| *** tpb has joined #melange | 00:00 | |
| -kornbluth.freenode.net- [freenode-info] channel trolls and no channel staff around to help? please check with freenode support: http://freenode.net/faq.shtml#gettinghelp | 00:00 | |
| *** ArneBab_ has joined #melange | 02:13 | |
| *** ArneBab has quit IRC | 02:17 | |
| *** mang0 has quit IRC | 02:17 | |
| *** AnujSharma has joined #melange | 03:04 | |
| *** madrazr has quit IRC | 03:17 | |
| *** AnujSharma has quit IRC | 03:45 | |
| *** variable is now known as constant | 04:16 | |
| *** Gentlecat has joined #melange | 04:16 | |
| *** AnujSharma has joined #melange | 04:35 | |
| *** Niharika has joined #melange | 05:06 | |
| *** Niharika has quit IRC | 05:40 | |
| *** madrazr has joined #melange | 05:46 | |
| *** ChanServ sets mode: +o madrazr | 05:46 | |
| *** raghunayyar has joined #melange | 07:31 | |
| *** tim_ has joined #melange | 07:46 | |
| *** tim_ has quit IRC | 08:02 | |
| *** penguinRaider has quit IRC | 08:13 | |
| *** penguinRaider has joined #melange | 08:28 | |
| *** samgtr has joined #melange | 08:50 | |
| *** rigelk has joined #melange | 08:57 | |
| *** robbyoconnor has quit IRC | 09:15 | |
| *** robbyoconnor has joined #melange | 09:16 | |
| *** robbyoconnor has quit IRC | 09:23 | |
| *** robbyoconnor has joined #melange | 09:23 | |
| *** robbyoconnor has quit IRC | 09:33 | |
| *** madrazr has quit IRC | 10:13 | |
| *** Niharika has joined #melange | 10:32 | |
| *** samgtr has quit IRC | 10:56 | |
| *** mang0 has joined #melange | 11:05 | |
| *** samgtr has joined #melange | 11:28 | |
| *** mang0 has quit IRC | 12:38 | |
| *** Abhishek_ has quit IRC | 13:08 | |
| *** sawrubh_ has joined #melange | 13:09 | |
| *** sawrubh has quit IRC | 13:09 | |
| *** sawrubh_ is now known as sawrubh | 13:10 | |
| *** Abhishek_ has joined #melange | 13:10 | |
| *** gsocbot has quit IRC | 13:20 | |
| *** AnujSharma has quit IRC | 13:25 | |
| *** gsocbot has joined #melange | 13:27 | |
| *** suranga has joined #melange | 14:04 | |
| *** AnujSharma has joined #melange | 15:16 | |
| *** samgtr has quit IRC | 15:26 | |
| *** AnujSharma has quit IRC | 15:28 | |
| *** teepee has quit IRC | 17:22 | |
| *** teepee has joined #melange | 17:23 | |
| *** Niharika has quit IRC | 17:42 | |
| *** mang0 has joined #melange | 17:49 | |
| *** mang0 has joined #melange | 17:49 | |
| *** Gentlecat has quit IRC | 18:01 | |
| *** madrazr has joined #melange | 19:30 | |
| *** ChanServ sets mode: +o madrazr | 19:30 | |
| *** Abhishek_ has quit IRC | 20:30 | |
| *** MatthewWilkes has quit IRC | 20:51 | |
| *** Abhishek_ has joined #melange | 21:07 | |
| * Robert_S looks for the switch to disable closure. | 21:07 | |
| madrazr | Robert_S: disable closure? | 21:30 |
|---|---|---|
| madrazr | Robert_S: disable where? | 21:31 |
| *** raghunayyar has quit IRC | 22:07 | |
| Robert_S | madrazr: in paver. i figured out how to do that. have a patch coming shortly. Next step is figuring out for a local version of interactive.py | 22:14 |
| madrazr | Robert_S: ah Ok. Isn't it just passing --skip-closure to bin/paver? | 22:15 |
| Robert_S | madrazr: possibly, but i didn't want to re-run the entire build process. | 22:15 |
| madrazr | Robert_S: what do you want to figure out for a local version of interactive.py? | 22:15 |
| Robert_S | madrazr: I want a local python interpreter where s = soc.logic.site.singleton() works and accesses the store shared by my dev_appserver. | 22:16 |
