Hello,
another update of the TODO list. Code freeze is on 2000-08-02 (Wednesday).
I changed the #include <> versus #include "" problem in most backends.
I hope nobody is offended by this. I think it's the easiest way to change
such global things. I have left out some backends because I know that their
maintainers are working on them at the moment or the fix was not too easy.
I will change these backends before code freeze, too, if there aren't any
objections and the maintainers haven't done this until Wednesday.
TODO for 1.0.3 (as of 2000-07-29)
* Use #include "sane/..." instead of #include <sane/...>.
It is not neccessary to search the system include directories for
sane header files. This may also cause problems if different SANE
versions are installed.
- artec
- avision
- hp
- plustek
- sp15c
- umax
* Bug in sane_control_option()
There was a possible segfault in this function. The pointer "info" was not
checked for NULL in all cases. This was found in the mustek, v4l and canon
backends and is now fixed. Please check your backends for this bug.
* check consistency in *.desc (e.g. "Parport" not "Parallel")
- hp (maintainer was contacted)
- umax (maintainer was contacted)
* Test SANE on different os/platforms, fix possible compilation problems
Platform static shared SCSI Net X gimp
Linux/x86 OK OK OK OK OK OK
Linux/Alpha
Solaris/Sparc OK OK ? ? ? ?
AIX/RS6000 OK NO(1) ? ? ? ?
Irix (2) OK OK ? OK OK OK
untested:
*BSD
Digital Unix
HP/UX
Linux/...
...
(1) I can't get to work the compilation of AIX shared libraries. The
first problem is that the filename of every library is
libsane.so.version not e.g. libsane-mustek.so.version so the
libraries overwrite each other. By temporarily removing Olivers
soname addition to ltmain.sh the right libraries are compiled
and installed. But starting scanimage -L results in a segmentation
fault. Someone with more experience in this field?
(2) I had to disable dc210 for Irix because the jpeg libraries on the
two Irix systems I tested produced all kinds of problems
(compilation and linking).
Test methods: scanimage -d pnm:0 --filename=file.pnm -T
scanimage -d net:localhost:pnm:0 --filename=file.pnm -T
(not possible if I'm not root on the system)
xscanimage / gimp with xscanimage plugin (where available)
All the above with shared and static libraries.
* use DBG(...) instead of fprintf(stderr, ...)
- hp
- microtek & microtek2
I won't fix these two backends. Fixing this is more complicated
than just replacing fprintf by DBG. At least this doesn't seem to
harm users of other backends. Not release critical but should
be fixed before 1.0.4.
I will try to fix everything I can by monday if the maintainers won't
have done this until then.
I consider the following points as not release-critical, but they
should be done nevertheless:
* use sanei_config_read() instead of fgets()
- m3096g (maintainer wants to fix this on Saturday)
- sp15c (maintainer wants to fix this on Saturday)
- umax (will be fixed with the next backend release)
* Documentation for sanei API
- any ideas for this?
To change before release:
* Change default to --disable-warnings
* increase version number
* remove freeze markers
* Update NEWS to the current backend versions
After release:
* Add news item on website
* update backend page on website
* update/add man pages on website
* update platforms list (sane-support.html) on website
* contact freshmeat etc.
Bye,
Henning
-- Source code, list archive, and docs: http://www.mostang.com/sane/ To unsubscribe: echo unsubscribe sane-devel | mail majordomo@mostang.com
This archive was generated by hypermail 2b29 : Sat Jul 29 2000 - 09:15:48 PDT