I initially didn't feel like renumbering all the entries and just
deleted some questions of no relevance in 2026. Then the lists
were skipping numbers, and I thought it necessary to explain that by
adding another FAQ entry. Then I decided to just renumber the things
after all and delete that new entry.
I left a dropping of that confused mess in the file in my last commit,
and am removing it now.
The FAQ had many entries that referred to ancient bugs in other
libraries that have long since stopped being the subject of frequently
asked questions.
Some entries had links to dead URLs or documented old solutions to
problems that have more modern answers.
Many entries referred to old issues with ImageMagick and how to fix
them, but there are so many NEW issues with ImageMagick that have
nothing to do with these FAQ entries that it is better to remove the
entries entirely.
All references to Lesstif have been removed. Lesstif ended as a
project in 2012 when OpenMotif was open sourced.
All entries have been renumbered after deleting many of them.
Make all copyright notices include this year.
Most say "2000-2026" but some files created later had later start
dates, and some created last year only had the start date. They now
all have "-2026" regardless.
Closes#340
README.MAPS is a mess of basic map help, advanced map tweaking, map
minutia, and outdated guidance, with a healthy dose of dead URLs.
I have been updating the Xastir wiki at Github to have all of this
information in updated form. My intent is that README.MAPS will
ultimately be nothing but a link pointing at the Github wiki.
To this end, I'm ripping out all parts of the README.MAPS file that
have been migrated to the wiki and updated.
There's still a bit to go, and the wiki doesn't have *ANY* pointers to
updated URLs for finding maps yet. But README.MAPS only had links
that are long gone.
While I was at it, I removed references to README.MAPS in the
testdbfawk man page, the FAQ, the INSTALL.md file, and a few other
places.
While I was at it, I removed some really ancient FAQ data --- like,
when someone asks "why is Xastir slow rendering maps," it is pointless
to tell them that the developers know that a 60MHz Pentium performs
pretty well. If someone asks "Why is my new version of lesstif not
working?" the answer is "there is no new version of lesstif, this is
2025 and that project died in 2012." And if someone complains that
Xastir doesn't work right on Redhat 9, the only right answer is to say
"don't run Redhat 9, it's 22 years old."
Part of my work on #285
Xastir went through a lot of work to make the code compatible with the
Linux Standard Build (LSB), but LSB is a dead project since 2015.
Keeping scripts and ifdefs and configure tests all over the place to
support a way of supporting LSB makes no sense and bloats the code and
repo.
Closes#324
Bootstrap.sh is an ancient technique that used to be recommended in
documentation about "autoconfiscating" projects for portability. It
was meant to run all the bits of the autotools (aclocal, autoheader,
autoconf, automake...) to create the configure script.
Some time much later than Xastir's bootstrap.sh was created, autotools
began to include a program called "autoreconf" that does basically the
same thing. As it happens, this approach *does* work with Xastir,
although for a period of about two years it would not have.
Our bootstrap.sh got a few non-autoconf steps added to it in 2008,
which were then quietly superseded by changes to config/Makefile.am
that *should* have led to those steps being removed from our
bootstrap, but didn't.
Our bootstrap.sh *now* has ONLY autoconf stuff in it, and so is
now basically identical to running "autoreconf -i".
This commit adds a little information about using autoreconf instead
to several of the document files (README.developers.md, README.GIT,
INSTALL, and FAQ).
The server that used to serve the census Tiger maps went down many
years ago, but all the code that enabled displaying it and configuring
the map features remained --- all of this code was unusable.
There is now a WMS server available that does the same thing, and it's
in the code base as a .geo file. It does NOT use any of the
configuration that was enabled by the code base put in place for the
old server.
I am removing all vestiges of this old code to resolve issue #12.
A user had a question today about Xastir building with ImageMagick
that appeared to be because he had Magick-config in /usr/local/bin,
but didn't have /usr/local/bin in his path.
The FAQ entry on "Why isn't ImageMagick recognized..." had a lot of
RPM-specific reasons why this might not happen, but nothing about this
extremely simple error. Now it has the simple answer first.
I have replaced all the "Copyright (C) .* The Xastir Group" lines in
the entire source tree with identical text, "Copyright (C) 2000-2018
The Xastir Group." By making every file have exactly the same text,
this simplifies the operation we have to perform in 2019 when we
update the copyrights.
I also found a few files with DOS-style CRLF line breaks. I have replaced
them with plain LF line breaks.
Also fixed a few "whitespace errors" in db2APRS.c
In commit 1d6e920 I removed "README.win32" because it was an ancient
relic that had ceased to be maintained -- and therefore ceased to be
correct.
I forgot to remove all the places that the build system makes
reference to it, and so Curt's Jenkins CI testing failed.
This should fix it all. While I was fixing that, I cleaned out other
references to README.win32 that were scattered around.
CVS keywords such as "$Id$" no longer serve any purpose now that we're
using git, and need to be removed. This commit removes all of the
"$Id$" keywords, and a few other CVS keywords of no current value.
I also had to change some lines of the scripts/lang*.pl files that were
looking for $Id$ tags in their input and replacing them with comments.
The entire intent of this was to add two lines of comments to the top of
the output, and now it just does that, instead of depending on the
presence of the $Id$ keyword.
