also updated the tutorial. I've simplified the tut2.c
by using comedi2phys instead of comedi2physical. The use of
polynomials as a 2nd comedi program is a bit too difficult
for a novice and might deter the user to dig deeper into
the coding. Also, I've added tut3.c which is a stripped
down version of cmd.c and of that again the relevant lines
for the handbook to save space.
just left udev as the only hotplug mechanism. I've also added
an option to disable the installation of the firmware. This
is useful for debian or ubuntu packages where the firmware
is now provided by a package accompanying the staging kernel.
All usbdux files are included in the kernel firmware packages.
Package maintainers should use the --disable-firmware option.
The PDF output of the table of configurations was awful. Fiddled with
the column widths a bit and simplified the layout "Meanings of data[1],..."
column to allow it to wrap.
Add --with-pdf-backend=[yes|no|dblatex|fop|default] option which
defaults to "yes". This controls the option passed to 'xmlto' to
generate PDFs.
If set to "yes" (the default) it looks for 'dblatex' and uses it as the
backend for 'xmlto' if found, otherwise 'xmlto' will use the default
backend.
If set to "no" (same as --without-pdf-backend), PDF generation is
disabled.
If set to "dblatex" it looks for 'dblatex' and uses it as the backend
for 'xmlto' if found, otherwise PDF generation is disabled.
If set to "fop" it looks for 'fop' and uses it as the backend
for 'xmlto' if found, otherwise PDF generation is disabled.
If set to "default", 'xmlto' will use the default backend.
Rebuilt "doc/drivers.txt" using "make Documentation/comedi/drivers.txt"
in the Comedi package. Note: this scans the driver sources included
with the comedi.org Comedi package, not the Linux "staging" Comedi
sources, so some newer, staging-only drivers are missing.
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
dblatex is passed the option `-o $(srcdir)/pdf`, but if this directory
does not exist, the PDF file is created with the name 'pdf' in
$(srcdir). Make the directory so that the -o option treats the
specified name as an output directory instead of an output filename.
rules were installed in the etc directory and not the lib.
See also:
http://lintian.debian.org/tags/udev-rule-in-etc.html
./configure --with-udev-hotplug=/lib places now all
files in /lib. The scripts were wrongly prefixed with the
sysconf directory which is /etc but should be /lib.
The directory is now also taken from --with-udev-hotplug=/lib.
In the early udev days the firmware had to be uploaded
by comedi_config. However now the driver requests the
firmware by itself and there is no need to have dedicated
scripts for every kernel module in udev.
dblatex -t pdf ./comedilib.xml
There were two major issues in the comedilb.xml:
1) there were too many nested sections so that
latex ran out of sub-sub-sub-sub-sections.
2) A section needs a title and can no longer
exist without one. The first line
of the textfiles contains now the title
and will be inserted by mkref into the xml file.
todo: switch from xmlto to dblatex for the pdf generation
and re-enable it in the makefile.
Generate the ChangeLog using git log in the dist-hook target of the main
Makefile. The git command used is:
git log --stat --name-only --date=iso
The format of the resulting ChangeLog is different to the old one that
was produced with cvs2cl.
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Use AS_HELP_STRING() to make the HELP-STRING arguments for the
AC_ARG_ENABLE() and AC_ARG_WITH() macro calls.
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
The SWIG Ruby bindings do not build cleanly from scratch. They need an
existing installation of comedilib. So disable them in the default
configuration. They can be enabled with the --enable-ruby-bindings
configure option.
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
I also added 2010 for:
commit 3fe8e6baac
Author: Ian Abbott <abbotti@mev.co.uk>
Date: Mon Dec 20 14:50:21 2010 +0000
Add sampl_t and lsampl_t array_class()es to comedi.i.
...
Patch from W. Trevor King.
The old method only worked out of the box if you were running Python
2.1 on a 586. I imagine that is too small a slice of current users to
be worth hard coding ;).
This avoids:
Traceback (most recent call last):
...
File ".../mmap.py", line 96, in prepare_cmd
C.scan_begin_arg = 1e9/freq
...
TypeError: in method 'comedi_cmd_struct_scan_begin_arg_set',
argument 2 of type 'unsigned int'
With the previous implementation, `mmap` was importing the demo
script, not the standard `mmap` library module. From the docs [1]:
The directory containing the script being run is placed at the
beginning of the search path, ahead of the standard library
path. This means that scripts in that directory will be loaded
instead of modules of the same name in the library directory.
I'm not sure when this changed in Python, since the demo script
presumably worked with an earlier version.
[1]: http://docs.python.org/tutorial/modules.html#the-module-search-path
This catches it up for the folowing commits to the comedi repository:
commit 311e9c540217d60bdf56a1dba9eafe474fd3b218
Author: Bernd Porr <Bernd.Porr@f2s.com>
Date: Sun Jan 13 23:07:02 2008 +0000
Added a proper PWM subdevice for comedi.
commit 8092ed57ff38e1fcaf75a7e0bf3b8c4f5372f3d7
Author: Frank Mori Hess <fmhess@speakeasy.net>
Date: Sat Jan 31 22:13:09 2004 +0000
serial dio (SCXI) support from Caleb Tennis
This catches the Python demos up to:
commit 003b9ec501
Author: Ian Abbott <abbotti@mev.co.uk>
Date: Mon Nov 8 11:01:42 2010 +0000
Add chanlist_len parameter to comedi_get_cmd_generic_timed.
This makes it possible to populate `comedi_insnslist_struct` for use
with `comedi_do_insnlist`.
I also removed my earlier re-typedefs of `sampl_array` and
`lsampl_array`. I didn't understand the reason for the underlying
`TypeError` when I added them. The proper way to avoid that error is
to use the `.cast()` and `.frompointer()` methods. See
`demo/python/insn.py` (coming soon) for details.
The comedi_calibrate.8 man page has been moved to the comedi_calibrate
package where it belongs. Thanks to W. Trevor King for pointing it out.
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Here's a small patch to make the configure script compatible with Python
3. The wrapper itself (swig/python/comedi_python_wrap.c) will have to be
generated with SWIG >= 2.0.0 for Python 3 compatibility.
The 'go' variable is only used by the 'while' loop that reads the data.
We don't need to set it in the 'for' loop that checks the data
afterwards.
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
If the command does not stop immediately after the last sample, there
can be more data to read than we have room for. Make sure the 'while'
loop 'read()' does not overrun the end of the malloc'ed 'buf'.
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
When doing an out-of-tree build, the comedi_version.h file is in
the build tree, not in the source tree. The CFLAGS needs an additional
-I option so the compiler can find it.
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
