Skip to end of metadata
Go to start of metadata

[08:22:52 CDT(-0500)] <jhung> hey justin_o: I'm going to be working on Training Materials documentation today and setting up a 12.04 VM for testing Deca 0.5

[08:23:36 CDT(-0500)] <Justin_o> jhung: cool.. i'm also doing some quick testing of 12.04.. just to see if the install works though

[08:24:10 CDT(-0500)] <Justin_o> jhung: for 12.04 32-bit version it seemed to install okay but the unit tests failed when calling genpdf … not sure what caused that though..

[08:24:16 CDT(-0500)] <Justin_o> i'm going to try in a 64-bit environment now

[08:24:25 CDT(-0500)] <jhung> k.

[08:29:06 CDT(-0500)] <Justin_o> jhung: did we need to talk more about the decapod roadmap today.. i guess we still need to do tasking

[08:29:53 CDT(-0500)] <jhung> Yes justin_o, but perhaps before we do that we should schedule time with Jessm to dicuss our changes.

[08:30:04 CDT(-0500)] <Justin_o> jhung: makes sense

[08:30:29 CDT(-0500)] <jhung> jessm: do you have time today to talk about our changes in Decapod roadmap? In particular we don't want to drop anything that was specified in the grant.

[08:32:46 CDT(-0500)] <jessm> jhung: sure

[08:32:53 CDT(-0500)] <jessm> jhung: give me 10 min?

[08:32:59 CDT(-0500)] <jhung> jessm: sure

[08:33:10 CDT(-0500)] <jhung> just ping us jessm.

[08:34:24 CDT(-0500)] <Justin_o> jhung: looking over the genpdf spec

[08:35:12 CDT(-0500)] <Justin_o> there it says that the input list should be specified by "-l" but in reality it isn't used

[08:35:13 CDT(-0500)] <jhung> justin_o: yep. I was looking it over earlier too.

[08:35:29 CDT(-0500)] <Justin_o> is that a bug in the spec or the script or should it be optional?

[08:35:51 CDT(-0500)] <jhung> justin_o: Yes I noticed that.

[08:36:54 CDT(-0500)] <jhung> justin_o: I prefer the -l parameter followed by a list. It's more strict. Less room for error. Currently you can put image files where ever in the command line and it would use it.

[08:37:24 CDT(-0500)] <jhung> It's a bit confusing, especially if the user did something unintentional.

[08:39:43 CDT(-0500)] <Justin_o> jhung: okay… so i'm thinking about this test plan i'm writing.. originally i was doing it for decapod-0.5 but if i follow the spec and know things work differently I don't know if it makes sense.. so i think we have a few options 1) make this for 0.5 still and we can just consider all these bugs and the tests will all fail 2) make an 0.5 and 0.6 test plan with different expectations on input and output 3) just make a test plan for a futu

[08:39:43 CDT(-0500)] <Justin_o> release.. possibly just against the spec and we can use it to determine when things are fully working

[08:39:50 CDT(-0500)] <Justin_o> jhung: any preference?

[08:41:15 CDT(-0500)] <jhung> justin_o: I think we should have a single test that tests the expected final version. The command line is nearly stable anyway.

[08:42:12 CDT(-0500)] <Justin_o> jhung: okay.. i'll go with that.. so i'll tests cases for all the various command line options

[08:43:40 CDT(-0500)] <jhung> Justin_o: yes. But perhaps we should chop off some of the features? i.e. inches and upscale?

[08:45:12 CDT(-0500)] <Justin_o> jhung: sure.. i guess we should drop them from the spec or have them in some future section

[09:15:53 CDT(-0500)] <Justin_o> jhung, jessm: i'm back now

[09:15:56 CDT(-0500)] <Justin_o> ready when you all are

[09:16:03 CDT(-0500)] <jessm> k, let's do it

[09:36:57 CDT(-0500)] <jhung> justin_o: how useful will it be to have the -f / --force flag to overwrite directories and files created by genpdf?

[09:38:11 CDT(-0500)] <Justin_o> jhung: i guess it prevents errors from old files lingering around

[09:38:23 CDT(-0500)] <Justin_o> i guess it depends on how permanent we want those things to be

[09:38:42 CDT(-0500)] <Justin_o> if for example we want the old ones hanging around, it would be useful for the server

[09:44:17 CDT(-0500)] <jhung> Yeah. I guess we can drop the feature since we can always write our own command to delete old files.

