IRC log for #brlcad on 20141114

00:28.51 starseeker clears out his BRL-CAD checkout...
00:37.30 maths22 brlcad: I don't use command line editors enough to have bothered learning them
00:37.36 maths22 Yes... I know I should
00:37.51 maths22 But nano is so user-friendly
00:37.59 maths22 I was using pico, but it doesn't have syntax hightlighting
01:05.02 Stragus uses some weird Motif-based editor that hasn't been updated in 10 years
01:05.22 Stragus Although I made various changes in the source for the version I use for myself
02:10.02 Notify 03BRL-CAD:starseeker * 63429 (brlcad/branches/qtged/src/qbrlcad/CMakeLists.txt brlcad/branches/qtged/src/qbrlcad/main_window.cxx): Get the fundamental pieces in place for a tool palette in the style of the gimp or krita. That type of widget doesn't seem to be standard, so combine the Qt FlowLayout example with some custom logic to define a 'QToolPalette' that will have the desired behavior. Needs a lot of
02:10.04 Notify refinement, but seems to be working.
02:12.28 *** join/#brlcad kintel (~kintel@unaffiliated/kintel)
02:20.53 starseeker Stragus: nedit?
02:23.56 Notify 03BRL-CAD:starseeker * 63430 brlcad/branches/qtged/src/qbrlcad/cadtreemodel.cxx: This needs to be checked, but I'm guessing the problem case is actually when aircode and region_id are set, rather than the region flag itself?
03:23.35 *** join/#brlcad kintel (~kintel@unaffiliated/kintel)
04:13.00 brlcad maths22: emacs isn't exactly a command line editor, heh
04:14.11 brlcad but yeah, I'd strongly recommend you learn one dev envrionment in detail, and of course recommend emacs for a long list of reasons
04:14.25 brlcad but really the trick is learning it in detail and becoming uber efficient with it
04:15.38 maths22 I agree that learning such a thing would be useful.
04:15.57 brlcad sublime, atom, emacs, all good choices
04:16.12 maths22 By the way, there will be one restriction on duplicating tasks
04:16.31 maths22 The duplicates cannot have their titles changed, as that would mess up the numbering
04:16.46 maths22 Actually, once duplicates are made, the master should not be able to have its title changed
04:17.58 brlcad yeah, that's understandable
04:18.19 brlcad btw, if you decide to take the plunge, I found this useful: http://www.gnu.org/software/emacs/refcards/pdf/refcard.pdf
04:19.02 brlcad granted a rather different version back when I learned .. :)
04:19.05 maths22 I could mirror the title, but I think that would cause more problems
04:19.22 maths22 more in users minds than in the app, but usability is important
04:22.53 brlcad fwiw, here's the melange form: http://ibin.co/1hAl212o59D5
04:27.14 maths22 Thanks
04:27.34 brlcad some annoyances there are that you have to fill out every field repeatedly, I found myself very freuqently using the browser back-arrow, changing the title and description a little, and submitting; just to avoid having to set the type, tags, time, and, mentor every single task
04:27.43 maths22 thanks - I can see that one as a mentor (probably because you would need to approve it any ways)
04:27.57 brlcad ah, good
04:28.14 maths22 I can make 2 copy functions-one that makes literal duplicates and one that pre-fills the edit form
04:28.18 brlcad so yeah, you can add some and see how it gets really tedius
04:29.06 brlcad default round-robin mentor assignment would be useful
04:30.24 maths22 OK-I will add a table for a list of mentors soon
04:30.39 maths22 Do you want the 2 types of copy?
04:30.50 brlcad possible to pull that list from melange?
04:31.22 brlcad usernames are listed on https://www.google-melange.com/gci/dashboard/google/gci2014#all_orgs_mentors
04:33.16 brlcad course if you can pull that csv you could pull the task list csv too... :)
04:34.34 maths22 Can't pull dynamically-That page is behing a login
04:36.44 brlcad surely cakehttps://github.com/corefactor/CakePHP-Oauth-Plugin
04:36.58 brlcad surely indeed
04:37.21 brlcad and a snippet that shows how: http://stackoverflow.com/questions/8741501/cakephp-oauth-with-google :)
04:37.55 maths22 I am not talking about authenticating into my app
04:38.17 maths22 I am saying that the app cannot pull that page without authentiation
04:38.34 brlcad yes..
