diff options
Diffstat (limited to 'doc/umax/umax.CHANGES')
| -rw-r--r-- | doc/umax/umax.CHANGES | 1172 |
1 files changed, 1172 insertions, 0 deletions
diff --git a/doc/umax/umax.CHANGES b/doc/umax/umax.CHANGES new file mode 100644 index 0000000..77d8d62 --- /dev/null +++ b/doc/umax/umax.CHANGES @@ -0,0 +1,1172 @@ +CHANGES FOR UMAX BACKEND FOR SANE +--------------------------------- + +umax -> sane-0.6: +----------------- + + * umax-sane-backend created: + - umax.c splitted in: + umax-cli.c, umax-scsi.c, umax-driver.c, umax-text.c + - umax.h splitted in: + umax-scanner.h, umax-scsi.h, umax-scsidef.h, umax-struct.h + - outputs done by DBG(level,msg,args...) defined in debug.h + - scsi-calls changed to sane-format (of sane-0.57) + - created umax.c and umax.h for sane-backend + + * changed search-path for scsi.h and sg.h, now needs /usr/include/scsi + + * removed mistake in bound-function + + * added "Vista S-6E" as supported + + * added PowerLook-II (PL-II) as supported + + * added Astra 600S as supported + + +sane-0.6 / 0.61 / 0.62 -> 0.63 +------------------------------ + + * added exception to gpl-licence: + > This program is free software; you can redistribute it and/or + > modify it under the terms of the GNU General Public License as + > published by the Free Software Foundation; either version 2 of the + > License, or (at your option) any later version. + > + > This program is distributed in the hope that it will be useful, but + > WITHOUT ANY WARRANTY; without even the implied warranty of + > MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + > General Public License for more details. + > + > You should have received a copy of the GNU General Public License + > along with this program; if not, write to the Free Software + > Foundation, Inc., 59 Temple Place - Suite 330, Boston, + > MA 02111-1307, USA. + > + > As a special exception, the authors of SANE give permission for + > additional uses of the libraries contained in this release of SANE. + > + > The exception is that, if you link a SANE library with other files + > to produce an executable, this does not by itself cause the + > resulting executable to be covered by the GNU General Public + > License. Your use of that executable is in no way restricted on + > account of linking the SANE library code into it. + > + > This exception does not, however, invalidate any other reasons why + > the executable file might be covered by the GNU General Public + > License. + > + > If you submit changes to SANE to the maintainers to be included in + > a subsequent release, you agree by submitting the changes that + > those changes may be distributed with this exception intact. + > + > If you write modifications of your own for SANE, it is your choice + > whether to permit this exception to apply to your modifications. + > If you do not wish that, delete this exception notice. + + * split resolution in x-res and y-res, flag bind-resolution + + * only sane_... functions will be exported from umax-backend, + all umax_... functions are made static + + * removed umax-emergency-routine for SANE + + * made program-flow (esp. with do_canel) SANE-specific + + * fixed missing 3-pass parts + + * made scanmodes, sources, analog gamma, gamma download, negative, + quality calibration, highlight, shadow only selectable if they are + available + + * calibration-mode selectable in advanced options if available + + * warmup selectable if available + + * changed umax-scsi.c like described from Gordon Matzigkeit + in his diff/patch-file from 21.06.97 so that the umax-backend + should compile on non-linux-systems now - thanks! + + * split highlight and shadow in R, G and B for color-mode. + + * reset adf and uta flag when not used (in sane_start) + + * added possibilty to change calibration-mode if available + + * added send_halftone_pattern to umax-scsi.c + + * changed umax-scsi.c like described from Gordon Matzigkeit + in his diff/patch-file from 21.06.97 so that the umax-backend + should compile on non-linux-systems now - thanks + + * added support for each color (R,G,B) for highlight and shadow + + * added "Astra 1200S" and "UC840" as supported + + * added bad bugfix for preview-problem with S6E + (in umax-scsi.c function start_scan) + + * moved resolution-check from check_values to bound_scanarea, + now no floatingpoint-exception will occur if no resolution is given! + + * changed range-check for width and height + + * removed bug in set_window for three-pass-rgb-mode + -selected wrong buffer for some green and blue options + -scanmode now is greyscale instead of rgb + + * changed start_scan(us) to better use with wid-list + and added missing 3 wid-bytes in scan.cmd + + * removed parameter color in three_pass_function_headers, + the color ist taken from us->three_pass_color now. + + +sane-0.63/0.64/0.65 -> 0.66 +--------------------------- + + * changed gamma-data-upload for 3-pass-color-scan + + * corrected wrong calculation for highlight and shadow + selection for rgb-values + + * changed gamma data to use 8 bit (256 bytes) or 10 bit (1024 bytes) + + * included 10/30 bits per pixel mode + + * changed bug in sane_exit (free()) that sometimes made a + segmentation-fault + + * made patches runtime-selectable + + * changed set_window_command to support 30 bits per pixel + + * changed send_gamma_data to support three_pass_gamma and 10/30 bpp + + * changed row_len in umax_bound_scanarea to 10/30 bit mode + + * added halftone-support for ug630, uc630 and uc840, + but I don't know if all values are ok! + + * changed CBHS-50-range to 78-178 (from 103-153) + + * changed rangecheck for highlight and shadow + + * fixed get_cbhs_range-bug in umax-scsidef.h + + * added ADF-inquiry-bits to umax-scsidef.h + + * added color_line_arrangement_mode-inquiry-byte to umax-scsidef.h + + * added CCD_line_distance-inquiry-byte to umax-scsidef.h + + * added all most unused bits and bytes in inquiry to umax-scsidef.h + + * made patches runtime-selectable + + +sane-0.66 -> 0.67 +----------------- + + * removed bug in sane_start: ADF and UTA flags have not been set correct + + * removed bug in init_options: value range for 10 gamma-input-bits corrected + + * made custom gamma available for lineart and halftone for nonlinear effects + + * added Vista-T630 as supported scanner + + * removed test_warmup in wait_scanner because adaptec 1542 has problems + with request_sense-command in test_warmup + it was only used for printing a message, so it was not important. + + * if a scanner has a full inquiry-block but is not known to be supported, + the driver will print a warning and continue instead of aborting, + option --continue (on umax-cli) will suppress the warning! + + * changed ugly definition + char gamma[x] -> char *gamma; gamma=alloca(x); + + * removed bug (wrong length for 10 bit-gamma) in send_gamma_data + + * changed all buffer-definitions to unsigned char and insterted + type conversions, so there are no more warnings because of + typemismatches. + + * corrected bad definition of set_DCF2_gamma_output_bits + + +sane-0.67 -> 0.68 +----------------- + + * changed open config to fp = sanei_config_open(UMAX_CONFIG_FILE) + + * changed all set OPT...BIND commands, endless-loop happend in sane-0.67 + =>ane-0.67-umax-fix.diff.gz + after that patch changed same for ten_bit_mode + + * changed some strdup() to (SANE_Char *)strdup() + + * added UMAX Astra 610S as supported scanner. it has the same problems + in color-mode like the Astra 600S. + + +sane-0.68 -> 0.69 +----------------- + + * changed maximum y-resolution from optical_res to y_res + + * changed sane_cancel (waitpid) like suggested by David Mosberger + + * added #include <lalloca.h> for AIX-support + + * replaced call to umax_wait_scanner with umax_do_calibration + after umax_start_scan in umax.c + + * moved sense_handler to umax-scsi.c + + * removed call to umax_get_data_buffer_status in reader_process + + * added umax_send_shading_data, umax_send_gain_data + and umax_read_shading_data, umax_read_gain_data + + * changed name: umax_read_data_block -> umax_read_image_data + + * changed length offest in do_inquiry from 5 to 4 + and changed set_inquiry_length-offset from 5 to 4 + + * added calib_lines to structure US and added it to + set_window - command in umax-scsi.c + + * changed in umax-scsi.c routine identify_scanner: + vendor and product is tested for each model instead of + vendor=="UMAX " and only model tested! + Because of that the format of scanner_str has changed + from product,product,... + to vendor,product, vendor,product,... + + * added support for Linotype Hell Jade (LinoHell Office) + + * added empty routine umax_do_calibration in umax-scsi.c, it shall be called + after umax_start_scan! + + * replaced all umax_*scsi with sanei_*scsi, so the umax-driver + now directly calls the sanei_scsi-commands. + + * moved sense-handler from umax.c to umax-scsi.c + and added detection for driver-calibration-request + sense_handler got a new parameter (void *arg). + + * changed wait_scanner for SUN: usleep(0.1sec) -> sleep(1s) + this is only a bad bugfix that will be replaced one day! + + +sane-0.69 -> 0.70 +----------------- + + * moved reader_process to umax-scsi.c + + * changed include <sane/config.h> for _AIX and non-_AIX + + * added missing "," in definition line for UMAX Astra 1200S in umax-scanner.h + + * added in umax-scsi.c