1commit d82847a6b4bbd2864388f2bf8e6a2b9d1aed05c4 2Author: Vasily Khoruzhick <anarsoul@gmail.com> 3Date: Sun May 14 09:26:30 2017 -0700 4 5 0.7.0 6 7commit 5a7e6e07ffa34fd4131f95fbfe5ca23d7eedc592 8Author: Vasily Khoruzhick <anarsoul@gmail.com> 9Date: Mon Nov 7 08:10:44 2016 -0800 10 11 etes603: fix compilation error caused by a0bbbd7d3 12 13commit 2162aa9f48a739cc8e3b25643f9f8cba979d545e 14Author: Vasily Khoruzhick <anarsoul@gmail.com> 15Date: Mon Jul 18 20:46:20 2016 -0700 16 17 uru4k: decrypt image if necessary 18 19 Add naive detection of image encryption: if stddev between two 20 adjacent rows is higher than threshold assume that image is encrypted 21 and decrypt it. 22 23 Fixes fd.o bug 88945 24 25commit a1f36c71c9063a5ec179acb2cfbcc56206ba2e4f 26Author: Vasily Khoruzhick <anarsoul@gmail.com> 27Date: Thu Feb 11 00:14:47 2016 -0800 28 29 upeksonly: decrease bz3_threshold for 147e:1001 30 31 Its width is only 216 pixels, and it appears not to be enough for matching at 32 default threshold. 33 34commit e4eedef27e0b8b9eb5a8f3ae68086e1013e90b85 35Author: Vasily Khoruzhick <anarsoul@gmail.com> 36Date: Tue Feb 9 20:15:38 2016 -0800 37 38 upeksonly: use fpi_std_sq_dev() and fpi_mean_sq_diff_norm() 39 40 Use fpi_mean_sq_diff_norm() for blank line detection and 41 fpi_mean_sq_diff_norm() for duplicate line detection. 42 Fixes finger presence and removal detection. 43 44commit 5e296959697643aab14feac709c8adda8726d582 45Author: Vasily Khoruzhick <anarsoul@gmail.com> 46Date: Tue Feb 9 17:19:39 2016 -0800 47 48 lib: move some functions from vfs5011.c into img.c 49 50commit 49a46668ad2f45d26d3b6d6168ada620c54290ab 51Author: Konstantin Semenov <zemen17@gmail.com> 52Date: Tue Jan 19 14:26:29 2016 +0300 53 54 lib: Add VFS0050 driver 55 56 New driver for 138a:0050 device 57 58 https://bugs.freedesktop.org/show_bug.cgi?id=91616 59 60commit 76269decdd405a0ae919a24c1ba3061c44bf80b2 61Author: Bastien Nocera <hadess@hadess.net> 62Date: Sat Nov 5 00:12:46 2016 +0100 63 64 aes4000: Fix possible crash if USB init fails 65 66 The code was indented, but only the first call was actually in the 67 conditional, which meant that initialisation carried on as normal on top 68 of a failed USB device. Exit early and remove the conditional to fix 69 this. 70 71 https://bugzilla.gnome.org/show_bug.cgi?id=98594 72 73commit 8454a25ecfc27c5d96867a9fda012d0a73ffeddb 74Author: Bastien Nocera <hadess@hadess.net> 75Date: Sat Nov 5 00:11:30 2016 +0100 76 77 aes3500: Fix possible crash if USB init fails 78 79 The code was indented, but only the first call was actually in the 80 conditional, which meant that initialisation carried on as normal on top 81 of a failed USB device. Exit early and remove the conditional to fix 82 this. 83 84commit a0bbbd7d32150ab6ed25f466cce05169fbbbaead 85Author: Bastien Nocera <hadess@hadess.net> 86Date: Sat Nov 5 00:10:17 2016 +0100 87 88 drivers: Print USB error when libusb_claim_interface() fails 89 90 https://bugs.freedesktop.org/show_bug.cgi?id=98594 91 92commit 12f6dae8cd81f363e02849acea40c69a8764f487 93Author: Bastien Nocera <hadess@hadess.net> 94Date: Thu Nov 3 14:05:11 2016 +0100 95 96 build: Fix 9570c36 on Debian 97 98 Debian's ash doesn't implement pushd/popd. 99 100commit 9570c36fd42bff6246de50ace398a457b9495b46 101Author: Bastien Nocera <hadess@hadess.net> 102Date: Tue Jul 19 11:18:31 2016 +0200 103 104 build: Fix running autogen.sh out-of-tree 105 106commit 487dae0d2fa5a7a1951581f17b55fa1f411e73e6 107Author: Bastien Nocera <hadess@hadess.net> 108Date: Tue Apr 12 16:10:53 2016 +0200 109 110 Work-around kernel's lack of USB PM 111 112 The device is already supported by the vfs5011 driver and non- 113 blacklisted, so will show up in the udev rules, but for ease of 114 backporting to older versions, add it to the whitelist anyway. 115 116 https://bugzilla.redhat.com/show_bug.cgi?id=1173367 117 118commit d71018bd8f478baea65cc48c59074abb66813b90 119Author: Vasily Khoruzhick <anarsoul@gmail.com> 120Date: Tue Nov 24 15:22:46 2015 -0800 121 122 vfs5011: add USB IDs of device found in some Toshiba laptops 123 124commit 61fa57b05dccda8ed855f66e650564e5da9acb1a 125Author: Vasily Khoruzhick <anarsoul@gmail.com> 126Date: Wed Oct 14 10:45:15 2015 -0700 127 128 aes1610: decrease bz3_threshold to 20 129 130commit fc1781e317f66eb5f9dd10afb765d721f8116f2f 131Author: Volkau Siarhei <lis82@mail.by> 132Date: Sat Nov 1 15:46:52 2014 +0300 133 134 vfs101: fix broken enrolling by prev commit to vfs101.c 135 136commit ffef6c2bcc87a40767b23891ec1e53a4d459e9c1 137Author: Vasily Khoruzhick <anarsoul@gmail.com> 138Date: Tue Sep 29 17:10:42 2015 -0700 139 140 vfs5011: add proper dev_deactivate() support 141 142 Driver is not able to cancel imaging process without this change 143 144commit 67d29f79368aab9604b23e1744907bfdbe9ec44c 145Author: Vasily Khoruzhick <anarsoul@gmail.com> 146Date: Tue Sep 29 17:08:51 2015 -0700 147 148 aes1610: improve gain settings 149 150commit 9437c98d54c33a5a752f4ffd28a761b7cbf32141 151Author: Vasily Khoruzhick <anarsoul@gmail.com> 152Date: Tue Sep 22 19:49:58 2015 -0700 153 154 lib: frame assembling: flip image for non-reverse direction 155 156 It was here for aes2501 and aes1610 before moving assembling routines 157 into common code. 158 159commit f7d00a828dd0af5db75a80415758c848d48d11f8 160Author: Vasily Khoruzhick <anarsoul@gmail.com> 161Date: Sat Sep 19 15:27:19 2015 -0700 162 163 lib: use normalized error to make decision about movement direction 164 165 Height is not a good determinant in movement direction, normalized 166 error is much better. Should fix aes1610 and aes2501 driver issues. 167 168commit 0f0a4b2da6704505471aa2797fac7c41de21a862 169Author: Vasily Khoruzhick <anarsoul@gmail.com> 170Date: Sat Sep 19 12:25:05 2015 -0700 171 172 aes2660: set partial flag on an image 173 174 Set partial flag to remove false minutiae at the perimeter of a scan 175 176commit c9cdbaf8803be8c9ca33029419f006deb55d7e88 177Author: Vasily Khoruzhick <anarsoul@gmail.com> 178Date: Sat Sep 19 12:24:48 2015 -0700 179 180 aes2550: set partial flag on an image 181 182 Set partial flag to remove false minutiae at the perimeter of a scan 183 184commit 83f29dad9f42f4cae53c4d130ca792d25e6468c3 185Author: Vasily Khoruzhick <anarsoul@gmail.com> 186Date: Sat Sep 19 12:24:21 2015 -0700 187 188 aes2501: set partial flag on an image 189 190 Set partial flag to remove false minutiae at the perimeter of a scan 191 192commit c2a11c5918684a842c34418fd7630f26492790d2 193Author: Vasily Khoruzhick <anarsoul@gmail.com> 194Date: Sat Sep 19 12:23:33 2015 -0700 195 196 aes1610: set partial flag on an image 197 198 Set partial flag to remove false minutiae at the perimeter of a scan 199 200commit 3746b2ad5c91e649488916179b44460609bab805 201Author: Vasily Khoruzhick <anarsoul@gmail.com> 202Date: Tue Sep 1 08:39:38 2015 -0700 203 204 upeksonly: use line assembling routines to account variable swiping speed 205 206commit b51fa446e38e621c1965998e1e404574599dc728 207Author: Vasily Khoruzhick <anarsoul@gmail.com> 208Date: Sat Sep 5 09:38:10 2015 -0700 209 210 lib: move line assembling routines out of vfs5011 into common code 211 212commit 6fc5293e8330e65ed21c0e43a18b3be061933e74 213Author: Vasily Khoruzhick <anarsoul@gmail.com> 214Date: Sat Sep 19 10:20:55 2015 -0700 215 216 upeksonly: add support for 147e:1001 device 217 218commit aab3daa28b4b45a94f7142eadfef76343c1b13d3 219Author: Vasily Khoruzhick <anarsoul@gmail.com> 220Date: Sat Sep 19 10:15:35 2015 -0700 221 222 upeksonly: move regwrite sequences into a header 223 224commit e40f7bd1f7aba96de11a5165a15b3b1783181ac3 225Author: Vasily Khoruzhick <anarsoul@gmail.com> 226Date: Mon Aug 31 21:25:20 2015 -0700 227 228 lib: move frame assembling routines into a separate file and make them usable by non-aes drivers 229 230 Frame assembling routines are not aes-specific, so move them into a separate file 231 and add an accessor for peeking a pixel. 232 233commit 6664f87d8f2650d7471a110b33839c878f97f8b4 234Author: Vasily Khoruzhick <anarsoul@gmail.com> 235Date: Sat Sep 19 10:28:49 2015 -0700 236 237 aes1610: fix memory corruption introduced by a457658f1b 238 239commit 6e8d5cd6a199f83fb5f862a8421f8582608032f5 240Author: Vasily Khoruzhick <anarsoul@gmail.com> 241Date: Sun Aug 30 16:38:44 2015 -0700 242 243 vfs5011: do duplicate line search for 30 lines 244 245 It seems that on faster devices, the driver can get more duplicate lines. 246 Without this change scan looks too stretched. 247 248 Suggested by: Thomas Rinsma <thomasrinsma@gmail.com> 249 250commit 9f7e1ecf40ec86bae0088c847ee2453be8a27ad5 251Author: Vasily Khoruzhick <anarsoul@gmail.com> 252Date: Sun Aug 30 16:36:58 2015 -0700 253 254 vfs5011: make '5 scans per enroll' work with this driver 255 256 Restart scan if core did not ask to terminate it explicitly. 257 258commit d9567002e458fca4f289d84bb3cb49d9d97585e7 259Author: Vasily Khoruzhick <anarsoul@gmail.com> 260Date: Wed Apr 1 13:39:54 2015 +0300 261 262 imgdev: reset action_result after reporting it 263 264 Otherwise we end up in reporting the same result on next iteration. 265 266commit a656a4a9f37dcf34cd1047647af7253fae89eb4c 267Author: Vasily Khoruzhick <anarsoul@gmail.com> 268Date: Thu Mar 19 19:34:08 2015 +0300 269 270 upektc_img: Handle scan status reported by sensor properly 271 272 Sensor can report "short scan" or "finger not centered" instead of image, 273 so it's necessary to handle them gracefully. 274 275commit 2944a35e74fdf1d91789b5715cce81974d6668a3 276Author: Vasily Khoruzhick <anarsoul@gmail.com> 277Date: Thu Mar 19 19:27:13 2015 +0300 278 279 imgdev: Add fpi_imgdev_abort_scan() to abort scan gracefully from within the driver. 280 281 Smart sensors like Upek TouchChip Coprocessor can provide scan status instead 282 of image, as result we need to report such status as "short scan" or "finger not centered" 283 from within the driver, since it's not a session error. 284 285commit 391373fb0c9e427ef93a71b056a999739297e428 286Author: Vasily Khoruzhick <anarsoul@gmail.com> 287Date: Thu Mar 19 19:33:26 2015 +0300 288 289 upektc_img: Fix copy/paste error 290 291 Replace upekts_img with upektc_img 292 293commit 985e8c4577cf6e79c6f99cf0c694f4e732daab27 294Author: Vasily Khoruzhick <anarsoul@gmail.com> 295Date: Sun Mar 15 21:21:37 2015 +0300 296 297 aes1660: set partial flag on an image 298 299 Sensor width seems to be only 128 pixels and that's not enough 300 for scanning whole finger surface. Lower bz3_threshold to 25, 301 since for wrong finger score never goes above 10, but sometimes 302 for right finger score is below 40. 303 304commit bd0d4258e4b27579ecdfa5e91bd6f228d28796f3 305Author: Vasily Khoruzhick <anarsoul@gmail.com> 306Date: Sun Mar 15 16:18:50 2015 +0300 307 308 upektc_img: set partial flag on an image 309 310 Sensor width seems to be only 128 pixels and that's not enough 311 for scanning whole finger surface. Lower bz3_threshold to 20, 312 since for wrong fingerprint score never goes above 10, but sometimes 313 for right finger score is below 40. 314 315commit 9bbd9b208a69282a7ede960711839b03b6ea4c8c 316Author: Vasily Khoruzhick <anarsoul@gmail.com> 317Date: Sun Mar 15 16:17:08 2015 +0300 318 319 lib: add partial image flag 320 321 And activate perimeter points removal if this flag is set. 322 323commit 59fe0fb699beca75d087ca146cfd413a9bb3da8a 324Author: Vasily Khoruzhick <anarsoul@gmail.com> 325Date: Sun Mar 15 16:13:32 2015 +0300 326 327 nbis: add one more step of false minutia removal 328 329 Remove minutiae that are located on scan perimeter 330 331commit f1fdd71613a483e054cd11e19bdcfb49c95fa2cb 332Author: Bastien Nocera <hadess@hadess.net> 333Date: Sun Mar 15 23:13:08 2015 +0100 334 335 rules: Add driver name to the USB properties 336 337 Makes it easier to see which driver is being used. 338 339commit bb66780cb58daf33d55524a9b434e218a991507c 340Author: Bastien Nocera <hadess@hadess.net> 341Date: Sun Mar 15 23:10:03 2015 +0100 342 343 build: Document why some drivers are disabled 344 345 This avoids confusion when looking at the build logs 346 347commit 45fb6d790815eec3d9e08a9e459a325fbb9addc0 348Author: Carlos Olmedo Escobar <carlos.olmedo.e@gmail.com> 349Date: Fri Feb 6 17:22:16 2015 +0100 350 351 vfs101: Logical conjunction always evaluates to false. 