collection as math/gp-fplsa.
This package is a GAP interface to a C program called fplsa for
dealing with finitely presented Lie super-algebras.
This GAP package installs an alternative method for the GAP
operation `IsomorphismSCTableAlgebra'. The method calls fplsa to do the
hard part of the computation. This speeds up the calculation and permits
larger problems to be attempted. The external program has much additional
functionality which is not used by the present version of the package.
collection as math/gp-fplsa.
This package is a GAP interface to a C program called fplsa for
dealing with finitely presented Lie super-algebras.
This GAP package installs an alternative method for the GAP
operation `IsomorphismSCTableAlgebra'. The method calls fplsa to do the
hard part of the computation. This speeds up the calculation and permits
larger problems to be attempted. The external program has much additional
functionality which is not used by the present version of the package.
* libexif/exif-data.h: Introduce an array of ExifContents. This
doesn't break binary compatibility, but it breaks compilation.
Do something like "%s/->ifd_0/->ifd[EXIF_IFD_0]" in your source
code to make it compile again.
Always check for ${OPSYS} == "IRIX" (and not once for != and once for ==)
Always use "-Wl,-rpath,/path" rather than sometimes "-Wl,-rpath -Wl,/path"
and sometimes "-Wl,-rpath,/path".
Lots of changes and new features: you want to update!
Changes since latest packaged version:
Version 1.6.1
* fixed a bug in the configure phase that used to make the compilation
fail with compilers not providing the function getop_long
(reported by Horacio Montenegro <horaciom@unicamp.br>).
Version 1.6
* scanners for flex and ChangeLog added by John Millaway <millaway@acm.org>
* default to stdout when no --output is given along with the -i option
and when --output="STDOUT" (as suggested by Hugh S. Myers,
<hsmyers@sdragons.com>)
* fixed a problem with string comments in Python scanner
* a separate executable is built, source-highlight-cgi, that can be
used as a CGI program
Version 1.5.1
* xhtml doctype generated correctly (thanks to Christian W. Zuckschwerdt
<zany@triq.net> and Josh Ghiloni <josh@joshghiloni.net>) reported
by Martin Gebert <Murphy.Gebert@gmx.de>
* Anonymous CVS Access is now available
http://savannah.gnu.org/projects/src-highlite
Version 1.5
* xhtml output format is now handled (thanks to Christian W. Zuckschwerdt
<zany@triq.net> and Josh Ghiloni <josh@joshghiloni.net>)
* assert is now recognized as a Java keyword (thanks to Fabio Calefato
<calefato@libero.it>)
* a scanner for Python has been written by Martin Gebert
<Murphy.Gebert@gmx.de>.
* when in Php3 strings can span more than one lines and line numbers are
correctly formatted.
Version 1.4
* fixed compilation error with some compilers due to headers istream.h
and ostream.h (thanks to Dan barthel <dbarthel@mac.com>)
* a scanner for Php3 has been written by Alain Barbet
<alian@alianwebserver.com>
* reading from stdin does not kill line numbers (notified by Keith
Robertson <keithr@nortelnetworks.com>)
* line number count is correctly reset when handling multiple files (bug
fixed)
* for Java sources, if, while, and other keywords that use '(' ')' are
now correctly highlighted as keywords instead of as functions (reported
by Manuel Kauers <manuel@kauers.de> and Fabio Calefato
<calefato@libero.it>)
* tabs are correctly handled for Prolog and Perl sources (reported by
Martin Gebert <Murphy.Gebert@gmx.de>)
Version 1.3
* now there is only one program, namely source-highlight, that accepts a
command line option, --src-lang or -s, specifying the source language
(for the moment java, cpp and prolog). Previous programs, java2html
and cpp2html are provided as shell scripts, just for compatibility, but
their use is not advised.
* --out-format, -f, command line option added for specifying the output
format (for the moment only html is accepted). Though only html is
handled, this option is required.
* a scanner for Prolog has been written by Martin Gebert
<Murphy.Gebert@gmx.de>. This scanner is still to be considered beta,
and actually it was written for Visual Prolog, so it should be tested
against other Prolog dialects.
* a scanner for Perl has been written by Alain Barbet
<alian@alianwebserver.com>
* Correctly check whether it is possible to write to a file (notified by
Martin)
* --tab option is correctly handled even for C/C++ files (bug fixed,
notified by Martin)
* Added option --line-number (-n) in order to print line numbers (as
suggested by Ed Scott and Andre Charbonneau).
* Man page added.
* Numerous bug fixes for most of the PAM modules, including several
string length checks and fixes (update recommended!).
* fix for legacy behavior of pam_setcred and pam_close_session in
the case that pam_authenticate and pam_open_session hadn't been
called
* pam_unix:
- don't zero out password strings during password changing function
* pam_wheel:
- feature: can use the module to provide wheel access to non-root
accounts.
* pam_limits:
- added '%' domain for maxlogins limiting, now '*' and @group
have the old meaning (every) and '%' the new one (all)
- handle negative priority limits (which can apply to the
superuser too).
* pam_userdb:
- require that all of typed password matches that in database
* pam_access:
- added the 'fieldsep=' argument, made a PAM_RHOST of ""
equivalent to NULL
Incidentally, cups-1.1.18 will once again do PAM authentication using
pam_unix.so if built against PAM-0.77.
3.2 released on Sat 13 Jan 2001
Added '-i' switch to display whether a file contains seek information
Cleaned up endian-conversion code
Fixed a file naming bug - now, when only an input file name is given,
output files are named according to the following rules:
If shortening a file:
If the input file name ends in .wav, then change
extension to .shn; otherwise, append .shn to the
input file name. Examples:
file.wav -> file.shn
file.ext -> file.ext.shn
If extracting a file:
If the input file name ends in .shn, then change
extension to .wav; otherwise, append .wav to the
input file name.
file.shn -> file.wav
file.ext -> file.ext.wav
When the caller specifies -v2 on the command line, seek tables are
not generated (neither appended nor created in separate files).
The default behavior is to generate/append seek tables.
3.3 released on Sun 12 Aug 2001
Converted to autoconf/automake build system
Made seek table code 64-bit clean
Changed seek table data structures to compensate for compilers that don't
properly implement #pragma pack(1)
Various minor cleanups/updates that don't impact the shorten algorithm
3.4 released on Sun 10 Feb 2002
Fix for reading/writing binary data on stdin/stdout for operating systems
that do not have a single '\x0A' as a line separator (Frank Klemm)
Shorten now refuses to input data from or output data to a tty
Fixed crash when trying to create seek tables from a non-shorten file
-s and -S options can now create seek table files from data read on stdin
(with -s, output filename is 'stdin.skt')
Seek tables are now appended to non-stdout output file when reading
uncompressed data from stdin (e.g. shorten - outfile.shn < data.wav)
3.5.0 released on Mon 18 Nov 2002
Support for compression of AIFF files contributed by Brian Willoughby of
Sound Consulting
Changed file naming convention during extraction - when only an input file
name is given, output files are named according to the following rules:
If the input file name contains more than one '.' and
ends in .shn, then drop the .shn; otherwise, if the input
file name ends in .shn, then change the extension to .wav;
otherwise, append .wav to the input file name.
file.aiff.shn -> file.aiff
file.aiff.ext -> file.aiff.ext.wav
file.shn -> file.wav
file.ext -> file.ext.wav
This change was made so that non-wav files would be properly named after
shortening and unshortening. For example, 'shorten file.aiff' creates
file.aiff.shn, and with the change above, 'shorten -x file.aiff.shn'
will now create file.aiff (instead of file.aiff.wav).
