| *** tpb has joined #melange | 00:00 | |
| *** r0bby has quit IRC | 01:50 | |
| *** solydzajs has quit IRC | 03:01 | |
| *** solydzajs has joined #melange | 03:21 | |
| *** ChanServ sets mode: +o solydzajs | 03:21 | |
| jamtoday1 | ajaksu: just pushed new changeset to github | 03:24 |
|---|---|---|
| *** Merio has joined #melange | 03:34 | |
| *** solydzajs has quit IRC | 04:55 | |
| *** dhans has joined #melange | 05:05 | |
| Merio | dhans: heya! :) | 05:07 |
| dhans | Merio: Hey! | 05:08 |
| dhans | Merio: I am reading your changes in wiki:) | 05:09 |
| Merio | dhans: ok let me know when you're finished :) | 05:10 |
| dhans | Merio: actually I just did finish:) | 05:11 |
| Merio | Oh great, so, have some thoughts on it? | 05:12 |
| dhans | I am thinking about it... or just trying to imagine how it is supposed to work... | 05:14 |
| Merio | So, until "jLinq", we thought about collecting ready-to-serve JSONs to the client | 05:15 |
| dhans | yes, I remember | 05:15 |
| Merio | So we collect data already ... ready, say "students per age" by a specific range of ages | 05:15 |
| *** MatthewWilkes has joined #melange | 05:16 | |
| Merio | In that way | 05:16 |
| dhans | so for those stats which have let's say some ranges, we will send 'raw json' so as user will be able to execute queries on it? | 05:17 |
| Merio | We only need to pass the dicts (IIRC) to gviz and serve it to Visualization... and the stat is ready | 05:17 |
| Merio | Exactly | 05:17 |
| Merio | So we might want to do your first students per age (don't know if it has changed), so just send "0" years: "0" students, "18" years: "200" students... and so on | 05:18 |
| Merio | This is obviously something that changes things, but it's more a JS change than a Python one | 05:19 |
| Merio | But we can go into more depth then | 05:19 |
| Merio | So the user might want to save the chart | 05:20 |
| dhans | so this ""0" years: "0" students, "18" years: "200" students." is the format of 'half-cooked' json you talked about? | 05:20 |
| Merio | yep | 05:21 |
| Merio | In that way, we need to have a JS Object and a Python one to remember which data is the source, which chart (piechart, table, barchart), the destination data (say, the steps of jLinq queries) and the options of the chart (3d, not 3d, eventually colors and so on) | 05:21 |
| dhans | of and if I understand correctly, for each stat we will send only one json: either created by gviz or half cooked one? | 05:22 |
| Merio | yes | 05:23 |
| Merio | But there is another one to be fair :) | 05:23 |
| Merio | The problem is: we need to connect the object in Python (say, chart.py) with the source data | 05:23 |
| Merio | But the host might want to save and let be public a chart which starts from data that a final user should not read | 05:24 |
| Merio | So it's not good to save the original JSON and serve it blindly | 05:24 |
| Merio | So we thought with James to let an host (which is the poweruser) save only a part of the original JSON with the original data... so when a common user watch for the chart, than there is no way to see the original JSON with the sensitive data | 05:25 |
| dhans | ok I understand, but I don't know exactly what kind of 'sensitive' data you have in mind...:) | 05:26 |
| Merio | Well for example locations, we mi | 05:27 |
| Merio | we might want to show that there are 2 GSoC students in Rome, but we don't want to connect the students to their personal data | 05:27 |
| Merio | Just an example | 05:27 |
| dhans | ok, I get it. I just thought that we will remove any (for example) personal data while collecting the stats, so that the final_json does not have it | 05:29 |
| Merio | Basically everything that it's not "sensitive" if shown by number of students/mentors but can be sensitive if shown person by person | 05:29 |
| Merio | This can be true perhaps even throughout GSoC, we might not have the time to accomplish all the flexibility, but say you want a graph with "students per age per degree" or something | 05:31 |
| Merio | In that way (I think) you can't aggregate data, you have to give a JSON with a big list for every student, with their age and degree | 05:32 |
| Merio | You can avoid displaying names and such things, but it's going more granular | 05:33 |
