Commit Graph

312 Commits

Author SHA1 Message Date
c59e23efda provide estimates, format dates better 2021-09-16 19:20:43 +10:00
b6f672b575 ai stats has divs not tables, so more compact/responsive, also stats now have clickable link to appropriate AI:<tag> search 2021-09-15 00:23:25 +10:00
10f8e4cc9d aria-label to make lighthouse report better 2021-09-14 22:17:23 +10:00
301b3d220b fixed BUG-54/55, viewer now works in PROD, current_file being set in run_ai_on job, added "new" bug-53 re video fname not being set properly when next/prev in viewer 2021-09-14 17:56:06 +10:00
56f3853bfc removed unneeded ref to font-awesome, added compression, ttl for static files, several small cleanups on labels, etc. for lighthouse improvements 2021-09-14 17:22:42 +10:00
6e96c83d9e put title on progress bars so on hover you get the details, useful when you have 1 of 10000 on a slow job 2021-09-14 17:06:47 +10:00
5bb2195239 fix BUG-51, its ENV not env and production not Production 2021-09-13 17:54:59 +10:00
f71e3d6144 fix bug with 2 captions when video caption has to be outside <video> tag 2021-09-12 18:28:37 +10:00
57281be3d1 use tooltips for settings, to explain path usage in particular 2021-09-06 19:16:24 +10:00
35a096c79d added name for videos 2021-09-06 19:15:58 +10:00
479c6179c2 fix icon url needing to be passed as a jinja2 var 2021-09-05 22:23:08 +10:00
7c192b5d66 with MoveDBox going into js file, need to pass jinja2 var as a param 2021-09-05 21:34:13 +10:00
4a9c3b4aa3 minor clean up of screen content 2021-09-04 19:50:27 +10:00
3472717d01 fixed BUG-44: quickly clicking in folder view can somehow end up with a view/<id> crash 2021-08-30 18:25:19 +10:00
7b672206d2 alert in form fails in bs5?, so made it a div 2021-08-30 18:24:55 +10:00
dc4ff64600 viewer hotkeys are in tooltip now 2021-08-30 17:47:33 +10:00
10fcda0d7c convert OPT from a dict to a class 2021-08-29 20:13:26 +10:00
f9bd558727 moved most of the js out to internal/js/view_support.js 2021-08-28 21:45:28 +10:00
c38929d884 viewer now deals with videos and images when next/prev, AND, fullscreen is maintained on videos and handles when escape gets you out as well... 2021-08-28 13:26:07 +10:00
1d3caf17de complete rewrite of viewer, videos need to be fixed. Otherwise, viewer now loads entry data for all "how_many" images on the screen, and allows prev/next movement between images without a DB load for the current page of images, then as required, it will retrieve the prev/next "how_many" from the database via /viewlist route, and go back into view/<id> of the appropriate new image of the new list. Also prevents going below 0 and beyond end of DB for the first time. 2021-08-27 23:38:31 +10:00
652a89161d include the js form the internal dir and viewer now has tooltips on buttons, and all buttons work, call the right javascript, send jobs to the pa_job_manager, and make image go gray with throbber, and on job completion, hide throbber, redraw the image in color and with the newly transformed image 2021-08-15 21:26:43 +10:00
59c72f9aee moving to /internal for a js dir, icons.svg Ive built, etc. and added route in files.py, updates all html to use new path 2021-08-15 00:26:27 +10:00
d49af24fee spacing 2021-08-08 16:18:28 +10:00
ae0d27b284 first pass of usable fullscreen, will need arrow keys to remember state, or realistically pre-load? but for now you can fullscreen the image you are on 2021-07-31 22:49:40 +10:00
902830ad95 viewer now has (non-functional) buttons for rotation, flip, fullscreen, and Ive put icons.svg into the repo as its hand-crafted upstream content so I cant really just download it anymore 2021-07-31 22:40:12 +10:00
