This class stores the raw information about a single production or
    grammar rule.
    # It has a few required attributes:
    #
    # name - Name of the production (nonterminal)
    # prod - A list of symbols making up its production
    # number - Production number.
    #
    # In addition, a few additional attributes are used to help with
    debugging or
    # optimization of table generation.
    #
    # file - File where production action is defined.
    # lineno - Line number where action is defined
    # func - Action function
    # prec - Precedence level
    # lr_next - Next LR item. Example, if we are ' E -> E . PLUS E'
    # then lr_next refers to 'E -> E PLUS . E'
    # lr_index - LR item index (location of the ".") in the prod
    list.
    # len - Length of the production (number of symbols on
    right hand side)
    #
    -----------------------------------------------------------------------------

    class Production:
    def __init__(self,**kw):
    for k,v in kw.items():
    setattr(self,k,v)
    self.lr_index = -1
    self.lr0_added = 0 # Flag indicating whether or not added
    to LR0 closure
    self.usyms = [ ]

    def __str__(self):
    if self.prod:
    s = "%s -> %s" % (self.name," ".join(self.prod))
    else:
    s = "%s -> <empty>" % self.name
    return s

    def __repr__(self):
    return str(self)

    # Compute lr_items from the production
    def lr_item(self,n):
    if n > len(self.prod): return None
    p = Production()
    p.name = self.name
    p.prod = list(self.prod)
    p.number = self.number
    p.lr_index = n
    p.prod.insert(n,".")
    p.prod = tuple(p.prod)
    p.len = len(p.prod)
    p.usyms = self.usyms

    # Precompute list of productions immediately following
    try:
    p.lrafter = Prodnames[p.prod[n+1]]
    except (IndexError,KeyError),e:
    p.lrafter = []
    try:
    p.lrbefore = p.prod[n-1]
    except IndexError:
    p.lrbefore = None

    return p

    class MiniProduction:
    pass

    # Utility function
    def is_identifier(s):
    for c in s:
    if not (c.isalnum() or c == '_'): return 0
    return 1

    #
    -----------------------------------------------------------------------------
    # add_production()
    #
    # Given an action function, this function assembles a production rule.
    # The production rule is assumed to be found in the function's
    docstring.
    # This rule has the general syntax:
    #
    # name1 ::= production1
    # | production2
    # | production3
    # ...
    # | productionn
    # name2 ::= production1
    # | production2
    # ...
    #
    -----------------------------------------------------------------------------

    def add_production(f,file,line,prodname,syms):

    if Terminals.has_key(prodname):
    print "#%s:%d: Illegal rule name '%s'. Already defined as a
    token." % (file,line,prodname)
    return -1
    if prodname == 'error':
    print "#%s:%d: Illegal rule name '%s'. error is a reserved
    word." % (file,line,prodname)
    return -1

    if not is_identifier(prodname):
    print "#%s:%d: Illegal rule name '%s'" % (file,line,prodname)
    return -1

    for s in syms:
    if not is_identifier(s) and s != '%prec':
    print "#%s:%d: Illegal name '%s' in rule '%s'" %
    (file,line,s, prodname)
    return -1

    # See if the rule is already in the rulemap
    map = "%s -> %s" % (prodname,syms)
    if Prodmap.has_key(map):
    m = Prodmap[map]
    print "#%s:%d: Duplicate rule %s." % (file,line, m)
    print "#%s:%d: Previous definition at %s:%d" % (file,line,
    m.file, m.line)
    return -1

    p = Production()
    p.name = prodname
    p.prod = syms
    p.file = file
    p.line = line
    p.func = f
    p.number = len(Productions)


    Productions.append(p)
    Prodmap[map] = p
    if not Nonterminals.has_key(prodname):
    Nonterminals[prodname] = [ ]

    # Add all terminals to Terminals
    i = 0
    while i < len(p.prod):
    t = p.prod
    if t == '%prec':
    try:
    precname = p.prod[i+1]
    except IndexError:
    print "#%s:%d: Syntax error. Nothing follows %%prec."
    % (p.file,p.line)
    return -1

    prec = Precedence.get(precname,None)
    if not prec:
    print "#%s:%d: Nothing known about the precedence of
    '%s'" % (p.file,p.line,precname)
    return -1
    else:
    p.prec = prec
    del p.prod
    del p.prod
    continue

    if Terminals.has_key(t):
    Terminals[t].append(p.number)
    # Is a terminal. We'll assign a precedence to p based on
    this
    if not hasattr(p,"prec"):
    p.prec = Precedence.get(t,('right',0))
    else:
    if not Nonterminals.has_key(t):
    Nonterminals[t] = [ ]
    Nonterminals[t].append(p.number)
    i += 1

    if not hasattr(p,"prec"):
    p.prec = ('right',0)

    # Set final length of productions
    p.len = len(p.prod)
    p.prod = tuple(p.prod)

    # Calculate unique syms in the production
    p.usyms = [ ]
    for s in p.prod:
    if s not in p.usyms:
    p.usyms.append(s)

    # Add to the global productions list
    try:
    Prodnames[p.name].append(p)
    except KeyError:
    Prodnames[p.name] = [ p ]
    return 0

    # Given a raw rule function, this function rips out its doc string
    # and adds rules to the grammar

    def add_function(f):
    line = f.func_code.co_firstlineno
    file = f.func_code.co_filename
    error = 0

    if f.func_code.co_argcount > 1:
    print "#%s:%d: Rule '%s' has too many arguments." %
    (file,line,f.__name__)
    return -1

    if f.func_code.co_argcount < 1:
    print "#%s:%d: Rule '%s' requires an argument." %
    (file,line,f.__name__)
    return -1

    if f.__doc__:
    # Split the doc string into lines
    pstrings = f.__doc__.splitlines()
    lastp = None
    dline = line
    for ps in pstrings:
    dline += 1
    p = ps.split()
    if not p: continue
    try:
    if p[0] == '|':
    # This is a continuation of a previous rule
    if not lastp:
    print "#%s:%d: Misplaced '|'." % (file,dline)
    return -1
    prodname = lastp
    if len(p) > 1:
    syms = p[1:]
    else:
    syms = [ ]
    else:
    prodname = p[0]
    lastp = prodname
    assign = p[1]
    if len(p) > 2:
    syms = p[2:]
    else:
    syms = [ ]
    if assign != ':' and assign != '::=':
    print "#%s:%d: Syntax error. Expected ':'" %
    (file,dline)
    return -1
    e = add_production(f,file,dline,prodname,syms)
    error += e
    except StandardError:
    print "#%s:%d: Syntax error in rule '%s'" %
    (file,dline,ps)
    error -= 1
    else:
    print "#%s:%d: No documentation string specified in function
    '%s'" % (file,line,f.__name__)
    return error


    # Cycle checking code (Michael Dyck)

    def compute_reachable():
    '''
    Find each symbol that can be reached from the start symbol.
    Print a warning for any nonterminals that can't be reached.
    (Unused terminals have already had their warning.)
    '''
    Reachable = { }
    for s in Terminals.keys() + Nonterminals.keys():
    Reachable = 0

    mark_reachable_from( Productions[0].prod[0], Reachable )

    for s in Nonterminals.keys():
    if not Reachable:
    print "#yacc: Symbol '%s' is unreachable." % s

    def mark_reachable_from(s, Reachable):
    '''
    Mark all symbols that are reachable from symbol s.
    '''
    if Reachable:
    # We've already reached symbol s.
    return
    Reachable = 1
    for p in Prodnames.get(s,[]):
    for r in p.prod:
    mark_reachable_from(r, Reachable)

    #
    -----------------------------------------------------------------------------
    # compute_terminates()
    #
    # This function looks at the various parsing rules and tries to detect
    # infinite recursion cycles (grammar rules where there is no possible
    way
    # to derive a string of only terminals).
    #
    -----------------------------------------------------------------------------
    def compute_terminates():
    '''
    Raise an error for any symbols that don't terminate.
    '''
    Terminates = {}

