ERRATA
-------------------------------------------------------------------

p.xvi -- Halfway down:

    severeral --> several 

p.6 -- 2nd line S.B.:

    short_regvals = filter(both(shortline, isRegDBVal), lines)

p.24 -- Final three lines of interactive shell S.B.:

    >>> ldn = LogDictNew()
    >>> ldn['this'] = 'that'
    Set: this->that

p.36 -- Definition of AutoFillingDict should contain:

    def __getitem__(self, key):
        return self.data.get(key, '')

p.36 -- Definition of ClosureDict should contain:

    def __getitem__(self, key):
        return self.data.get(key, '%('+key+')s')

p.60 -- 2nd function explained should be:

    filecmp.dircmp.report_full_closure()

p.77 -- First and second line:

    os.mkdirs -> os.makedirs

p.145 -- 2nd example in swapcase(), should read:

    >>> "mary HAD a little lamb!".swapcase()
    'MARY had A LITTLE LAMB!'

p.162 -- 2nd paragraph, 2nd sentence should read:

    Unlike uu, quopri adds no content headers to encoded data; ...

p.202 -- Last sentence should read:

    Does it do better than SlicedZPickleIndexer (the best variant
    included in both speed and space)?

p.213 -- Third line should read:
 
    >>> re_show(r'th[^s]*s', s2)

p.219 -- 1st paragraph, 2nd sentence should read:

    These are similar to regular grouped subexpressions, ...

p.262 -- Last complete paragraph, second sentence, should read:

    ...since the latter can form a subsequence of the former."

p.271 -- Code sample near bottom, second to last line should be:

    map(splitnames, useful(parts(txt,'\n['))) )

p.273 -- The .add_state() method should read:

    def add_state(self, handler, end_state=0):
        self.handlers.append(handler)
        if end_state:
            self.endStates.append(handler)

p.313 -- mx.TextTools.prefix, first sentence should read:

    ...that matches the start of s.

p.364 -- Second listing of 'email.Utils.formataddr' should be 
         'email.Utils.formatdate'

p.385 -- First paragraph under HTMLParser.HTMLParser()

    ...is fairly useful... --> ...is fairly useless...

p.392 -- First sentence in urlparse should read:

    The module urlparse supports just one...
 
p.426 -- 1st full paragraph, 1st sentence should read:

    Each argument to the print statement...

p.426 -- last sentence of same paragraph:

    String interpolation is also performed...

p.448 -- in 'def hypotenuse':

    'a,b = ab[:]' --> 'a,b = ab'

p.448 -- 3/4 down:

    "substitue" should read "substitute"

p.482 -- 2nd sentence should read:

    Complexities one often sees are:

Back cover:

    Book URL incorrectly listed as http://gnosis.cx/TpiP/ in the
    main text (but correct at lower right).  A redirect mitigates 
    the error.


------------------------------------------------------------------------
Thanks to:

    Glenn R Williams <gloonie@earthlink.net>
    Bill Scherer <Bill.Scherer@verizonwireless.com>
    Greg Lee <glee@pharsight.com>
    John Hazen <john@hazen.net>
    Bob Kimble <bob@iplicity.com>
    Gary Duncan <gmduncan@netspace.net.au>
    Tim Churches <tchur@optushome.com.au>
    Andrew Purshottam <andy@andypurshottam.com>
    Nils Barth <nils_barth@post.harvard.edu>
    John W. Warren, CAA <John.Warren@caa.army.mil>
    Chen Levy <gnosis.cx@chenlevy.com>