in function identify_scanner a test that only + warnings and errors are printed if vendor="UMAX " + + * added routine umax_order_line_to_pixel and umax_order_line + + * moved reader_process from umax.c to umax-scsi.c + and added routine for color ordering. + + * reset length offest in do_inquiry from 4 to 5 + and set_inquiry_length-offset from 4 to 5 + because umax-S6E didn't work with it! + + * corrected error in identify-scanner in vendor-test + + * changed end-detection of list of known scanners from + #define known_scanners 15 to last item "END_OF_LIST" + + * added routine that corrects wrong inquiry-data from + Astra 600S and Astra 610S + + * added test and selection for do_color_ordering in check_values + + * changed __sun-code in umax_wait_scanner because the scsi-driver + answers with IOERROR instead of DEVICE_BUSY + (this is only a bugfix for a little time) + + +sane-0.70/0.71 -> 0.72 +---------------------- + + * inactivated selection of quality calibration (until it is supported + by the driver) + + * changed maximum x_resolution from optical_res to max_x_res + + * reordered DBG-levels and defined them by symbols + + * changed all sizes that used type int to type size_t + + * changed wait_scanner for testing !sane_status_good, + because testing sane_status_device_busy does not work with + all scsi-drivers! + + * changed ccd-order in umax_order_line, the order is now + specified by 9 values (before by 3 values), changed + order tables and added a table for Astra 600S and + another table for Astra 610S + + * changed ccd-order, now based on optical_resolution + and added new routine umax_forget_line + + * added inquiry-block for UC1200S: NOT TESTED! + + * added inquiry-block for UC1260: NOT TESTED! + + * included print_inquiry into umax-scsi.c + + * added correction-routine for x_res > 600 dpi, phys_width is + corrected in umax-scsi.c: bound_scanarea + + * changed umax-scsi: sense_handler: + changed test for valid sense-block + added test for button pressed + + * added output of sense-code and error-code in sense_handler + + * added options to select calibratin and scan exposure times + + * added options to select calibratin and scan lamp density + + * moved all option-names/titles/descr. to saneopts.h + + * combined bind for highlight, shadow, analog-gamma and new + exposure time functions to rgb-bind + + * added set_WD_line_arrangement(buffer_r, WD_line_arrengement_by_driver); + if driver is doing color-ordering from line to pixel + + * included umax-scsi.c into umax.c + + * included umax-struct.h into umax.h + + * included US-structure into Umax_Device-structure + + * added free(scanner->device->buffer) to sane_close + + * added free(scanner->device->pixelbuffer) to sane_cancel + + * added free(scanner->device->pixelbuffer) to do_cancel + + * added SANE_UNIT_MICROSECOND to exposure-times + + * added output of status-string in wait_scanner + + * moved some options for Scan-mode-block to Enhancemant-block + +sane-0.72 -> 0.73 +----------------- + + * disabled RGB_BIND if no such value is available + + * included readproc + + * included bound_scanarea into check_values, problem + with wrong scanarea for UTA should be solved now! + + * all scanmode- and scansource-strings are #DEFINEd now + +sane-0.73 -> 0.74 +----------------- + + * added support for 12/36 bits per pixel (for Astra 1220S) + + * added support for Astra 1220S + + * included inquiry data for UG80 (Escom Image Scanner 256) + + * moved WAIT_SCANNER from umax_read_data to umax_read_image_data, + umax_read_shading_data now works! + + * added umax_start_scan at the end of calibration by driver, + the scanner does not hang any more if quality calibration is enabled! + There must'nt be a "wait_scanner" in calibration by driver routine!!! + + * added usage of driver internal values for calibration by driver if + the scanner does not give back the informations about the shading data: + width = dev->width * dev->inquiry_optical_res ( * 3 in RGB-mode ) + + * added 2 calibration-routines: + - use last line for shading data + - calculate average value for shading data + - prepared routine for complex calculation + + * added gamma download curve format type 0 and 1 + + * defined gamma DCF = 1 for UC1200S and UC1260 + + * defined gamma DCF = 0 for UC1200SE and ARCUS PLUS + + * changed sense_handler: if bad sense_key => DEVICE_BUSY + problem with black area at top of scan for some SCSI-adapters is solved now + + * changed wait_scanner: waits 0.5 sec now + + * disabled quality calibration in preview mode + + * made upper_left_x and upper_left_y even values + it looks like even resolutions > 600 dpi work now! + (not ok!!!) + + * added sigterm_handler