    # Terminals:
    for t in Terminals.keys():
    Terminates[t] = 1

    Terminates['$'] = 1

    # Nonterminals:

    # Initialize to false:
    for n in Nonterminals.keys():
    Terminates[n] = 0

    # Then propagate termination until no change:
    while 1:
    some_change = 0
    for (n,pl) in Prodnames.items():
    # Nonterminal n terminates iff any of its productions
    terminates.
    for p in pl:
    # Production p terminates iff all of its rhs symbols
    terminate.
    for s in p.prod:
    if not Terminates:
    # The symbol s does not terminate,
    # so production p does not terminate.
    p_terminates = 0
    break
    else:
    # didn't break from the loop,
    # so every symbol s terminates
    # so production p terminates.
    p_terminates = 1

    if p_terminates:
    # symbol n terminates!
    if not Terminates[n]:
    Terminates[n] = 1
    some_change = 1
    # Don't need to consider any more productions for
    this n.
    break

    if not some_change:
    break

    some_error = 0
    for (s,terminates) in Terminates.items():
    if not terminates:
    if not Prodnames.has_key(s) and not Terminals.has_key(s)
    and s != 'error':
    # s is used-but-not-defined, and we've already warned
    of that,
    # so it would be overkill to say that it's also non-
    terminating.
    pass
    else:
    print "#yacc: Infinite recursion detected for symbol
    '%s'." % s
    some_error = 1

    return some_error

    #
    -----------------------------------------------------------------------------
    # verify_productions()
    #
    # This function examines all of the supplied rules to see if they seem
    valid.
    #
    -----------------------------------------------------------------------------
    def verify_productions(cycle_check=1):
    error = 0
    for p in Productions:
    if not p: continue

    for s in p.prod:
    if not Prodnames.has_key(s) and not Terminals.has_key(s)
    and s != 'error':
    print "#%s:%d: Symbol '%s' used, but not defined as a
    token or a rule." % (p.file,p.line,s)
    error = 1
    continue

    unused_tok = 0
    # Now verify all of the tokens
    if yaccdebug:
    _vf.write("Unused terminals:\n\n")
    for s,v in Terminals.items():
    if s != 'error' and not v:
    print "#yacc: Warning. Token '%s' defined, but not used."
    % s
    if yaccdebug: _vf.write(" %s\n"% s)
    unused_tok += 1

    # Print out all of the productions
    if yaccdebug:
    _vf.write("\nGrammar\n\n")
    for i in range(1,len(Productions)):
    _vf.write("Rule %-5d %s\n" % (i, Productions))

    unused_prod = 0
    # Verify the use of all productions
    for s,v in Nonterminals.items():
    if not v:
    p = Prodnames[0]
    print "#%s:%d: Warning. Rule '%s' defined, but not used."
    % (p.file,p.line, s)
    unused_prod += 1


    if unused_tok == 1:
    print "#yacc: Warning. There is 1 unused token."
    if unused_tok > 1:
    print "#yacc: Warning. There are %d unused tokens." %
    unused_tok

    if unused_prod == 1:
    print "#yacc: Warning. There is 1 unused rule."
    if unused_prod > 1:
    print "#yacc: Warning. There are %d unused rules." %
    unused_prod

    if yaccdebug:
    _vf.write("\nTerminals, with rules where they appear\n\n")
    ks = Terminals.keys()
    ks.sort()
    for k in ks:
    _vf.write("%-20s : %s\n" % (k, " ".join([str(s) for s in
    Terminals[k]])))
    _vf.write("\nNonterminals, with rules where they appear\n\n")
    ks = Nonterminals.keys()
    ks.sort()
    for k in ks:
    _vf.write("%-20s : %s\n" % (k, " ".join([str(s) for s in
    Nonterminals[k]])))

    if (cycle_check):
    compute_reachable()
    error += compute_terminates()
    # error += check_cycles()
    return error

    #
    -----------------------------------------------------------------------------
    # build_lritems()
    #
    # This function walks the list of productions and builds a complete
    set of the
    # LR items. The LR items are stored in two ways: First, they are
    uniquely
    # numbered and placed in the list _lritems. Second, a linked list of
    LR items
    # is built for each production. For example:
    #
    # E -> E PLUS E
    #
    # Creates the list
    #
    # [E -> . E PLUS E, E -> E . PLUS E, E -> E PLUS . E, E -> E PLUS
    E . ]
    #
    -----------------------------------------------------------------------------

    def build_lritems():
    for p in Productions:
    lastlri = p
    lri = p.lr_item(0)
    i = 0
    while 1:
    lri = p.lr_item(i)
    lastlri.lr_next = lri
    if not lri: break
    lri.lr_num = len(LRitems)
    LRitems.append(lri)
    lastlri = lri
    i += 1

    # In order for the rest of the parser generator to work, we need
    to
    # guarantee that no more lritems are generated. Therefore, we
    nuke
    # the p.lr_item method. (Only used in debugging)
    # Production.lr_item = None

    #
    -----------------------------------------------------------------------------
    # add_precedence()
    #
    # Given a list of precedence rules, add to the precedence table.
    #
    -----------------------------------------------------------------------------

    def add_precedence(plist):
    plevel = 0
    error = 0
    for p in plist:
    plevel += 1
    try:
    prec = p[0]
    terms = p[1:]
    if prec != 'left' and prec != 'right' and prec !=
    'nonassoc':
    print "#yacc: Invalid precedence '%s'" % prec
    return -1
    for t in terms:
    if Precedence.has_key(t):
    print "#yacc: Precedence already specified for
    terminal '%s'" % t
    error += 1
    continue
    Precedence[t] = (prec,plevel)
    except:
    print "#yacc: Invalid precedence table."
    error += 1

    return error

    #
    -----------------------------------------------------------------------------
    # augment_grammar()
    #
    # Compute the augmented grammar. This is just a rule S' -> start
    where start
    # is the starting symbol.
    #
    -----------------------------------------------------------------------------

    def augment_grammar(start=None):
    if not start:
    start = Productions[1].name
    Productions[0] =
    Production(name="S'",prod=[start],number=0,len=1,prec=('right',
    0),func=None)
    Productions[0].usyms = [ start ]
    Nonterminals[start].append(0)


    #
    -------------------------------------------------------------------------
    # first()
    #
    # Compute the value of FIRST1(beta) where beta is a tuple of symbols.
    #
    # During execution of compute_first1, the result may be incomplete.
    # Afterward (e.g., when called from compute_follow()), it will be
    complete.
    #
    -------------------------------------------------------------------------
    def first(beta):

    # We are computing First(x1,x2,x3,...,xn)
    result = [ ]
    for x in beta:
    x_produces_empty = 0

    # Add all the non-<empty> symbols of First[x] to the result.
    for f in First[x]:
    if f == '<empty>':
    x_produces_empty = 1
    else:
    if f not in result: result.append(f)

    if x_produces_empty:
    # We have to consider the next x in beta,
    # i.e. stay in the loop.
    pass
    else:
    # We don't have to consider any further symbols in beta.
    break
    else:
    # There was no 'break' from the loop,
    # so x_produces_empty was true for all x in beta,
    # so beta produces empty as well.
    result.append('<empty>')

    return result


    # FOLLOW(x)
    # Given a non-terminal. This function computes the set of all symbols
    # that might follow it. Dragon book, p. 189.

    def compute_follow(start=None):
    # Add '$' to the follow list of the start symbol
    for k in Nonterminals.keys():
    Follow[k] = [ ]

    if not start:
    start = Productions[1].name

    Follow[start] = [ '$' ]

    while 1:
    didadd = 0
    for p in Productions[1:]:
    # Here is the production set
    for i in range(len(p.prod)):
    B = p.prod
    if Nonterminals.has_key(B):
    # Okay. We got a non-terminal in a production
    fst = first(p.prod[i+1:])
    hasempty = 0
    for f in fst:
    if f != '<empty>' and f not in Follow:
    Follow.append(f)
    didadd = 1
    if f == '<empty>':
    hasempty = 1
    if hasempty or i == (len(p.prod)-1):
    # Add elements of follow(a) to follow(b)
    for f in Follow[p.name]:
    if f not in Follow:
    Follow.append(f)
    didadd = 1
    if not didadd: break

    if 0 and yaccdebug:
    _vf.write('\nFollow:\n')
    for k in Nonterminals.keys():
    _vf.write("%-20s : %s\n" % (k, " ".join([str(s) for s in
    Follow[k]])))

    #
    -------------------------------------------------------------------------
    # compute_first1()
    #
    # Compute the value of FIRST1(X) for all symbols
    #
    -------------------------------------------------------------------------
    def compute_first1():

    # Terminals:
    for t in Terminals.keys():
    First[t] = [t]

    First['$'] = ['$']
    First['#'] = ['#'] # what's this for?