04:38.39 maths22 and the authentication would be logging into google
04:38.49 maths22 Which can only happen on the client side
04:39.17 brlcad oauth authenticates with google, once you do that youll have auth tokens and can get at the dashboard
04:39.58 maths22 I have auth tokesn for my app
04:40.20 maths22 That would allow me to make each google user a user in my app
04:40.39 maths22 That does not allow me to make arbitrary reqeuests to other google apps using their credentials
04:44.12 brlcad I get what you're saying, but I recall there being a means to basically do that (at least with google login apps), basically feeding them the cookie they expect
04:44.28 brlcad "basically"
04:45.14 Notify 03BRL-CAD Wiki:122.102.45.251 * 7797 /wiki/ARL_Technical_Reports:
04:45.52 brlcad heh, and here's a gsoc project to try and prevent it: http://www.google-melange.com/gsoc/proposal/public/google/gsoc2014/anuroopkuppam/5629499534213120
04:47.39 maths22 That also would add significant latency to the app.
04:49.54 maths22 Take a look at the clone button
04:50.24 brlcad totally worth it if it just talked to melange ... otherwise that latency still exists, it's just on the mentors/admins clicking on and waiting on melange, copy-pasting data periodically back and forth
04:54.24 maths22 I see what you are saying. If I can find a way to easily do it, I will.
04:57.52 brlcad hm, just tried cloning something and it didn't seem to work
04:58.36 brlcad looks like import messed up on at least one of them too, sort by title, the "Alex" task
04:59.04 maths22 I see that task
04:59.16 maths22 What did you try to clone?
04:59.22 maths22 Also, the copy button should work
05:00.55 brlcad tried to clone promo video #2
05:01.01 maths22 try again
05:01.10 brlcad and #1
05:01.13 maths22 It may have happened when I was in the middle of updating the file ore something like that
05:01.33 brlcad just did, still no
05:02.05 brlcad tried missing docs too
05:02.07 maths22 I see why that is failing
05:02.15 maths22 CHeck if copy works, and I will fix that
05:05.08 brlcad yes, worked
05:05.24 brlcad few usability suggestions
05:07.04 brlcad unless it's going to seriously make it sluggish, I'd rather see all tasks so I can search the page to jump to tasks instead of hunting where it's on page 9 of 16 or 11 or 13 or 12, ah, there it is...
05:07.33 maths22 OK-I will remove the paginator
05:07.34 brlcad no point in seeing the tags in the summary unless I can filter on them, just clutter
05:07.48 maths22 THe nice thing about cake is that these sorts of changes are fairly easy
05:08.28 brlcad length seems inconsistent ... might be mixing hours and days
05:08.59 maths22 That is an import artifact
05:09.19 maths22 I manually converted stuff, and was not that careful about it, as I was just trying to create usable test data
05:10.26 brlcad suggest eliminating view and edit button, make selecting the titel go to that task ideally to an edit mode with a button to preview the rendered state or display it side-by-side or something so it's not a separate thing
05:10.42 brlcad okay, makes sense
05:11.57 brlcad might as well display your unique ID if delete is going to reference it, and it helps for talking about them
05:12.15 maths22 Everything is now on one page
05:13.23 brlcad a filtering option would be nice, that part does work well on melange, though it is very laggy
05:13.36 maths22 I will see about filtering
05:13.45 maths22 There are some jquery table plugins that might help
05:14.05 brlcad huh, I still get pages
05:14.22 maths22 Now you shouldn't
05:14.24 maths22 saving helps
05:14.30 maths22 also, the id links to the edit page
05:14.32 brlcad ah, yes
05:15.33 brlcad need a better moniker of clone vs copy if other orgs use this, unclear which is which
05:16.19 maths22 Yep
05:16.44 brlcad especially considering the clone button asks how many copies ;)
05:17.20 maths22 This is true
05:17.29 maths22 Do you have any ideas?