for umax_reader_process + changed sane_canel, do_canel and do_eof(=close_pipe), + cancel now works! + + * changed trim_rowbufsize that made problems with SGI, + there has been an overflow because of big buffersize + + +sane-0.74 -> 1.00 (build 0) +--------------------------- + + * added output of umax-backend-version in sane_init + + * added support for Linotype Hell Jade2 (Vobis ScanBoostar Premium) + SCSI-ID "LinoHell","Office2" + + * changed base of calculation of scanwidth and scanlegth from + pixel to resolutionbase (1/300, 1/600 or 1/1200 inch) + problem with wrong pixelcount at higher resolutions should be + solved now + + * changed selection for calibration mode + + * changed calculation for width in pixels for lineart and halftone mode + + * added inquiry for UMAX UC1200SE + + * corrected 12/36-bit-mode in send_gamma_data and array in umax.h + + * added support for Gemini D-16 + + +sane-1.00-build-0 -> build-1 +----------------------------- + + * added support for Nikon AX-210 + + * changed all grey to gray (David Mosberger asked for it) + + * changed bug in gray gamma table conversion + + * added Linotype Hell Saphir 2 as supported scanner + + * changed umax-uc1260.h + + * added "- 1" in umax_order_line: + else if (dev->pixelline_opt_res < length * 3 - CCD_distance * 3 - 1) + else if (dev->pixelline_opt_res < length * 3 - CCD_distance - 1) + + * changed all allloca() to malloc() + free() + + * removed options speed and smear, I got the info from umax that + these values should never be changed! + + * removed ten_bit_mode and twelve_bit_mode options, added option bit_depth + and added support for 9, 14 and 16 bpp + + * replaced u10_range and u12_range by dynamically set scanner.output_range + + * removed static allocation for gamma_table[0..3], this is done dynamically + in sane_open and freed in sane_close now. + + * added an empty string to the end of scan_mode_list and source_list + +build-1 -> build-2 +------------------ + + * added "*info |= SANE_INFO_RELOAD_PARAMS;" in control_option set bit_depth + + * added inverted calibration modes for Astra 6X0S scanners. + + +build-2 -> build-3 (sane-1.0.1) +------------------------------- + + * inversion of calibration is automatically set for Astra 6X0S and Vista S6E + + * check for machine endianess is done only at backend start now! + + * made umax_do request_sense static + + +build-3 -> build-4 +------------------ + + * Added option name and parameter in debug output in option set value + + * Added option name in debug output in option get value + + * Changed calculation of pixels/line and lines, may be the problem + with shaered images at resolutions>600 dpi is solved now + later Remark: This error does still occur, but not so often! + + +build-4 -> build-5 +------------------ + + * Changed calculation in umax_order_line, color images should + have the correct length now for Astra 6X0S and Vista S6E + + * Added DOR resolutions and area + + +build-5 -> build-6 +------------------ + + * Bits for ADF and UTA in driver internal inquiry block (old UMAX scanners) + are set dynamically now. + + * Added Genius ColorPage-HR5 (vendor KYE) (=Astra 1200S) as supported scanner + + * Added Astra 2400S as experimentally supported scanner + + * Added pauses for Astra 2400S + + * Removed reposition_scanner at start of scan, Astra 2400S does not work with it! + + * changed debug output in correct_inquiry + + * for UMAX S12 the driver does not wait any more until scanner is repositioned + this is experimental, it may work with other scanners too! + + +build-6 -> build-7 +------------------ + + * changed some int -> unsigned int (resolution and scanwidth etc) + + * calculate_pixel is called with right values in UTA mode now. + + * if DOR resolution is not defined and DOR is available + the DOR resolution is set to double standard resolution + + * removed wait_scanner after set_window-command + + * added Mirage D16-L as supported scanner + + * Astra 2400S works stable now + + +build-7 -> build-8 +------------------ + + * for Mirage D-16L the driver does not wait any more until scanner is repositioned + this is experimental, it may work with other scanners too! + + * Mirage D-16L does not give back informations for calibration by driver, + added new modus that "guesses" the values for Mirage D-16L. + + * Added read_shading_data to 16 bit calibration routine and made a routine for + averaging with low byte first and one for high byte first + + * Byte order (HBT) is set always (also in lineart and halftone) + because it is used in driver calibration + + * Added relevant_*_res the is set to normal or double resolution values + + * Added test if scsi buffer is