    # Nonterminals:

    # Initialize to the empty set:
    for n in Nonterminals.keys():
    First[n] = []

    # Then propagate symbols until no change:
    while 1:
    some_change = 0
    for n in Nonterminals.keys():
    for p in Prodnames[n]:
    for f in first(p.prod):
    if f not in First[n]:
    First[n].append( f )
    some_change = 1
    if not some_change:
    break

    if 0 and yaccdebug:
    _vf.write('\nFirst:\n')
    for k in Nonterminals.keys():
    _vf.write("%-20s : %s\n" %
    (k, " ".join([str(s) for s in First[k]])))

    #
    -----------------------------------------------------------------------------
    # === SLR Generation ===
    #
    # The following functions are used to construct SLR (Simple LR)
    parsing tables
    # as described on p.221-229 of the dragon book.
    #
    -----------------------------------------------------------------------------

    # Global variables for the LR parsing engine
    def lr_init_vars():
    global _lr_action, _lr_goto, _lr_method
    global _lr_goto_cache

    _lr_action = { } # Action table
    _lr_goto = { } # Goto table
    _lr_method = "Unknown" # LR method used
    _lr_goto_cache = { }

    # Compute the LR(0) closure operation on I, where I is a set of LR(0)
    items.
    # prodlist is a list of productions.

    _add_count = 0 # Counter used to detect cycles

    def lr0_closure(I):
    global _add_count

    _add_count += 1
    prodlist = Productions

    # Add everything in I to J
    J = I[:]
    didadd = 1
    while didadd:
    didadd = 0
    for j in J:
    for x in j.lrafter:
    if x.lr0_added == _add_count: continue
    # Add B --> .G to J
    J.append(x.lr_next)
    x.lr0_added = _add_count
    didadd = 1

    return J

    # Compute the LR(0) goto function goto(I,X) where I is a set
    # of LR(0) items and X is a grammar symbol. This function is written
    # in a way that guarantees uniqueness of the generated goto sets
    # (i.e. the same goto set will never be returned as two different
    Python
    # objects). With uniqueness, we can later do fast set comparisons
    using
    # id(obj) instead of element-wise comparison.

    def lr0_goto(I,x):
    # First we look for a previously cached entry
    g = _lr_goto_cache.get((id(I),x),None)
    if g: return g

    # Now we generate the goto set in a way that guarantees uniqueness
    # of the result

    s = _lr_goto_cache.get(x,None)
    if not s:
    s = { }
    _lr_goto_cache[x] = s

    gs = [ ]
    for p in I:
    n = p.lr_next
    if n and n.lrbefore == x:
    s1 = s.get(id(n),None)
    if not s1:
    s1 = { }
    s[id(n)] = s1
    gs.append(n)
    s = s1
    g = s.get('$',None)
    if not g:
    if gs:
    g = lr0_closure(gs)
    s['$'] = g
    else:
    s['$'] = gs
    _lr_goto_cache[(id(I),x)] = g
    return g

    # Compute the kernel of a set of LR(0) items
    def lr0_kernel(I):
    KI = [ ]
    for p in I:
    if p.name == "S'" or p.lr_index > 0 or p.len == 0:
    KI.append(p)

    return KI

    _lr0_cidhash = { }

    # Compute the LR(0) sets of item function
    def lr0_items():

    C = [ lr0_closure([Productions[0].lr_next]) ]
    i = 0
    for I in C:
    _lr0_cidhash[id(I)] = i
    i += 1

    # Loop over the items in C and each grammar symbols
    i = 0
    while i < len(C):
    I = C
    i += 1

    # Collect all of the symbols that could possibly be in the
    goto(I,X) sets
    asyms = { }
    for ii in I:
    for s in ii.usyms:
    asyms = None

    for x in asyms.keys():
    g = lr0_goto(I,x)
    if not g: continue
    if _lr0_cidhash.has_key(id(g)): continue
    _lr0_cidhash[id(g)] = len(C)
    C.append(g)

    return C

    #
    -----------------------------------------------------------------------------
    # slr_parse_table()
    #
    # This function constructs an SLR table.
    #
    -----------------------------------------------------------------------------
    def slr_parse_table():
    global _lr_method
    goto = _lr_goto # Goto array
    action = _lr_action # Action array
    actionp = { } # Action production array (temporary)

    _lr_method = "SLR"

    n_srconflict = 0
    n_rrconflict = 0

    print "#yacc: Generating SLR parsing table..."
    if yaccdebug:
    _vf.write("\n\nParsing method: SLR\n\n")

    # Step 1: Construct C = { I0, I1, ... IN}, collection of LR(0)
    items
    # This determines the number of states

    C = lr0_items()

    # Build the parser table, state by state
    st = 0
    for I in C:
    # Loop over each production in I
    actlist = [ ] # List of actions

    if yaccdebug:
    _vf.write("\nstate %d\n\n" % st)
    for p in I:
    _vf.write(" (%d) %s\n" % (p.number, str(p)))
    _vf.write("\n")

    for p in I:
    try:
    if p.prod[-1] == ".":
    if p.name == "S'":
    # Start symbol. Accept!
    action[st,"$"] = 0
    actionp[st,"$"] = p
    else:
    # We are at the end of a production. Reduce!
    for a in Follow[p.name]:
    actlist.append((a,p,"reduce using rule %d
    (%s)" % (p.number,p)))
    r = action.get((st,a),None)
    if r is not None:
    # Whoa. Have a shift/reduce or reduce/
    reduce conflict
    if r > 0:
    # Need to decide on shift or
    reduce here
    # By default we favor shifting.
    Need to add
    # some precedence rules here.
    sprec,slevel =
    Productions[actionp[st,a].number].prec
    rprec,rlevel = Precedence.get(a,
    ('right',0))
    if (slevel < rlevel) or ((slevel
    == rlevel) and (rprec == 'left')):
    # We really need to reduce
    here.
    action[st,a] = -p.number
    actionp[st,a] = p
    if not slevel and not rlevel:
    _vfc.write("shift/reduce
    conflict in state %d resolved as reduce.\n" % st)
    _vf.write(" ! shift/
    reduce conflict for %s resolved as reduce.\n" % a)
    n_srconflict += 1
    elif (slevel == rlevel) and (rprec
    == 'nonassoc'):
    action[st,a] = None
    else:
    # Hmmm. Guess we'll keep the
    shift
    if not slevel and not rlevel:
    _vfc.write("shift/reduce
    conflict in state %d resolved as shift.\n" % st)
    _vf.write(" ! shift/
    reduce conflict for %s resolved as shift.\n" % a)
    n_srconflict
    +=1
    elif r < 0:
    # Reduce/reduce conflict. In
    this case, we favor the rule
    # that was defined first in the
    grammar file
    oldp = Productions[-r]
    pp = Productions[p.number]
    if oldp.line > pp.line:
    action[st,a] = -p.number
    actionp[st,a] = p
    # print "#Reduce/reduce conflict
    in state %d" % st
    n_rrconflict += 1
    _vfc.write("reduce/reduce conflict
    in state %d resolved using rule %d (%s).\n" % (st,
    actionp[st,a].number, actionp[st,a]))
    _vf.write(" ! reduce/reduce
    conflict for %s resolved using rule %d (%s).\n" %
    (a,actionp[st,a].number, actionp[st,a]))
    else:
    print "#Unknown conflict in state
    %d" % st
    else:
    action[st,a] = -p.number
    actionp[st,a] = p
    else:
    i = p.lr_index
    a = p.prod[i+1] # Get symbol right after the
    "."
    if Terminals.has_key(a):
    g = lr0_goto(I,a)
    j = _lr0_cidhash.get(id(g),-1)
    if j >= 0:
    # We are in a shift state
    actlist.append((a,p,"shift and go to state
    %d" % j))
    r = action.get((st,a),None)
    if r is not None:
    # Whoa have a shift/reduce or shift/
    shift conflict
    if r > 0:
    if r != j:
    print "#Shift/shift conflict
    in state %d" % st
    elif r < 0:
    # Do a precedence check.
    # - if precedence of reduce
    rule is higher, we reduce.
    # - if precedence of reduce is
    same and left assoc, we reduce.
    # - otherwise we shift
    rprec,rlevel =
    Productions[actionp[st,a].number].prec
    sprec,slevel = Precedence.get(a,
    ('right',0))
    if (slevel > rlevel) or ((slevel
    == rlevel) and (rprec != 'left')):
    # We decide to shift here...
    highest precedence to shift
    action[st,a] = j
    actionp[st,a] = p
    if not slevel and not rlevel:
    n_srconflict += 1
    _vfc.write("shift/reduce
    conflict in state %d resolved as shift.\n" % st)
    _vf.write(" ! shift/
    reduce conflict for %s resolved as shift.\n" % a)
    elif (slevel == rlevel) and (rprec
    == 'nonassoc'):
    action[st,a] = None

    else:
    # Hmmm. Guess we'll keep the
    reduce
    if not slevel and not rlevel:
    n_srconflict +=1
    _vfc.write("shift/reduce
    conflict in state %d resolved as reduce.\n" % st)
    _vf.write(" ! shift/
    reduce conflict for %s resolved as reduce.\n" % a)

    else:
    print "#Unknown conflict in state
    %d" % st
    else:
    action[st,a] = j
    actionp[st,a] = p

    except StandardError,e:
    raise YaccError, "Hosed in slr_parse_table", e

    # Print the actions associated with each terminal
    if yaccdebug:
    for a,p,m in actlist:
    if action.has_key((st,a)):
    if p is actionp[st,a]:
    _vf.write(" %-15s %s\n" % (a,m))
    _vf.write("\n")
    for a,p,m in actlist:
    if action.has_key((st,a)):
    if p is not actionp[st,a]:
    _vf.write(" ! %-15s [ %s ]\n" % (a,m))

