Permalink
Commits on Feb 16, 2013
  1. Timestamp v3.25.

    theory committed Feb 16, 2013
  2. Remove useless code.

    theory committed Feb 16, 2013
Commits on Feb 14, 2013
  1. Increment to v3.25.

    theory committed Feb 14, 2013
  2. Timestamp v3.24.

    theory committed Feb 14, 2013
  3. Skip X<> tags when searching for titles in the pull parser.

    theory committed Feb 14, 2013
    Also strip out trailing whitespace that may result from ignoring X<> tags.
    Resolves RT#74389, and per discussion on pod-people, archived here:
    
      http://code.activestate.com/lists/perl-pod-people/1698/
Commits on Feb 13, 2013
  1. Add pre and tt to HTML tagmap.

    theory committed Feb 13, 2013
  2. Loosen up man page matching.

    theory committed Feb 13, 2013
  3. Note fix for C<>s in the TOC.

    theory committed Feb 13, 2013
  4. Merge pull request #46 from tsibley/code-in-head-regression

    theory committed Feb 13, 2013
    Fix code (C<>) in head regression
  5. Merge pull request #44 from rjbs/master

    theory committed Feb 13, 2013
    Issue a warning when =item types mismatch within one =over/=back.
  6. Merge pull request #45 from ilmari/speedup-parse

    theory committed Feb 13, 2013
    Parsing speedups from @ilmari.
Commits on Jan 23, 2013
  1. Fix regression with C<> formatting codes inside =headN sections

    tsibley committed Jan 23, 2013
    With 29306b7, headers stash unformatted text in $_[0]{htext} for use in
    the TOC and generating section IDs.  Code sections (C<>) were handled
    separately inside handle_text and never appended to $_[0]{htext},
    meaning headers lacked those pieces of text.  Adjust handle_text to
    always append to the htext stash if inside a header.
    
    The regression tests in t/xhtml10.t now pass.
    
    Resolves [rt.cpan.org #82706].
Commits on Jan 22, 2013
  1. Close over encoding object in transcoder

    ilmari committed Jan 22, 2013
    No need to look it up every time, since the encoding is fixed.
  2. Inline parse_characters and _try_encoding_guess

    ilmari committed Jan 22, 2013
    This is called for every line we parse, so each µs counts.
Commits on Jan 18, 2013
  1. suppress errors in item tests

    rjbs committed Jan 18, 2013
    These tests now trigger the mixed-type =item warning, new in the
    preceding commit.  Rather than test for the exact error here, I have
    suppressed error generation from these three specific tests only.
    
    I think a good strategy for fixing these tests further will be to
    write a JSON emitter to replace the XML one.  This will be much
    easier to use in tests, as we can use JSON::PP to parse the output
    into a structure.  It will make tests like this easier to update
    to match errors, rather than to tweak the output to suppress them.
Commits on Dec 4, 2012
  1. Eliminate warnings.

    theory committed Dec 4, 2012
  2. Credit @demerphq.

    theory committed Dec 4, 2012
  3. Merge pull request #43 from rjbs/master

    theory committed Dec 4, 2012
    fix hash order dependency
Commits on Nov 27, 2012
Commits on Oct 23, 2012
Commits on Oct 16, 2012
  1. Merge pull request #42 from dolmen/fix/github-https

    theory committed Oct 16, 2012
    Fix Github URLs to https://
  2. Fix Github URLs to https://

    dolmen committed Oct 16, 2012
Commits on Oct 8, 2012
  1. Merge pull request #41 from Smylers/no_trailing_punc_id

    theory committed Oct 8, 2012
    Don't end XHTML IDs with punctuation
  2. Don't end XHTML IDs with punctuation

    Smylers committed Oct 8, 2012
    An ID ending with a punctuation character means that a URL with a fragment
    ID linking to that section ends with a punctuation character. Such URLs
    are awkward to extract from plain text, for example when used in a plain
    text e-mail and being converted to a link by a mailing list archive or
    e-mail client: a punctuation character at the end of a URL may be
    interpreted as punctuation in the e-mail message following the URL.
    
    Avoid the problem by ensuring that IDs never end with punctuation
    characters. (Underscore is allowed, since it isn't used for punctuation in
    human text.)
Commits on Aug 28, 2012
  1. Merge pull request #40 from grantm/encoding-heuristic

    theory committed Aug 28, 2012
    Encoding heuristic should be applied to first non-ASCII byte sequence.
Commits on Aug 27, 2012
  1. add tests for encoding heuristic corner cases

    grantm committed Aug 27, 2012
    These tests use undeclared CP1252 since a plausible Latin-1 POD source
    that can be detected as UTF-8 is much harder to contrive.  This also
    documents our expectation that 'smart quote' symbols in CP1252 will be
    rendered as control character if no =encoding was declared.