352 353commit 9f408bf51b7740e87c3d0b90e5adc5de3c37c052 354Author: Timo Teräs <timo.teras@iki.fi> 355Date: Tue Oct 7 08:18:41 2014 +0300 356 357 imgdev: fix enum mismatch for dev_change_state() call 358 359 This bug has existed long time, but it was uncovered by commit 360 e215b0509448 which enabled multiple enrollment rounds. 361 362 In practice this broke (at least) URU4000 driver state machine 363 causing it to enter indefinite loop - due to the invalid state 364 change callback. 365 366 Patch originally posted at: 367 http://lists.freedesktop.org/archives/fprint/2014-June/000603.html 368 369 Test and verification results: 370 http://lists.freedesktop.org/archives/fprint/2014-June/000607.html 371 372 Signed-off-by: Timo Teräs <timo.teras@iki.fi> 373 374commit ee32166267323ad6c66c022e18ef6af76a3a5423 375Author: Bastien Nocera <hadess@hadess.net> 376Date: Tue Feb 3 17:36:32 2015 +0100 377 378 0.6.0 379 380commit a3c90f2b24434aa36f782aca3950fd89af01fce0 381Author: Vasily Khoruzhick <anarsoul@gmail.com> 382Date: Fri Sep 6 16:31:15 2013 +0300 383 384 upektc_img: handle 0x28 message better 385 386commit dc537ef2c9b1ee62e88db02560abcead4f9486f9 387Author: Arseniy Lartsev <arseniy@chalmers.se> 388Date: Tue Nov 5 16:32:15 2013 +0800 389 390 lib: Add VFS5011 driver 391 392 New driver for VFS5011 138a:0011 and 138a:0018 393 394 https://bugs.freedesktop.org/show_bug.cgi?id=61692 395 396 [vasilykh]: 397 - use g_get_real_time() instead of non-portable time() 398 - use g_free() instead of free() 399 - comment out "RECV(VFS5011_IN_ENDPOINT_CTRL2, 8)" 400 401commit 061a457658f1b39fe04ceb3c104c46b19a13799f 402Author: Vasily Khoruzhick <anarsoul@gmail.com> 403Date: Mon Dec 30 12:07:53 2013 +0300 404 405 aeslib: improve frame-assembling routines 406 407 Some scanners provide hardware assistance in frame assemling, i.e. 408 horizontal and vertical offset to previous frame is provided. This 409 commit improves code to utilise that assistance. Sensors without 410 hardware assistance will use software algorithm, which was also 411 improved to do search in horizontal direction. 412 413commit 82ae7c1c0909dba765ed8c42d26c22913841f206 414Author: Vasily Khoruzhick <anarsoul@gmail.com> 415Date: Mon Dec 30 12:05:08 2013 +0300 416 417 aesx660: fix spelling of define name 418 419commit 79d79c3c875d41379eae418157a8254e87641f85 420Author: Rex Dieter <rdieter@math.unl.edu> 421Date: Wed Dec 17 07:27:15 2014 -0600 422 423 lib: Test before applying power saving udev rules 424 425 add TEST=="power/control"... to test for the presence of the file 426 before setting it. 427 428 See https://bugzilla.redhat.com/show_bug.cgi?id=950205 429 430 https://bugs.freedesktop.org/show_bug.cgi?id=87414 431 432commit 35e356f625d254f44c14f720c0eb9216297d35c2 433Author: Vasily Khoruzhick <anarsoul@gmail.com> 434Date: Mon Aug 12 11:45:17 2013 +0300 435 436 build: add subdir-objects option to make automake happy 437 438 Without subdir-object automake throws warnings like this one: 439 440 Makefile.am:x: warning: source file 'drivers/aes2550.c' is in a subdirectory, 441 Makefile.am:x: but option 'subdir-objects' is disabled 442 443commit 948ab02d1a9d321e31dfd501646c29b880c1e5d0 444Author: Vasily Khoruzhick <anarsoul@gmail.com> 445Date: Fri May 24 12:34:45 2013 +0300 446 447 lib: use pixman for imaging utils 448 449 pixman is very lightweight library for pixel manipulation, and it 450 has no dependencies except glibc, so using it instead of gdkpixbuf/imagemagick 451 makes list for libfprint dependencies a bit shorter. 452 453commit a6101026d2ddff76efe9f46669e24db3a0a6a0e4 454Author: Patrick Marlier <patrick.marlier@gmail.com> 455Date: Wed Dec 12 20:43:09 2012 +0100 456 457 Add EgisTec ES603 driver 458 459 This driver handles EgisTec ES603 device, ID 1c7a:0603 460 461commit e0966cb20fb7cf1c9ee8165428f5d3810e2e8c4e 462Author: Vasily Khoruzhick <anarsoul@gmail.com> 463Date: Mon Mar 4 17:50:54 2013 +0300 464 465 vfs101: drop nr_enroll_stages override 466 467 Imaging drivers aren't allowed to change number of enroll stages anymore, 468 all that imaging driver does is providing image to the upper layer. 469 Upper layer decides if it needs more scans. 470 471commit e278e8321cb91bcbf577d05778fe6553a14b4f14 472Author: Vasily Khoruzhick <anarsoul@gmail.com> 473Date: Mon Feb 18 14:37:48 2013 +0300 474 475 vfs301: drop nr_enroll_stages override 476 477 Imaging drivers aren't allowed to change number of enroll stages anymore, 478 all that imaging driver does is providing image to the upper layer. 479 Upper layer decides if it needs more scans. 480 481commit e1728e7c25d6bd3fdbebf5bc9cc2040cf9063979 482Author: Vasily Khoruzhick <anarsoul@gmail.com> 483Date: Mon Feb 18 13:41:13 2013 +0300 484 485 aes1660: remove unneeded lowering of bz3_threshold 486 487 Since 5 scans for enroll was introduced it's not necessary to lower 488 bz3_threshold anymore, there's a good probability that scan to verify matches 489 with at least one enrolled sample. 490 491commit e215b0509448e05bf7a352317bc3282a9d5fd437 492Author: Vasily Khoruzhick <anarsoul@gmail.com> 493Date: Mon Feb 18 12:27:33 2013 +0300 494 495 imgdev: perform 5 scans for enrollment 496 497 This feature dramatically improves matching rate on devices with small 498 sensors. 499 500commit fe3fdd1f5061d3b6fa172758c07705f8bfcb6d07 501Author: Vasily Khoruzhick <anarsoul@gmail.com> 502Date: Sat Mar 9 14:09:11 2013 +0300 503 504 upeke2: disable by default, device is now handled by upektc_img driver 505 506commit 5ff45658c0921f250e3c24538037b80024287ea7 507Author: Vasily Khoruzhick <anarsoul@gmail.com> 508Date: Sat Mar 9 14:08:50 2013 +0300 509 510 upektc_img: add support for Upek Eikon 2 devices 511 512commit d12b29478305da18918b2cb0df98c93516cd77d9 513Author: Vasily Khoruzhick <anarsoul@gmail.com> 514Date: Fri Feb 8 19:04:30 2013 +0300 515 516 Add upektc_img driver 517 518 Imaging driver, handles UPEK 147e:2020 device 519 520commit 2bba4fb0733e97a4f3aca44b170159c5b92cf5b5 521Author: Vasily Khoruzhick <anarsoul@gmail.com> 522Date: Fri Feb 8 14:38:13 2013 +0300 523 524 Implement image capture (both synchronous and asynchronous variants) 525 526 fp_dev_img_capture() is not implemented and returns -ENOTSUPP for all devices 527 since migration to asynchronous model. This commit implement missing functionality 528 529commit aeca32fc12a4fd98ae39334b301f94d371a34dcf 530Author: Juvenn Woo <machese@gmail.com> 531Date: Tue May 21 13:40:36 2013 +0800 532 533 aes3k: add aes3500 driver 534 535 This driver supports AES3500 device (08ff:5731) and utilizes common routines from 536 AES4000. 537 538commit 12c1088777b3d700d465f5dca961934837ae9013 539Author: Juvenn Woo <machese@gmail.com> 540Date: Thu May 16 18:25:10 2013 +0800 541 542 aes3k: extract common routines aes3k from aes4000 543 544 AES3500 and AES4000 are pretty similar devices, have same 545 command send, the only difference is in image size and init sequence. 546 Extract common routines from AES4K to be used later in AES3500 driver 547 548 https://bugs.freedesktop.org/show_bug.cgi?id=64351 549 550commit 9e2f8b5e754bf57fe0379428e7a382dacda165fe 551Author: Bastien Nocera <hadess@hadess.net> 552Date: Sun Aug 11 17:53:21 2013 +0200 553 554 0.5.1 555 556commit c14ebd8b636574cbe74aa05db0d0c1f6da9e2410 557Author: Bastien Nocera <hadess@hadess.net> 558Date: Thu Jul 25 12:29:53 2013 +0200 559 560 Add MS keyboard to the suspend blacklist 561 562 045e:00bb isn't a stand-alone fingerprint reader, but a keyboard 563 with a reader integrated. 564 565 https://bugs.freedesktop.org/show_bug.cgi?id=66659 566 567commit a6339a30ef8abea6c4dc90acbc04c49c07906abe 568Author: Bastien Nocera <hadess@hadess.net> 569Date: Thu Jul 25 12:27:53 2013 +0200 570 571 Fix udev rules printing 572 573 The blacklisted devices weren't correctly checked for past the first 574 item, as we weren't using the right index to get the product ID 575 from the ID table. 576 577commit f3dd55815ec1ef357e343ae1a810c6728d3ffd9a 578Author: Bastien Nocera <hadess@hadess.net> 579Date: Wed Jun 26 14:21:33 2013 +0200 580 581 lib: Use g_malloc0 instead of g_malloc+memset 582 583commit 6d65bfcf809d9b9be08027472ab1884c709d65c0 584Author: Patrick Marlier <patrick.marlier@gmail.com> 585Date: Tue Dec 4 09:06:06 2012 +0100 586 587 lib: g_malloc never fails 588 589 Or rather, it never returns errors and aborts instead if 590 memory cannot be allocated, so remove code that handled 591 failures. 592 593 https://bugs.freedesktop.org/show_bug.cgi?id=57869 594 595commit 1acd647b29c50b73d147ccb8c3530198ba44db70 596Author: Jeremy Bicha <jbicha@ubuntu.com> 597Date: Sat Apr 20 11:33:17 2013 -0400 598 599 build: Fix underlinking against glib 600 601 https://bugs.freedesktop.org/show_bug.cgi?id=63755 602 603commit 0e843ad6b36e370f9b7150a285a901075657cc8d 604Author: Vasily Khoruzhick <anarsoul@gmail.com> 605Date: Mon Feb 18 12:58:28 2013 +0300 606 607 build: fix autoreconf warnings 608 609 Replace AC_LANG_PROGRAM with AC_LANG_SOURCE and INCLUDES with AM_CFLAGS to fix 610 autoreconf warnings 611 612 https://bugs.freedesktop.org/show_bug.cgi?id=62748 613 614commit 7eafca7babe40bc7dddd27ce9051b13b90ada021 615Author: Vasily Khoruzhick <anarsoul@gmail.com> 616Date: Thu Jan 24 12:05:36 2013 +0300 617 618 build: glib-2.28 or newer is a requirement 619 620 We're using g_slist_free_full which was introduced in glib-2.28 621 622 https://bugs.freedesktop.org/show_bug.cgi?id=59790 623 624commit 3b3679c900f6739f7067f8d720e15d548bb39be9 625Author: Vasily Khoruzhick <anarsoul@gmail.com> 626Date: Sun Jan 13 16:43:38 2013 +0300 627 628 upeke2: Add support for 147e:2020 ID 629 630 https://bugs.freedesktop.org/show_bug.cgi?id=59320 631 632commit 43eca622cd49b58c87157e1ff1a2fcfdfba0934e 633Author: Vasily Khoruzhick <anarsoul@gmail.com> 634Date: Sun Jan 13 16:42:13 2013 +0300 635 636 configure.ac: Use AC_CONFIG_HEADERS 637 638 AM_CONFIG_HEADER is obsolete, use AC_CONFIG_HEADERS instead of it. 639 640 https://bugs.freedesktop.org/show_bug.cgi?id=59320 641 642commit 02509e10731b503b88ce0896ba6bb4c6c9e38679 643Author: Bastien Nocera <hadess@hadess.net> 644Date: Mon Jan 7 12:36:45 2013 +0100 645 646 build: udev rules must be created when building from git 647 648 https://bugs.freedesktop.org/show_bug.cgi?id=59076 649 650commit 0b2d33c712d6917ed71068188788a258a4dce61e 651Author: Timo Teräs <timo.teras@iki.fi> 652Date: Mon Dec 3 09:57:25 2012 +0200 653 654 imgdev: fix cancelling of enrollment from stage_completed callback 655 656 Re-check device state after fpi_drvcb_enroll_stage_completed(). 657 If enrollment was cancelled after non-completing stage, we must 658 not restart acquire as it would confuse the internal state machine. 659 660 https://bugs.freedesktop.org/show_bug.cgi?id=57829 661 662commit 7751fcb375f81796473074763625e7e948417b4e 663Author: Timo Teräs <timo.teras@iki.fi> 664Date: Mon Dec 3 09:59:30 2012 +0200 665 666 AUTHORS: Update 667 668 https://bugs.freedesktop.org/show_bug.cgi?id=57829 669 670commit 8a87ba448c62eb14fa51d70ce870dd10b67ba76b 671Author: Timo Teräs <timo.teras@iki.fi> 672Date: Mon Dec 3 09:51:27 2012 +0200 673 674 uru4000: fix cancelling of imaging from error callback 675 676 Call error callback before resetting img_transfer to NULL. This 677 variable is internally used to detect if we are still in imaging 678 loop and the call to execute_state_change() needs to be postponed. 679 Since this is the final thing imaging_complete() we can't reset 680 img_transfer until just before this call. 681 682 https://bugs.freedesktop.org/show_bug.cgi?id=57829 683 684commit 7e1646c382bbd6dc21a167bf7f0e45afa5ea217e 685Author: Timo Teräs <timo.teras@iki.fi> 686Date: Tue Dec 4 10:50:30 2012 +0200 687 688 uru4000: fix race condition on waiting power up irq 689 690 It can come before we finish reading the status register on some 691 cases. Arm the irq handler early, and fix the state machine to 692 handle early irq properly. 