05:19.50 brlcad could merge them, just one button
05:20.06 maths22 How would you choose which you want?
05:20.09 maths22 I like the idea
05:22.22 brlcad Clone -> How many defaulting to 1 -> clicking okay taking you to the edit page where the info is all filled in and the title includes #next
05:23.13 brlcad if they enter > 1, it creates the N copies like you have it now
05:23.52 maths22 OK
05:27.33 brlcad something odd with the sorting (try title sort, look at first two)
05:30.20 brlcad feels like the task title should be selectable -- it is the locus of attention when searching
05:31.29 brlcad always wanted to try cakephp out but never had a reason .. this is pretty impressive progress for one day
05:31.37 brlcad good salespitch ;)
05:36.47 maths22 Will make the title clickable too.
05:37.02 brlcad have you played at all with cappuccino? that's another I've been interested in
05:37.05 maths22 The title sorting issue is in the data-I'm guessing there are some hidden characters
05:37.12 maths22 I have not looked at or heard of cappuccino
05:37.21 brlcad http://www.cappuccino-project.org
05:37.36 brlcad different style, more about building web GUIs
05:38.17 maths22 I will have to take a look later
05:43.20 brlcad there's a space before the word Design in that one title
05:43.29 maths22 OK
05:43.32 brlcad easily found and fixed
05:43.34 maths22 That would do it
05:43.35 maths22 Yep
05:43.45 maths22 I just hadn't bothered looking
05:54.35 maths22 if you copy once, should the title be mutable
05:57.55 maths22 I also have the choice logic working
05:58.03 maths22 1=let you edit
05:58.10 maths22 many = make a bunch
06:12.06 brlcad yeah, mutable is fine, even if there are >1 copies
06:12.40 brlcad just changing that title is going to mess with future cloning
06:13.25 brlcad I might want to make 10 copies, and then tweak all 10 in some small way
06:13.37 brlcad either in the title or in the description
06:14.31 brlcad another common pattern is creating follow-on tasks, but I don't think there's a usability improvement possibility there without live bidirectional syncing because you want the melange url embedded in the description
08:13.19 *** join/#brlcad ries (~ries@D979EA84.cm-3-2d.dynamic.ziggo.nl)
09:15.54 *** join/#brlcad bugport (~bugport@91.103.85.100)
11:24.21 Notify 03BRL-CAD Wiki:NyahCh3ck20 * 7798 /wiki/Google_Code_In/2014: /* Overview */
11:28.42 Notify 03BRL-CAD:starseeker * 63431 brlcad/trunk/misc/CMake/BRLCAD_CMakeFiles.cmake: Add INTERFACE to the TARGET_FLAGS list.
11:29.59 Notify 03BRL-CAD Wiki:NyahCh3ck20 * 7799 /wiki/Community_Publication_Portal: /* Geometry Viewer */
11:31.13 Notify 03BRL-CAD Wiki:NyahCh3ck20 * 7800 /wiki/Community_Publication_Portal: /* What is a geometry viewer? */
11:33.08 Notify 03BRL-CAD Wiki:NyahCh3ck20 * 7801 /wiki/Community_Publication_Portal: /* Geometry Viewer */
11:34.29 Notify 03BRL-CAD Wiki:NyahCh3ck20 * 7802 /wiki/Community_Publication_Portal: /* It’s the equivalent to? */
11:35.31 Notify 03BRL-CAD Wiki:NyahCh3ck20 * 7803 /wiki/Community_Publication_Portal: /* Why is it used? */
11:36.40 Notify 03BRL-CAD:starseeker * 63432 (brlcad/branches/qtged/src/qbrlcad/QAccordianWidget.cxx brlcad/branches/qtged/src/qbrlcad/cadattributes.cxx): Clear a couple warnings.