| dhans | do I understand this stat correctly? it would show that there is for example 200 students that are masters and have 23 years? | 05:33 |
| Merio | yes, something like dividing the students in ranges: "18-25"-"25-30"-">30" and for each range you want their degrees | 05:34 |
| Merio | This can be easily divided in three charts | 05:35 |
| Merio | But also can be shown in a table with different intensity of colors for each cell | 05:35 |
| Merio | (if for example you want it also divided by country) | 05:36 |
| dhans | hmm ok, i think I am starting to get it | 05:36 |
| Merio | I mean, there are lots of possibilities with this kind of design, and it's very difficult to think of everything beforehand | 05:36 |
| Merio | But even if for more complex stats we will need in a future to redesign something, still we should implement the basic concepts | 05:37 |
| dhans | yes | 05:37 |
| Merio | Another thing that we should think beforehand (and here I ask you how we can achieve this, as I'm going this week to dig further into Python but I don't have an idea now ^__^) | 05:38 |
| Merio | is the possibility to have every saved chart with its link_id and its ACL, so we can ask for them in the dashboard as well as in a (for example) iframe into documents | 05:39 |
| Merio | We thought with James it should be enough to inherit from document.py (for the moment, I don't know where the actual access rights discussion will go) | 05:41 |
| Merio | But I have no real idea on how this is gonna work for the moment, so I ask you if it's ok :) | 05:41 |
| Merio | that link_id should retrieve the data from chart.py and also have the JavaScript loaded in it | 05:42 |
| dhans | ok, so for a given user you want to get charts that he saved for himself and he has access for? | 05:42 |
| Merio | well also if the user is the host then the user can save the "computed" JSON and let the chart.py be public | 05:44 |
| dhans | I am not sure in 100%, but I think it is possible what you say | 05:44 |
| Merio | Basically each user in Melange will have a list of available charts, saved for example by other users. So we might want to have a link_id dependent on users and name of the chart or something like that | 05:45 |
| dhans | IIRC only program admins are able to save stats? or something chanegd? | 05:46 |
| Merio | Apart from the saved charts, the user has also a list of "full-cooked" (or "full-baked" don't know which is correct :P) and "half-baked" JSONs (which are "pool of data" to the user) to play with | 05:46 |
| Merio | I think that the users should have less flexibility on how they can play with the data, but still it would be great if they can play in someway | 05:47 |
| Merio | The classical "students per age", for example, it's fine for them to play with the ranges | 05:47 |
| dhans | by play with data you mean 'collect' or change for example ranges? | 05:48 |
| Merio | You want the students divided age per age? or do you want them shown by ranges? which ranges? which colors? slices of the piechart under 1% should be displayed? and so on | 05:48 |
| Merio | No, collect queries should be raised only by hosts/developers I think | 05:49 |
| dhans | yes I agree | 05:49 |
| dhans | with your last 2 messages:) | 05:49 |
| Merio | eheh :) | 05:49 |
| Merio | Ok, so... there is much work to be accomplished | 05:50 |
| Merio | But we're two, so we should go for teamplay and be clever in what we do, when we do and how we do | 05:51 |
| dhans | exactly | 05:51 |
| Merio | That means for example, put IRC in autorun in your own computer ;) | 05:51 |
| dhans | I promise I will:) | 05:52 |
| Merio | ^__^ | 05:52 |
| Merio | In that way we can chat everytime we want about the project and what we're doing | 05:52 |
| Merio | I would propose to have a daily irc meeting planned | 05:53 |
| dhans | good idea | 05:53 |
| Merio | Just in case we don't have much to say during the day, but still we're able to get in sync | 05:53 |
| Merio | For me the best hour range for a planned daily meeting is 15-19 (not 20 as we might overlap with Melange conference calls) | 05:54 |
| Merio | What about you? | 05:54 |
| dhans | so in my timezone it's 16-20...:P | 05:55 |
| Merio | yes, sorry, you're right ^_^ | 05:55 |
| dhans | ok, it's also ok | 05:55 |
| Merio | I think 15 mine/16 yours should be fine, as we're in the middle of the day | 05:56 |
| dhans | yes | 05:56 |