[09:44:42 CDT(-0500)] <jhung> I recall Hasan saying that we can resume the pipeline if we knew where it failed. So having the old files could be useful.

[09:44:47 CDT(-0500)] <jhung> ^justin_o

[09:44:59 CDT(-0500)] <Justin_o> true

[09:45:52 CDT(-0500)] <jhung> okay. Let's drop the -f feature.

[09:45:55 CDT(-0500)] <jhung> Cool?

[09:46:02 CDT(-0500)] <Justin_o> jhung: sure

[09:46:05 CDT(-0500)] <Justin_o> anything else?

[09:46:31 CDT(-0500)] <jhung> So I guess we'll drop the -f, upscale, and inches features.

[09:46:37 CDT(-0500)] <Justin_o> jhung: what about trim

[09:46:40 CDT(-0500)] <jhung> disabling deskew may still be useful.

[09:46:53 CDT(-0500)] <jhung> yes. Trim to fit is important.

[09:46:54 CDT(-0500)] <Justin_o> jhung: trim seems similar to upscale

[09:47:21 CDT(-0500)] <jhung> Not exactly justin_o.

[09:47:34 CDT(-0500)] <Justin_o> does trim happen before processing?

[09:48:07 CDT(-0500)] <jhung> trim-to-fit ensure there is no empty space generated when placing the image on say an A4 page. So we effectively crop that page.

[09:48:15 CDT(-0500)] <Justin_o> jhung: as for deskew i guess so for cases where the text is supposed to be slightly skewed

[09:48:40 CDT(-0500)] <Justin_o> jhung: now i'm confused again.. so what does upscale do

[09:49:05 CDT(-0500)] <jhung> Upscaling is just increasing width and height to fit the page while maintaining the original aspect ratio. Can still result in empty space.

[09:50:28 CDT(-0500)] <jhung> Like scaling a 2x3 wallet to fit into a 4x5 frame. It isn't going to upscale exactly to fit the 4x5.

[09:52:03 CDT(-0500)] <jhung> justin_o: about deskew, yes. Purely to match the nature of the original.

[09:52:23 CDT(-0500)] <jhung> turning deskew off will have ramifications for OCR though I think.

[09:53:13 CDT(-0500)] <Justin_o> jhung: hmm

[09:53:40 CDT(-0500)] <Justin_o> jhung: so it's only really usefull for image pdf?

[09:57:56 CDT(-0500)] <jhung> Well, it may be useful for other types too if somehow we can OCR on a deskew to get the processing done, and then rotate the genpdf results to match the original?

[09:58:03 CDT(-0500)] <jhung> I don't know if that's useful.

[09:58:51 CDT(-0500)] <Justin_o> depends on the rotation i guess.. since they won't line up at all

[09:58:57 CDT(-0500)] <jessm> hey everyone: i'm going to miss standup today

[09:59:00 CDT(-0500)] <Justin_o> unless we skew the overlay back

[09:59:06 CDT(-0500)] <Justin_o> jhung: ^

[09:59:15 CDT(-0500)] <Justin_o> jessm: thanks for the warning

[10:00:29 CDT(-0500)] <jhung> justin_o: yeah. Seems more trouble than it's worth.

[10:00:38 CDT(-0500)] <Justin_o> jhung: a bit yes

[10:00:47 CDT(-0500)] <Justin_o> that could be a more advanced feature to add later

[10:01:05 CDT(-0500)] <jhung> justin_o: I guess since the user disables deskew, they would probably want to preserve the original nature of content. So type 3+ doesn't make sense.

[10:01:25 CDT(-0500)] <jhung> But not sure if someone expects to do type 2 with deskew disabled.

[10:01:36 CDT(-0500)] <Justin_o> jhung: hard to tell

[10:01:45 CDT(-0500)] <Justin_o> we could provide that option only for type 1

[10:02:04 CDT(-0500)] <jhung> Yeah. I guess let's file it low for now.

[10:02:37 CDT(-0500)] <Justin_o> jhung: can we put it in the future section.. i don't want to add a test for it if we don't think we'll get to it

[10:03:10 CDT(-0500)] <jhung> sure

[10:03:50 CDT(-0500)] <jhung> justin_o: updated.