    # Construct the goto table for this state
    if yaccdebug:
    _vf.write("\n")
    nkeys = { }
    for ii in I:
    for s in ii.usyms:
    if Nonterminals.has_key(s):
    nkeys = None
    for n in nkeys.keys():
    g = lr0_goto(I,n)
    j = _lr0_cidhash.get(id(g),-1)
    if j >= 0:
    goto[st,n] = j
    if yaccdebug:
    _vf.write(" %-15s shift and go to state %d\n" %
    (n,j))

    st += 1

    if n_srconflict == 1:
    print "#yacc: %d shift/reduce conflict" % n_srconflict
    if n_srconflict > 1:
    print "#yacc: %d shift/reduce conflicts" %
    n_srconflict
    if n_rrconflict == 1:
    print "#yacc: %d reduce/reduce conflict" % n_rrconflict
    if n_rrconflict > 1:
    print "#yacc: %d reduce/reduce conflicts" % n_rrconflict


    #
    -----------------------------------------------------------------------------
    # ==== LALR(1) Parsing ====
    # **** UNFINISHED! 6/16/01
    #
    -----------------------------------------------------------------------------


    # Compute the lr1_closure of a set I. I is a list of tuples (p,a)
    where
    # p is a LR0 item and a is a terminal

    _lr1_add_count = 0

    def lr1_closure(I):
    global _lr1_add_count

    _lr1_add_count += 1

    J = I[:]

    # Loop over items (p,a) in I.
    ji = 0
    while ji < len(J):
    p,a = J[ji]
    # p = [ A -> alpha . B beta]

    # For each production B -> gamma
    for B in p.lr1_after:
    f = tuple(p.lr1_beta + (a,))

    # For each terminal b in first(Beta a)
    for b in first(f):
    # Check if (B -> . gamma, b) is in J
    # Only way this can happen is if the add count
    mismatches
    pn = B.lr_next
    if pn.lr_added.get(b,0) == _lr1_add_count: continue
    pn.lr_added = _lr1_add_count
    J.append((pn,b))
    ji += 1

    return J

    def lalr_parse_table():

    # Compute some lr1 information about all of the productions
    for p in LRitems:
    try:
    after = p.prod[p.lr_index + 1]
    p.lr1_after = Prodnames[after]
    p.lr1_beta = p.prod[p.lr_index + 2:]
    except LookupError:
    p.lr1_after = [ ]
    p.lr1_beta = [ ]
    p.lr_added = { }

    # Compute the LR(0) items
    C = lr0_items()
    CK = []
    for I in C:
    CK.append(lr0_kernel(I))

    print CK

    #
    -----------------------------------------------------------------------------
    # ==== LR Utility functions ====
    #
    -----------------------------------------------------------------------------

    #
    -----------------------------------------------------------------------------
    # _lr_write_tables()
    #
    # This function writes the LR parsing tables to a file
    #
    -----------------------------------------------------------------------------

    def lr_write_tables(modulename=tab_module):
    filename = modulename + ".py"
    try:
    f = open(filename,"w")

    f.write("""
    # %s
    # This file is automatically generated. Do not edit.

    _lr_method = %s

    _lr_signature = %s
    """ % (filename, repr(_lr_method), repr(Signature.digest())))

    # Change smaller to 0 to go back to original tables
    smaller = 1

    # Factor out names to try and make smaller
    if smaller:
    items = { }

    for k,v in _lr_action.items():
    i = items.get(k[1])
    if not i:
    i = ([],[])
    items[k[1]] = i
    i[0].append(k[0])
    i[1].append(v)

    f.write("\n_lr_action_items = {")
    for k,v in items.items():
    f.write("%r:([" % k)
    for i in v[0]:
    f.write("%r," % i)
    f.write("],[")
    for i in v[1]:
    f.write("%r," % i)

    f.write("]),")
    f.write("}\n")

    f.write("""
    _lr_action = { }
    for _k, _v in _lr_action_items.items():
    for _x,_y in zip(_v[0],_v[1]):
    _lr_action[(_x,_k)] = _y
    del _lr_action_items
    """)

    else:
    f.write("\n_lr_action = { ");
    for k,v in _lr_action.items():
    f.write("(%r,%r):%r," % (k[0],k[1],v))
    f.write("}\n");

    if smaller:
    # Factor out names to try and make smaller
    items = { }

    for k,v in _lr_goto.items():
    i = items.get(k[1])
    if not i:
    i = ([],[])
    items[k[1]] = i
    i[0].append(k[0])
    i[1].append(v)

    f.write("\n_lr_goto_items = {")
    for k,v in items.items():
    f.write("%r:([" % k)
    for i in v[0]:
    f.write("%r," % i)
    f.write("],[")
    for i in v[1]:
    f.write("%r," % i)

    f.write("]),")
    f.write("}\n")

    f.write("""
    _lr_goto = { }
    for _k, _v in _lr_goto_items.items():
    for _x,_y in zip(_v[0],_v[1]):
    _lr_goto[(_x,_k)] = _y
    del _lr_goto_items
    """)
    else:
    f.write("\n_lr_goto = { ");
    for k,v in _lr_goto.items():
    f.write("(%r,%r):%r," %
    (k[0],k[1],v))
    f.write("}\n");

    # Write production table
    f.write("_lr_productions = [\n")
    for p in Productions:
    if p:
    if (p.func):
    f.write(" (%r,%d,%r,%r,%d),\n" % (p.name, p.len,
    p.func.__name__,p.file,p.line))
    else:
    f.write(" (%r,%d,None,None,None),\n" % (p.name,
    p.len))
    else:
    f.write(" None,\n")
    f.write("]\n")
    f.close()

    except IOError,e:
    print "#Unable to create '%s'" % filename
    print e
    return

    def lr_read_tables(module=tab_module,optimize=0):
    global _lr_action, _lr_goto, _lr_productions, _lr_method
    try:
    exec "import %s as parsetab" % module

    if (optimize) or (Signature.digest() ==
    parsetab._lr_signature):
    _lr_action = parsetab._lr_action
    _lr_goto = parsetab._lr_goto
    _lr_productions = parsetab._lr_productions
    _lr_method = parsetab._lr_method
    return 1
    else:
    return 0

    except (ImportError,AttributeError):
    return 0

    #
    -----------------------------------------------------------------------------
    # yacc(module)
    #
    # Build the parser module
    #
    -----------------------------------------------------------------------------

    def yacc(method=default_lr, debug=yaccdebug, module=None,
    tabmodule=tab_module, start=None, check_recursion=1, optimize=0):
    global yaccdebug
    yaccdebug = debug

    initialize_vars()
    files = { }
    error = 0

    # Add starting symbol to signature
    if start:
    Signature.update(start)

    # Try to figure out what module we are working with
    if module:
    # User supplied a module object.
    if not isinstance(module, types.ModuleType):
    raise ValueError,"Expected a module"

    ldict = module.__dict__

    else:
    # No module given. We might be able to get information from
    the caller.
    # Throw an exception and unwind the traceback to get the
    globals

    try:
    raise RuntimeError
    except RuntimeError:
    e,b,t = sys.exc_info()
    f = t.tb_frame
    f = f.f_back # Walk out to our calling function
    ldict = f.f_globals # Grab its globals dictionary

    # If running in optimized mode. We're going to

    if (optimize and lr_read_tables(tabmodule,1)):
    # Read parse table
    del Productions[:]
    for p in _lr_productions:
    if not p:
    Productions.append(None)
    else:
    m = MiniProduction()
    m.name = p[0]
    m.len = p[1]
    m.file = p[3]
    m.line = p[4]
    if p[2]:
    m.func = ldict[p[2]]
    Productions.append(m)

    else:
    # Get the tokens map
    tokens = ldict.get("tokens",None)

    if not tokens:
    raise YaccError,"module does not define a list 'tokens'"
    if not (isinstance(tokens,types.ListType) or
    isinstance(tokens,types.TupleType)):
    raise YaccError,"tokens must be a list or tuple."