693 694 https://bugs.freedesktop.org/show_bug.cgi?id=57834 695 696commit a5ec0b30e18275a4fae632c9382d01d5493d5dc6 697Author: Bastien Nocera <hadess@hadess.net> 698Date: Mon Dec 3 16:16:41 2012 +0100 699 700 0.5.0 701 702commit 29cf86a02e8cef9363fd247f40c17710a88bd57e 703Author: Bastien Nocera <hadess@hadess.net> 704Date: Mon Dec 3 16:22:33 2012 +0100 705 706 build: Fix configure-time warning 707 708 Due to missing quotes 709 710commit 7892c943e68478b9ef4efaa83b5795ce1d45311d 711Author: Bastien Nocera <hadess@hadess.net> 712Date: Mon Dec 3 16:18:07 2012 +0100 713 714 build: Create .tar.xz distribution by default 715 716commit a7d6b7c30a7c8de7c1bd306d3766033e1f49397e 717Author: Vasily Khoruzhick <anarsoul@gmail.com> 718Date: Fri Nov 30 15:17:32 2012 +0300 719 720 nbis: Fix crash in gen_initial_maps() 721 722 If the image is too small some coordinates can become negative. 723 Handle this to avoid memory corruption. 724 725 https://bugs.freedesktop.org/show_bug.cgi?id=57730 726 727commit 258ac2d4da1f7a3beecbeb257e85e19965f4fd58 728Author: Bastien Nocera <hadess@hadess.net> 729Date: Sat Dec 1 15:10:13 2012 +0100 730 731 build: Build all the drivers by default 732 733commit 31cf7a93833f99f3eccc70c198538e975e732fac 734Author: Vasily Khoruzhick <anarsoul@gmail.com> 735Date: Fri Nov 30 09:38:57 2012 +0300 736 737 build: Add common routines to the dist 738 739 Fixes dist when aesX660 drivers are disabled from the build. 740 741 https://bugs.freedesktop.org/show_bug.cgi?id=57724 742 743commit f335256cbe9743e405851a001f494a72536e2471 744Author: Vasily Khoruzhick <anarsoul@gmail.com> 745Date: Fri Nov 30 12:26:51 2012 +0300 746 747 build: Don't build unneeded aesX660 routines 748 749 When neither of the AES drivers are built, no need to build the common 750 routines. 751 752 https://bugs.freedesktop.org/show_bug.cgi?id=57724 753 754commit c02cb3083d90099f8993d2531ffa1ff357cfc963 755Author: Vasily Khoruzhick <anarsoul@gmail.com> 756Date: Thu Nov 29 17:16:33 2012 +0300 757 758 AES-drivers: drop redudant type cast in g_slist_free_full call 759 760 https://bugs.freedesktop.org/show_bug.cgi?id=57688 761 762commit 2084724115fd470c2e5f446b25829636fa17e493 763Author: Vasily Khoruzhick <anarsoul@gmail.com> 764Date: Sat Nov 24 16:05:58 2012 +0300 765 766 upektc: add support for Eikon Touch 300 767 768 https://bugs.freedesktop.org/show_bug.cgi?id=45197 769 770commit 59925d20270d2297bdebf14512f8586e2b707822 771Author: Vasily Khoruzhick <anarsoul@gmail.com> 772Date: Sun Nov 25 21:13:01 2012 +0300 773 774 Update AUTHORS file 775 776 https://bugs.freedesktop.org/show_bug.cgi?id=57426 777 778commit 118c610e293653132d891caab361f7d15fdbb007 779Author: Vasily Khoruzhick <anarsoul@gmail.com> 780Date: Sun Nov 25 20:51:32 2012 +0300 781 782 lib: add AES2660 driver 783 784 https://bugs.freedesktop.org/show_bug.cgi?id=57426 785 786commit 313bfede7781540dfad831a520a5ffa254945533 787Author: Vasily Khoruzhick <anarsoul@gmail.com> 788Date: Sun Nov 25 20:50:11 2012 +0300 789 790 lib: add AES1660 driver 791 792 https://bugs.freedesktop.org/show_bug.cgi?id=57426 793 794commit c1da647aedfb604704670d0a0fab4033e067e6d4 795Author: Vasily Khoruzhick <anarsoul@gmail.com> 796Date: Sun Nov 25 20:47:01 2012 +0300 797 798 lib: add AES1660/AES2660 common routines 799 800 https://bugs.freedesktop.org/show_bug.cgi?id=57426 801 802commit 22d204cc68a244baacb079d434fdf227addca272 803Author: Vasily Khoruzhick <anarsoul@gmail.com> 804Date: Thu Nov 22 10:12:38 2012 +0300 805 806 lib: Split fpi_im_resize factor into width/height 807 808 https://bugs.freedesktop.org/show_bug.cgi?id=57426 809 810commit bc497f1b263d6d0e7efafba1568a79b644ecc875 811Author: Vasily Khoruzhick <anarsoul@gmail.com> 812Date: Sun Nov 18 11:30:34 2012 +0300 813 814 aeslib: prevent integer overflow 815 816 AuthenTec devices send 4bpp images, but current code assumes 3bpp for 817 some reason. 818 819 https://bugs.freedesktop.org/show_bug.cgi?id=57426 820 821commit 8c5f2e6434275603da6a5eb21468a43497692575 822Author: Vasily Khoruzhick <anarsoul@gmail.com> 823Date: Sat Nov 17 15:32:25 2012 +0300 824 825 aes: unify image processing code for AuthenTec devices 826 827 Move overlap detection and assembling code into aeslib to prevent 828 code duplication 829 830 https://bugs.freedesktop.org/show_bug.cgi?id=57426 831 832commit 5b20892dd43ca67071b1e113dd3b2c8c2b310168 833Author: Vasily Khoruzhick <anarsoul@gmail.com> 834Date: Thu Nov 22 00:19:59 2012 +0300 835 836 Thanks to Greg and Martin from AuthenTec 837 838 https://bugs.freedesktop.org/show_bug.cgi?id=57426 839 840commit b2a53a459cc4294dec049d8d7f1b92ebb704f983 841Author: Colin Walters <walters@verbum.org> 842Date: Fri Nov 16 13:28:15 2012 -0500 843 844 autogen.sh: Honor NOCONFIGURE=1 845 846 See http://people.gnome.org/~walters/docs/build-api.txt 847 848commit 45ac0eefb0b19a2e37df4065dbf4efeda461edca 849Author: Timo Teräs <timo.teras@iki.fi> 850Date: Fri Nov 16 18:38:52 2012 +0100 851 852 uru4000: Fix image capture error on ARM 853 854 http://www.mail-archive.com/fprint@lists.freedesktop.org/msg00294.html 855 856commit ea6d5ba6d65aa7ed4b900b23263bc4d8abe4a868 857Author: Vasily Khoruzhick <anarsoul@gmail.com> 858Date: Sat Nov 10 18:37:41 2012 +0300 859 860 lib: Fix mess with driver IDs 861 862 ID is just a some magic number to make fingerprint from one scanner model 863 incompatible with another scanner model. Get rid of "magic", declare enum 864 and use it. 865 866 https://bugs.freedesktop.org/show_bug.cgi?id=56956 867 868commit ed2c75842a4542b0dbd9e4f188a40057bb537ea6 869Author: Vasily Khoruzhick <anarsoul@gmail.com> 870Date: Sat Nov 10 18:22:32 2012 +0300 871 872 upektc: Port to asynchronous model 873 874 https://bugs.freedesktop.org/show_bug.cgi?id=56955 875 876commit b307dd1a6abef8f87062d35dcbe2f2c22b959638 877Author: Vasily Khoruzhick <anarsoul@gmail.com> 878Date: Sat Nov 10 18:21:19 2012 +0300 879 880 aes2550: fix bug introduced by last commit 881 882 ssm->priv is not aesdev, but fp_img_dev 883 884 https://bugs.freedesktop.org/show_bug.cgi?id=56782 885 886commit 39902374ce0f7ca21cf3412a7295ad82198ef053 887Author: Vasily Khoruzhick <anarsoul@gmail.com> 888Date: Sat Nov 3 18:30:43 2012 +0300 889 890 aes2550: Harden against low finger pressure 891 892 Stop scan only after 3rd heartbeat message 893 894 https://bugs.freedesktop.org/show_bug.cgi?id=56782 895 896commit 9e92d4cf2c8ae4fad0b49576b0b6f8f4e5f2daa1 897Author: Bastien Nocera <hadess@hadess.net> 898Date: Tue Nov 6 09:30:40 2012 +0100 899 900 examples: Fix compile-time warnings 901 902 A few signedness problems. 903 904commit 6b84c6664fe9da98e013caab99145d8f0279b9bb 905Author: Vasily Khoruzhick <anarsoul@gmail.com> 906Date: Mon Nov 5 22:39:49 2012 +0300 907 908 AES1610: driver cleanup, part #2 909 910 Remove dead code 911 912 https://bugs.freedesktop.org/show_bug.cgi?id=56439 913 914commit f569d0bf4479c4d6ebf1972b59cd7450ec9a3258 915Author: Vasily Khoruzhick <anarsoul@gmail.com> 916Date: Sat Oct 20 12:26:38 2012 +0300 917 918 AES1610: driver cleanup, part #1 919 920 Fix warnings reported by gcc 921 922 https://bugs.freedesktop.org/show_bug.cgi?id=56439 923 924commit 1cfd14b7fe8f9e68c5bd74461bcd17e25b9f1d15 925Author: Vasily Khoruzhick <anarsoul@gmail.com> 926Date: Fri Oct 26 16:40:53 2012 +0300 927 928 nbis: fix rest of warnings reported by gcc 929 930 https://bugs.freedesktop.org/show_bug.cgi?id=56439 931 932commit 5d32102efefe955e9f5bdbf5809a5fb637d77e89 933Author: Vasily Khoruzhick <anarsoul@gmail.com> 934Date: Fri Oct 26 16:36:21 2012 +0300 935 936 nbis: prefix global variables with "g_" 937 938 It fixes a lot of warnings about shadowing global variable 939 940 https://bugs.freedesktop.org/show_bug.cgi?id=56439 941 942commit 9e10edd422e5130e210e17d90c5db9135029f165 943Author: Bastien Nocera <hadess@hadess.net> 944Date: Mon Nov 5 19:52:46 2012 +0100 945 946 aes2501: Add year 2008 to the copyright. 947 948 https://bugs.freedesktop.org/show_bug.cgi?id=56715#c7 949 950commit fecf6d6fe5d17f4d0cb5cc8e19d0f51c58fcc712 951Author: Vasily Khoruzhick <anarsoul@gmail.com> 952Date: Sat Nov 3 16:29:51 2012 +0300 953 954 aes2501: Update copyright line 955 956 https://bugs.freedesktop.org/show_bug.cgi?id=56715 957 958commit e32fa8cc3825a6ce5db558094e81134d433293e3 959Author: Vasily Khoruzhick <anarsoul@gmail.com> 960Date: Sat Nov 3 16:13:05 2012 +0300 961 962 aes2501: Improve image contrast 963 964 When scanning, check the histogram sum, and increase the 965 ADREFHI register value if the sum is too low, or decrease 966 it if it's too high. 967 968 https://bugs.freedesktop.org/show_bug.cgi?id=56715 969 970commit d8aae30a672485de747e5a58ff7520128b718515 971Author: Vasily Khoruzhick <anarsoul@gmail.com> 972Date: Sat Nov 3 16:08:04 2012 +0300 973 974 aes2501: Harden against low finger pressure 975 976 Wait for 3 empty frames before stopping the scan 977 978 The driver used to stop the scan immediately after an empty frame 979 (by checking for hist sum == 0), but it is possible to get empty 980 frames in the middle of the scan due to low finger pressure. 981 Waiting for 3 empty frames stop the driver wrongly aborting the scan 982 too early. 983 984 https://bugs.freedesktop.org/show_bug.cgi?id=56715 985 986commit 84b97ea15b519ac26af3a71e2a4d929cb140f0ff 987Author: Vasily Khoruzhick <anarsoul@gmail.com> 988Date: Fri Oct 26 16:44:37 2012 +0300 989 990 lib: Fix warning when debug logs are disabled 991 992 Fix warning in sync.c when debug logs are not enabled 993 994 https://bugs.freedesktop.org/show_bug.cgi?id=56439 995 996commit 8f987438573d38581039a98a1178a7fd25bba4ad 997Author: Vasily Khoruzhick <anarsoul@gmail.com> 998Date: Sat Oct 20 12:28:31 2012 +0300 999 1000 upeksonly: Fix build-time warnings 1001 1002 Fix warnings reported by gcc 1003 1004 https://bugs.freedesktop.org/show_bug.cgi?id=56439 1005 1006commit be29f27e254317b713c7a0f007d6e959ee2a3905 1007Author: Vasily Khoruzhick <anarsoul@gmail.com> 1008Date: Sat Oct 20 12:27:52 2012 +0300 1009 1010 upeke2: Fix build-time warnings 1011 1012 Fix warnings reported by gcc 1013 1014 https://bugs.freedesktop.org/show_bug.cgi?id=56439 1015 1016commit d003f08855ba34be88775e4281e7f686b00f6d2f 1017Author: Timo Teräs <timo.teras@iki.fi> 1018Date: Wed Sep 26 13:41:13 2012 +0300 1019 1020 lib: Add support for DigitalPersona URU4500 1021 1022 By adding native encryption support, rather than poking at 1023 the firmware to disable it. 1024 1025 This also makes the URU4000B use native encryption. 1026 1027 https://bugs.freedesktop.org/show_bug.cgi?id=55351 1028 1029commit 0f7ad00fc49023ff7f637eb740e7120d0e16b47e 1030Author: Bastien Nocera <hadess@hadess.net> 1031Date: Mon Oct 15 11:46:03 2012 +0200 1032 1033 HACKING: Update with bugzilla link 1034 1035commit 83333bce3f06ca55651a7516666ee63a5892f09f 1036Author: Vasily Khoruzhick <anarsoul@gmail.com> 1037Date: Tue Oct 9 01:20:44 2012 +0300 1038 1039 lib: Add AES2550/AES2810 driver 1040 1041 Initial implementation of AES2550/AES2810 driver. 1042 Does not support AES2810 crypto engine and external flash. 1043 1044 https://bugs.freedesktop.org/show_bug.cgi?id=50859 1045 1046commit 132b17830492187b80c8442b1e6e981343b1d27e 1047Author: Bastien Nocera <hadess@hadess.net> 1048Date: Tue Aug 28 20:02:17 2012 +0100 1049 1050 lib: Add driver names to the udev rules 1051 1052commit c3689665dbf3cd2126f3729792aa71efe13aab8d 1053Author: Andrej Krutak <dev@andree.sk> 1054Date: Sun Jul 22 21:44:34 2012 +0200 1055 1056 lib: Add VFS300/VFS301 driver 1057 1058 https://bugs.freedesktop.org/show_bug.cgi?id=37574 1059 1060commit 3d222ddda7f6acfadd40fcd06b3ab40192829d5a 1061Author: Bastien Nocera <hadess@hadess.net> 1062Date: Tue Aug 28 19:42:05 2012 +0100 1063 1064 lib: Fix whitespace errors 1065 1066commit 080c414ce9586d7b4b92e786705e9b716b91aa33 1067Author: Bastien Nocera <hadess@hadess.net> 1068Date: Tue Aug 28 19:41:41 2012 +0100 1069 1070 build: Whitespace fixes in configure.ac 1071 1072commit ff02115b0fa8b72a1d8c07b8e2c9ad7e90459427 1073Author: Vasily Khoruzhick <anarsoul@gmail.com> 1074Date: Mon Aug 13 20:41:42 2012 +0300 1075 1076 lib: Fix global variables collision with libusb 1077 1078commit 3b409c767cb5c0c27229994ac47edf47c1c59804 1079Author: Bastien Nocera <hadess@hadess.net> 1080Date: Tue Aug 14 15:19:35 2012 +0100 1081 1082 build: Allow disabling the update of udev rules 1083 1084 We can only create the udev rules file when we're not cross-compiling, 1085 so allow people to disable updating the file when building from 1086 tarballs. 