11:37.43 Notify 03BRL-CAD Wiki:NyahCh3ck20 * 7804 /wiki/Community_Publication_Portal: /* What is needed to access this service? */
11:39.38 Notify 03BRL-CAD Wiki:NyahCh3ck20 * 7805 /wiki/Community_Publication_Portal: /* Advantages */
11:43.13 Notify 03BRL-CAD Wiki:NyahCh3ck20 * 7806 /wiki/Community_Publication_Portal: /* How it works? */
11:54.11 Notify 03BRL-CAD Wiki:NyahCh3ck20 * 7807 /wiki/Deuces: /* Create Numerics library (LIBBN) API unit tests */
11:54.55 Notify 03BRL-CAD:starseeker * 63433 (brlcad/trunk/CMakeLists.txt brlcad/trunk/src/other/tcl/CMakeLists.txt brlcad/trunk/src/other/tk/CMakeLists.txt): Work on getting ready for CMake 3.1. Particularly need to check distcheck with this new setup, with both old and new CMake. Should investigate CMP0053 issues at some point...
11:58.01 *** join/#brlcad deepak_ (~chatzilla@117.199.105.189)
12:01.06 Notify 03BRL-CAD:starseeker * 63434 brlcad/trunk/src/other/stepcode/CMakeLists.txt: Set policy 54 for stepcode as well
13:21.35 *** join/#brlcad teepee- (bc5c2133@gateway/web/freenode/ip.188.92.33.51)
13:52.57 Notify 03BRL-CAD:starseeker * 63435 (brlcad/branches/qtged/src/qbrlcad/QFlowLayout.cxx brlcad/branches/qtged/src/qbrlcad/QFlowLayout.h brlcad/branches/qtged/src/qbrlcad/QToolPalette.cxx): Add vertical and horizontal spacing controls to QFlowLayout
13:55.39 ``Erik heh, ronja mentioned in a hn comment https://news.ycombinator.com/item?id=8606345
14:05.24 Notify 03BRL-CAD:starseeker * 63436 brlcad/branches/qtged/src/qbrlcad/QToolPalette.cxx: More tweaks to the tool palette
14:24.51 *** join/#brlcad ries (~ries@D979EA84.cm-3-2d.dynamic.ziggo.nl)
14:43.19 *** join/#brlcad teepee- (bc5c2133@gateway/web/freenode/ip.188.92.33.51)
14:56.14 *** join/#brlcad kintel (~kintel@unaffiliated/kintel)
15:01.55 *** join/#brlcad clock (~clock@212.203.58.127)
15:08.52 *** join/#brlcad teepee- (bc5c2133@gateway/web/freenode/ip.188.92.33.51)
15:13.33 Notify 03BRL-CAD:carlmoore * 63437 brlcad/trunk/TODO: remove trailing blanks/tabs
15:17.33 Notify 03BRL-CAD:carlmoore * 63438 brlcad/trunk/doc/history.txt: fix a spelling, and what is the meaning of the brackets in ' December 21]], 2004'?
15:18.52 *** join/#brlcad teepee- (bc5c2133@gateway/web/freenode/ip.188.92.33.51)
15:22.54 *** join/#brlcad gaganjyot (~gaganjyot@124.253.230.195)
15:25.28 *** join/#brlcad gaganjyot (~gaganjyot@124.253.230.195)
15:25.37 brlcad mm, lots of interesting news bits on HN today
15:27.39 *** join/#brlcad gaganjyot (~gaganjyot@124.253.230.195)
15:30.09 *** join/#brlcad gaganjyot (~gaganjyot@124.253.230.195)
15:31.29 Notify 03BRL-CAD:brlcad * 63439 brlcad/trunk/doc/history.txt: it was 21st, but doesn't need to be
15:31.39 brlcad nifty: http://tanzer.io
15:31.53 brlcad tcl web server
15:34.22 *** join/#brlcad gagan (~gagan@124.253.230.195)
15:46.20 Notify 03BRL-CAD:starseeker * 63440 (brlcad/branches/qtged/src/qbrlcad/QAccordianWidget.cxx brlcad/branches/qtged/src/qbrlcad/QFlowLayout.h and 3 others): Improve QToolPalette resizing behavior
15:46.34 *** join/#brlcad gagan (~gaganj@124.253.230.195)
15:54.27 Notify 03BRL-CAD:starseeker * 63441 brlcad/branches/qtged/src/qbrlcad/QToolPalette.cxx: Use control heights.
