[13:45:45] <tonythomas>	 Niharika: last time I put something less rude, some applicant reported it to Marina :P
[13:45:58] <Niharika>	 :D
[13:45:59] <tonythomas>	 telling that I just kicked them out of Outreachy or something like that ;\
[13:46:19] <Niharika>	 Meh. I don't care.
[13:46:42] <tonythomas>	 Niharika: true. they might think I am kinda softey :\
[13:46:52] <Niharika>	 tonythomas: You are! :D
[13:47:10] <tonythomas>	 Niharika: comm'on. :\ I think you are reading my mails to the Mailing List
[13:47:47] <Niharika>	 The one about having a program to recruit volunteer devs before GSoC/Outreachy?
[13:48:07] <tonythomas>	 Niharika: exactly :D
[13:48:32] <tonythomas>	 Niharika: I really want that to happen, but it require more of me.
[13:48:54] <Niharika>	 tonythomas: I completely agree that there is a gap there we need to fill but I don't think anyone has the bandwidth to take on that extra work that needs though.
[13:48:59] <Niharika>	 Yeah.
[13:49:29] <Niharika>	 Also it will need some sort of incentive for people to get involved in it.
[13:49:34] <tonythomas>	 Niharika: true. I was just shaken at the way FOSSASIA does it - they just do not put in like a lot of extra effort
[13:49:50] <tonythomas>	 yeah - like maybe a grand prize and couple of T-shirts :D
[13:50:40] <Niharika>	 Yeah. I guess the main problem is organizing resources.
[13:51:33] <tonythomas>	 Niharika: true. once we have that, a simple website, and vroooom.
[13:52:11] <tonythomas>	 its kinda far fetched, but might be a huge hit (specially for Universities like ours who are hunting here and there for programs like this :P)
[13:52:57] <tonythomas>	 Anyway, lets see how things roll
[13:53:31] <tonythomas>	 Niharika: and did you hear that they are changing the GSoC structure -- like the payments are going to be cost-of-living dependant
[13:53:39] <Niharika>	 tonythomas: I think it sounds way simpler than it will be. Like, there will be students trying to cheat the system. And also this program will run just before GSoC/Outreachy which means it will coincide with exams most likely.
[13:53:46] <tonythomas>	 and the first 500$ do not go to the student after getting selected
[13:53:52] <Niharika>	 tonythomas: Oh? I didn't hear about that.
[13:54:04] <Niharika>	 That's a good change?
[13:54:06] <tonythomas>	 they are making it like 3 payments. after each month of work - and that is really good.
[13:54:12] <Niharika>	 Yeah.
[13:54:28] <Niharika>	 Will Indians be getting lesser money though? It seems likely...
[13:54:35] <tonythomas>	 Niharika: the cost of living based - I think so - they were telling something like 1300 USD for Indians :-( there were loads of unreset in the crowd
[13:54:46] <tonythomas>	 I dont know, what are the effects.
[13:54:46] <Niharika>	 Woah.
[13:54:55] <Niharika>	 It's still a lot of money though.
[13:55:00] <tonythomas>	 probably a lot of IITians and IIITians would be heartbroken
[13:55:17] <tonythomas>	 I hope there is none from ^^ here in this channel
[13:55:26] <tonythomas>	 btw, is this logged ? :o :\\ what did I do
[13:55:33] <Niharika>	 :D
[13:55:49] <Niharika>	 poke wm-bot2
[13:55:58] <Niharika>	 Not sure if it logs.
[13:56:40] <tonythomas>	 Niharika: well. I cannot agree with 1300, but maybe 2250 (1.5 Lakh INR) is a nice bet!
[13:56:45] <tonythomas>	 like its 3 months work, right
[13:57:56] <tonythomas>	 anyway. we can see a possible shift from money to code in the future Summer of Code rounds, which is great
[13:58:28] <Niharika>	 Yeah.
[14:00:18] <andre__>	 It logs.
[14:00:21] <andre__>	 http://bots.wmflabs.org/~wm-bot/logs/%23wikimedia-devrel/20161109.txt
[14:00:23] <tonythomas>	 Niharika: oh - coinciding with exams were something I never found out
[14:01:02] <tonythomas>	 andre__: oh no :\
[14:01:36] <andre__>	 I don't see anything bad you said, sorry.