| Merio | Two other questions: | 05:57 |
| Merio | 1) what about you in Saturdays and Sundays? Just to know. | 05:57 |
| Merio | 2) More practical, I think our best next step. What can we do to setup the demo instance? | 05:58 |
| dhans | 1) if I am in Warsaw, it's ok. Sometimes I am leaving, but unfortunately it does not happen often lately | 05:58 |
| dhans | 2) I have to read mails about the demo instance, because I think I got lost with this 'demo instance' idea | 06:00 |
| Merio | Ok it's not a problem, I'll probably need a day sometime just to not to break with my girlfriend ;) :D | 06:00 |
| dhans | :) | 06:00 |
| Merio | We have a demo instance up on http://dhansmelange.appspot.com/ | 06:01 |
| tpb | Title: Home Page (at dhansmelange.appspot.com) | 06:01 |
| Merio | We need to have it working even for non developers to view our work anytime | 06:01 |
| dhans | ah, ok | 06:02 |
| Merio | So we might want to work today to have it working. It would also be great to start joining JS and Python stuff, at least to show a simple stat | 06:02 |
| Merio | In that way, it will be also easier for me to start doing Python related stuff | 06:02 |
| Merio | but it's a personal concern :P | 06:02 |
| dhans | in the final version, if I understand correctly, there will be a dashboard page with some kind of widgets where each widget will represent a stat, ok? | 06:04 |
| Merio | yes, but each stat will be retrieved by a link_id or something, so we can put it in a dashboard widget and embed it in a document as well | 06:05 |
| Merio | I think we need for first to start joining Python and JS to have a demo, then we go on improving it a bit, it would be great to think about the UI as well | 06:06 |
| dhans | ok, but I am asking about something else:) | 06:06 |
| Merio | So we have a clear idea on what the user can do and how the user will do those things. It would help designing the rest of the module | 06:06 |
| Merio | oh sorry :) | 06:06 |
| dhans | and this dashboard page will retrieve its data from chart.py, won't it? | 06:06 |
| Merio | basically yes, this was the thought | 06:06 |
| Merio | But feel free to object if there is something wrong with it obviously :) | 06:07 |
| Merio | Also there will be some communication with statistic.py to retrieve the JSON if there is nothing saved in my dashboard | 06:08 |
| dhans | ok so now, you are right it's better to just join python and JS code to have at least some simple stats shown | 06:08 |
| Merio | Basically chart.py is where we will save the user specific statistic (think of it like a sort of wiki collection) | 06:08 |
| Merio | and statistic.py is where all the data is collected (or saved by the host) and where they will be retrieved anytime | 06:09 |
| Merio | Yes, forgetting for the moment about chart.py | 06:09 |
| Merio | Just use for example gviz to send the JSON and having a simple simple JS to show a simple graph | 06:09 |
| dhans | ok, so now for each statistic you need a 'half-cooked' json or gviz object... | 06:10 |
| Merio | It's more Python stuff than JS, so we might want to work together on it in Python | 06:10 |
| dhans | yes | 06:10 |
| Merio | Yes or a "saved-by-the-host" json | 06:10 |
| Merio | I think gviz is fine as it's simpler now | 06:11 |
| dhans | I already added one function that creates a simple DataTable object... | 06:12 |
| dhans | it's getDataTableSample. I just wanted to make sure that this gviz works:) | 06:13 |
| Merio | eheh you've done right :) | 06:13 |
| Merio | is it actually working? :P | 06:13 |
| dhans | yes. I even displayed a simple table:) | 06:14 |
| Merio | you rock ^_^ | 06:14 |
| dhans | :) | 06:14 |
| Merio | ok so, just give me 10 minutes, I need to do some stuff one moment... in the meantime I'll pull your edits and we will go for it :) | 06:16 |
| dhans | ok | 06:17 |
| dhans | I'll go to eat some breakfast:P | 06:17 |
| Merio | are you available all day or you've to go away, say, in the afternoon or something? | 06:17 |
| Merio | Yes, please do, it's important to do work, but to keep us breathing is needed as well :D | 06:18 |
| Merio | bbs :) | 06:19 |
| dhans | I need to leave in about an hour... I am purchasing a new laptop today and have to go to a bank. But I should be back at about 3 PM (your time) | 06:19 |
