Txt2tags Changelog


- 20021220 v1.3:
  - new Txt2tags User Guide document
  - new mark `` for RAW strings (pass-thru txt2tags parsing)
    {thanks Leslie Harlley Watter @ .br for requesting}
  - new %!encoding: command to specify the document charset
  - new 'contrib' dir for user contributed stuff
  - improvements on SGML target: now using <toc> and <descrip>
  - added '$' and '@' chars to the URL matcher, so http://this.is/valid@$99
    {thanks Hubert Chan @ .ca for pointing}

  - Sanity: removed from code structures marked as obsoleted on v1.1
  - Sanity: removed \email{} tag from LaTeX headers. Using \url{}.
  - Sanity: \usepackage[latin1]{inputenc} is not default anymore on
    LaTeX headers. If needed, use new encoding command.

  - Bugfix: LaTeX target added on the Web Interface menu
  - Bugfix: \n, \t and other special pairs escaped under `inline mono` 
  - Bugfix: TOC anchor respecting maximum TOC level
  - Bugfix: beautifiers not parsed on TOC items
  - Bugfix: special chars not double escaped on TOC items

- 20021203 v1.2:
  - new 'tex' target (for LaTeX documents)
    {thanks Leslie Harlley Watter @ .br for making it}
  - now multiple source files can be specified on the command line
    as txt2tags -t html *.t2t
    {thanks Maksim Ischenko @ malva.ua for the idea}
  - URL matcher was improved and now gets valid insane paths as
    http://this.com/// and http://this.com?var=abc#anchor
    {thanks Hubert Chan @ .ca for pointing}
  - added a COPYING file on the tarball, with the GPL license
  - added a handy unhtml.vim script to the extras directory, to
    convert by brute force an HTML file to a txt2tags file on Vim.
  - Bugfix: special chars escaped on link label

- 20021106 v1.1:
  - images can now point to links, as: [[img.gif] www.abc.com]
  - new foldmethod=syntax rules on the Vim syntax file
    {thanks Peter Valach @ gmx.net for requesting}
  - now any non-table line closes a table (blank line or not)
  - begin of the major code-cleanup (aka complete rewrite)
    {thanks EVERYBODY ;) for requesting}

  - Sanity: new RULE for headers: if the very first line of the file
    is blank, this means 'this file has no header information'
  - Sanity: new comment char: % at the line begining (no leading spaces!)
  - Sanity: marks are not parsed on title lines
    {thanks Maksim Ischenko @ malva.ua for pointing}
  - Sanity: // as comment mark is obsoleted, due conflicts with
    italic mark. will be removed on version 1.3
  - Sanity: TAB-made tables are marked as obsoleted, will be removed on
    version 1.3. Use the PIPE-made tables instead.
  - Sanity: removed support for filename with spaces on the image mark,
    due conflicts with named URL mark, [like this.gif]
  - new temporary option --fixme to update obsoleted structures and
    automatically fix all this Sanity changes

  - Bugfix: --noheaders now act just as 'supress headers from output',
    and not 'treat headers as plain text'
  - Bugfix: trailing . is now part of the URL when it contains anchor
    location or form data (as #abc. and ?var=abc.)
  - Bugfix: trailing / added to the URL matcher (as www.abc.com/)
    {thanks Rubens Queiroz @ unicamp.br for requesting}
  - Bugfix: title with \ char now is correctly handled

- 20020925 v1.0:
  - new Graphical Tk Interface
  - fixed target file format on Windows and Mac platforms
  - TOC deepness now is limited to level 3
  - RULES file sync'ed with actual rules (sorry!)
    {thanks Grigory Bakunov @ asplinux.ru for requesting}
  - ChangeLog (this file) is now a txt2tags file
    {thanks Grigory Bakunov @ asplinux.ru for suggesting}
  - added underscore char _ to anchor on URL regex

  - Bugfix: Closing any open list or table at EOF
  - Bugfix: HTML anchor name have no #
    {thanks Grigory Bakunov @ asplinux.ru for pointing}