[14:02:03] <tonythomas>	 andre__: yeah. thats the point. Btw, one of the prospective GCI student actually started working on one of the tasks
[14:02:29] <tonythomas>	 he told he wanted to make himself cool with the project so that he can fix more during the program
[14:03:29] <andre__>	 I'm aware of a problem here and currently thinking how to deal with tasks that might have been already worked on before GCI starts and handed in when GCI starts. :-/
[14:03:40] <andre__>	 Not an easy one as we strive to work and plan in the open.
[14:03:57] <tonythomas>	 yeah. this is hard.
[14:04:19] <tonythomas>	 like they would have the pathc somewhere ready, and then just push it
[14:04:27] <tonythomas>	 I removed the GCI tag from that task though
[14:04:35] <tonythomas>	 to make sure ^^ do not happen
[14:04:37] <andre__>	 If students claim a task in Phab before GCI I'd remove it from GCI. If students don't claim in Phab and only claim in GCI they might have the race condition (hah, literally used) that someone else claimed it earlier.
[14:04:42] <andre__>	 I saw that, thanks.
[14:04:56] <andre__>	 So for the latter case, it would be their risk.
[14:05:09] <tonythomas>	 andre__: but I need to double check if you havent uploaded the thing to GCI
[14:06:18] <andre__>	 Plus I'm expecting some usual complaints "why didn't you review this in 2h? I want to claim more tasks!" and my usual "mentors have 36h to review" answer. And we are totally free to slow down things a little bit, by using 24-36h, if we wanted to.
[14:06:53] <andre__>	 (pinging srishakatux as the last 5min are relevant for GCI ^ )
[14:07:11] <andre__>	 tonythomas, I myself have not uploaded anything yet. I'm lazy, err, busy, I mean! :P
[14:08:07] <tonythomas>	 andre__: that is great.
[14:08:15] <tonythomas>	 so do the mentors know they are supposed to upload the tasks ?
[14:09:57] <andre__>	 tonythomas, not explicitly. It's very welcome though
[14:10:09] <andre__>	 (I rather see myself doing that on a weekend in 10days for those tasks not in yet)
[14:10:42] <andre__>	 it's the tradeoff regarding making it comfortable for mentors and having 'good' tasks also left after GCI
[14:10:58] <tonythomas>	 andre__: well, you can tell them that - maybe ask people to create a Google-Code-In header in their project, and import it themeselves (is it just me, or because I'm in Germany - where all is Do-It-Yourself)
[14:12:20] <tonythomas>	 btw, we have this API now right :o
[14:13:52] <andre__>	 yeah... I've never played with it so far :(
[14:14:06] <tonythomas>	 andre__: I found the export thingy!
[14:14:10] <tonythomas>	 check here https://phabricator.wikimedia.org/maniphest/query/authored/
[14:14:12] <tonythomas>	 bottom
[14:14:15] <andre__>	 for the logs, I refer to https://developers.google.com/open-source/gci/api
[14:14:19] <tonythomas>	 Export (Disabled)
[14:14:36] <andre__>	 That's Excel/CSV export in Phabricator
[14:14:50] <andre__>	 not sure if relevant
[14:14:57] <tonythomas>	 andre__: once we have the CSV, querying through each row and pushing via the API is like 15 lines of python
[14:15:29] <tonythomas>	 if we can get the CSV of the google-code-in project, then its a go go
[14:15:55] <andre__>	 we investigated enabling it, but it requires a not-too-small package to install, with hundreds of lines of code that would need to get security reviewed
[14:16:23] <tonythomas>	 gawd. thats gross
[14:17:04] <tonythomas>	 in the worst case, we will have to write our own script which can take individual stuffs in https://phabricator.wikimedia.org/project/view/2304/
[14:17:15] <tonythomas>	 I'm entirely jobless right now.
[14:18:18] <andre__>	 Tss. Do you want to become a GCI admin? :D
[14:19:05] <tonythomas>	 andre__: I would love to, but not this time andre__, I do not have a stable internet connection working at my hostel yet :-( getting a contract takes like 3 weeks in this place
[14:19:13] <tonythomas>	 but I would love to try out the script
[14:19:40] <andre__>	 (well, GCI starts in 3 weeks :D )
[14:19:53] <andre__>	 script: I don't have a good idea or concept currently :-/
[14:20:15] <tonythomas>	 andre__: dont worry. let me just see what my little python expertise can do for you
[14:20:21] <andre__>	 thanks
[14:20:57] <tonythomas>	 andre__: I will let you know in another week, as we have applied for a contract like 2 weeks before.