    # Check to see if a requires dictionary is defined.
    requires = ldict.get("require",None)
    if requires:
    if not (isinstance(requires,types.DictType)):
    raise YaccError,"require must be a dictionary."

    for r,v in requires.items():
    try:
    if not (isinstance(v,types.ListType)):
    raise TypeError
    v1 = [x.split(".") for x in v]
    Requires[r] = v1
    except StandardError:
    print "#Invalid specification for rule '%s' in
    require. Expected a list of strings" % r


    # Build the dictionary of terminals. We a record a 0 in the
    # dictionary to track whether or not a terminal is actually
    # used in the grammar

    if 'error' in tokens:
    print "#yacc: Illegal token 'error'. Is a reserved word."
    raise YaccError,"Illegal token name"

    for n in tokens:
    if Terminals.has_key(n):
    print "#yacc: Warning. Token '%s' multiply defined." %
    n
    Terminals[n] = [ ]

    Terminals['error'] = [ ]

    # Get the precedence map (if any)
    prec = ldict.get("precedence",None)
    if prec:
    if not (isinstance(prec,types.ListType) or
    isinstance(prec,types.TupleType)):
    raise YaccError,"precedence must be a list or tuple."
    add_precedence(prec)
    Signature.update(repr(prec))

    for n in tokens:
    if not Precedence.has_key(n):
    Precedence[n] = ('right',0) # Default, right
    associative, 0 precedence

    # Look for error handler
    ef = ldict.get('p_error',None)
    if ef:
    if not isinstance(ef,types.FunctionType):
    raise YaccError,"'p_error' defined, but is not a
    function."
    eline = ef.func_code.co_firstlineno
    efile = ef.func_code.co_filename
    files[efile] = None

    if (ef.func_code.co_argcount != 1):
    raise YaccError,"%s:%d: p_error() requires 1
    argument." % (efile,eline)
    global Errorfunc
    Errorfunc = ef
    else:
    print "#yacc: Warning. no p_error() function is defined."

    # Get the list of built-in functions with p_ prefix
    symbols = [ldict[f] for f in ldict.keys()
    if (isinstance(ldict[f],types.FunctionType) and
    ldict[f].__name__[:2] == 'p_'
    and ldict[f].__name__ != 'p_error')]

    # Check for non-empty symbols
    if len(symbols) == 0:
    raise YaccError,"no rules of the form p_rulename are
    defined."

    # Sort the symbols by line number
    symbols.sort(lambda x,y:
    cmp(x.func_code.co_firstlineno,y.func_code.co_firstlineno))

    # Add all of the symbols to the grammar
    for f in symbols:
    if (add_function(f)) < 0:
    error += 1
    else:
    files[f.func_code.co_filename] = None

    # Make a signature of the docstrings
    for f in symbols:
    if f.__doc__:
    Signature.update(f.__doc__)

    lr_init_vars()

    if error:
    raise YaccError,"Unable to construct parser."

    if not lr_read_tables(tabmodule):

    # Validate files
    for filename in files.keys():
    if not validate_file(filename):
    error = 1

    # Validate dictionary
    validate_dict(ldict)

    if start and not Prodnames.has_key(start):
    raise YaccError,"Bad starting symbol '%s'" % start

    augment_grammar(start)
    error = verify_productions(cycle_check=check_recursion)
    otherfunc = [ldict[f] for f in ldict.keys()
    if (isinstance(ldict[f],types.FunctionType) and
    ldict[f].__name__[:2] != 'p_')]

    if error:
    raise YaccError,"Unable to construct parser."

    build_lritems()
    compute_first1()
    compute_follow(start)

    if method == 'SLR':
    slr_parse_table()
    elif method == 'LALR1':
    lalr_parse_table()
    return
    else:
    raise YaccError, "Unknown parsing method '%s'" %
    method

    lr_write_tables(tabmodule)

    if yaccdebug:
    try:
    f = open(debug_file,"w")
    f.write(_vfc.getvalue())
    f.write("\n\n")
    f.write(_vf.getvalue())
    f.close()
    except IOError,e:
    print "#yacc: can't create '%s'" % debug_file,e

    # Made it here. Create a parser object and set up its internal
    state.
    # Set global parse() method to bound method of parser object.

    p = Parser("xyzzy")
    p.productions = Productions
    p.errorfunc = Errorfunc
    p.action = _lr_action
    p.goto = _lr_goto
    p.method = _lr_method
    p.require = Requires

    global parse
    parse = p.parse

    # Clean up all of the globals we created
    if (not optimize):
    yacc_cleanup()
    return p

    # yacc_cleanup function. Delete all of the global variables
    # used during table construction

    def yacc_cleanup():
    global _lr_action, _lr_goto, _lr_method, _lr_goto_cache
    del _lr_action, _lr_goto, _lr_method, _lr_goto_cache

    global Productions, Prodnames, Prodmap, Terminals
    global Nonterminals, First, Follow, Precedence, LRitems
    global Errorfunc, Signature, Requires

    del Productions, Prodnames, Prodmap, Terminals
    del Nonterminals, First, Follow, Precedence, LRitems
    del Errorfunc, Signature, Requires

    global _vf, _vfc
    del _vf, _vfc


    # Stub that raises an error if parsing is attempted without first
    calling yacc()
    def parse(*args,**kwargs):
    raise YaccError, "yacc: No parser built with yacc()"
    mike, Jan 11, 2012
    #3
  4. On Wed, 11 Jan 2012 12:28:52 +0100, Laurent Claessens
    <> wrote:

    >
    >Python 2.6.6 (r266:84292, Sep 15 2010, 16:22:56)
    >[GCC 4.4.5] on linux2
    >Type "help", "copyright", "credits" or "license" for more information.
    >>>> import hashlib
    >>>> h=hashlib.new("md5")
    >>>> h.update("Hello")
    >>>> h.hexdigest()

    >'d992641f1b2b9c08b569c0a17c4e7cb8'
    >

    Interesting... I get a different result here...

    PythonWin 2.7.1 (r271:86832, Feb 7 2011, 11:33:02) [MSC v.1500 64 bit
    (AMD64)] on win32.
    Portions Copyright 1994-2008 Mark Hammond - see 'Help/About PythonWin'
    for further copyright information.
    >>> import hashlib
    >>> h = hashlib.new("md5")
    >>> h.update("Hello")
    >>> h.hexdigest()

    '8b1a9953c4611296a827abf8c47804d7'
    >>>


    PythonWin 2.7.1 (r271:86832, Feb 7 2011, 11:30:38) [MSC v.1500 32 bit
    (Intel)] on win32.
    Portions Copyright 1994-2008 Mark Hammond - see 'Help/About PythonWin'
    for further copyright information.
    >>> import hashlib
    >>> h = hashlib.new("md5")
    >>> h.update("Hello")
    >>> h.hexdigest()

    '8b1a9953c4611296a827abf8c47804d7'
    >>>

    --
    Wulfraed Dennis Lee Bieber AF6VN
    HTTP://wlfraed.home.netcom.com/
    Dennis Lee Bieber, Jan 11, 2012
    #4
  5. mike

    Terry Reedy Guest

    On 1/11/2012 6:19 AM, mike wrote:
    > Hi,
    >
    > We are running are running Python program on Redhat 5.5.
    >
    > When executing our program we get the following error ( see below).
    >
    > Any ideas what this is due to?
    >
    > br,
    >
    > //mike
    >
    > /pysibelius/lib/common/


    > DataTypes.py
    > Overwriten ...
    > ERROR:root:code for hash md5 was not found.


    These 3 lines do not come from the interpreter.
    It might come from the hashlib module.

    > Traceback (most recent call last):
    > File "/python/lib/python2.7/hashlib.py", line 139, in<module>
    > globals()[__func_name] = __get_hash(__func_name)
    > File "/python/lib/python2.7/hashlib.py", line 91, in
    > __get_builtin_constructor
    > raise ValueError('unsupported hash type %s' % name)
    > ValueError: unsupported hash type md5


    This is a standard traceback from the interpreter.
    The manual says that md5 should always be available.
    On the other hand, hashlib depends on the OpenSSL library on your
    system. Perhaps RedHat removed md5 because it has been broken.
    I suggest you ask them or check your OpenSSL doc. The Python manual may
    need to be changed.