1087 1088 If you run with --enable-udev-rules=no and the 1089 60-fprint-autosuspend.rules doesn't exist, you're on your own. 1090 1091commit d05c69698c6eb1aadcd5a7cbdb883b7accb35eb2 1092Author: Bastien Nocera <hadess@hadess.net> 1093Date: Tue Aug 14 15:12:20 2012 +0100 1094 1095 build: Fix distcheck'ing not working 1096 1097 We need to be able to override the udev rules directory for 1098 distchecking to work. 1099 1100commit 2d09b10a2795528134fb0699903160545e153188 1101Author: Bastien Nocera <hadess@hadess.net> 1102Date: Tue Aug 14 14:52:58 2012 +0100 1103 1104 build: Include udev rules in tarball 1105 1106commit bc03d5618619c3236cd61df77342ede97186622a 1107Author: Bastien Nocera <hadess@hadess.net> 1108Date: Tue Aug 14 14:51:23 2012 +0100 1109 1110 build: Add list of drivers to include when distchecking 1111 1112commit 96e7224a236d0a7d515fde5593c77cf6eb8fd942 1113Author: Bastien Nocera <hadess@hadess.net> 1114Date: Tue Aug 14 14:48:43 2012 +0100 1115 1116 build: Fix distchecking of examples directory 1117 1118commit 0f4a75ff61cc156b6cabdb1f6c1e22ea314a9aff 1119Author: Benedikt Morbach <benedikt.morbach@googlemail.com> 1120Date: Mon May 28 14:13:03 2012 +0200 1121 1122 update udev rules dir location again 1123 1124 Instead of hardcoding whatever the new fashion is every few months, which might 1125 even differ between distributions, just put the rules where udev says we should. 1126 1127 https://bugs.freedesktop.org/show_bug.cgi?id=50426 1128 1129commit dfff16f3e32519ad071bdb1233c75bfe3ffa5557 1130Author: Bastien Nocera <hadess@hadess.net> 1131Date: Thu Feb 2 05:28:17 2012 +0000 1132 1133 Don't print duplicate udev rules 1134 1135 https://bugs.freedesktop.org/show_bug.cgi?id=45513 1136 1137commit 3d2e54526454ed1436b6436dffb0e8769fdbff1d 1138Author: Patrick Marlier <patrick.marlier@gmail.com> 1139Date: Wed Jan 18 18:52:02 2012 -0500 1140 1141 Fix possible crash without a deactivate callback 1142 1143 We were checking for ->activate existing instead of ->deactivate. 1144 1145commit c96fa32da4a3d93950039fdc176571376dafc8cb 1146Author: Benedikt Morbach <benedikt.morbach@googlemail.com> 1147Date: Thu Jan 5 21:49:13 2012 +0100 1148 1149 Move udev rules to libdir 1150 1151 This moves the dir from /etc/udev/rules.d to /lib/udev/rules.d, 1152 which is encouraged by upstream, as /etc is for user overrides only 1153 1154 https://bugs.freedesktop.org/show_bug.cgi?id=44507 1155 1156commit 48ec64f68330452b011d14be67653f1d48fb1405 1157Author: Matthew Garrett <mjg@redhat.com> 1158Date: Tue Oct 4 15:11:08 2011 +0100 1159 1160 Fix udev control path 1161 1162 Newer kernels uses power/control for USB suspend setup, not power/level. 1163 Fix the udev rules generation. 1164 1165commit f8aa82a554a0862eb9b83869d42dbefbde666543 1166Author: Bastien Nocera <hadess@hadess.net> 1167Date: Mon Apr 18 18:02:37 2011 +0100 1168 1169 0.4.0 1170 1171commit 1451a2dde2ca0651ec3e5711ee6278d89f3cc32a 1172Author: Bastien Nocera <hadess@hadess.net> 1173Date: Fri Apr 8 15:38:33 2011 +0100 1174 1175 build: Quiet by default 1176 1177commit d2e957683f5b9bd4781e2216c259e9b10fa3508d 1178Author: Sergio Cerlesi <sergio.cerlesi@gmail.com> 1179Date: Wed Mar 30 14:20:22 2011 +0200 1180 1181 Added support for Validity VFS101 1182 1183 USB ID 138a:0001 1184 1185commit a3ae96c2143ee7837225772008872ed0627336e1 1186Author: Sergio Cerlesi <sergio.cerlesi@gmail.com> 1187Date: Tue Feb 22 14:39:11 2011 +0100 1188 1189 Fix function fpi_im_resize on create new image 1190 1191 The function fpi_im_resize copy the new resized image into old fp_img 1192 instead of newimg so, new image is empty and often it generated a 1193 segmentation fault error. 1194 1195commit 3dd905d4e92962a970b877285fa6f39ed997a2ea 1196Author: Sergio Cerlesi <sergio.cerlesi@gmail.com> 1197Date: Wed Feb 16 09:34:55 2011 +0100 1198 1199 Fix return timeout of fp_get_next_timeout 1200 1201 On function fp_get_next_timeout if exist fprint timeout or libusb 1202 timeout the function return the smaller of they. 1203 1204 But if one of that not exist and have a smaller value the function 1205 return a timeout that not exist. 1206 1207commit aab031ce371e690eb550e8414282180190880ae3 1208Author: Bastien Nocera <hadess@hadess.net> 1209Date: Wed Sep 8 11:11:31 2010 +0100 1210 1211 0.3.0 1212 1213commit 6bfc516553b8db1ed3a3d275eb3ff3f0bde24b48 1214Author: Bastien Nocera <hadess@hadess.net> 1215Date: Wed Sep 8 11:03:58 2010 +0100 1216 1217 Add test program for C++ support 1218 1219 So that we don't regress 1220 1221commit ff842125c59ec87028bf53618b0a6acb310702e0 1222Author: Kunal Gangakhedkar <kunal.gangakhedkar@gmail.com> 1223Date: Wed Sep 8 12:13:17 2010 +0530 1224 1225 Add C++ guards to take care of C++ name-mangling. 1226 1227 The libfprint library is generally built with C calling conventions. 1228 Which makes it difficult to link to C++ programs as it is. 1229 1230 This patch adds the support for linking with C++ code - by telling the C++ 1231 compiler to use C calling conventions for libfprint functions. 1232 1233 Reported-by: Guus Ellenkamp <guus@activediscovery.net> 1234 Signed-off-by: Kunal Gangakhedkar <kunal.gangakhedkar@gmail.com> 1235 1236commit 712ebb6d44c421543a999227c92d6e08003c7a96 1237Author: Hugo Grostabussiat <dw23.devel@gmail.com> 1238Date: Tue Aug 31 16:57:26 2010 +0200 1239 1240 Namespace functions and structs for 147e:2016 1241 1242 Inserted "_2016" in structures, enums and functions names 1243 that are related with device 147e:2016. 1244 1245commit a18f318446df01d27f3a8dc677c2a644fb086a60 1246Author: Hugo Grostabussiat <dw23.devel@gmail.com> 1247Date: Tue Aug 31 16:09:22 2010 +0200 1248 1249 Added support for UPEK TCS4C (USB ID 147e:1000) 1250 1251 https://bugs.freedesktop.org/show_bug.cgi?id=29719 1252 1253commit 15b8e8f37617306cbfb03679475bb714a4f28a73 1254Author: Bastien Nocera <hadess@hadess.net> 1255Date: Wed Aug 25 11:20:55 2010 +0100 1256 1257 Remove HAL fdi generation 1258 1259 HAL is deprecated, and the rules are broken when multiple 1260 drivers use the same USB IDs. 1261 1262commit 9f8dd29a553ce61999cd0cee8c81013cbeb07926 1263Author: Bastien Nocera <hadess@hadess.net> 1264Date: Wed Aug 25 11:17:42 2010 +0100 1265 1266 Make sure all sources are dist'ed 1267 1268 So that you can compile against ImageMagick even if the 1269 distter doesn't. 1270 1271commit f28c5797791edd8b17f05ed3671b45223326dd21 1272Author: Bastien Nocera <hadess@hadess.net> 1273Date: Wed Aug 25 11:17:19 2010 +0100 1274 1275 Print out summary of build options 1276 1277 At the end of the configure run. 1278 1279commit 375575b701fd1a1187b75c472ea77fb1e89ee7a2 1280Author: Bastien Nocera <hadess@hadess.net> 1281Date: Wed Aug 25 00:42:01 2010 +0100 1282 1283 Port OpenSSL using code to NSS 1284 1285 As OpenSSL cannot really be used in an LGPLv2+ piece of software 1286 without an exception. 1287 1288 Adapted from the example code at: 1289 http://www.mozilla.org/projects/security/pki/nss/sample-code/sample2.html 1290 and: 1291 http://github.com/bagder/curl/commit/f3b77e5 1292 1293commit 25161286f5c9bde3a03eb181e8c4db8029f94001 1294Author: Alexia Death <alexiadeath@gmail.com> 1295Date: Sat Dec 26 17:02:03 2009 +0200 1296 1297 Only complete scan with MIN_ROWS in print 1298 1299 Don't consider the scan complete unless theres at least 1300 MIN_ROWS recorded or very long blank read occurred. 1301 1302 Typical problem spot: one brief touch before starting the 1303 actual scan. Happens most commonly if scan is started from 1304 before the first joint resulting in a gap after the inital touch. 1305 1306 http://lists.reactivated.net/pipermail/fprint/2009-December/001406.html 1307 1308commit c575afba9ae9c518f208d5834e2ee72f8aede72c 1309Author: Alexia Death <alexiadeath@gmail.com> 1310Date: Fri Dec 25 00:04:55 2009 +0200 1311 1312 Make the +2 right shift happen on image handoff. 1313 1314 Rather than when reading the rows. Doing it there seems a lot 1315 more sensible and does not get in the way of minimizing the 1316 impact of lost USB packets. 1317 1318 http://lists.reactivated.net/pipermail/fprint/2009-December/001404.html 1319 1320commit 8cd31f6ad81bc2d8e8d3ffdea7cb4f13a6914684 1321Author: Alexia Death <alexiadeath@gmail.com> 1322Date: Thu Dec 24 21:42:01 2009 +0200 1323 1324 Try to correct missing packets in image stream 1325 1326 If a packet goes missing, use the data from previous row 1327 to minimize distortion, or use some neutral dummy data. 1328 1329 http://lists.reactivated.net/pipermail/fprint/2009-December/001404.html 1330 1331commit 4b518bd230ef0f3a26c8eb89736815ed9334b6bb 1332Author: Alexia Death <alexiadeath@gmail.com> 1333Date: Sat Dec 26 01:00:59 2009 +0200 1334 1335 Fix a segfault if a scan was shorter than 8 lines. 1336 1337 This segfault was caused by proceeding to free the conversion result 1338 without checking for an error condition in image to print 1339 data conversion. 1340 1341 http://lists.reactivated.net/pipermail/fprint/2009-December/001405.html 1342 1343commit daee067c4ffabff9836617b90c547873130087f1 1344Author: Bastien Nocera <hadess@hadess.net> 1345Date: Thu Aug 19 14:43:30 2010 +0100 1346 1347 Release 0.2.0 1348 1349commit 9608ab370763b54c778b27c5b2015d1dda2ecec0 1350Author: Bastien Nocera <hadess@hadess.net> 1351Date: Wed Aug 18 11:17:20 2010 +0100 1352 1353 Don't crash when no devices are present 1354 1355commit 73561989dd485560e4c62b46a9af31f9fce1bc4e 1356Author: Bastien Nocera <hadess@hadess.net> 1357Date: Tue Aug 17 23:41:23 2010 +0100 1358 1359 Implement discover() methods for upeke2 and upeksonly 1360 1361 upeksonly handles revision 1 devices, upeke2 revision 2 devices. 1362 1363 https://bugzilla.redhat.com/show_bug.cgi?id=504399 1364 1365commit df4a6f53d2b74f43d453940cebd43d0cd802584d 1366Author: Bastien Nocera <hadess@hadess.net> 1367Date: Tue Aug 17 23:14:18 2010 +0100 1368 1369 Rework "discover" code for drivers 1370 1371 Check all the drivers for one that'll drive our device in 1372 question, and prefer ones that have a discover() method that runs 1373 successfully to a driver without such a method. 1374 1375 This allow drivers to both reject devices (if 2 drivers handle 1376 the same USB IDs), or a single driver to handle variants of the 1377 same device (through the devtype out value of the discover method). 1378 1379commit 48af46a2c504b43854a3877d962a301921ef7f86 1380Author: Jorge Suarez de Lis <yo@jorgesuarezdelis.name> 1381Date: Tue Aug 17 19:55:48 2010 +0100 1382 1383 upeke2: Add support for 0x2e status code 1384 1385 As reported working in: 1386 http://www.mail-archive.com/fprint@reactivated.net/msg01426.html 1387 1388commit 5a1926bb416fa4e5a38fa10b9ac3da04ce9f3f72 1389Author: Jorge Suarez de Lis <yo@jorgesuarezdelis.name> 1390Date: Tue Aug 17 19:36:56 2010 +0100 1391 1392 Add (disabled) UPEK EikonII (TCRD4C) driver 1393 1394 The device driver is always disabled as it uses the same Device ID 1395 as another working reader: 1396 https://bugzilla.redhat.com/show_bug.cgi?id=504399#c38 1397 1398commit 7c7402a697eb7f4d1c2c929922cbd88945d2616f 1399Author: Guido Grazioli <guido.grazioli@gmail.com> 1400Date: Tue Aug 17 11:36:34 2010 +0100 1401 1402 Finish implementation of AES1610 driver 1403 1404 Ported to the new internal libfprint API. 1405 1406 https://bugzilla.redhat.com/show_bug.cgi?id=499732 1407 1408commit a99e813bbf0cfaa454feda47d11debb98ce6de39 1409Author: Matthew Garrett <mjg@redhat.com> 1410Date: Thu Oct 1 15:08:41 2009 +0100 1411 1412 Add udev rules to set devices to autosuspend 1413 1414 Most fingerprint readers can be suspended safely, so use 1415 the udev rules to autosuspend them when they appear. 