[14:24:19] <andre__>	 good luck!
[14:25:56] <tonythomas>	 andre__: danke.
[16:34:55] <tonythomas>	 andre__: the script is half ready, btw :D
[16:35:01] <tonythomas>	 phabricator is making it more and more diff
[16:35:13] <tonythomas>	 I have all the links in the first column now
[16:35:20] <tonythomas>	 now need to open up each one, and extract the details
[16:37:35] <tonythomas>	 andre__: it would be here - https://github.com/tonythomas01/phabgciscrapbot
[16:42:43] <andre__>	 whou. thank you!
[16:43:27] <tonythomas>	 andre__: not yet. dont know if the API would coperate
[16:44:13] <tonythomas>	 andre__: so there are two approaches - tell me what you want
[16:44:52] <andre__>	 did you intentional upload a pyc file?
[16:45:12] <tonythomas>	 I can manually scrap them all, and export them into a CSV file, from where someone should add these fields (mentors, tags, is_beginner, time_to_complete)
[16:45:20] <tonythomas>	 andre__: I was lazy to remove that :\
[16:45:22] <tonythomas>	 damn
[16:45:24] <andre__>	 haha
[16:45:32] <andre__>	 interesting, so you really scrap instead of using the conduit api
[16:45:55] <tonythomas>	 andre__: yeah. I do not have access, so I am scarpping :\
[16:46:09] <andre__>	 tonythomas, how do you mean you do not have access?
[16:46:19] <tonythomas>	 I think that is the only way, as there is no way to automatically get those things from teh phab task
[16:46:56] <andre__>	 https://phabricator.wikimedia.org/conduit/method/maniphest.info/ ?
[16:47:11] <andre__>	 after https://phabricator.wikimedia.org/conduit/method/maniphest.query/
[16:48:12] <tonythomas>	 argh. I couldnt find a proper documentation of these things anywhere
[16:48:38] <tonythomas>	 so my plan was -> download all datat to a sheet, from where these extra details can be added
[16:48:50] <tonythomas>	 and then write a second script, which would parse that sheet, and put that data to GCI
[16:48:51] <tonythomas>	 what say ?
[16:49:23] <andre__>	 I'm not a programmer. :D All I can say is that there is the Conduit API if you want to parse JSON.
[16:49:39] <andre__>	 and it allows you to pass the queryKey of a query that you could construct in the web UI
[16:49:47] <andre__>	 see https://secure.phabricator.com/book/phabricator/article/conduit_search/
[16:51:04] <tonythomas>	 andre__: https://secure.phabricator.com/T5024  -- this one
[16:51:10] <tonythomas>	 we need specifically data in first column
[16:51:43] <tonythomas>	 or https://secure.phabricator.com/T7907
[16:51:52] <tonythomas>	 I think that feature was not implemented
[16:52:03] <andre__>	 tonythomas, you could manually exclude those few task IDs that are not in the first column?
[16:52:19] <andre__>	 and that can already be done when constructing the query in the web ui
[16:52:53] <tonythomas>	 andre__: wait a min, are we talking about the first part (getting all the tasks in first column) here ?
[16:53:01] <andre__>	 yes
[16:53:05] <andre__>	 ah darrrn no. you can only include certain IDs but not exclude certain IDs
[16:53:21] <tonythomas>	 andre__: yeah. that was why I had to manually go through :\
[16:53:40] <tonythomas>	  now once I have this ID, I might be able to use this Conduit API
[16:53:44] <tonythomas>	 to actually get the data
[16:53:58] <tonythomas>	 (I have the ID AFAIK)
[16:54:54] <andre__>	 still: exclude those few IDs that are not in the first column?