| madrazr | Robert_S: that is easy, just run bin/stats dev~your-app-name localhost:8000 | 22:17 |
| madrazr | Robert_S: one caveat is, to use bin/stats (aka scripts/stats.py) you will have to switch to madhusudancs/stats-fix branch | 22:18 |
| madrazr | Robert_S: Nathaniel and I had quite some discussion about this | 22:18 |
| Robert_S | madrazr: bin/stats doesn't work because it's not updates for ndb. | 22:18 |
| Robert_S | Aha. | 22:18 |
| madrazr | Robert_S: it is | 22:18 |
| Robert_S | Ok, any chance of getting that merged? | 22:18 |
| madrazr | if you use madhusudancs/stats-fix | 22:18 |
| madrazr | Robert_S: there is a reason why it is not merged | 22:18 |
| madrazr | continuing | 22:19 |
| madrazr | Robert_S: Nathaniel and I had discussions about this. Nathaniel had concerns about 800 or so lines of code having no tests at all and he had said that it would bother him to have that code checked in and being updated often | 22:20 |
| madrazr | Robert_S: I, on the other hand, did not see (and still don't see) a value in writing tests for stats.py | 22:20 |
| madrazr | Robert_S: it is supposed to be a bunch of adhoc scripts | 22:20 |
| madrazr | Robert_S: so we decided to keep it out of master where I could keep updating it | 22:21 |
| Robert_S | sigh. | 22:21 |
| madrazr | Robert_S: I update it once in a while | 22:21 |
| madrazr | Robert_S: and that once in a while is generally when I desperately need it | 22:21 |
| Robert_S | Sigh. | 22:22 |
| Robert_S | So, interactive.py is a subset of stats.py. | 22:22 |
| Robert_S | At this point, I'd rather a working stats.py without tests to a non-working stats.py without tests. | 22:22 |
| madrazr | Robert_S: sort of. Or I would say stats.py is interactive.py + some imports/names in the namespace + some small adhoc functions | 22:23 |
| madrazr | Robert_S: fair point | 22:23 |
| madrazr | I would be happy to merge it to master if that is the stand we are going to take | 22:23 |
| Robert_S | I would find that useful. We can't fix everything all at once, and no reason to suffer. | 22:25 |
| madrazr | Robert_S: I agree | 22:25 |
| Robert_S | If you rebase it and send a review request, I'll review it. It should be straightforward. | 22:26 |
| Robert_S | In other news, bin/stats dev~robert-melange localhost:8000 results in a 404 error. | 22:28 |
| Robert_S | +1's inbound. | 22:30 |
| madrazr | Robert_S: Oh! Thanks! That was quick | 22:31 |
| madrazr | Robert_S: let me take a look at 404 | 22:31 |
| Robert_S | I can also get a 401 if I use the app port. | 22:32 |
| Robert_S | If I use the API port, I get an AttributeError. I'm sure I'm doing something stupid. | 22:33 |
| madrazr | Robert_S: Oh oops, sorry. This was my bad. App Engine folks changed what they serve on different ports | 22:33 |
| madrazr | Robert_S: your command line should be localhost:8080 instead of localhost:8000 | 22:33 |
| madrazr | or more specifically it should be the port on which your app is configured to run | 22:33 |
| Robert_S | And which auth does it use? | 22:33 |
| madrazr | Robert_S: it does not matter. It takes arbit values. | 22:34 |
| madrazr | Robert_S: I give [email protected] as username | 22:34 |
| Robert_S | Username: | 22:34 |
| Robert_S | Password: | 22:34 |
| Robert_S | Invalid username or password. | 22:34 |
| Robert_S | Username:[email protected] | 22:34 |
| Robert_S | Password: | 22:34 |
| Robert_S | Invalid username or password. | 22:34 |
| madrazr | and use blank as the password | 22:34 |
| madrazr | Robert_S: may be try to sign in on the web ui with [email protected] and see what happens? | 22:35 |
