in this package has split into py-clearsilver and clearsilver-base.
Release Notes for Clearsilver 0.9.13 11/24/2004
-------------------------------------
bugfix release
- one incompatible change: semi-public function cgi_html_ws_strip now
takes a second argument to set the level of stripping
- white space stripper now has two levels: 1 is the old debug level
(which doesn't strip whitespace at the beginning of a line) and the
second strips all duplicate whitespace
- make java/perl/python/ruby wrappers for CS include the html strfuncs
(ie, filters)
- make hdf destroy less recursive (so we don't blow the stack on
really large data sets)
Release Notes for Clearsilver 0.9.12 09/29/2004
-------------------------------------
bugfix release
- fix bugs in Java wrapper and add white space stripper option
- fix some allocation bugs in cs parser
- add single quotes to the list of characters to escape in url_escape
- fix core dump in find_context (hit during parse error in evar)
- fix some operator precedence bug and some bad expression handling of
empty strings
Release Notes for Clearsilver 0.9.11 08/03/2004
-------------------------------------
bugfix release
- fix bugs in Java wrapper
- update copyright notice
- fixes to get rid of gcc3 warnings
- fix bug in cgi/html.c
Release Notes for Clearsilver 0.9.10 06/02/2004
-------------------------------------
bugfix release
- Parenthesis in the wrong place in the last patch
- fix some memory leaks in template parsing and rendering
Release Notes for Clearsilver 0.9.9 05/14/2004
-------------------------------------
A small release.
INCOMPATIBLE CHANGE - The convert_text_html_alloc_options()
HTML_CONVERT_OPTS struct now has another member, link_name, which (if
not NULL) will be used instead of the link as the title of the link
which gets created. If you don't zero out your HTML_CONVERT_OPTS
struct (using either memset or calloc, for instance), you'll need to
initialize the link_name member (probably to NULL). Users of the
older convert_text_html_alloc() function or the python wrapped version
don't have to worry about this.
Also, added a ne_file_read_len() for loading binary files since it
returns the length of the loaded data.
Release Notes for Clearsilver 0.9.8 04/22/2004
-------------------------------------
A bit more than a bugfix release, we have some new features and
abilities.
New Features
- Internal functions in ClearSilver can now take multiple arguments.
This includes support for the comma ',' operator in CS, which works
like the comma operator in C.
- added string.slice function as an example of a multiple argument
function.
- added string.length function to replace old len function
- added subcount function to replace bad len function overloading
- added gettext support from Gerald Dachs <gda at gmx.de>
- INCOMPATIBLE CHANGE: added secure flag to cgi_cookie_set function
- Add new allocating sprintf routines which take the char * as an
argument and return the size of the allocated string
- Dave added a bunch of my public python code to the python example
code (we need to define a real package for this stuff at some point)
- support for sqlite in odb, and various odb improvements
Bug fixes
- gzip compression was broken, thanks to the guys at OmniWeb for
pointing out my mistake.
- handle redirects when using https correctly
- handle invalid form data without erroring out (or segfaulting)
Clearsilver is a fast, powerful, and language-neutral HTML template
system. In both static content sites and dynamic HTML applications,
it provides a separation between presentation code and application
logic which makes working with your project easier.