[16:55:53] <andre__>	 (locally, after receiving the results of that api query, of course,)
[16:55:59] <andre__>	 at leastr that would be my idea
[16:56:53] <tonythomas>	 andre__: yeah - I am doing that actually, my script actually throws out only results from first column
[16:57:02] <tonythomas>	 and I think I found a conduit API thingy to get data too
[16:57:05] <tonythomas>	 let me try running that
[16:57:06] <andre__>	 heh :)
[16:58:22] <tonythomas>	 wtf. there was https://pypi.python.org/pypi/phabricator
[16:59:55] <tonythomas>	 let me try calling it.
[17:01:37] <tonythomas>	 andre__: can you reveal the conduit API url to me from your settings page ? maybe PM ?
[17:03:07] <andre__>	 tonythomas, huh?!
[17:03:54] <andre__>	 tonythomas, have you scrolled to the very bottom of e.g. https://phabricator.wikimedia.org/conduit/method/project.query/ ? :)
[17:04:18] <tonythomas>	 I see that, they are using some kinda package. and sadly there is no python thingy there
[17:04:34] <tonythomas>	 let me try with https://phabricator.wikimedia.org/api/maniphest.query
[17:04:39] <andre__>	 tonythomas, what "kinda package"?
[17:04:50] <andre__>	 Do you refer to using Arcanist? If so, follow https://www.mediawiki.org/wiki/Phabricator/Differential
[17:04:52] <tonythomas>	 like see that PHP - require_once 'path/to/libphutil/src/__phutil_library_init__.php';
[17:05:05] <tonythomas>	 ah. I see now
[17:05:12] <tonythomas>	 the CURL makes it clear
[17:05:12] <andre__>	 libphutil is a core library required by Phabricator. but that's on the server. :)
[17:05:21] <tonythomas>	 https://phabricator.wikimedia.org/api/maniphest.query  is the one I am looking for
[17:05:28] <tonythomas>	 true
[17:05:30] <andre__>	 ah :)
[17:10:48] <tonythomas>	 andre__: I found that I should be giving ["150286"] instead of ["T150286"] -- and I have the data :)
[17:10:48] <stashbot>	 T150286: Integrate SwaggerUI with the service template - https://phabricator.wikimedia.org/T150286
[17:17:17] <andre__>	 shaweeet!
[17:23:53] <tonythomas>	 andre__: well, I'm having trouble with the conduit key now
[17:23:53] <tonythomas>	 few more minutes :\
[17:24:22] <andre__>	 if I can help, provide more details :)
[17:24:39] <andre__>	 Last time I played with arc, following https://www.mediawiki.org/wiki/Phabricator/Differential was extremely smooth for me at least
[17:24:48] <andre__>	 if it's not smooth for you, maybe we can improve the docs? :)
[17:26:42] <tonythomas>	 andre__: fixed.
[17:26:45] <tonythomas>	 damn
[17:26:54] <tonythomas>	 andre__: https://secure.phabricator.com/T11538#193548 helped
[17:27:14] <tonythomas>	 andre__: now we got all the stuff
[17:27:22] <tonythomas>	 we just need to import them to maybe a csv
[17:27:28] <tonythomas>	 and you can add the other details
[17:27:32] <tonythomas>	 and then we can write one more script
[17:27:32] <tonythomas>	 :D
[17:27:34] <tonythomas>	 looks good ?
[17:28:00] <tonythomas>	 I can run that on all that on the first column, and maybe make a google doc and share with you, if you want
[17:28:28] <andre__>	 as far as I understand and can follow: looks good, yeah. Thanks for playing with this!
[17:28:54] <tonythomas>	 andre__: it was fun.
[17:29:09] <tonythomas>	 now let me pick only the required fields, as its returning too much
[17:35:28] * andre__ has to prepare a meeting and then do 'real life' stuff
[17:36:09] <tonythomas>	 andre__: alright. I will share this thing with you soon.
[18:02:04] <andre__>	 srishakatux: around?
[18:02:40] <tonythomas>	 andre__: btw. it worked out, I have the csv. sharing with you now
[18:03:44] <andre__>	 yay! *hug*
[18:04:13] <tonythomas>	 :) it was a brush up in python for me too
[18:13:33] <tonythomas>	 and https://github.com/tonythomas01/phabgciscrapbot is complete. I got to leave my lab. ciao
[18:13:38] <tonythomas>	 mailed to you, btw
[19:05:58] <andre__>	 wh00t. Awesome!
[19:06:10] * andre__ needs to take a look tomorrow