| madrazr | Robert_S: what is surprising is, it works for me even if I type | 22:36 |
| madrazr | Username:[email protected] | 22:36 |
| madrazr | Password: | 22:36 |
| madrazr | those are some random characters I typed | 22:36 |
| Robert_S | seems to work fine, although i have an empty db, so hard to tell. | 22:36 |
| madrazr | Robert_S: easy test is to run /seed_db on the Web UI | 22:37 |
| Robert_S | I also noticed /site/edit, which is also broken. | 22:37 |
| madrazr | and to try to look up for that | 22:37 |
| Robert_S | There we go. | 22:37 |
| Robert_S | This solves my short term issue. | 22:37 |
| madrazr | Robert_S: cool! | 22:40 |
| Robert_S | Ok, I now have a dev instance running on my desktop with uncompiled javascript and in theory, google analytics. | 22:40 |
| madrazr | Robert_S: /site/edit is broken? What is the error? | 22:40 |
| Robert_S | /site/edit is broken before you run seed_db | 22:40 |
| Robert_S | I still don't have stats or interactive that runs locally. | 22:40 |
| madrazr | Robert_S: Oh! I assume by broken you see a login error or something related to auth? Not exactly an exception/HTTP 500? | 22:41 |
| madrazr | Robert_S: what happened to stats that you could run just now? Doesn't it work? | 22:42 |
| Robert_S | Er, no. | 22:42 |
| Robert_S | Let me re-state everything: | 22:42 |
| Robert_S | 1. If I use /site/edit before /seed_db, there's an exception. Probably because there's nothing for it to enumerate. Not a huge deal right now. | 22:43 |
| Robert_S | 2. My goal was to update the Google Analytics code. I can do this with /site/edit now. | 22:43 |
| Robert_S | 3. I'm still getting the auth error with stats and interactive. | 22:43 |
| Robert_S | <done> | 22:43 |
| madrazr | 1. Oh! New breakage! Don't know when this has happened though. I don't even remember when I remember running it last time before running /seed_db | 22:44 |
| Robert_S | I wonder if it's because I'm running the appserver on non-localhost. | 22:44 |
| madrazr | Should be fixed though. I assume, by your usage of the word enumeration, it is related to the active programs that we set in site settings | 22:45 |
| Robert_S | It was iterating through something, I didn't pay too much attention to what at the time. | 22:45 |
| madrazr | Robert_S: shouldn't related | 22:45 |
| madrazr | Robert_S: Oh Ok. I think it is "active programs" setting | 22:45 |
| madrazr | Robert_S: That's one of the very few things that has changed there in years | 22:46 |
| * Robert_S nods. | 22:47 | |
| madrazr | Robert_S: about 3. you get auth error with stats/interactive after you see the Python prompt? | 22:47 |
| madrazr | or even before you see the prompt? | 22:47 |
| Robert_S | Before. | 22:47 |
| Robert_S | Something on the server is rejecting it. | 22:47 |
| madrazr | Robert_S: and your second argument is host:app-port (and not host:API/Admin-port)? | 22:48 |
| Robert_S | Yes. | 22:48 |
| madrazr | Robert_S: let me think | 22:49 |
| madrazr | Robert_S: is there a way you can share your STDOUT/STDERR containing the auth permission error? | 22:50 |
| Robert_S | There's nothing of value. | 22:52 |
| Robert_S | INFO 2014-08-15 22:50:07,804 module.py:639] default: "GET /_ah/remote_api HTTP/1.1" 401 57 | 22:52 |
| Robert_S | hrm. | 22:52 |
| Robert_S | And now it works. | 22:53 |
| Robert_S | Somewhere, something is keeping state. | 22:54 |
| Robert_S | - it only works for [email protected] if it's the first thing you authenticate with. | 22:56 |