16:30.02 ``Erik hn is the new slashdot, though it's already going the way of slashdot... *cough*
16:52.14 *** join/#brlcad gagan (~gaganj@124.253.230.195)
16:53.11 *** join/#brlcad kintel (~kintel@unaffiliated/kintel)
16:54.28 *** join/#brlcad gagan (~gaganj@124.253.230.195)
16:57.03 *** join/#brlcad gagan (~gaganj@124.253.230.195)
16:58.57 *** join/#brlcad gagan (~gaganj@124.253.230.195)
17:00.03 *** join/#brlcad sofat (~sofat@202.164.45.204)
17:19.45 Notify 03BRL-CAD:carlmoore * 63442 brlcad/trunk/include/icv.h: use 2 semicolons to provide the (presumably-correct) meaning, because commas were not quite as clear a separation
17:21.53 *** join/#brlcad Stragus (~alexis@modemcable090.29-19-135.mc.videotron.ca)
17:26.18 *** join/#brlcad deepak_ (~chatzilla@117.199.97.85)
17:33.56 starseeker ``Erik: so have you develped the successer to hn yet? :-P
17:41.35 *** join/#brlcad kintel (~kintel@unaffiliated/kintel)
18:19.59 brlcad neat: http://setosa.io/ev/
18:20.20 brlcad especially the markov chain code you can modify
18:27.23 *** join/#brlcad kintel (~kintel@unaffiliated/kintel)
18:47.23 Notify 03BRL-CAD:starseeker * 63443 (brlcad/branches/qtged/src/qbrlcad/CMakeLists.txt brlcad/branches/qtged/src/qbrlcad/cadapp.cxx and 2 others): QTextBrowser is able to display the html man pages.
18:50.13 Notify 03BRL-CAD:starseeker * 63444 brlcad/branches/qtged/src/qbrlcad/cadhelp.cxx: Man page viewers are not blockers for the rest of the app
18:52.29 deepak_ brlcad: Hi
18:55.06 brlcad n_reed: starseeker: good read regarding recent discussion about std::string performance: http://www.drdobbs.com/cpp/c-string-performance/184405453
18:57.43 brlcad would be interesting to see where bu_vls compares, but it's conceptually in the middle minimizing heap activity
18:59.00 brlcad his results were an optimized approach was 35x faster than std::string for single threaded and a whooping 2500x faster for a 10-thread multithreaded comparison (avoids locking entirely)
19:22.11 starseeker brlcad: interesting. What's the license?
19:29.31 Notify 03BRL-CAD:starseeker * 63445 (brlcad/branches/qtged/src/qbrlcad/CMakeLists.txt brlcad/branches/qtged/src/qbrlcad/main.cxx): Move the command registration to its own file.
19:32.15 brlcad starseeker: it's just code snippets to teach/test, not a library to be used .. you apply the techniques to ones own code
19:33.49 brlcad more just food for thought, as with any stats .. lots of caveats and gotchas that undboutedly affect this
19:35.34 brlcad different libc's will undoubtedly show hugely different results, wouldn't be surprised if a modern gcc /clang libc weren't faster now (article is 10 years ago)
19:37.32 deepak_ I'm trying to work with OGV. After cloning it from the BRL-CAD repo I don't find any instructions in ReadME file to install and use it. Please guide me with the steps to do the same.
19:40.55 brlcad deepak_: I don't know the installation steps either, did you try to just put it on a server and see if it loads?
19:41.42 brlcad deepak_: you should ask on the mailing list, there are a couple people that have it set up successfully ... you should write a README or INSTALL file ;)
19:41.52 brlcad great patch submission ;)
19:44.36 deepak_ brlcad: Okay, I'm proceeding and try to install it on my server. If something went wrong I'll be reporting on our mailing list :).
19:52.14 Notify 03BRL-CAD:starseeker * 63446 (brlcad/branches/qtged/src/qbrlcad/cadapp.h brlcad/branches/qtged/src/qbrlcad/cadcommands.cxx): Add the commands listed in mged's setup.c. Most of them won't do what they're supposed to, of course, and Archer has some that are missing from here, but it's a start.