c0a8ab8f79 allowed view prev/next to remember settings (fname, faces, distance) and use new CreateSelect() for AI menu, also make fname toggle actually do that 2021-07-29 21:44:54 +10:00
e0802bc573 move to form toggles, made them bg-info colours give or take, made the fname, face & distance toggles work. The model drop-down is not set correctly, or taking action, but mara needs more rel mem to run cnn model anyway 2021-07-28 21:42:40 +10:00
b7d346c206 model_used is now per file, not per face, implemented split of default_model to default_refimg_model and default_scan_model in settings, and default_refimg_model actualy works when creating refimgs in person.py. The model_used in face_file_link is based on default settings model and will scan with cnn if chosen and store that in DB as needed. Need viewer to allow changing per file / not just default for future scans 2021-07-27 17:14:03 +10:00
cd51ea21bf quick add of default_{model|threshold} to settings, face_refimg_link now stores model_used and face_distance AND working implementation of own face_distance algorithm rather than compareAI(), removed older AI code it would no longer work with DB structures anyway, tweaked viewer to remove coords of unmatched faces for now 2021-07-26 18:39:36 +10:00
b40ecbf208 put svgs in icons.svg and reference appropriately, still not happy with pa_logo, but otherwise all working 2021-07-25 13:19:01 +10:00
c6a31f24e5 added ai model settings 2021-07-25 13:18:18 +10:00
d25c4fae62 removed last font-awesome and obv. bootstrap5 items 2021-07-22 21:28:54 +10:00
ef6759c5ed use svg / remove -lg- / bootstrap 5 2021-07-22 20:44:10 +10:00
4f471ffa5a remove -lg- / bootstrap 5 2021-07-22 20:43:11 +10:00
bd51f86d82 remove -lg- / bootstrap 5 2021-07-22 20:42:50 +10:00
3757c942e6 more bootstrap 5 fixes 2021-07-22 19:54:18 +10:00
f3b0cc5ecc more bootstrap 5 fixes 2021-07-22 19:53:21 +10:00
07408278b1 more bootstrap 5 fixes 2021-07-22 19:53:09 +10:00
085dbcdf3a more bootstrap 5 fixes 2021-07-22 19:51:54 +10:00
e9020373c6 more bootstrap 5 fixes 2021-07-22 19:51:07 +10:00
915063a7a7 semi fixed move dbox, the input group is still not right, but it will suffice, more use of svg and ditching font awesome 2021-07-22 00:50:45 +10:00
6b6e7bc901 bootstrap 5 grid "fixups" for viewer 2021-07-20 22:04:52 +10:00
f52ba689b7 more bootstrap 5 fixes, badge colors, data-bs-target, padding 2021-07-20 22:04:31 +10:00
678e18b89d disable next button when we have no more int he list -- doesnt deal with when we exactly land on the last image, and it seems we have 1 too many in the first set of images (have 10 on select, but see 11 on page - after next button though its 10 on the page 2021-07-20 19:08:32 +10:00
b9f05d64f8 okay this now works with bootstrap 5, I dont love the sz buttons / mixed black and white hover text needs love - probably own class with own styles, but I dont want to hard-code info color, need to work out how to do style / css using the right bs variable for info color 2021-07-20 19:03:14 +10:00
44620f1e18 viewer now has toggle to turn on face bounding box + nametag 2021-07-19 18:48:39 +10:00
7e360270a7 make use of row / col consistent to fix up odd title/buttons wrong left-hand margin / sizing 2021-07-19 18:45:57 +10:00
eddc00d4a9 fixup missing explicit px; 2021-07-19 18:45:23 +10:00
1bd30e106e arrow keys now switches viewed image 2021-07-17 23:55:27 +10:00
24c4b962e5 now draws green rectangle and person.tag for matches on faces -- need to fix interface, but it works - lots of "plumbing" across database/classes and this viewer.html to get this to work 2021-07-17 16:37:43 +10:00