1416 1417 The script itself contains whitelists and blacklists. 1418 1419commit 014b933b2bf635dab4ccbb9beb15be8d55a9d427 1420Author: Bastien Nocera <hadess@hadess.net> 1421Date: Wed Oct 29 12:53:24 2008 +0000 1422 1423 Add gdk-pixbuf support 1424 1425 When imaging support is required, we prefer to use gdk-pixbuf, 1426 as it's already on things like Live CDs. 1427 Also fix the examples building against the system libfprint. 1428 1429commit 911c6bf102760ba49bcc2f851a7bd7d9cebff46e 1430Author: Daniel Drake <dsd@gentoo.org> 1431Date: Sat Jun 20 14:50:35 2009 +0100 1432 1433 v0.1.0-pre2 release 1434 1435commit a51763bd9993da7de78f19701cd62c7b7d147bf4 1436Author: Daniel Drake <dsd@gentoo.org> 1437Date: Sat Jun 20 14:50:23 2009 +0100 1438 1439 New sourceforge file upload system 1440 1441commit e5da34e80a1cc6f5af480cc5343dd48d52c60a96 1442Author: Bastien Nocera <hadess@hadess.net> 1443Date: Tue Nov 25 15:54:56 2008 +0000 1444 1445 Fix fp_get_pollfds() 1446 1447 It was increasing i too early, and writing past the allocated 1448 structures. 1449 1450commit d816fed5c384e6d45ae11e2530883d54aacedf4b 1451Author: Bastien Nocera <hadess@hadess.net> 1452Date: Tue Nov 25 15:54:25 2008 +0000 1453 1454 Disable finger_num_to_str when debugging is disabled 1455 1456 One less warning. 1457 1458commit 749bb3e30ebfc08be26bd849753a0e169d46b504 1459Author: Bastien Nocera <hadess@hadess.net> 1460Date: Mon Nov 24 15:31:21 2008 +0000 1461 1462 Export scan type through HAL 1463 1464 Also export the scan type through HAL. 1465 1466commit d4822283256e153b1d8b61c4126617abcf291f43 1467Author: Daniel Drake <dsd@gentoo.org> 1468Date: Thu Nov 20 19:59:43 2008 +0000 1469 1470 upekts: Don't choke on non-zero bytes in command response 1471 1472 Alon Bar-Lev's device behaves differently, but seems to work fine with 1473 upekts even if we ignore the difference. 1474 1475commit 7c44c86d4c8838a69d93225bbe3e31f4e7427445 1476Author: Daniel Drake <dsd@gentoo.org> 1477Date: Thu Nov 20 17:23:04 2008 +0000 1478 1479 Logging control 1480 1481 Don't log anything by default, and let the applications change the log 1482 level. 1483 1484commit 29044d9ca95646596c8c817d470c01381f4b8f9f 1485Author: Bastien Nocera <hadess@hadess.net> 1486Date: Thu Nov 20 14:40:01 2008 +0000 1487 1488 Export scan type for each driver 1489 1490 For each driver, export the scan type supported by the devices it 1491 handles. This allows front-ends to direct the user better. 1492 1493commit 1fd247ecd7fb000260f2a457e264dcb6f1d09d98 1494Author: Daniel Drake <dsd@gentoo.org> 1495Date: Tue Nov 18 17:10:49 2008 +0000 1496 1497 v0.1.0-pre1 release 1498 1499 because Bastien asked nicely. No guarantees about API or stability. 1500 1501commit b93f038dbc35b8b8875a26d01811dd4e3dabeac4 1502Author: Daniel Drake <dsd@gentoo.org> 1503Date: Tue Nov 18 17:18:05 2008 +0000 1504 1505 distcheck fix for fdi file 1506 1507commit 2d30e40ed750d4ccd2a1eea6b2398bbf3275b39f 1508Author: Daniel Drake <dsd@gentoo.org> 1509Date: Sun Nov 2 14:49:39 2008 +0000 1510 1511 aes2501: Add AES2500 ID 1512 1513 Reported to work with Motion Computing M1400 tablet by Daniel Hazelbaker 1514 and Alex Van Deusen. 1515 1516commit 7c83b6d825f27b1afb7e61dfed7a4ead2e506699 1517Author: Bastien Nocera <hadess@hadess.net> 1518Date: Thu Oct 30 15:02:52 2008 +0000 1519 1520 Tag fingerprint devices in HAL 1521 1522 Tag all the fprint supported devices in HAL. Patch from 1523 myself and Timo Hoenig <thoenig@suse.de>. 1524 1525commit 7672c43cea603f55e04eceab1d85b9a850400af4 1526Author: Daniel Drake <dsd@gentoo.org> 1527Date: Sun Nov 2 11:03:32 2008 +0000 1528 1529 distcheck "" fix 1530 1531 Pointed out by Bastien Nocera 1532 1533commit d94ffa22d5c2e05c4fe1a296720d3c01151983bf 1534Author: Gordon Stratton <gordon.stratton@gmail.com> 1535Date: Fri Oct 31 23:36:55 2008 +0000 1536 1537 Missing parameter in fp_err call 1538 1539commit 6b8b17f575f5932ebe865d1db93a2807a49b961b 1540Author: Daniel Drake <dsd@gentoo.org> 1541Date: Sun Aug 10 14:34:44 2008 -0500 1542 1543 Split imagemagick hack from main library 1544 1545 Cleans up the conditional compilation system 1546 1547commit bba1c1085fa678e856a3b1e482d68dcc81664d11 1548Author: Pavel Herrman <morpheus.ibis@gmail.com> 1549Date: Sun Aug 10 14:21:58 2008 -0500 1550 1551 Allow drivers to be included/excluded at compile time 1552 1553commit b6dabaacc8d27c7f3ea54fcce7bf0d3ccc0f97c3 1554Author: Daniel Drake <dsd@gentoo.org> 1555Date: Wed Jul 9 03:25:23 2008 +0100 1556 1557 Enable libusb debug messages 1558 1559 While we're a noisy library we might as well ask libusb to be as well 1560 1561commit 3915cb262269a32844b3672d8bc3edc188b2f402 1562Author: Daniel Drake <dsd@gentoo.org> 1563Date: Wed Jul 9 03:23:21 2008 +0100 1564 1565 Add upeksonly driver for UPEK TouchStrip sensor-only devices 1566 1567 The long awaited 147e:2016 reverse engineering efforts. 1568 Image quality is OK, not brilliant - a little fuzzy, but it works 1569 quite well. 1570 1571commit 59c62c7d744a9b5e4c7355932889cd0342044e09 1572Author: Daniel Drake <dsd@gentoo.org> 1573Date: Wed Jul 2 23:27:37 2008 -0500 1574 1575 Update for libusb-0.9.1 1576 1577commit e1691a29dd21475264134c42d929d90b3d26ceca 1578Author: Daniel Drake <dsd@gentoo.org> 1579Date: Sun Jun 15 10:54:07 2008 -0500 1580 1581 update for new libusb_open API 1582 1583commit aa7c8ae81eb3eae1afbbae982a6617db8e79e29d 1584Author: Vasily Khoruzhick <anarsoul@gmail.com> 1585Date: Tue May 13 01:24:05 2008 +0300 1586 1587 Adapting to latest libusb-1.0 changes 1588 1589 Signed-off-by: Vasily Khoruzhick <anarsoul@gmail.com> 1590 [dsd: remove config const qualifier and add freeing of config] 1591 1592commit 63421e3fd61d0779a5cc4c11a59ea7318cdb8f72 1593Author: Daniel Drake <dsd@cs.manchester.ac.uk> 1594Date: Wed May 7 23:46:06 2008 +0100 1595 1596 uru4000: fix compile against latest libusb 1597 1598commit 76cbf212239e5561a11877a2a3a486844107ec2e 1599Author: Daniel Drake <dsd@cs.manchester.ac.uk> 1600Date: Fri May 2 11:55:50 2008 +0100 1601 1602 uru4000: Add support for Digital Persona U.are.U Fingerprint Keyboard 1603 1604 Reported by Tudor Panaitescu 1605 1606commit f540ae5d5bc6ef6af3016759f6c9ca26056bec7c 1607Author: Daniel Drake <dsd@cs.manchester.ac.uk> 1608Date: Sat Apr 19 23:09:53 2008 +0100 1609 1610 Fix memory leak in fpi_print_data_new() 1611 1612 Pointed out by Krystian (yoB) on FS#24 1613 1614commit ebcabc1aded5043e0f5abf08e97ae14cc9ff0a60 1615Author: Daniel Drake <dsd@gentoo.org> 1616Date: Sat Apr 12 16:58:49 2008 +0100 1617 1618 uru4000: avoid cancelling a completed transfer 1619 1620commit 870ddfb9054d00341ebf87d1e39153feaa96cbc3 1621Author: Daniel Drake <dsd@cs.manchester.ac.uk> 1622Date: Sun Apr 6 23:55:41 2008 +0100 1623 1624 some libusb pointers are now const 1625 1626commit 40e7ac195a4be0e6951b7a8b0388bdbc88cfc8d7 1627Author: Vasily Khoruzhick <anarsoul@gmail.com> 1628Date: Sun Apr 6 21:56:33 2008 +0300 1629 1630 Adapting to latest libusb 1631 1632 Signed-off-by: Vasily Khoruzhick <anarsoul@gmail.com> 1633 1634commit 1ab52617c7d857acb01368938b713595af078b4e 1635Author: Azzie <azzie@astronet.pl> 1636Date: Sat Mar 29 18:11:17 2008 +0000 1637 1638 Update for latest libusb constant names 1639 1640commit 981f2eb0aa133669523938ac17f095d82777fd33 1641Author: Daniel Drake <dsd@gentoo.org> 1642Date: Thu Mar 20 15:31:03 2008 +0000 1643 1644 vcom5s: implement finger detection 1645 1646 Uses thresholding of average pixel intensity of the center region of 1647 the image. 1648 1649 Also document guessed meanings of magic numbers for registers. 1650 1651commit d6d4e56ccffc8e0b704f4778058f677ad17c2439 1652Author: Daniel Drake <dsd@gentoo.org> 1653Date: Thu Mar 20 13:34:48 2008 +0000 1654 1655 uru4000: add fwenc search location for UareU4000 1656 1657 ticket #14 includes a device with a different firmware, this should get 1658 it up and running. 1659 1660commit d12ad6dea905758c76558ee66a48bb0fd3c485f3 1661Author: Daniel Drake <dsd@gentoo.org> 1662Date: Thu Mar 20 13:19:44 2008 +0000 1663 1664 Update version number 1665 1666 I'm about to do another 0.0 release, so change the tag in this branch to 1667 avoid confusion 1668 1669commit c9ff2c5541a522224137beb1fd4ecef453ba8955 1670Author: Daniel Drake <dsd@gentoo.org> 1671Date: Wed Mar 19 23:29:30 2008 +0000 1672 1673 Add Veridicom 5th Sense driver 1674 1675 Based on bus traffic analysis of the windows driver. This device is 1676 working well, but finger presence detection is not yet implemented. 1677 1678commit 617ca8cd1567091263f77fb37d422290c920c72e 1679Author: Daniel Drake <dsd@gentoo.org> 1680Date: Wed Mar 19 23:28:44 2008 +0000 1681 1682 fix imaging action state for repeated enroll attempts 1683 1684 Enroll attempts may retry in some situations, so we need to reset the 1685 state to accept more scans. 1686 1687commit a11ac7d8f8e1dc1f04b4721247ef24ca4cd8f7d6 1688Author: Daniel Drake <dsd@gentoo.org> 1689Date: Wed Mar 19 17:46:59 2008 +0000 1690 1691 uru4000: search for encryption byte 1692 1693 New DigitalPersona devices have a new firmware. We now use a search 1694 scheme to find the encryption byte, by examining known locations 1695 in the firmware until we find something that we recognise. 1696 1697commit fb49c24ae24bc296bb123f00bda1f80f50587eaf 1698Author: Daniel Drake <dsd@gentoo.org> 1699Date: Wed Mar 19 17:25:53 2008 +0000 1700 1701 uru4000: Unify register I/O 1702 1703commit d3d68be49033f17c9933674d6edc52aea09d9b60 1704Author: Daniel Drake <dsd@gentoo.org> 1705Date: Wed Mar 19 16:19:07 2008 +0000 1706 1707 uru4000: fix register widths 1708 1709 Encryption byte address was being truncated in the sm_fix_fw_read_cb 1710 path 1711 1712commit 156e5e7d7c691c32e56eecfb0b272d3dbdf6e0fd 1713Author: Daniel Drake <dsd@cs.manchester.ac.uk> 1714Date: Thu Mar 13 12:54:27 2008 +0000 1715 1716 Update for new libusb_get_pollfds API 1717 1718commit f546fa2555ee180f4fef9a4bb2cf3a8f15edaa35 1719Author: Daniel Drake <dsd@cs.manchester.ac.uk> 1720Date: Mon Mar 10 13:33:32 2008 +0000 1721 1722 Fix synchronous enrollment for multi-stage devices 1723 1724commit 4e884807a793091e3410ea0ba10e3e314e21818f 1725Author: Daniel Drake <dsd@cs.manchester.ac.uk> 1726Date: Mon Mar 10 11:35:58 2008 +0000 1727 1728 Update for libusb API rework 1729 1730commit 81a5d6d9664ef9084213cb536218b31ab46d5cd9 1731Author: Daniel Drake <dsd@cs.manchester.ac.uk> 1732Date: Fri Mar 7 00:11:56 2008 +0000 1733 1734 Only use -fgnu89-inline when supported 1735 1736 This only works for GCC 4.3 and newer. GCC 4.2 and older already have 1737 the semantics introduced by using this argument. 1738 1739commit 647d1408343b5a14ee70984865535aa5d02019a4 1740Author: Vasily Khoruzhick <anarsoul@gmail.com> 1741Date: Wed Mar 5 22:15:32 2008 +0200 1742 1743 Fixed fp_identify_finger_img didn't return result of operation 1744 1745 Signed-off-by: Vasily Khoruzhick <anarsoul@gmail.com> 1746 1747commit 07a42d5f91c072bcb4639744c1e79d03cfd9f9dc 1748Author: Daniel Drake <dsd@cs.manchester.ac.uk> 1749Date: Mon Mar 3 23:24:00 2008 +0000 1750 1751 uru4000: Retry initialization 3 times 1752 1753 Sometimes we don't get the powerup interrupt, so retry the whole process 1754 up to 3 times if we time out waiting for the interrupt. 1755 1756commit 2d27df0268002eb55d41fbf384d200e7b0418ad2 1757Author: Daniel Drake <dsd@cs.manchester.ac.uk> 1758Date: Mon Mar 3 23:23:50 2008 +0000 1759 1760 Allow timeouts to be cancelled 1761 1762commit c29c3fa01493bd03e218cf846adb90932f4c3c04 1763Author: Daniel Drake <dsd@cs.manchester.ac.uk> 1764Date: Mon Mar 3 13:47:59 2008 +0000 1765 1766 Fix low minutiae handling 1767 1768 Fixes a crash and adds in a missing break statement 1769 1770commit b167dd02fc691a9760e80a5848762fd47f060843 1771Author: Daniel Drake <dsd@cs.manchester.ac.uk> 1772Date: Sat Mar 1 12:16:31 2008 +0000 1773 1774 Add API to determine next timeout 1775 1776commit 3ea28e334d2acfa0f988dbc155b4f9618a5b5fb2 1777Author: Daniel Drake <dsd@cs.manchester.ac.uk> 1778Date: Tue Feb 26 22:40:38 2008 +0000 1779 1780 Provide access to fd set 1781 1782 Includes notifications when the set changes. 