Added support for the Avision AV210D2+ scanner. This is the first scanner
to require the use of the gray filter bits, so code was added to enable them.
Regenerated description file to reflect this change.
Moved device init code from sane_init() to the newly created
sane_reload_devices() function, which is now called from sane_get_devices().
This means that device initialization now happens when sane_get_devices() is
called, not when sane_init() is called.
Replaced the clever, yet totally confusing, code using the "adf_mode"
variable. This variable was used with a complex interaction of bit shifts
and subtraction indirectly influnced by the value of s->source_mode. The
replacement code direct references s->source_mode instead, simplifying
things greatly. This change is functionally identical.
As a side note, the adf_mode code allowed "ADF Rear" scanners, in addition to
"ADF Duplex" scanners, to enter the section of code normally used for (HP)
duplex scanners. It's not clear if this was the intended effect, or if, as
the comments suggest, this was an unnoticed side effect caused by the
confusing code. The point is moot, however, because prior in the code, the
ADF Duplex flag is set anytime the ADF Rear flag is set. As long as all
ADF Rear scanners are duplex scanners this shouldn't be a problem.
Changed the AVISION_FILTER_* defines to their final values rather than
bit-shifting them prior to use. This make reading the code much easier
since no mental calculation is necessary to determine what the final
bit-pattern/value will be.
Wrap scanner entries with duplicate USB ids with ifdefs so they will not be
compiled. This shortens the search for new scanners because we don't try
the same USB id more than once.
the same and already work, but this change will reflect that fact on the
website.
- Generated a new version of doc/descriptions/avision.desc from avision.c.
the AV610, and added a comment for the @V5100 in avision.c.
- Generated a new version of doc/descriptions/avision.desc from avision.c.
This restores information lost with my last update. I've intentionally
seperated these changes so that they can be more easily understood.
Source code was inconsistent on how they were referencing
include files in include/sane. In certain cases, sane/sane.h
and sane/saneopts.h was really including external copies (with
some preprocessor implementations anyways).
Global replace of u_int??_t with C9x standard based uintxx_t
in order to remove some tricky and overlapping portability logic
from acinclude.m4 and leave it in only m4/stdint.m4.
issues, by Mattias Ellert.
* backend/fujitsu.c: backend v89, fi-4750 has no serial number support
* doc/descriptions/fujitsu.desc: backend v89, remove 'MAC/TWAIN' text
* doc/sane-fujitsu.man: backend v89
* backend/.cvsignore: ignore *.loT
with partial matches on USB ID, but vendor / product string
matches as mostly only the HP5300 vs. HP5370, fixed non-color
calibration for devices not filling all RGB fields of calibration
format information, do not send 3x3 color matrix to older devices
(ASIC versions) as some HP53xx does not correctly handle it,
improved calibration accuracy and fixed transparency adapter
detection to not use non-zero as present, but just 1 (some
devices now set 0xff as (-1) - not present ...
Implemented support for latest Avision ASICs and features,
including overscan, background raster, software scaling for
cheaper ASICs used in HP scanners. Changed types used to
more portable uint* ones, without underscore (_), removed
in-file ChangeLog history, as the files became big enough.
Implemented early calibration for easier handling of
Avision-based film scanners and implemented support for
2-pass duplex scanners.
options from the parser, example avision.conf and manual page.
Fixed 12 Bit gray and color modes to actually work as well as
software CCD line-difference correction. Minimal scan resolutions
have been slightly adapted for some ASICs.
from BUILD 167 to 179, including support for HP 5370, AV600U,
AV210C2, AV220C2, et. al. The USB i/o paths got hardened, and a
lot of fixes as well es enhancements and optimizations where added
and a lot of typos, mostly in debug output and comments, got fixed.
The "Line Art" mode was renamed to "Lineart" to match the other
backends.
fixed segmentation faults for some modes, many many new devices, two
different duplex modes are supported, vast speedup of multi page batch scans,
buttons as well a 7-segment LED readout and some quirks for old scanners not
filling all fields
wait_4_light bugfixes, fixed color packing, fixed 16bit modes
(especially for big-endian systems), avoid 16bit modes for default
modes, fixed gamma-table for some models as well as spelling on the
way
* backend/avision.c doc/descriptions/avision.desc doc/sane-avision.man: substituted my mail address and web-site to a new one - due to
leaving the rocklinux project
Mike Kelly