large enough for one shading line + + * selection for Calibration mode, Shading type and Speed/Smear + are only compiled in if define is set in umax.h + + +build-8 -> build-9 +------------------ + + * slow speed options is preset for Vista T-630 + + * speed/smear can be set now (forgotten #ifdef) + + * selection of exposure times now is only available if it is enabeld + + * gamma table is disabled in lineart and halftone modes + + * if image is transmitted completly sane_read (close_pipe) + returns with SANE_STATUS_GOOD in adf mode and in any other case + with SANE_STATUS_EOF + + +build-9 -> build-10 +------------------- + + * removed change about sane_read returns SANE_STATUS_GOOD in ADF mode + + * transformed all gif files into jpeg format because of all the + trouble about the gif licence. Will never use gif again! + + * added correction for inquiry-ADF-bit for Astra 2400S + + * added second test method to detect ADF error in sane_start: + inquiry block is read and cover open, jammed and no paper is tested + + +build-10 -> build-11 +-------------------- + + * test for adf errors (sane_start) is now only done if adf is selected + + +build-11 -> build-12 +-------------------- + + * moved adf error tests from sane_start to umax_start_scan + + * sane_control_option: printing option name is set to "(none)" if pointer is NULL + + * added error message after each call of sanei_scsi_cmd + + * changed manipulation of ADF bit for Astra 2400S in umax_correct_inquiry + from dev->adf to set_inquiry_sc_adf() + + +build-12 -> build-13 +-------------------- + + * changed inquiry data of UC1200SE + + * added set_inquiry_CBHS(50) in correct_inquiry for UC1200SE + + * correct_inquiry is called for scanners with short inquiry now + + * dev->inquiry_max_calib_lines is preset to 66, + problem with Vist S6E V1.6 should be solved this way. + + * inquiry block parameters > 0x93 are only used if block is long enough + + +build-13 -> build-14 +-------------------- + + * changed inquiry_cbhs_[min,max] -> inquiry_[contrast, brightness, threshold, + highlight, shadow]_[min,max] and set correct values for CBHS_255 and CBHS_50 + + * changed cbhs_str[CBHS_50] from "78-178" to "0-50" + + * changed umax-uc1200se.h inquiry byte 0x64 bit 5 to 0 (gamma download format != 2) + and wdb len from 0x30 to 0x31 + + * disabled halftone pattern download, and halftone dimension selection - not ready + preselected is a 8x8 built in halftone pattern + + * removed set_inquiry_CBHS(50) from correct_inquiry for UC1200SE + + * added "Astra 2200 " as supported scanner + + * added cast from (void *) to (unsigned char*) in umax_send_gamma_data + + +build-14 -> build-15 +-------------------- + + * add "(unsigned)" to option number test in sane_control_option + + * for Astra 1200 the driver does not wait any more until scanner is repositioned + this is experimental, it may work with other scanners too! + + +build-15 -> build-16 +-------------------- + + * OPT_SOURCE: removed change of width/height and "*info |= SANE_INFO_RELOAD_PARAMS;" + from get_value to set_value and added "*info |= SANE_INFO_RELOAD_OPTIONS;" + + +build-16 -> build-17 +-------------------- + + * changed calibration mode from SHADING_TYPE_AVERAGE_INVERT to SHADING_TYPE_AVERAGE + for UMAX S-6E, UMAX S-6EG and Vista-S6E + + * changed several int -> unsigned int + + * added support for sanei_scsi_open_extended and variable scsi buffer size + + +build-17 -> build-18 (sane-1.0.2) +--------------------------------- + + * some cosmetic changes of the form + if (..) {...} + => + if (..) + { + ... + } + + * removed unnecessary SHADING_TYPE-handling routines for UMAX S-6E, UMAX S-6EG and Vista-S6E + + * scsi buffer is only reallocated when size has changed + + +build-18 -> build-19 +-------------------- + + * added UMAX PowerLook III as supported scanner + + * added EPSON Perfection600 as supported scanner (is a Astra 1200S) + + * type of option 0 (Option numbers) is set to SANE_TYPE_INT now (undefined before) + + * removed bug that mixed 12 bit and 14/16 bit gamma table when 14 or 16 bit are supported + and changed value range of grey custom gamma table: + - maximum output value is 255 for 8 bits/pixel + - maximum output value is 65535 for 9-16 bits/pixel + the custom gamma table does work now with 9-16 bits/pixel output mode + + * added routine that reverses bit order in lineart mode when scanner sends LSB first in lineart + + +build-19 -> build-20 +-------------------- + + * added variable calibration_width_offset, Mirage D-16L offset of 308 is defined there + and added offset of 28 for PowerLook III + + * umax_do_calibration: guessed calibration data vaules use 2 bytes/color if scanner + supports