1783 1784commit 88e9f4a5f80cf78aaf448a7fd6cd60beba7abd5b 1785Author: Daniel Drake <dsd@cs.manchester.ac.uk> 1786Date: Tue Feb 26 18:04:54 2008 +0000 1787 1788 Public async API implementation 1789 1790 Involved some internal overhaul/reorganisation. Nice side effect is that 1791 the synchronous API is now expressed purely in terms of the public async 1792 API. 1793 1794commit 83f9da1b874ce31ea69d21a2079b909c0f1f5343 1795Author: Daniel Drake <dsd@cs.manchester.ac.uk> 1796Date: Sun Feb 17 00:08:56 2008 +0000 1797 1798 uru4000: port to asynchronous model 1799 1800commit 754e56aec1e78d02d9f984402be33df931350d1b 1801Author: Daniel Drake <dsd@cs.manchester.ac.uk> 1802Date: Mon Feb 18 17:54:15 2008 +0000 1803 1804 Allow execution of sub-SSMs 1805 1806 A subssm will be executed, and upon completion the parent state machine 1807 will either be advanced to the next state or aborted with error, depending 1808 on the outcome of the subssm. 1809 1810 Also add some debug messages. 1811 1812commit b8dc77b56e89194a185a4d3db1e830459e551c5d 1813Author: Daniel Drake <dsd@cs.manchester.ac.uk> 1814Date: Mon Feb 18 17:50:48 2008 +0000 1815 1816 Fix verify/identify error handling 1817 1818 If verify_start failed, we need to run verify_stop. Same for 1819 identification. 1820 1821commit 866dd941cc85b28ff5f0ec80b7f3696150efa50f 1822Author: Daniel Drake <dsd@cs.manchester.ac.uk> 1823Date: Sat Feb 16 12:44:50 2008 +0000 1824 1825 Add timing and polling infrastructure 1826 1827 Add timeout mechanism as an asynchronous equivalent of sleeping (uru4000 1828 needs this). 1829 1830 Start implementing polling infrastructure which also accounts for pending 1831 timeouts. We don't expose file descriptors yet, but this is a start. 1832 1833commit eff26bf0be5c1ea22469958abcdb65ce6e9ec428 1834Author: Daniel Drake <dsd@cs.manchester.ac.uk> 1835Date: Fri Feb 15 18:11:01 2008 +0000 1836 1837 aes2501: port to asynchronous model 1838 1839commit 0aaffcee33015163d79320433cfd9d9c5941143a 1840Author: Daniel Drake <dsd@cs.manchester.ac.uk> 1841Date: Fri Feb 15 18:09:14 2008 +0000 1842 1843 Allow jumping to arbitrary state in a SSM 1844 1845 This is looking less and less like a simple state machine now, but it's 1846 proving a useful model! 1847 1848commit 77b5dab04b1305ba1e65eb7459349f99f36eba73 1849Author: Daniel Drake <dsd@cs.manchester.ac.uk> 1850Date: Fri Feb 15 18:06:24 2008 +0000 1851 1852 aeslib: introduce user_data for write registers callback 1853 1854 aes2501 needs this for state tracking 1855 1856commit 6f08a3666d71a4992b4af037164493ad633b8b1b 1857Author: Daniel Drake <dsd@cs.manchester.ac.uk> 1858Date: Fri Feb 15 18:03:03 2008 +0000 1859 1860 aeslib: fix register write boundaries 1861 1862commit d914712437a63f9671233bd0a29bef6adb16402c 1863Author: Daniel Drake <dsd@cs.manchester.ac.uk> 1864Date: Sun Feb 10 18:36:10 2008 +0000 1865 1866 aes4000: port to asynchronous model 1867 1868commit d731d5f3a3970aa8a8560096c137b2eda1b462e1 1869Author: Daniel Drake <dsd@cs.manchester.ac.uk> 1870Date: Sun Feb 10 18:35:19 2008 +0000 1871 1872 aeslib: port to asynchronous model 1873 1874commit 3048b37176263201cde7e754abb33feff47eafd8 1875Author: Daniel Drake <dsd@cs.manchester.ac.uk> 1876Date: Sun Feb 10 18:35:01 2008 +0000 1877 1878 Port imaging layer to asynchronous model 1879 1880commit aeeec97fb2efafdc22d55e505e28c41ff9d117ed 1881Author: Daniel Drake <dsd@cs.manchester.ac.uk> 1882Date: Mon Feb 4 10:24:27 2008 +0000 1883 1884 upekts: port to asynchronous model 1885 1886commit 5b1f6a0df7eb7c8fa84ced7177188a5a1b845e2c 1887Author: Daniel Drake <dsd@cs.manchester.ac.uk> 1888Date: Mon Feb 4 10:23:11 2008 +0000 1889 1890 Port primitive driver layer to asynchronous model 1891 1892commit 69760547df424710b9c92e935aebac4107a37e2a 1893Author: Daniel Drake <dsd@cs.manchester.ac.uk> 1894Date: Wed Jan 30 14:17:08 2008 +0000 1895 1896 Disable all drivers while internal API is being reworked 1897 1898commit bf7093d6077131031e89e447082c4f21352aa46d 1899Author: Daniel Drake <dsd@cs.manchester.ac.uk> 1900Date: Wed Jan 30 14:16:35 2008 +0000 1901 1902 Port to libusb-0.9.0 (synchronous I/O) 1903 1904 Straightforward port to libusb-0.9.0 sync I/O functions, to give libusb-1.0 1905 a bit of a test drive and to create a reference point for later. 1906 1907commit 6f456f399d99c49c48d28441d471a55c1d94081a 1908Author: Gustavo Chain <g@0xff.cl> 1909Date: Thu Dec 27 01:09:32 2007 +0000 1910 1911 Add Secugen FDU2000 driver 1912 1913 [dsd@gentoo.org: some trivial changes: 1914 fixed some warnings 1915 fixed fp_img memory leak on error 1916 renamed endpoint constants (calling a bulk endpoint 'ctrl' is confusing, 1917 as a control endpoint is something different) 1918 ] 1919 1920commit 659d8346beb4a0506d5c2f20c918a57804b4f591 1921Author: Daniel Drake <dsd@cs.manchester.ac.uk> 1922Date: Thu Dec 27 00:46:11 2007 +0000 1923 1924 Compiler flags update 1925 1926 Dropped -Wextra as it is not that useful. 1927 Set compiler flags to be more like the kernel's. 1928 Set std=gnu99 1929 Fixed some warnings. 1930 1931commit c8aa33da0bd2467594a92eeda641fe7749b9f33d 1932Author: Daniel Drake <dsd@cs.manchester.ac.uk> 1933Date: Fri Dec 7 15:03:27 2007 +0000 1934 1935 v0.0.5 release 1936 1937commit 642010643d04a5e43eac6fcb3d41b18003cde132 1938Author: Daniel Drake <dsd@cs.manchester.ac.uk> 1939Date: Fri Nov 16 17:25:41 2007 +0000 1940 1941 uru4000: Add support for Microsoft Fingerprint Reader v2 1942 1943 After lot 713, Microsoft fingerprint readers changed. The new version 1944 comes with a new USB product ID and a challenge-response authentication 1945 scheme where the device challenges the authenticity of the driver. 1946 1947 An independent third party produced documentation on the computations 1948 needed to convert a challenge into the correct response, and I then used 1949 this documentation to produce a clean-room reimplementation of the 1950 authentication scheme. 1951 1952commit 474da5f2c779a1c4a1309fd9f05972372ced7568 1953Author: Daniel Drake <dsd@cs.manchester.ac.uk> 1954Date: Fri Dec 7 14:59:38 2007 +0000 1955 1956 Add a disclaimer for the university 1957 1958commit c1f8581ec93d4e7d32f3880524cb1d775020d5b9 1959Author: Daniel Drake <dsd@cs.manchester.ac.uk> 1960Date: Sat Dec 1 13:00:33 2007 +0000 1961 1962 Cleanup discover_device in examples 1963 1964 Andrei Tchijov pointed out that the discover_device() function is written 1965 in a strange way. 1966 1967commit 3479dee0e4f6e0a2e779289378cb90c7a7710f88 1968Author: Daniel Drake <dsd@cs.manchester.ac.uk> 1969Date: Mon Nov 26 23:28:14 2007 +0000 1970 1971 aeslib: image assembly 1972 1973 All 3 authentec drivers now use a shared image assembly function. 1974 1975commit e228f2c7dab4960ed008d2170f07b6a53fa57eb3 1976Author: Daniel Drake <dsd@cs.manchester.ac.uk> 1977Date: Tue Nov 20 12:23:27 2007 +0000 1978 1979 Introduce shared functions for authentec drivers 1980 1981 Currently provides register writing functionality only, more common ground 1982 will be moved here soon. 1983 1984commit 8e0e8e43867dbd214149edeb83ad5688b5d9e8f2 1985Author: Anthony Bretaudeau <wxcover@gmail.com> 1986Date: Mon Nov 26 17:51:31 2007 +0000 1987 1988 aes1610: imaging improvements 1989 1990 bz threshold is now 10 (instead of 15). I think it is ok for the moment. 1991 If we manage to get higher image quality in the future, we'll change that. 1992 1993 max_frame is set to 350. 150 was too low => incomplete fingerprints 1994 1995 the driver was designed to stop the acquisition as soon as it gets a blank 1996 frame (=> incomplete fingerprints). Now it waits to have at least 50 blank 1997 frames before stopping. 1998 1999commit 4e2be56db7f977672fbf7a4cb5d260cbdc55a58c 2000Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2001Date: Thu Nov 22 17:59:33 2007 +0000 2002 2003 v0.0.4 release 2004 2005commit 1ee6210be320c471e448f7e5aa772e63b7ae85e8 2006Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2007Date: Thu Nov 22 17:25:27 2007 +0000 2008 2009 Hookup AES1610 driver 2010 2011 An oversight pointed out by Michele B. 2012 2013commit 69700855e27c655262de0d908a0447100238c5e4 2014Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2015Date: Tue Nov 20 22:13:18 2007 +0000 2016 2017 Implement identification for imaging devices 2018 2019commit 3f70d6f79410f483a9b34945dfe5b1cf5bfc23f6 2020Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2021Date: Tue Nov 20 21:52:10 2007 +0000 2022 2023 Identification infrastructure 2024 2025commit dbcd293ab31cabd44b6c840d6701d4f79a9aeee2 2026Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2027Date: Mon Nov 19 18:41:03 2007 +0000 2028 2029 v0.0.3 release 2030 2031commit c5a550e42c716be0fe3aa61e135fd62b1e861ed7 2032Author: Anthony Bretaudeau <wxcover@users.sourceforge.net> 2033Date: Mon Nov 19 18:39:03 2007 +0000 2034 2035 Add Authentec AES1610 driver 2036 2037 This is an active capacitance swipe-type device similar to but smaller 2038 than the AES2501. 2039 2040 Image processing performance is good but matching performance is not so 2041 good. The bozorth3 matcher needs tweaking in order to better cope with 2042 small sets of minutiae. 2043 2044 With a lowered threshold, matching performance is good enough for now. 2045 2046commit 6245acea848392861c2cd175d0292ef6b4f6638e 2047Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2048Date: Mon Nov 19 18:29:26 2007 +0000 2049 2050 Add API to delete enroll data 2051 2052commit 887e0e6acf8767c1f3bf571fc1ef8f5223c4c33b 2053Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2054Date: Mon Nov 19 00:27:27 2007 +0000 2055 2056 Add API to access minutiae 2057 2058commit fa742a2142ec969e9392823d6113f046c336cfb8 2059Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2060Date: Sat Nov 17 23:47:06 2007 +0000 2061 2062 Use G_N_ELEMENTS macro 2063 2064commit af945b2738ee3f8a50e8aafcce91a8f1d356b512 2065Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2066Date: Sat Nov 17 12:48:26 2007 +0000 2067 2068 Version 0.0.2 2069 2070commit fa24d51304316f8d9901368164a133232834ead4 2071Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2072Date: Sat Nov 17 13:15:25 2007 +0000 2073 2074 Fix variable get_img_width/height return value 2075 2076 Fix the functions to conform to the documentation: -1 means non-imaging 2077 device, 0 means variable. Internally, -1 is used to represent variable 2078 height (to be noticably different from the memset-imposed default of zero). 2079 2080commit ba24c0884a6c193854104bc234d981928ca33c32 2081Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2082Date: Sat Nov 17 12:46:00 2007 +0000 2083 2084 Rework some image handling; add public binarization API 2085 2086 I want to offer the ability for an application to view a binarized 2087 version of a scanned print. This lead onto a few changes: 2088 2089 1. Store minutiae and binarized data inside fp_img 2090 2. Move resize code to the capture path, it previously happened much 2091 later. 2092 3. Add fp_img_binarize() to return a new image in binarized form. 2093 4. Add a BINARIZED_FORM flag to prevent an image being binarized again. 2094 2095 In future, it would be nice to be able to binarize without detecting 2096 minutiae, but this involves some work on the NBIS interaction. 2097 2098commit 72c304999a07af03f0bad9485ebee6f9d0fcb520 2099Author: Jan-Michael Brummer <buzz2@gmx.de> 2100Date: Fri Nov 16 18:15:43 2007 +0000 2101 2102 Add UPEK TouchChip driver 2103 2104 The UPEK TouchChip is an active capacitance imaging device with a 2105 press-type sensor. It also has image storage capabilities which will 2106 hopefully be accessible through libfprint in the near future. 2107 2108 This device can be found in the Samsung P35 laptop. 2109 2110commit 791eeab85991f513d437809b83863f400105db13 2111Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2112Date: Fri Nov 16 15:44:09 2007 +0000 2113 2114 aes2501: restore default bozorth3 match threshold 2115 2116 With the multiple register writing code, the image quality is much 2117 better. It's trivially easy to get a match score of 100, and 200 is 2118 possible with a little effort. Remove the lowered match threshold. 