| Robert_S | - if you fail once, it will always fail. | 22:56 |
| Robert_S | - it only works on localhost, not if i specify a different host (to get a different interface) | 22:56 |
| Robert_S | (network interface) | 22:57 |
| madrazr | Robert_S: that's interesting | 23:01 |
| madrazr | that is a problem similar to the one that I am aware of. But I hoped, that wouldn't affect the SDK | 23:01 |
| Robert_S | fun times | 23:03 |
| madrazr | Robert_S: yeah, I would ignore it for now if it runs on localhost | 23:03 |
| Robert_S | Yeah, I have several workarounds. | 23:04 |
| madrazr | Robert_S: Oh Ok, cool | 23:04 |
| Robert_S | (see above where they're listed) | 23:04 |
| Robert_S | I *now* have several workarounds. | 23:04 |
| Robert_S | I didn't an hour ago. | 23:04 |
| madrazr | Robert_S: is that a good thing or a bad thing? :P | 23:08 |
| Robert_S | A good thing. | 23:08 |
| Robert_S | Now I can try and make analytics work. | 23:08 |
| madrazr | Robert_S: ah Ok. I still like the idea of building a base-image/container that we can just pull and use. It should have all the workarounds needed for local dev and/or builds | 23:09 |
| Robert_S | I agree. | 23:09 |
| madrazr | at least then we can call those workarounds as configuration :P | 23:09 |
| Robert_S | Although right now, at least on ubuntu/debian, there aren't a lot of systematic workarounds. | 23:10 |
| madrazr | agreed | 23:11 |
| Robert_S | FYI: I've been dumping all my changes into Gerrit for today. It slows down the review process a little, but lets us play with things in a not-time-sensitive fashion. | 23:18 |
| madrazr | Robert_S: sure! | 23:26 |
| *** MatthewWilkes has joined #melange | 23:27 | |
| Robert_S | I've been enjoying it as a creator, haven't had the opportunity to play with it much as a reviewer. | 23:27 |
| madrazr | Robert_S: I could have pushed my stats-fix branch there I guess | 23:27 |
| madrazr | Robert_S: hopefully I will have some changes for you by Monday | 23:27 |
| Robert_S | madrazr: Not worth it, didn't have much to say. :) | 23:27 |
| madrazr | Robert_S: if I make any reasonable progress on the security bug you pinged me on :P | 23:28 |
| Robert_S | That would be appreciated. If only because it's a security bug. | 23:28 |
| madrazr | Robert_S: heh, also those changes were sort of trivial :P | 23:28 |
| madrazr | Robert_S: yeah! I really wanted to give Go and App Engine modules a shot | 23:29 |
| madrazr | and this seemed like a perfect opportunity | 23:29 |
| Robert_S | madrazr: I'm a little concerned about adding new languages right now, I want to do that eventually, but I'm concerned about adding complexity before we remove some. But lets take a look. | 23:30 |
| madrazr | Robert_S: yeah | 23:30 |
| madrazr | it won't be a giant piece of code | 23:30 |
| madrazr | we can always throw it away, if we don't like it and write it in Python | 23:30 |
| * Robert_S nods. | 23:30 | |
| madrazr | Robert_S: but it will have to live as a separate module anyway by design | 23:31 |
| Robert_S | That helps with the logic to do it in Go. | 23:31 |
| madrazr | yeah | 23:31 |
| Robert_S | More changes in gerrit. | 23:47 |
| Robert_S | Now where are those emails... | 23:47 |
| Robert_S | Oh right, they go into my Sent folder. :) | 23:48 |
| *** suranga has quit IRC | 23:51 | |
| madrazr | Robert_S: wait, what!? Notificaton emails go to "Sent"? | 23:56 |
Generated by irclog2html.py 2.12.1 by Marius Gedminas - find it at mg.pov.lt!