20:06.01 deepak_ brlcad: One thing more, I have read certain html files of OGV and I found that at some places coding standard are violated eg: after <div> next <div> is represented by the space of 2 instead of 4. Is it the violation of coding standard? I couldn't find this in hacking file.
20:06.02 brlcad deepak_: okie dokie
20:06.59 brlcad deepak_: we don't yet have published coding standards for web apps, but it's more important that it be consistent within that file
20:14.51 deepak_ brlcad: Okay :D.
20:29.49 Notify 03BRL-CAD:carlmoore * 63447 brlcad/trunk/doc/docbook/system/man1/en/pix-bw.xml: touch up man page for pix-bw; why were hardwired 1.0's present for -R, -G-, -B?
20:34.10 Notify 03BRL-CAD:starseeker * 63448 brlcad/branches/qtged/src/qbrlcad/cadcommands.cxx: Make a stab at listing commands in Archer but not MGED. For this sort of thing to be really comprehensive, we'll also have to check the various tcl script directories.
20:36.32 brlcad that list isn't quite accurate .. many/most of those are also in mged
20:36.49 brlcad either as built-in or tclscript commands, just not in the callback table list
20:37.40 brlcad cd and pwd are tcl system commands
20:42.28 brlcad maths22: this is the form I set up last year: https://docs.google.com/forms/d/1l5berjknWtj29zkHJ0q8jLUzPMBYCWYfkv97qAGirZI/viewform
20:43.09 brlcad by breaking up the description into four separate fields, consistent formatting can be applied
20:43.13 brlcad auto-applied
21:10.47 Notify 03BRL-CAD:starseeker * 63449 (brlcad/branches/qtged/src/qbrlcad/cadapp.cxx brlcad/branches/qtged/src/qbrlcad/cadapp.h): Rework command execution flow a bit.
21:10.54 starseeker brlcad: yeah, figured
21:11.20 starseeker mostly was trying to get a sense of how many commands were/are "wired" into the GUI in various ways
21:12.09 starseeker hasn't even looked at Tcl yet - dealing in straight libged argc/argv calls for the moment to avoid introducing complexities
21:13.41 maths22 brlcad: thanks
21:13.49 starseeker really, if we added a uniform ability to handle matching characters (ls name*, for example) at the libged level things would just about work until the user wants to actually write tcl scripts...
21:14.02 maths22 I can make a custom add view (something like tasks/add-brlcad)
21:14.08 maths22 taht would have these sors of reqtures
21:37.41 brlcad starseeker: yep, globbing and such should definitely happen lower-level imo too
21:57.50 starseeker brlcad: do we still have that list of missing man pages? I'm thinking for GCI tasks we might have them stub in basic pages (command name, maybe args, and hook it into the build) to get some templates set up, if that still needs doing...
21:58.22 *** join/#brlcad andrei_ (~andrei@188.25.162.16)
22:10.50 *** join/#brlcad KimK (~Kim__@ip68-102-30-143.ks.ok.cox.net)
22:53.38 Notify 03BRL-CAD:starseeker * 63450 (brlcad/branches/qtged/src/qbrlcad/cadapp.cxx brlcad/branches/qtged/src/qbrlcad/cadapp.h and 7 others): Make the search command full-path outputs active links into the tree
22:58.18 Notify 03BRL-CAD:starseeker * 63451 brlcad/branches/qtged/src/qbrlcad/console.cxx: Make sure the command isn't empty
23:00.52 Notify 03BRL-CAD:carlmoore * 63452 (brlcad/trunk/doc/docbook/system/man1/en/pix-bw.xml brlcad/trunk/src/util/pix-bw.c): fix an item in usage; fix wrong use of 'pixfade'; add 'no color scheme specified' (this can be changed later to some sort of default); add -o to manpage, but that is subject to removal
23:04.38 *** join/#brlcad FreezingCold (~FreezingC@135.0.41.14)
23:58.28 *** join/#brlcad ries (~ries@D979EA84.cm-3-2d.dynamic.ziggo.nl)

Generated by irclog2html.pl Modified by Tim Riker to work with infobot.