2119 2120commit be0e9debd90592525cc135a22fc92e08be4b0274 2121Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2122Date: Fri Nov 16 15:35:48 2007 +0000 2123 2124 aes2501: write multiple registers simultaneously 2125 2126 Instead of writing each register in a separate USB transaction, we now 2127 write up to 16 at once. 2128 2129 This drastically improves scan image quality due to reducing the amount of 2130 time needed per iteration of the sampling loop (sending 1 USB transaction 2131 per iteration instead of 7). 2132 2133commit bfc55c4f0f4ff75b52c9d9c2d7b2fffb01c5dc8e 2134Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2135Date: Fri Nov 16 12:49:29 2007 +0000 2136 2137 Use older ImageMagick API 2138 2139 Debian ship an ImageMagick that is more than 2 years old. Switch to using 2140 older APIs so that we have compatibility all-round. 2141 2142commit 3fa3c0daf7f909187870f4f8136a9e830d18bdfb 2143Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2144Date: Fri Nov 16 13:14:30 2007 +0000 2145 2146 Initialize enrolled image pointer to NULL 2147 2148 Currently this is a dangling pointer if the image capture fails, and then 2149 we try to standardize an invalid pointer --> crash 2150 2151commit 069b4a30ebdac8723a1f7a3e2d880924a6be9552 2152Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2153Date: Thu Nov 15 17:23:36 2007 +0000 2154 2155 Doc and build updates 2156 2157 Add ChangeLog generation makefile target. 2158 Group all copyright notices in AUTHORS. 2159 Update some doxygen comments, and allow fp_dev_close(NULL) 2160 2161commit 06e8040956013fcba3c36d58356fb281b3888a58 2162Author: Vasily Khoruzhick <anarsoul@gmail.com> 2163Date: Thu Nov 15 16:09:38 2007 +0000 2164 2165 aes2501: Detect reversed scans 2166 2167 aes2501 can be mounted 180 degrees rotated (this happens on most part of 2168 laptops), so driver should detect whether sensor is 180degrees rotated 2169 and assemble frames in right order. 2170 2171commit 6866ea626e22a861ade880c9eb1c1e2fd48e3510 2172Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2173Date: Thu Nov 15 09:56:14 2007 +0000 2174 2175 v0.0.1 release 2176 2177commit 2ad14c021299fd362072a1fc36e634fc491df255 2178Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2179Date: Wed Nov 7 17:49:10 2007 +0000 2180 2181 Add AES2501 driver 2182 2183 This driver works quite nicely. Seems a little too sensitive though (too 2184 much black in the standardized image, not enough ridge gap definition). 2185 2186 Processing results are quite good, but you need a good enrollment image 2187 (i.e. long!). It's best to get such images by pressing harder than you 2188 might think necessary and swiping slowly. 2189 2190commit 5485f8031d97f26a88ae51e83872c6c79edafc3e 2191Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2192Date: Tue Nov 13 20:46:15 2007 +0000 2193 2194 Return images through enroll/verify path 2195 2196 Added new API functions to obtain images, even when scans are bad, perhaps 2197 a useful way to show the user just how good/bad the scan actually was. 2198 2199 Drivers and examples updated accordingly. 2200 2201commit a86cd51959cdb6e34ba0ddee4eed07880ac0fa2c 2202Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2203Date: Tue Nov 13 14:32:31 2007 +0000 2204 2205 Clean up on exit, and convert to singly-linked-lists 2206 2207 This adds fp_exit() to the public API, intended to be called while the 2208 driving app is shutting down. 2209 2210commit a76f55db0d668278e1a8108a203e685b65400458 2211Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2212Date: Mon Nov 12 23:32:17 2007 +0000 2213 2214 upekts relicensing 2215 2216 Pavel Machek and Timo Hoenig agreed to relicense their work. Thanks guys, 2217 I look forward to working together. 2218 2219commit 71e4bb39ec214883bc707b5ba2b95423f1fd05b6 2220Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2221Date: Fri Nov 9 14:57:42 2007 +0000 2222 2223 Allow drivers to specify custom bz3 match threshold 2224 2225 aes4000 detects fewer minutiae and hence returns lower scores. 2226 2227commit e3451158e9e1d0d2cb1f81197a8e392ac4294ec0 2228Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2229Date: Thu Nov 8 17:04:45 2007 +0000 2230 2231 Custom image resizing 2232 2233 mindtct appears to completely ignore the pixels-per-mm input parameter 2234 (ippmm). When processing AES4000 images, the binarized image is 2235 completely mangled and a lot of ridge information is lost. 2236 2237 Resizing the AES4000's small images results in a huge imaging performance 2238 gain. 2239 2240 We use imagemagick for the resizing, as it's resizing code resamples the 2241 image too (smoothing it out), which further improves performance. 2242 2243commit a73cbc10fb2ab3ded4551707b772a4c8b88ec8c4 2244Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2245Date: Fri Nov 9 13:47:32 2007 +0000 2246 2247 aes4000: image quality improvements 2248 2249 The windows driver takes one sample at the previous register settings 2250 and then changes some registers before resamping (and getting much better 2251 images). The exact changes that it makes seem to vary, perhaps based on 2252 the histogram. 2253 2254 Anyway, this is an approximation of the settings used for the 2nd sample 2255 which should help matching results. 2256 2257commit b9238e8b8a308f15fc0cadd2217ed105aab0cdde 2258Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2259Date: Thu Nov 8 13:56:21 2007 +0000 2260 2261 Support images with variable dimensions 2262 2263 For example, AES2501 returns images that vary in height. 2264 2265commit e1a25eeb672fdc0de2f1bb019b234587138b37f1 2266Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2267Date: Thu Nov 8 13:41:52 2007 +0000 2268 2269 Doc update 2270 2271commit 8e7afef736711d81a4f33b17800b8e132ac251d7 2272Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2273Date: Sun Nov 4 21:12:48 2007 +0000 2274 2275 uru4000: add Digital Persona U.are.U 4000 support 2276 2277 My UareU4000 device isn't working properly -- only seeing an all-white 2278 image and no finger-on detection, however I think this code should work 2279 for working devices. 2280 2281commit f498fcd1eac49b2d5cee6285d6e16e0cd1498f7c 2282Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2283Date: Sun Nov 4 19:10:23 2007 +0000 2284 2285 Fix img_capture_continuous build 2286 2287 Also enable it in the default development build 2288 2289commit f733ddcc3b3748a73a3c63865be681af4fc5ed12 2290Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2291Date: Sun Nov 4 19:01:15 2007 +0000 2292 2293 uru4000: retry scanner powerup 3 times 2294 2295 When we include the await_finger_off procedure (i.e. set mode 12), the 2296 current power-up code is unreliable - no 56aa interrupt arrives sometimes. 2297 Comparing the F0 data and hwstat at this point between a working and a 2298 non-working run yields no differences. 2299 2300 Work around this issue by retrying power-on 3 times. 2301 2302commit cbe1111e4e68702ac3d1b23c4a65b81052f0656c 2303Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2304Date: Sun Nov 4 18:57:44 2007 +0000 2305 2306 uru4000: handle images without headers 2307 2308 While testing the new code on my MS keyboard, I found some strange 2309 conditions where occasionally, the device will stop providing images 2310 headers until it is replugged. 2311 2312 Detect and work around this when it happens. 2313 2314commit 2e6c3b940cf30705e7a2aa4bbb5c1c191fa94fdc 2315Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2316Date: Thu Nov 1 23:11:29 2007 +0000 2317 2318 API documentation 2319 2320commit 7e6f25908bbd160f7f2a44618f29bbaded748bf9 2321Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2322Date: Fri Nov 2 17:04:19 2007 +0000 2323 2324 remove imgdev abstraction from public API 2325 2326 Make it easier for application developers, don't have to care about this 2327 level of abstraction which we're intending to make superfluous to them. 2328 2329commit fc77786e46f8e7e8912426e13e858eb041b8ecc2 2330Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2331Date: Wed Oct 31 22:42:09 2007 +0000 2332 2333 uru4000: interrupt handling fixes 2334 2335 This should improve driver stability somewhat. Powerup problems seem to 2336 happen when we receive 3 unrelated interrupts while waiting for the 2337 power-on interrupt, so let's assume that the device can only buffer 3 2338 interrupts, and when the buffer is full, it discards new ones. 2339 2340 When we detect a possible buffer overflow, ask the caller to retry the 2341 operation. 2342 2343 Also, saw the interrupt of death a few times, so add in a warning when 2344 this happens. Haven't seen it since handling interrupt overflows though. 2345 2346commit 07a81b07becb597d919c706e9894c31b2e95432f 2347Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2348Date: Wed Oct 31 21:51:37 2007 +0000 2349 2350 upekts: initialise and deinitialise device during verify/capture 2351 2352 Otherwise it heats up, and verify retries fail because we didn't end the 2353 session. 2354 2355commit fc0a97a801377fb5d9666d81a18b2c93c4c37f61 2356Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2357Date: Wed Oct 31 13:18:30 2007 +0000 2358 2359 Build system tweaks 2360 2361 Install fprint.h header 2362 Include headers in distribution 2363 distcheck fixes 2364 Include and install pkgconfig file 2365 2366commit 447b1f01b3163bebf640f2f44d09a7fbf0c4e79c 2367Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2368Date: Wed Oct 31 13:12:04 2007 +0000 2369 2370 Print discovery 2371 2372 Add concept of "discovered prints" and flesh out the API. A discovered print 2373 was found based on file name but has not been checked for file integrity, 2374 and of course there is no guarantee that it is usable when you try to 2375 use it later (might have been deleted, etc). 2376 2377 The .fprint directory structure was changed - instead of using mnemonics 2378 like "rind" we use their numeric value in the fp_finger enum as the filename. 2379 2380commit b96aade69ca93c28ed20bdac8be432d4feb0f9fd 2381Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2382Date: Tue Oct 30 22:24:31 2007 +0000 2383 2384 Flesh out driver_id/devtype API 2385 2386 Add accessor functions to the various structures 2387 2388commit cc3e976042e49587ecca87a7fdb3d6dbdcf76d0e 2389Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2390Date: Tue Oct 30 22:15:45 2007 +0000 2391 2392 Add 'discover' hook for drivers 2393 2394 Drivers can now offer a discover function, and are expected to offer the 2395 device devtype during discovery, without accessing the USB device. Hmm. 2396 Maybe we'll have to change that later... 2397 2398commit f0ef386f43d944430654f3ef486464db5717b4c9 2399Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2400Date: Mon Oct 29 23:37:28 2007 +0000 2401 2402 More NBIS cleanups 2403 2404commit 6f633cc771c5290593880c79304234eafb5e8e83 2405Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2406Date: Mon Oct 29 23:06:06 2007 +0000 2407 2408 Use glib's byte order functions 2409 2410commit 294f9ad44730400800f9de0cd213a7050f6ef2a5 2411Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2412Date: Sun Oct 28 22:02:04 2007 +0000 2413 2414 Improve print_data structure and on-disk format 2415 2416 Drivers now have an ID number. These will be assigned by me and documented 2417 on the wiki. 0 cannot be used. 2418 2419 Drivers now define a devtype for each device they initialise. This is to 2420 cope with the situation where a driver can support varying devices where 2421 their print data is incompatible (i.e. image scaling is totally changed). 2422 This doesn't apply to any existing supported devices. 2423 2424 Print data no longer includes driver name, and includes driver ID instead. 2425 2426 Paths to saved print data now include driver ID and devtype, and no longer 2427 include driver name. 2428 2429 APIs exposed for converting a print_data into a blob which can then 2430 be loaded back again later. Useful for systems who don't want to use 2431 my simple storage system (which is only aimed at a single user). 