Hi,
Just a quick update on where I am with XSANE development.
A number of issues have been addressed, including:
- A large number of compiler warnings have been addressed and issues
raised by static analysis tools, enabling the fixing of some real bugs.
- Upstreaming of patches from a number of sources, including FreeBSD and
Debian
We have had a few interesting and useful enhancements from a number of
people:
- Fixes/enhancements for the preview window, courtesy of Maxim Ushakov
- Fixes for the largely broken histogram, from Jonny Robbie
- Fix to dpi preferences suggested by Lukas Beck
So far, a lot of really positive things coming in from users, and we
have a number of functionality-based issues to work through, many
suggestions from Ulf Zibis and others.
I have decided to put them aside until more of the primary work has been
accomplished.
AutoTools
=============
Currently, I am trying to fixup the autotools setup. This turned out to
be quite a bit bigger job than I anticipated.
The repo currently uses autoconf but it does not use libtool or automake
(for sources).
There are some suggestive signs that automake has been used at some
point, but there are Makefile.in files but no Makefile.am sources.
The Makefile.in files look to have been autogenerated at some point.
It might be that Makefile.am files did exist but were missing from
checkin. It is difficult to say.
I am in the process of learning a lot about autotools and rebuilding
this infrastructure so that we can remove a lot of that boilerplate from
the repo.
One of the problems with having static Makefiles is that they are pretty
much stuck in time. Better to have these things generated and benefit
from platform and infrastructure improvements since the 10+ years ago
they were last substantially looked at.
I know Olaf offered to give some assistance in this area, and it might
be that I have to call in that offer at some point :D
For the moment though, I am learning a lot and making mistakes as is
typical of this process.
I thank you for your patience and hope that I can reward it with a more
resilient and up-to-date set of sources to move forward with.
Regards,
Ralph Little