gamma input bit depth supports 2 bytes/color + + * calibration by driver now does work for PowerLook III + + * added function umax_set_rgb_bind() that sets the sensitivity of some options + that make available different settings for RGB. This function now is called + when scanmode is selected. + + * changed some int to unsigned int + + * made dev->pause_after_reposition = -1; as default, so as default sane-umax does not wait + any more while the scanhead is moving back - this may break the support of some scanners + so it may have to be changed (dev->pause_after_reposition = 0;) for some scanners. + reported to work for: + - Supervista S12 + - PowerLook III + - Astra 2200 + + +build-20 -> build-21 (sane-1.0.3) +--------------------------------- + + * Added Linotype "JADE" as supported scanner + + * Removed bug in sane_close that caused segfault when + sane_close was called twice with the same scanner handle + +build-21 -> build-22 +-------------------- + + * replaced fgets() by sanei_config_read() + + * added command queueing for scsi read commands + - removed WAIT_SCANNER scsi command call after read command + - the additional buffers are allocated in the reader process + - the number of buffers can be defined in umax.conf + with option scsi-maxqueue [1..8] default:2 + (did some tests with Supervista S12 and PowerLook III + on the same scsi controller (ncr53c810): + command queueing works, it is possible to start + both scanners at the same time. The scanner with the + higher scsi-ID (=priority) does scan without stops, + the other scanner has to wait long time between the + scsi-read commands, it also is possible to give + bothe scanners different scsi command queueing depth). + + * prepared replacement of pipe by sanei_ipc_* routines + + * added routines umax_get_lamp_status and umax_set_lamp_status + + * updated inquiry output (new bits from powerlook scsi protocol 1.0 X101) + + * added new (powerlook scsi protocol 1.0 X101) window descriptor block entries to umax-scsidef.h + + * added button protocol (3 buttons, from powerlook scsi protocol 1.0 X101) + + * added button "lamp off" to turn off scanner lamp + (this option may be remoeved again in later version, may be + it will be replaced by an autmatic lamp turn off function) + + * added options to umax.conf + lamp-off-at-exit [01] default:0 + lamp-off-time [10..720] (in minutes) default:30 + + * if lamp_off_at_exit == 1 and lamp_control_available == 1 + the scanner lamp is turned off in sane_close + + * renamed umax-scanner.h and umax-"scannername".h to umax*.c + because these files create data code + + +build-22 -> build-23 +-------------------- + + * scsi commands set/get lamp control are send with a command length of + 10 bytes instead of 6 bytes and control bit #7 is in byte 0x03 + and not in byte 0x04 (it is described incorrect in the umax doc) + + * enabled lamp control for Astra 2200 + + * removed umax.conf options lamp-off-at-exit and lamp-off-time + + * added frontend option lamp-off-at-exit + + * added frontend option lamp-on + + * forcing calibration_bytespp = 1 for Astra 2200 - quality calibration does work now + + * for Astra 2200 selection for different x/y resolutions is disabled + + +build-23 -> build-24 (sane-1.0.4) +-------------------- + + * forcing calibration_bytespp = 1 for Astra 2100S + + * Added Astra 2100S as supported scanner + + * enabled lamp control for Astra 2100S + + * Added warning in sane_control_option if option could not be set + + +build-24 -> build-25 +-------------------- + + * dev->lamp_control_available is preset to 0 + + * dev is erased (0) after it is allocated + + * added special options for PowerLook 3000: + - calibration is done for each CCD pixel + - calibration_bytespp is set to 1 + the scanner still is not listed as supported scanner because + the changes are not tested + + * added options that are read from umax.conf: + + - execute-request-sense: + values: 0=disabled, 1=enabled, default=0 + If set to 1 umax_do_request_sense is called in umax_do_calibration. + This can hang the system (but has been enabled until this version) + + - scsi-buffer-size-min, scsi-buffer-size-max: + values: 4096-1048576, default min=65536, max=131072 + + - scan-lines, preview-lines: + values: 1-65535, default scan-lines=40, preview-lines=10 + define the maximum number of lines that are scanned into one buffer + + - force-preview-bit-rgb: + values: 0=disabled, 1=enabled, default=0 + set preview bit in rgb real scan + + - slow-speed, care-about-smearing: + values: -1=auto, 0=disabled, 1=enabled, default=-1 + dangerous options, needed for some scanners + + - calibration-full-ccd: + values: -1=auto, 0=disabled, 