- 20020823 v0.9:
  - now txt2tags is a 100% pure python script. the bash part has gone
    so now it runs nicely on MS Windows and other python aware platforms
    {thanks Ramon Felciano @ ingenuity.com for requesting}
  - new --toc and --toconly options to generate Table Of Contents
  - defined .t2t as the oficial txt2tags file extension
    {thanks Leslie Harlley Watter @ .br for requesting}
  - txt.vim file renamed to txt2tags.vim, and added instructions
  - now titles are underlined on target txt
    {thanks Ramon Felciano @ ingenuity.com for requesting}
  - cmdline used to generate the document is inserted as a comment on it
  - tarball reorganized, adding 'extras' and 'samples' directories

  - Bugfix: when all ok, force system exit status to 0
    {thanks Marcelo Subtil Marcal @ conectiva.br for pointing}

- 20020703 v0.8:
  - new abuseme.txt complex sample file (test-suite)
  - new smart image align feature for HTML (see abuseme.t2t)
    {thanks Rubens Queiroz @ unicamp.br for requesting}
  - new --maskemail option to hide email from SPAM robots
    {thanks Rubens Queiroz @ unicamp.br for requesting}
  - table now can have border or no on HTML
  - improvements on txt.vim syntax file, now colors works on gui also
  - image mark can't accept filename with spaces (conflicts named link)

  - Bugfix: parse more than one %%date on the same line
  - Bugfix: special chars now escaped inside preformatted
  - Bugfix: closing quote mark was kinda broken

- 20020620 v0.7:
  - new Emacs syntax highlight file for txt2tags rules
    {thanks Leslie Harlley Watter @ .br for making it}
  - new mark for definition lists (<DL> on html)
  - now adding protocol to guessed link like www.abc.com
  - explicit link mark changed from ["label" url] to [label url]
    in other words, quotes are not necessary anymore
  - image mark now correctly handles filenames with space for html
  - corrected typos on sample.txt file
    {thanks Rodrigo Stulzer @ conectiva.br for pointing}
  - named links now can point to local links as file.html, #anchor
    and file.html#anchor
    {thanks Rubens Queiroz @ unicamp.br for requesting}

- 20020410 v0.6:
  - tables are now supported for sgml and moin targets
  - new --enumtitle option to enumerate all titles as 1, 1.1, 1.1.1, etc
  - added numbered list type for all targets, the mark is '+ '
  - better pre-formatted font escaping (won't parse marks)
  - URL matcher now supports ftp://user:passwd@domain.com login URLs
    and user@domain.com?subject=test&cc=me@domain.com filled emails
    {thanks Rubens Queiroz @ unicamp.br for requesting}

- 20020322 v0.5:
  - new handy Web interface to use it online (internet or intranet)
  - new --noheaders option to suppress headers information
  - now it can read the marked text from STDIN (specify - as file)
  - adapted to work on python old v1.5 also

- 20020311 v0.4:
  - new simple table support (just for HTML by now)
  - fixed lots of bugs on the man target, now it's kinda usable
  - the preformatted line mark must have a space after the dashes: '--- '
  - the preformatted line now has leading spaces
  - added the classic -h, --help, -V and --version options
  - URL matcher is smarter, suporting #local_anchors and ?form=data
    {thanks Rubens Queiroz @ unicamp.br for requesting}

- 20011109 v0.3:
  - new explicit URL/email mark with label: ["my label" http://duh.com]
    {thanks Mark Stankus @ calpoly.edu for requesting}
  - date macro now supports format string like %% date(%m/%d/%Y)
    {thanks Mark Stankus @ calpoly.edu for requesting}

- 20011001 v0.2:
  - new 'man' target
  - nice shell wrapper to deal with files/directories/options.
    in fact, the python code is now "embeded" on the sh script.
  - new --lang, --split options (for sgml2html)

- 20010726 v0.1:
  - initial release