[10:04:52 CDT(-0500)] <Justin_o> jhung: did you update the wiki.. it still looks the same to me?

[10:05:04 CDT(-0500)] <jhung>

[10:05:05 CDT(-0500)] <jhung> ?

[10:05:37 CDT(-0500)] <jhung> List at top has been updated. I will update the tables now.

[10:05:43 CDT(-0500)] <Justin_o> oh okay

[10:05:46 CDT(-0500)] <Justin_o> i was looking at the table

[10:05:49 CDT(-0500)] <Justin_o> jhung: ^

[10:08:00 CDT(-0500)] <Justin_o> avtar: are all the vms created with vagrant headless?

[10:08:34 CDT(-0500)] <avtar> Justin_o: by default, yes

[10:08:45 CDT(-0500)] <jhung> justin_o: updated tables.

[10:08:48 CDT(-0500)] <Justin_o> avtar: is there a way to change that

[10:09:44 CDT(-0500)] <jhung> justin_o: I'll send an email to Hasan about dropping those features.

[10:10:12 CDT(-0500)] <Justin_o> jhung: thanks

[10:11:58 CDT(-0500)] <avtar> Justin_o: you'll need to add this to your Vagrantfile

[10:11:59 CDT(-0500)] <avtar> config.vm.boot_mode = :gui

[10:12:41 CDT(-0500)] <Justin_o> avtar: that seems pretty simple thanks

[10:13:07 CDT(-0500)] <Justin_o> avtar: also am i supposed to shut these vm's down or just exit the ssh session

[10:13:45 CDT(-0500)] <avtar> if you want to save the state of the vm then you should type 'vagrant halt'

[10:14:30 CDT(-0500)] <avtar> otherwise 'vagrant destroy' can be used to kill the process and wipe out the vm's disk

[10:15:07 CDT(-0500)] <avtar> you'll need to type that in the same directory where your Vagrantfile is located

[10:17:09 CDT(-0500)] <Justin_o> avtar: okay.. so vagrant destroy will restore it to a stock instance?

[10:17:21 CDT(-0500)] <Justin_o> or remove it completely and i'll have to vagrant init again?

[10:22:13 CDT(-0500)] <Justin_o> avtar: hmm.. it's asking for a login and password when i boot with the gui.. any idea what that will be

[10:22:26 CDT(-0500)] <avtar> Justin_o: vagrant/vagrant

[10:23:24 CDT(-0500)] <Justin_o> avtar: that was obvious.. i should have tried that..

[10:23:47 CDT(-0500)] <avtar> also, 'vagrant destroy' won't exactly restore it to a stock instance, just remove the newly created environment off your drive. to bring it up again you'll have to type 'vagrant up'

[10:24:10 CDT(-0500)] <avtar>

[10:24:23 CDT(-0500)] <Justin_o> but it looks like it doesn't have a window manager running.. i might just try the port forwarding for a quick test

[10:25:06 CDT(-0500)] <avtar> i haven't used suspend because i'm usually testing small changes but looks like that might be useful if you want to avoid booting up

[10:26:12 CDT(-0500)] <Justin_o> avtar: thanks.. that cleared things up for me

[11:17:44 CDT(-0500)] <logiclord_> yura2 : I wanted to discuss about hierarchy ?

[11:17:50 CDT(-0500)] <yura2> logiclord_: sure

[11:18:19 CDT(-0500)] <logiclord_> yura2 : I sent you an email

[11:18:50 CDT(-0500)] <yura2> logiclord_: ya , let me open it up

[11:20:31 CDT(-0500)] <yura2> logiclord_: ok got it

[11:20:46 CDT(-0500)] <yura2> so what in particular did you want to discuss ?

[11:21:15 CDT(-0500)] <logiclord_> yura2 : User Interface Options and WYSIWYG Wrapper

[11:21:47 CDT(-0500)] <logiclord_> I have kept User Interface Options to change reader UI and not epub file

[11:22:08 CDT(-0500)] <logiclord_> while WYSIWYG for editing book content

[11:23:35 CDT(-0500)] <logiclord_> this is a bit different to mockups that I have created earlier on where I used WYSIWYG only

[11:24:23 CDT(-0500)] <logiclord_> in #4 region

[11:25:47 CDT(-0500)] <logiclord_> so is it okay to keep these too separately ? ... user might think that he is doing extra work

[11:26:30 CDT(-0500)] <logiclord_> e.g. first changing viewing font in interface option and then in WYSIWYG to edit book

[11:26:31 CDT(-0500)] <logiclord_> ?