    > ERROR:root:code for hash sha1 was not found


    --
    Terry Jan Reedy
    Terry Reedy, Jan 11, 2012
    #5
  6. On Wed, 11 Jan 2012 05:54:29 -0800, mike wrote:

    > I did some more digging and found that our class imports a "yacc.py"
    > that uses
    >
    > import re, types, sys, cStringIO, hashlib, os.path
    >
    > so it has hashlib.
    >
    > yacc.py seems to be an old version 1.3 ( I found 2.3 -->).
    >
    > Reading about hashlib it seems to be dependent on os installation of
    > OpenSSL but I cannot find out how.


    It shouldn't be. It will use OpenSSL if available, otherwise it will fall
    back on its own code.


    > br,
    >
    > //mike
    >
    > yacc.py
    > ===============
    >

    #-----------------------------------------------------------------------------
    > # ply: yacc.py

    [snip over TWO THOUSAND lines of code]


    Mike, what lead you to believe that an error in hashlib could be solved
    by posting the ENTIRE two thousand lines of yacc.py? Please do not post
    such huge chunks of code unless asked. It is unnecessary and annoying.

    Try this. Open a terminal window and enter "python" at the prompt to
    start a clean interactive session. Then enter the following commands:

    import sys
    print(sys.version)
    import hashlib
    print(hashlib.__file__)
    print(hashlib.md5)
    import _md5
    print(_md5.__file__)

    and copy and paste (do not retype) the full output of these commands.




    Thank you.




    --
    Steven
    Steven D'Aprano, Jan 11, 2012
    #6
  7. On Wed, 11 Jan 2012 13:25:16 -0500, Terry Reedy <>
    wrote:

    Given

    >> /pysibelius/lib/common/

    >


    I suspect some library for accessing Sibelius music notation files...

    >> DataTypes.py
    >> Overwriten ...
    >> ERROR:root:code for hash md5 was not found.

    >
    >These 3 lines do not come from the interpreter.
    >It might come from the hashlib module.
    >

    .... might be responsible for those lines.

    >
    >> ERROR:root:code for hash sha1 was not found


    Note that whatever reported the MD5 failure appears to have tried a
    follow-over to SHA1 and isn't finding that either.

    Makes me suspect the entire hashlib library may be, uhm, corrupted?
    --
    Wulfraed Dennis Lee Bieber AF6VN
    HTTP://wlfraed.home.netcom.com/
    Dennis Lee Bieber, Jan 12, 2012
    #7
  8. mike

    mike Guest

    On Jan 12, 12:28 am, Steven D'Aprano <steve
    > wrote:
    > On Wed, 11 Jan 2012 05:54:29 -0800, mike wrote:
    > > I did some more digging and found that our class imports a "yacc.py"
    > > that uses

    >
    > > import re, types, sys, cStringIO, hashlib, os.path

    >
    > > so it has hashlib.

    >
    > > yacc.py seems to be an old version 1.3 ( I found 2.3 -->).

    >
    > > Reading about hashlib it seems to be dependent on os installation of
    > > OpenSSL but I cannot find out how.

    >
    > It shouldn't be. It will use OpenSSL if available, otherwise it will fall
    > back on its own code.
    >
    > > br,

    >
    > > //mike

    >
    > > yacc.py
    > > ===============

    >
    > #-----------------------------------------------------------------------------> # ply: yacc.py
    >
    > [snip over TWO THOUSAND lines of code]
    >
    > Mike, what lead you to believe that an error in hashlib could be solved
    > by posting the ENTIRE two thousand lines of yacc.py? Please do not post
    > such huge chunks of code unless asked. It is unnecessary and annoying.
    >
    > Try this. Open a terminal window and enter "python" at the prompt to
    > start a clean interactive session. Then enter the following commands:
    >
    > import sys
    > print(sys.version)
    > import hashlib
    > print(hashlib.__file__)
    > print(hashlib.md5)
    > import _md5
    > print(_md5.__file__)
    >
    > and copy and paste (do not retype) the full output of these commands.
    >
    > Thank you.
    >
    > --
    > Steven


    Hi,

    Sorry for posting huge file. I added your copy-paste snippet.

    esekilx5030 [7:09am] [roamFroBl/pysibelius/bin] -> python
    Python 2.7.2 (default, Jun 16 2011, 15:05:49)
    [GCC 4.5.0] on linux2
    Type "help", "copyright", "credits" or "license" for more information.
    >>> import sys

    print(sys.version)
    import hashlib
    print(hashlib.__file__)
    print(hashlib.md5)
    import _md5
    print(_md5.__file__) >>> 2.7.2 (default, Jun 16 2011, 15:05:49)
    [GCC 4.5.0]
    >>> ERROR:root:code for hash md5 was not found.

    Traceback (most recent call last):
    File "/vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/
    hashlib.py", line 139, in <module>
    globals()[__func_name] = __get_hash(__func_name)
    File "/vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/
    hashlib.py", line 91, in __get_builtin_constructor
    raise ValueError('unsupported hash type %s' % name)
    ValueError: unsupported hash type md5
    ERROR:root:code for hash sha1 was not found.
    Traceback (most recent call last):
    File "/vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/
    hashlib.py", line 139, in <module>
    globals()[__func_name] = __get_hash(__func_name)
    File "/vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/
    hashlib.py", line 91, in __get_builtin_constructor
    raise ValueError('unsupported hash type %s' % name)
    ValueError: unsupported hash type sha1
    ERROR:root:code for hash sha224 was not found.
    Traceback (most recent call last):
    File "/vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/
    hashlib.py", line 139, in <module>
    globals()[__func_name] = __get_hash(__func_name)
    File "/vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/
    hashlib.py", line 91, in __get_builtin_constructor
    raise ValueError('unsupported hash type %s' % name)
    ValueError: unsupported hash type sha224
    ERROR:root:code for hash sha256 was not found.
    Traceback (most recent call last):
    File "/vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/
    hashlib.py", line 139, in <module>
    globals()[__func_name] = __get_hash(__func_name)
    File "/vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/
    hashlib.py", line 91, in __get_builtin_constructor
    raise ValueError('unsupported hash type %s' % name)
    ValueError: unsupported hash type sha256
    ERROR:root:code for hash sha384 was not found.
    Traceback (most recent call last):
    File "/vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/
    hashlib.py", line 139, in <module>
    globals()[__func_name] = __get_hash(__func_name)
    File "/vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/
    hashlib.py", line 91, in __get_builtin_constructor
    raise ValueError('unsupported hash type %s' % name)
    ValueError: unsupported hash type sha384
    ERROR:root:code for hash sha512 was not found.
    Traceback (most recent call last):
    File "/vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/
    hashlib.py", line 139, in <module>
    globals()[__func_name] = __get_hash(__func_name)
    File "/vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/
    hashlib.py", line 91, in __get_builtin_constructor
    raise ValueError('unsupported hash type %s' % name)
    ValueError: unsupported hash type sha512
    >>> /vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/hashlib.pyc
    >>> Traceback (most recent call last):

    File "<stdin>", line 1, in <module>
    AttributeError: 'module' object has no attribute 'md5'
    >>> Traceback (most recent call last):

    File "<stdin>", line 1, in <module>
    ImportError: No module named _md5

    Thanks for support!

    //mike
    mike, Jan 12, 2012
    #8
  9. mike

    mike Guest

    On Jan 12, 7:13 am, mike <> wrote:
    > On Jan 12, 12:28 am, Steven D'Aprano <steve
    >
    >
    >
    >
    >
    >
    >
    >
    >
    > > wrote:
    > > On Wed, 11 Jan 2012 05:54:29 -0800, mike wrote:
    > > > I did some more digging and found that our class imports a "yacc.py"
    > > > that uses

    >
    > > > import re, types, sys, cStringIO, hashlib, os.path

    >
    > > > so it has hashlib.

    >
    > > > yacc.py seems to be an old version 1.3 ( I found 2.3 -->).

    >
    > > > Reading about hashlib it seems to be dependent on os installation of
    > > > OpenSSL but I cannot find out how.

    >
    > > It shouldn't be. It will use OpenSSL if available, otherwise it will fall
    > > back on its own code.