1=enabled, default=-1 + do calibration for each pixel of ccd instead of selected image + + - calibration-width-offset: + values: -1=disabled, >=0 set value + add an offset width to the calculated with for image/ccd + + - calibration-bytes-pixel: + values: -1=disabled, 0=not set, 1=1 byte/pixel, 2=x bytes/pixel + use # bytes per pixel for calibration + + - invert-shading-data: + values: -1=auto, 0=disabled, 1=enabled, default=-1 + invert shading data before sending it back to the scanner + + - lamp-control-available: + values: 0=automatically set by driver - if known, 1=available, default=0 + + * updated html-documentation + + * updated sane-umax.man + + * added pre focus options (PowerLook 3000): + - FP - fix focus position + - FDC - lens calibration in doc position + - PF - disable pre focus + - LCL - use 0.6/0.0mm holder focus position + + * added #define SANE_I18N(text) text to umax.c. This is to mark the backend + specific texts that shall be translated by the frontend + + +build-25 -> build-26 +-------------------- + + * added umax.conf option handle-bad-sense-error # + + * added Mirage II as supported scanner + + * added routine umax_set_max_geometry that defines available surface + and resolution in dependance of scansource and double optical resolution + (before this bugfix the change of dor mode overwrote the settings of + scansource and vice versa) + + * added ESCORT Galleria 600S as supported scanner, is an Astra 600S + + * removed if (config_line[len - 1] == '\n')... while reading config file + + * changed some static unsigned int to static int + + * sane_exit does free devlist + + * added some options for Powerlook 3000 - still is not completly supported + + +build-26 -> build-27 +-------------------- + + * added Linotype Saphir4 as supported scanner (is a Powerlook III) + + * increased length of vendor, product and version by 1 to make sure the + string always ends with '\0' + + * changed default minimum scsi buffer size from 64KB to 32 KB + + +build-27 -> build-28 +-------------------- + + * removed "#include <assert.h>" from umax.c, assert() is never called + + * added UMAX Powerlook 3000 as supported scanner + + * bugfix for 3 pass scanning: + reposition_scanner is called in sane_start for 2nd and 3rd pass of 3 pass scan + + + * bugfix for UTA and DOR geometry offsets: + - added inquiry_uta_x_off and inquiry_uta_y_off + - geometry ranges are defined from xoff..xoff+dx, yoff..yoff+dx now + - check_values also does test for this range + - removed dev->maxwidth and dev->maxlength, values are calculated in place now + + * bugfix for DOR: + - bottom-right edge keeps in position if possible when DOR is switched on + + * bugfix for Astra 600S and 610S color mode: + - scale_y is set to 0.5 if y_resolutions is smaller than optical_res/2 + this does fix the color scan problems for Astra 610S. + + * added Heidelberg Linoscan 1400 "HDM LS4H1S" as supported scanner + + +build-28 -> build-29 +-------------------- + + * removed bug in umax_set_max_geomentry: now left y keeps in position when possible + + * rgb bind keeps unchanged when scanmode != color (only gray selection active in this case) + + * in function static int umax_give_scanner(Umax_Device *dev) moved umax_reposition_scanner(dev); + to the end of the function + + * set calibration area for astra 2200 to full ccd and calibration_bytespp to 2 + + * UC840 does wait in reposition_scanner until scan head stops moving + + * set minimum resolution to 5 (was 1 before what could make problems) + + +build-29 -> build-30 +-------------------- + + * added EDGE KTX-9600US as supported scanner + + * Changing scansource does not change lower left selection edge + if not necessary + + +build-30 -> build-31 (sane-1.0.6) +--------------------------------- + + * sane-umax.man uses @LIBDIR@ and @CONFIGFIR@ + + * changed automatic calibration_width_offset marker from -1 to -99999 + + * changed + dev->calibration_area to 1 and + calibration_width_offset to -1 and + inquiry_max_calibration_data_lines to 66 + for UMAX Supervista S-12, LinoHell Office and LinHell JADE + + +build-31 -> build-32 (sane-1.0.7) +--------------------------------- + + * added Nikon AX-110 as supported scanner (Vista S6E) + + * added umax.conf option "gamma-lsb-padded" with options: + -1 = auto + 0 = inactive (msb padded) + 1 = active (lsb padded) + activated for astra 1220s, internal 12 bit gamma correction does work now + + +build-32 -> build-33 +-------------------- + + * added USB support for Astra 2200 + thanks to Frank Zago <fzago@greshamstorage.com> + for creating umax-usb.c and suggesting several changes to umax.c + + * added umax.conf option "connection-type", values: 1=scsi, 2=usb + + * added umax.conf device selection "usb 0x1606 0x0230" (usb vendor_id product_id) + + * removed #ifdef HAVE_SANEI_IPC parts (SANEI_IPC was never defined) + + +build-33 -> build-34 (sane-backends-1.0.8) +-------------------- + + * removed bug in reader_process / umax_queue_read_image_data: + - the integer value that keeps the length information is kept valid + as long a queued scsi command is still running now + - when the scsi command returns with less data than requested + then the routine does handle this correct now + (before the routine handled this as if the requested amount of + data was returned) + + * backend waits until scanhead has repositioned after scan for UMAX UC1260 + + * y_dpi quanitsation is set to 5 (same as x_dpi quantisation - was 1) + + +build-34 -> build-35 +-------------------- + + * changed ID string "LS4H1S " to "LS4H1S " (remove one space) + + * activated lamp control for Astra 1220S V1.5 + + +build-35 -> build-36 +-------------------- + + * added PowerLook 2100XL as supported scanner + + * settings for PowerLook 2100XL: + calibration_width_offest = 52 + (calibration_bytespp = 2, default) + (calibration area = image, default) + + * force_quality_calibration defined for PowerLook 2100XL: + (normal calibration mode expects different calibration_width_offsets + for PowerLook 2100XL, so it is not possible to calculate the correct + number of calibration pixels in normal calibration mode) + - disable quality calibration option for frontend + - always set quality calibration bit when scanning + + * changes for exposure times: + - Inquiry does print correct maximum value now + - changed variable type from SANE_FIXED to SANE_INT + - exposure times setting for colormode does work correct now + - as default there is only one common setting for exposure times in rgb mode, + added umax.conf option exposure-time-rgb-bind to enable different settings + for each color + - added option select-calibration-exposure-time, disabled by default. + if disabled then the calibration expousre times are the same as the scan exposure times + + * added UMAX PS-2400X as supported device + + * added batch scanning support: + Options: + --batch-scan-start: calibration, scan, scanhead is not moved into home position + --batch-scan-loop: no calibration, scan, scanhead is not moved into home position + --batch-scan-end: no calibration, scan, scanhead is moved into home position + --batch-next-tl-y: defines the y start position of the bext batch scan + + - calibration for batch scan needs a different offset and always is done for full ccd. + PowerLook 2100XL: calibration_width_offset_batch = 1052 + Powerlook III: calibration_width_offset_batch = 828 + + +build-36 -> build-37 +-------------------- + + * added Linotype Hell SAPHIR3 as supported scanner + + * if batch-next-tl-y is not set then the backend uses br_y + + +build-37 -> build-38 (sane-backends-1.0.9) +-------------------- + + * added Power Look 2000 as supported scanner + + +build-38 -> build-39 (sane-backends-1.0.10) +-------------------- + + * Power Look 2000: corrected wrong calibration-width-offset from 52 to 22 + and added calibration-width-offset-batch to 24 + + * initialized some global pointers and num_devices in sane_init + + * replaced fork() by thread for OS2 + +build-39 -> build-40 (sane-backends-1.0.12) +-------------------- + + * UC630: dev->pause_after_reposition = 0 + + +build-40 -> build-41 (sane-backends-1.0.13) +-------------------- + + * Astra 2200 SU via USB connection: quality calibration is reset and not selectable + + +build-41 -> build-42 +-------------------- + + * replaced fork() routines by sanei_thread routines + + * added call sanei_scsi_req_flush_all(); to do_cancel - in threaded case this is not done when reader_process is cancled + + * in reader_process: marked freed scsi buffers as freed(NULL). This is necessary when reader_process is run as thread! + + * Backend is compatible to Mac OS-X now. + + +build-42 -> build-43 +-------------------- + + * DOR mode does test for non negative upper_left_x values, set to 0 if negative + + +build-43 -> build-44 +-------------------- + + * the sense_hanlder stores the sense data in buffer[0] when valid sense data is avialable + + * do_calibration: + - corrected bug in reading calibration width from scanner (added /bytespp) + - when additional request sense is not called then the buffer is not erased, + the sense data from the sense_handler is used + - when additional request sense is called then the buffer is erased at first + +build-44 -> build-45 +-------------------- + + * added defualt options for Linotype Opal2 + * changed bug in highlight-blue function (using green value) + |