[11:26:45 CDT(-0500)] <yura2> logiclord_: i actually like that you separated those into different things

[11:26:52 CDT(-0500)] <yura2> logiclord_: one is clearly about presentation of the book

[11:27:06 CDT(-0500)] <yura2> logiclord_: the other one is for editing the content of the book

[11:27:55 CDT(-0500)] <logiclord_> yes that was idea but from user's perspective ?

[11:28:27 CDT(-0500)] <yura2> ya, exactly, i think it is clearer for the user

[11:28:43 CDT(-0500)] <logiclord_> okay

[11:29:23 CDT(-0500)] <yura2> logiclord_: parser, in my opinion, is a bit cluttered, parsing indexing, rendering the book will probably be separated into its own respected subcomponents

[11:30:18 CDT(-0500)] <logiclord_> I just wanted to confirm about allowing a sub-component like index/bookmarks/notes using parser and navigator ?

[11:31:32 CDT(-0500)] <logiclord_> yura2 : btw I have found some useful parsing libraries at

[11:31:54 CDT(-0500)] <logiclord_> can we use them.. in fact we can divide parser on similar lines

[11:32:15 CDT(-0500)] <logiclord_> like unzip, inflate, etc ?

[11:33:04 CDT(-0500)] <yura2> logiclord_: yes , you can totally start using them (make sure you evaluate them first though (smile))

[11:36:36 CDT(-0500)] <logiclord_> yura2 : and about allowing a component down in hierarcy invoking higher one ? I mean about interaction of index with parser ?

[11:38:01 CDT(-0500)] <yura2> logiclord_: yes once you start developing using Infusion you will see that this is exactly the kind of thing that Inversion of Control is useful for exactly such things

[11:39:17 CDT(-0500)] <logiclord_> okay then I suppose I should finish hierarchy by tomorrow and create a development plan

[11:40:40 CDT(-0500)] <yura2> logiclord_: that would be great, given that you created a github repo, you can open tickets for the tasks you will be workin on, ill try to comment on them as you start adding them. Also I pasted your preliminary architecture sketch into github's wiki

[11:46:09 CDT(-0500)] <logiclord_> yura2 : okay I will get acquainted with Github and open tickets

[11:46:39 CDT(-0500)] <yura2> logiclord_: sweet

[11:47:19 CDT(-0500)] <logiclord_> I was brushing up my js concepts by reading JS : good parts and I found

[11:48:02 CDT(-0500)] <logiclord_> any idea about how strictly should we follow its rules ?

[11:49:08 CDT(-0500)] <yura2> jslint is a bit too strict

[11:49:11 CDT(-0500)] <yura2> try jshint (smile)

[11:49:16 CDT(-0500)] <yura2> should be good enough

[11:49:18 CDT(-0500)] <yura2> i use it

[11:50:27 CDT(-0500)] <logiclord_> okay then I will update hierarchy and post on blog and wiki.. open tickets as well (smile)

[11:53:09 CDT(-0500)] <yura2> logiclord_: thanks

[13:10:27 CDT(-0500)] <clown> jameswy: FYI:

[13:14:21 CDT(-0500)] <jameswy> clown: Thanks

[13:14:54 CDT(-0500)] <clown> wlcm

[13:41:03 CDT(-0500)] <Justin_o> fluid-everyone: the community meeting will be starting shortly, does anyone want to join remotely?

[14:14:44 CDT(-0500)] <greggy1> Justin_o: is there a jquery update planned for the next release?

[14:19:10 CDT(-0500)] <greggy1> Justin_o: of Infusion, that is

[15:42:28 CDT(-0500)] <Justin_o> greggy1: hello.. if i'm not mistaken we've already started to make some changes in the codebase to accommodate some of the changes in the newer versions of jQuery. Also we typically pick up jQuery updates when we release. That being said, we do not yet have a date for our next release.

  • No labels