2432 2433 File format is now defined and will be documented on the wiki. The header 2434 is larger as we can no longer rely on directory paths in all scenarios. 2435 2436 Print data compat check now checks devtype and driver ID. 2437 2438commit 4e5cfdf92aa305a5d8ed58fb650ccdad028742ac 2439Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2440Date: Sun Oct 28 15:45:28 2007 +0000 2441 2442 NBIS cleanup 2443 2444 Part 1 of many. Remove some unused/pointless code, and made some code 2445 static. 2446 2447commit 6b8d17ef2654ad7f42fa9941074d132fb4d6e324 2448Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2449Date: Sun Oct 28 00:21:49 2007 +0100 2450 2451 Add bozorth3 from NBIS; implement verify for image devices 2452 2453 uru4000 works splendiferously (YAY!) 2454 aes4000 is going to need some work though :( 2455 2456commit 41b25f28a4dd980b88fef2e9ff9f243390793639 2457Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2458Date: Sat Oct 27 22:48:09 2007 +0100 2459 2460 Add mindtct from NBIS; implement enroll for image devices 2461 2462 mindtct is mostly as-is for now, with just a couple of bits ripped out. 2463 2464commit be67f85caa1c65c4c55d4e969d757e7f2790bd7e 2465Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2466Date: Sat Oct 27 18:12:16 2007 +0100 2467 2468 Add Authentec AES4000 driver 2469 2470commit 3bbb7b8b51829acd5b8efaf54d60338aec0c4717 2471Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2472Date: Sat Oct 27 18:10:06 2007 +0100 2473 2474 Internal image allocation API changes 2475 2476 Creating by dimensions is likely not useful now that we can infer image 2477 size from a device in question. 2478 2479commit 351ccb4ab7f60ce2225f10bab13fd0862c1d17de 2480Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2481Date: Sat Oct 27 15:17:19 2007 +0100 2482 2483 Add XV-based img_capture_continuous example 2484 2485commit da071ff8a2a20984b5f3ea5c355c8a7063f51860 2486Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2487Date: Sat Oct 27 15:16:54 2007 +0100 2488 2489 Add API for freeing images 2490 2491commit c2a83ec948ffb5604e30cb9cba4f6b71b5ad2b43 2492Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2493Date: Sat Oct 27 14:45:14 2007 +0100 2494 2495 Improved image dimension handling 2496 2497 Drivers now specify the size of the image they provide, and theres an API 2498 so that you can get the size of an image before you capture it. 2499 2500commit 9960a7ff3450fe6df892f794904c8fa2ac398115 2501Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2502Date: Sat Oct 27 12:07:11 2007 +0100 2503 2504 Image standardization 2505 2506 Initially supports flipping and colour inversion 2507 2508commit 0ddfef6c21e9944e1ed34e4cbb895e105b82359d 2509Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2510Date: Sat Oct 27 00:10:32 2007 +0100 2511 2512 Add uru4000 driver 2513 2514commit f367ae8f4be9dc504d5526d4e74fab497f9f1f59 2515Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2516Date: Sat Oct 27 00:09:46 2007 +0100 2517 2518 Zero out fprint_data fields on allocation 2519 2520commit f3a838e856423af802de1d777a19163843cdf635 2521Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2522Date: Sat Oct 27 00:06:53 2007 +0100 2523 2524 Initial implementation for image device access and capture 2525 2526 Also added new example: img_capture 2527 2528commit 0aac3706d707b26f478656282d2d7dce47e3adbf 2529Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2530Date: Thu Oct 25 15:35:10 2007 +0100 2531 2532 upekts: release interface on error condition and normal exit 2533 2534commit 0ac74ef649802000da82bb1bddf8c1637c515de6 2535Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2536Date: Thu Oct 25 15:00:30 2007 +0100 2537 2538 Pass driver_data from ID table to driver init function 2539 2540commit 2709c6dc8c2429c3202b88b709593376d69392e5 2541Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2542Date: Tue Oct 23 23:24:21 2007 +0100 2543 2544 Basic image driver foundations 2545 2546 The basic model is that image drivers declare a fp_img_driver structure 2547 rather than a fp_driver struct. fp_img_driver will contain primitive imaging 2548 operations such as 'scan finger and return image'. The imgdev layer will 2549 generically implement the primitive fp_driver operations, and the imgdev 2550 layer will fix up the enroll/verify/etc pointers at driver registration 2551 time. 2552 2553 Removed const from all fp_driver declarations, as these are now modified 2554 dynamically in the case of imaging drivers. 2555 2556commit c97f4bb1a721c7cee3acd8dec081a3f0fb86a5bb 2557Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2558Date: Tue Oct 16 14:35:33 2007 +0100 2559 2560 Fix verify debug messages 2561 2562 Teach it about the various error conditions. 2563 2564commit 801fcfc07ce3cee36ce8c16e8e4fa26d22ae8db3 2565Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2566Date: Tue Oct 16 14:34:47 2007 +0100 2567 2568 upekts: fix memory corruption 2569 2570 Found by valgrind 2571 2572commit 680142f268ad8bdfd0cbcbd6654f2e0dc2d20187 2573Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2574Date: Tue Oct 16 14:23:30 2007 +0100 2575 2576 storage: add save/load functionality 2577 2578 Prints can now be saved to disk (but you currently must classify which 2579 finger they are) and you can load them later. 2580 2581 Added 2 simple example programs to demonstrate this. 2582 2583commit 3b8f8c195cc49532669cce859be7969ff5813611 2584Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2585Date: Tue Oct 16 13:41:23 2007 +0100 2586 2587 upekts: support large messages 2588 2589 Add support for messages with a length greater than 255. 2590 Also picked out a few more retry codes. 2591 2592commit 10bb96a0dfd9dfb50e870499661f2adf8050f11f 2593Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2594Date: Mon Oct 15 22:16:22 2007 +0100 2595 2596 Rename verify to verify_live 2597 2598 ...to make way for an enroll example which stores on disk, and 2599 a verify example that loads from disk. 2600 2601commit bf9ce44f43c013a3822fe63a5214faf946a2ce8b 2602Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2603Date: Mon Oct 15 22:09:46 2007 +0100 2604 2605 upekts: verify implementation 2606 2607 We can now verify live fingerprints using the verify example. 2608 Error handling during verification needs to be improved. 2609 2610commit f59112d57cf008969d074ff4ffa5f31c5d29a099 2611Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2612Date: Mon Oct 15 22:07:22 2007 +0100 2613 2614 Unhide fprint_data buffer from drivers 2615 2616 May as well keep things out in the open. 2617 2618commit 68bdfc7868f8a0467da58cba4796a44569f806b9 2619Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2620Date: Sun Oct 14 00:58:45 2007 +0100 2621 2622 Verification infrastructure 2623 2624 Similar model to enrollment. 2625 2626commit 39271b4fe50b5f3ceb880e47c951f906b9bc8ff2 2627Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2628Date: Sun Oct 14 00:45:49 2007 +0100 2629 2630 Enrollment: improved error reporting 2631 2632 Convert enrollment function to return a signed integer, which is negative 2633 on error or corresponds into fp_enroll_result otherwise. 2634 Now we can treat a 'FAIL' condition differently from an actual error: fail 2635 means that enrollment didn't complete because the data was nonsense or 2636 whatever (e.g. scanned a different finger for each stage?). 2637 2638 Updated upekts accordingly. 2639 2640commit ab5b1ca8e9815599a5e5034736db7bcabebc9101 2641Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2642Date: Sat Oct 13 16:24:20 2007 +0100 2643 2644 upekts: communication handling and enrollment implementation 2645 2646 This completes an enrollment implementation for the upekts driver. 2647 It's based on code from libthinkfinger, but as there seemed to be many 2648 unknown details there, I spent some time analysing the bus messages in 2649 more detail and have effectively rewritten almost all of the borrowed code. 2650 2651 Note that libthinkfinger is GPL licensed, so that reflects on this code too. 2652 At release time I will ask them if they are OK for me to relicense it. 2653 2654commit 299514431073828585c10b27af34a43936014703 2655Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2656Date: Sat Oct 13 15:52:50 2007 +0100 2657 2658 Build system tweaks 2659 2660 Add configure-time controls for logging 2661 Set some default compile-time warnings 2662 2663commit 50e2de0730c5d93dd1ad5f0f07e010ee1d92ad01 2664Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2665Date: Fri Oct 12 15:27:23 2007 +0100 2666 2667 Enrollment tweaks 2668 2669 upekts will need to know when the first enrollment stage is attempted for 2670 the first time, so add an __enroll_stage counter which actually indicates 2671 the next stage to enroll. -1 is a special value and it means 0 is next *and* 2672 it is the initial attemt. 2673 2674 Added more debug output to the enroll handler. 2675 Added new fp_enroll_status codes for too short or uncentered scans. 2676 2677 Changed the print_data allocator to consider the device rather than the 2678 driver, this feels more natural. Added missing return value. 2679 2680 Make fp_enroll_status codes start at 1. 0 can now be used as a 2681 special/temporary value by the drivers. Also check that we aren't exceeding 2682 the number of enroll stages. 2683 2684 Also add a missing exit() call to the verify example and update for the 2685 above. 2686 2687commit e6624fe5086dc3d86a3947e396ab7c52c0cdddb8 2688Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2689Date: Fri Oct 12 15:16:51 2007 +0100 2690 2691 Add 16 bit endianness functions 2692 2693commit ec91736ac4fb77e5d852959820d0070d1b4fc56f 2694Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2695Date: Wed Oct 10 15:51:50 2007 +0100 2696 2697 Add logging infrastructure 2698 2699 Also added some debug/error messages to existing code. For now debugging is 2700 always on, this will be made optional later. 2701 2702commit 016ff33533bf4930e05ed5c818b83a5610e6b219 2703Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2704Date: Mon Oct 8 19:53:50 2007 +0100 2705 2706 Enrolment infrastructure 2707 2708commit b071f3cdb928117401d173aa5ea0bb9555295618 2709Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2710Date: Mon Oct 8 18:24:29 2007 +0100 2711 2712 Initial fingerprint data representation ideas 2713 2714commit 59b73af909162c5fa7d57d3c14f66508a7589873 2715Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2716Date: Mon Oct 8 17:37:58 2007 +0100 2717 2718 Documentation update 2719 2720commit f81aa47a19a8ab7c1e53d9b483ead064fdd85884 2721Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2722Date: Mon Oct 8 17:15:21 2007 +0100 2723 2724 Add 'verify' example skeleton 2725 2726 This will become an example program to enroll a fingerprint and then 2727 verify a finger against it. 2728 2729commit 2c06a4ebfaed93505bb7daa7d9a49bb394bdbfb4 2730Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2731Date: Mon Oct 8 17:05:58 2007 +0100 2732 2733 Expose driver info to library users 2734 2735commit 711bb1151b161f3696d97b78de975221ccee82cd 2736Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2737Date: Mon Oct 8 17:01:08 2007 +0100 2738 2739 open/close device functions 2740 2741commit 7d31cb4d456185a951fc98276b1244039317a0a2 2742Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2743Date: Mon Oct 8 16:55:50 2007 +0100 2744 2745 Basic device discovery infrastructure 2746 2747commit 046bdc0bdab54d1890c5a5445b6ca711e4a13f19 2748Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2749Date: Mon Oct 8 16:46:42 2007 +0100 2750 2751 Initial driver registration mechanism 2752 2753 Also added the basis of the upekts driver, which is the first I'll 2754 implement. 2755 2756commit 7d3612d9f6ebc5ecbdc400acbb97333c7bb402c9 2757Author: Daniel Drake <dsd@cs.manchester.ac.uk> 2758Date: Sun Oct 7 23:20:12 2007 +0100 2759 2760 Create repo structure 2761