    >
    > > > br,

    >
    > > > //mike

    >
    > > > yacc.py
    > > > ===============

    >
    > > #-----------------------------------------------------------------------------> # ply: yacc.py

    >
    > > [snip over TWO THOUSAND lines of code]

    >
    > > Mike, what lead you to believe that an error in hashlib could be solved
    > > by posting the ENTIRE two thousand lines of yacc.py? Please do not post
    > > such huge chunks of code unless asked. It is unnecessary and annoying.

    >
    > > Try this. Open a terminal window and enter "python" at the prompt to
    > > start a clean interactive session. Then enter the following commands:

    >
    > > import sys
    > > print(sys.version)
    > > import hashlib
    > > print(hashlib.__file__)
    > > print(hashlib.md5)
    > > import _md5
    > > print(_md5.__file__)

    >
    > > and copy and paste (do not retype) the full output of these commands.

    >
    > > Thank you.

    >
    > > --
    > > Steven

    >
    > Hi,
    >
    > Sorry for posting huge file. I added your copy-paste snippet.
    >
    > esekilx5030 [7:09am] [roamFroBl/pysibelius/bin] -> python
    > Python 2.7.2 (default, Jun 16 2011, 15:05:49)
    > [GCC 4.5.0] on linux2
    > Type "help", "copyright", "credits" or "license" for more information.>>>import sys
    >
    > print(sys.version)
    > import hashlib
    > print(hashlib.__file__)
    > print(hashlib.md5)
    > import _md5
    > print(_md5.__file__) >>> 2.7.2 (default, Jun 16 2011, 15:05:49)
    > [GCC 4.5.0]>>> ERROR:root:code for hash md5 was not found.
    >
    > Traceback (most recent call last):
    >   File "/vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/
    > hashlib.py", line 139, in <module>
    >     globals()[__func_name] = __get_hash(__func_name)
    >   File "/vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/
    > hashlib.py", line 91, in __get_builtin_constructor
    >     raise ValueError('unsupported hash type %s' % name)
    > ValueError: unsupported hash type md5
    > ERROR:root:code for hash sha1 was not found.
    > Traceback (most recent call last):
    >   File "/vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/
    > hashlib.py", line 139, in <module>
    >     globals()[__func_name] = __get_hash(__func_name)
    >   File "/vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/
    > hashlib.py", line 91, in __get_builtin_constructor
    >     raise ValueError('unsupported hash type %s' % name)
    > ValueError: unsupported hash type sha1
    > ERROR:root:code for hash sha224 was not found.
    > Traceback (most recent call last):
    >   File "/vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/
    > hashlib.py", line 139, in <module>
    >     globals()[__func_name] = __get_hash(__func_name)
    >   File "/vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/
    > hashlib.py", line 91, in __get_builtin_constructor
    >     raise ValueError('unsupported hash type %s' % name)
    > ValueError: unsupported hash type sha224
    > ERROR:root:code for hash sha256 was not found.
    > Traceback (most recent call last):
    >   File "/vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/
    > hashlib.py", line 139, in <module>
    >     globals()[__func_name] = __get_hash(__func_name)
    >   File "/vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/
    > hashlib.py", line 91, in __get_builtin_constructor
    >     raise ValueError('unsupported hash type %s' % name)
    > ValueError: unsupported hash type sha256
    > ERROR:root:code for hash sha384 was not found.
    > Traceback (most recent call last):
    >   File "/vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/
    > hashlib.py", line 139, in <module>
    >     globals()[__func_name] = __get_hash(__func_name)
    >   File "/vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/
    > hashlib.py", line 91, in __get_builtin_constructor
    >     raise ValueError('unsupported hash type %s' % name)
    > ValueError: unsupported hash type sha384
    > ERROR:root:code for hash sha512 was not found.
    > Traceback (most recent call last):
    >   File "/vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/
    > hashlib.py", line 139, in <module>
    >     globals()[__func_name] = __get_hash(__func_name)
    >   File "/vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/
    > hashlib.py", line 91, in __get_builtin_constructor
    >     raise ValueError('unsupported hash type %s' % name)
    > ValueError: unsupported hash type sha512>>> /vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/hashlib.pyc
    > >>> Traceback (most recent call last):

    >
    >   File "<stdin>", line 1, in <module>
    > AttributeError: 'module' object has no attribute 'md5'>>> Traceback (mostrecent call last):
    >
    >   File "<stdin>", line 1, in <module>
    > ImportError: No module named _md5
    >
    > Thanks for support!
    >
    > //mike


    Hi,

    I did some more testing.

    The python 2.7.2 lib is stored in a clearcase vob. When I set the env.
    and run suggested lines ( with slight modification), on Redhat 5.5,
    ( by Steven) I get the following:

    -> python
    Python 2.7.2 (default, Jun 16 2011, 15:05:49)
    [GCC 4.5.0] on linux2
    Type "help", "copyright", "credits" or "license" for more information.
    >>> import sys

    print(sys.version)
    import hashlib
    print(hashlib.__file__)
    print(hashlib.md5)
    import md5
    print(md5.__file__)>>> 2.7.2 (default, Jun 16 2011, 15:05:49)
    [GCC 4.5.0]
    >>> ERROR:root:code for hash md5 was not found.

    Traceback (most recent call last):
    File "/vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/
    hashlib.py", line 139, in <module>
    globals()[__func_name] = __get_hash(__func_name)
    File "/vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/
    hashlib.py", line 91, in __get_builtin_constructor
    raise ValueError('unsupported hash type %s' % name)
    ValueError: unsupported hash type md5
    ERROR:root:code for hash sha1 was not found.
    Traceback (most recent call last):
    File "/vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/
    hashlib.py", line 139, in <module>
    globals()[__func_name] = __get_hash(__func_name)
    File "/vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/
    hashlib.py", line 91, in __get_builtin_constructor
    raise ValueError('unsupported hash type %s' % name)
    ValueError: unsupported hash type sha1
    ERROR:root:code for hash sha224 was not found.
    Traceback (most recent call last):
    File "/vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/
    hashlib.py", line 139, in <module>
    globals()[__func_name] = __get_hash(__func_name)
    File "/vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/
    hashlib.py", line 91, in __get_builtin_constructor
    raise ValueError('unsupported hash type %s' % name)
    ValueError: unsupported hash type sha224
    ERROR:root:code for hash sha256 was not found.
    Traceback (most recent call last):
    File "/vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/
    hashlib.py", line 139, in <module>
    globals()[__func_name] = __get_hash(__func_name)
    File "/vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/
    hashlib.py", line 91, in __get_builtin_constructor
    raise ValueError('unsupported hash type %s' % name)
    ValueError: unsupported hash type sha256
    ERROR:root:code for hash sha384 was not found.
    Traceback (most recent call last):
    File "/vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/
    hashlib.py", line 139, in <module>
    globals()[__func_name] = __get_hash(__func_name)
    File "/vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/
    hashlib.py", line 91, in __get_builtin_constructor
    raise ValueError('unsupported hash type %s' % name)
    ValueError: unsupported hash type sha384
    ERROR:root:code for hash sha512 was not found.
    Traceback (most recent call last):
    File "/vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/
    hashlib.py", line 139, in <module>
    globals()[__func_name] = __get_hash(__func_name)
    File "/vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/
    hashlib.py", line 91, in __get_builtin_constructor
    raise ValueError('unsupported hash type %s' % name)
    ValueError: unsupported hash type sha512
    >>> /vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/hashlib.pyc
    >>> Traceback (most recent call last):

    File "<stdin>", line 1, in <module>
    AttributeError: 'module' object has no attribute 'md5'
    >>> Traceback (most recent call last):

    File "<stdin>", line 1, in <module>
    File "/vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/
    md5.py", line 10, in <module>
    from hashlib import md5
    ImportError: cannot import name md5

    But when I execute the same lines on a Suse 10.4 I get the following
    output:

    python
    Python 2.7.2 (default, Jun 16 2011, 15:05:49)
    [GCC 4.5.0] on linux2
    Type "help", "copyright", "credits" or "license" for more information.
    >>> import sys
    >>> print(sys.version)

    2.7.2 (default, Jun 16 2011, 15:05:49)
    [GCC 4.5.0]
    >>> import hashlib
    >>> print(hashlib.__file__)

    /vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/hashlib.pyc
    >>> print(hashlib.md5)

    <built-in function openssl_md5>
    >>> import md5
    >>> print(md5.__file__)

    /vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/md5.pyc

    That looks correct to me.

    It really confuses me!?

    br,

    //mike

    Ps. Thanks for all the replies.
    mike, Jan 12, 2012
    #9
  10. On Wed, 11 Jan 2012 22:13:16 -0800, mike wrote:
    [...]
    > esekilx5030 [7:09am] [roamFroBl/pysibelius/bin] -> python Python 2.7.2
    > (default, Jun 16 2011, 15:05:49) [GCC 4.5.0] on linux2
    > Type "help", "copyright", "credits" or "license" for more information.
    >>>> import sys

    > print(sys.version)
    > import hashlib
    > print(hashlib.__file__)
    > print(hashlib.md5)
    > import _md5
    > print(_md5.__file__) >>> 2.7.2 (default, Jun 16 2011, 15:05:49) [GCC
    > 4.5.0]
    >>>> ERROR:root:code for hash md5 was not found.

    > Traceback (most recent call last):
    > File "/vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/
    > hashlib.py", line 139, in <module>
    > globals()[__func_name] = __get_hash(__func_name)
    > File "/vobs/rnc/rrt/roam1/roamSs/roamFroBl/python/lib/python2.7/
    > hashlib.py", line 91, in __get_builtin_constructor
    > raise ValueError('unsupported hash type %s' % name)



    It looks like pysibelius comes with its own Python installation, which is
    *seriously* broken.

    What is pysibelius? I can't find it on the web. Does it have anything to
    do with Sibelius the music composition software?

    It looks like the installation you are trying to use is missing modules.
    You might need to consult the pysibelius forums or mailing lists, if they
    have any, or the author, or possibly re-install it and see if the problem
    goes away.



    --
    Steven
    Steven D'Aprano, Jan 13, 2012
    #10
  11. mike

    alex23 Guest

    On Jan 13, 1:34 pm, Steven D'Aprano <steve
    > wrote:
    > What is pysibelius? I can't find it on the web. Does it have anything to
    > do with Sibelius the music composition software?


    Yes, please provide more information about the pysibelius package,
    especially if this is the case.

    The few tenuous Python/Sibelius links I found didn't have anything on
    pysibelius, unfortunately.
    alex23, Jan 13, 2012
    #11
  12. mike

    mike Guest

    On Jan 13, 5:41 am, alex23 <> wrote:
    > On Jan 13, 1:34 pm, Steven D'Aprano <steve
    >
    > > wrote:
    > > What is pysibelius? I can't find it on the web. Does it have anything to
    > > do with Sibelius the music composition software?

    >
    > Yes, please provide more information about the pysibelius package,
    > especially if this is the case.
    >
    > The few tenuous Python/Sibelius links I found didn't have anything on
    > pysibelius, unfortunately.


    Hi,

    pysibelius is a lib that we use.

    I am not sure that is the problem since the python program works on
    SuSE but not on RH server. And AFAIK
    the only difference ( well that I can see) is the OpenSSL version.

    According to code it uses openssl:

    <built-in function openssl_md5>

    So I need to find a way to convince the linux sys admin to install
    same version of openssl on both servers.

    Thanks a lot for your valuable time.

    //mike
    mike, Jan 13, 2012
    #12
  13. On 01/13/2012 07:14 AM, mike wrote:
    > pysibelius is a lib that we use.
    >
    > I am not sure that is the problem since the python program works on
    > SuSE but not on RH server. And AFAIK
    > the only difference ( well that I can see) is the OpenSSL version.
    >
    > According to code it uses openssl:
    >
    > <built-in function openssl_md5>
    >
    > So I need to find a way to convince the linux sys admin to install
    > same version of openssl on both servers.


    Unfortunately that's going to be fairly difficult, as they would have to
    install a new version of openssl alongside the existing system version,
    which you can't just replace. Or it might be possible to copy the
    openssl libraries from SuSE over. Either way you will have to use the
    LD_LIBRARY_PATH or the LD_LIBRARY_PRELOAD environment variables to get
    PySibelius to load that version instead of the system one.
    Michael Torrie, Jan 13, 2012
    #13
  14. On Fri, 13 Jan 2012 06:14:50 -0800, mike wrote:

    > On Jan 13, 5:41 am, alex23 <> wrote:
    >> On Jan 13, 1:34 pm, Steven D'Aprano <steve
    >>
    >> > wrote:
    >> > What is pysibelius? I can't find it on the web. Does it have anything
    >> > to do with Sibelius the music composition software?

    >>
    >> Yes, please provide more information about the pysibelius package,
    >> especially if this is the case.
    >>
    >> The few tenuous Python/Sibelius links I found didn't have anything on
    >> pysibelius, unfortunately.

    >
    > Hi,
    >
    > pysibelius is a lib that we use.
    >
    > I am not sure that is the problem since the python program works on SuSE
    > but not on RH server. And AFAIK
    > the only difference ( well that I can see) is the OpenSSL version.


    OpenSSL is irrelevant. If it isn't available, or doesn't provide md5,
    then the hashlib library will use its own implementation. But the _md5
    module is missing in the pysibelius Python on your RedHat system.

    As I said, your Python installation is seriously broken. Required modules
    are just *gone*.

    pysibelius appears to have patched Python in some way, because strange
    unexpected error messages are being printed that do not happen on a
    normal unpatched Python, e.g.:

    ERROR:root:code for hash sha224 was not found.

    That is not a normal Python error message. That looks like something
    added by pysibelius.



    --
    Steven
    Steven D'Aprano, Jan 14, 2012
    #14
  15. mike

    Klaus Guest

    I had that problem after moving my Python installation into another directory. Reinstalling Python helped.
    Klaus, Sep 27, 2012
    #15
  16. mike

    Guest

    On Wednesday, January 11, 2012 12:28:52 PM UTC+1, Laurent Claessens wrote:
    > Le 11/01/2012 12:19, mike a �crit :
    >
    > > Hi,

    >
    > >

    >
    > > We are running are running Python program on Redhat 5.5.

    >
    > >

    >
    > > When executing our program we get the following error ( see below).

    >
    > >

    >
    > > Any ideas what this is due to?

    >
    >
    >
    > On my computer hashlib has "md5" :
    >
    >
    >
    >
    >
    >
    >
    > Python 2.6.6 (r266:84292, Sep 15 2010, 16:22:56)
    >
    > [GCC 4.4.5] on linux2
    >
    > Type "help", "copyright", "credits" or "license" for more information.
    >
    > >>> import hashlib

    >
    > >>> h=hashlib.new("md5")

    >
    > >>> h.update("Hello")

    >
    > >>> h.hexdigest()

    >
    > 'd992641f1b2b9c08b569c0a17c4e7cb8'
    >
    >
    >
    > While if I ask for a hash method that hashlib does not know, I get the
    >
    > same error as you :
    >
    >
    >
    > >>> g=hashlib.new("bla")

    >
    > Traceback (most recent call last):
    >
    > File "<stdin>", line 1, in <module>
    >
    > File "/usr/lib/python2.6/hashlib.py", line 101, in __hash_new
    >
    > return __get_builtin_constructor(name)(string)
    >
    > File "/usr/lib/python2.6/hashlib.py", line 80, in
    >
    > __get_builtin_constructor
    >
    > raise ValueError, "unsupported hash type"
    >
    > ValueError: unsupported hash type
    >
    >
    >
    >
    >
    > You should give us a more comprehensive example of your problem.
    >
    >
    >
    > Laurent


    Dear Laurent,

    I have encountered similar problem and your suggestion was helpful. Thanks a lot!

    Sincerely,

    Bidzina Kapanadze,
    Abastumani Astrophysical Observatory,
    Ilia State University, Tbilisi, Georgia
    , Feb 5, 2013
    #16
    1. Advertising

Want to reply to this thread or ask your own question?

It takes just 2 minutes to sign up (and it's free!). Just click the sign up button to choose a username and then you can ask your own questions on the forum.
Similar Threads
  1. Replies:
    9
    Views:
    17,048
    John Salerno
    May 8, 2006
  2. rp
    Replies:
    1
    Views:
    478
    red floyd
    Nov 10, 2011
  3. Peter Woodsky

    create a md5 / md5 passwd with a salt

    Peter Woodsky, Nov 20, 2008, in forum: Ruby
    Replies:
    6
    Views:
    189
    Brian Candler
    Nov 21, 2008
  4. Ibad Kureshi U0850037

    Re: ERROR:root:code for hash md5 was not found

    Ibad Kureshi U0850037, Feb 3, 2013, in forum: Python
    Replies:
    0
    Views:
    204
    Ibad Kureshi U0850037
    Feb 3, 2013
  5. dieter
    Replies:
    0
    Views:
    208
    dieter
    Feb 4, 2013
Loading...

Share This Page