| Merio | that's perfectly fine, if you want we can just start this afternoon, I'll be reading some Python guides to refresh my (little) knowledge in the meantime :) | 06:20 |
| Merio | dhans: I forgot to mention: Lennie told me about the task API, which is going to be released for the Google App Engine, it would mean changing Jobs, so you might need to be aware of it too | 06:31 |
| dhans | ok I am back | 06:32 |
| dhans | jobs is just something that let us schedule some task in certain time intervals, so now we are not using jobs, but collecting stats manually in batches (by a dev or program admin) by clicking collect | 06:34 |
| dhans | the problem is that the jobs have the same limitations as regular requests so they are not very helpful, but I will try to find something about the task api or will ask Lennie | 06:36 |
| dhans | thanks for info btw:) | 06:36 |
| Merio | dhans: ok :) So.. meeting at 16(your time), just ping me whenever you're ready, I'll be here :) | 06:43 |
| dhans | ok | 06:43 |
| *** jamtoday1 has quit IRC | 07:43 | |
| *** Merio has quit IRC | 08:06 | |
| *** Merio has joined #melange | 08:16 | |
| *** arun_ has joined #melange | 08:24 | |
| *** arun has quit IRC | 08:25 | |
| *** mithro has quit IRC | 08:35 | |
| *** arun_ is now known as arun | 08:53 | |
| *** florinciu has quit IRC | 09:39 | |
| Merio | dhans: ping | 10:21 |
| *** jamtoday has quit IRC | 10:53 | |
| dhans | Merio: pong :) | 11:25 |
| Merio | dhans: wb ^__^ | 11:25 |
| Merio | All ok for the laptop? | 11:25 |
| dhans | Merio: I think so... :) | 11:26 |
| Merio | wonderful then :) | 11:26 |
| Merio | I was hitting my head on the wall to let dynamic loading working for google APIs | 11:26 |
| dhans | and I have some problems with shipping company and it resulted in a situation that I had to meet with the delivery guy in the city, so that's why I am late ;-/ | 11:28 |
| dhans | I'm sorry :) | 11:28 |
| Merio | No problems :) | 11:29 |
| Merio | So, in this very moment I'm trying to solve this problem, is there something difficult on pushing a working demo (only about your work for the moment) in the demo instance? | 11:30 |
| dhans | what do you mean by 'working demo'? | 11:31 |
| Merio | something that can be visible in our demo instance even if you aren't a developer | 11:32 |
| Merio | So, let's say, James Crook or LH can use their google account simply to enter our demo instance and clicking into the proper link in the sidebar just show, for example, a simple visualization (as you already have) | 11:34 |
| dhans | yes, I get it... | 11:34 |
| dhans | so the simplest way, to make it visible for everyone is to just change access | 11:35 |
| Merio | this way is also easier for us to work on it in the meantime | 11:35 |
| Merio | dhans: can you work on it while I'm solving this problem? I mean I can join the effort if helps :) | 11:41 |
| dhans | ok I just changed those rights and will push it to the demo instance | 11:45 |
| Merio | great man! | 11:45 |
| dhans | ok committed and pushed on our instance | 11:55 |
| Merio | dhans: great ^__^ it's cool! I think that "visualize" link is somewhat broken, is it so? | 11:59 |
| dhans | yes, indeed :-) | 11:59 |
| dhans | it worked for like a very little amount of time ;-) | 12:00 |
| Merio | eheheh ;) is it fixable in the demo instance or we need to work on the backend before using it? | 12:00 |
| Merio | I mean it's only a broken link or there is nothing ready? Just to know what to do to help ^_^ | 12:01 |
| dhans | it was not flexible. actually no matter which 'visualize' button I'd have pressed, it'd show you visualization for students per degree :) | 12:02 |
| Merio | eheh ok ^__^ | 12:02 |
| Merio | dhans: need a 10 minute break now, I'm looking at the screen 3 hours without breaks :) So, is there a way I can help to fix that, or are you going to work on it now and I'll join after? | 12:04 |
| Merio | Just because I would like to have a demo instance working this afternoon so I can join my JS code and see what's going on :) | 12:04 |
| dhans | what we need is to have, for each stat that we want to visualize, a function that returns a DataTable object for this stat | 12:05 |
| dhans | because I prepared only one simple function that returns DataTable for students per degree | 12:06 |
| Merio | ok so I can join in and help developing that | 12:07 |
| dhans | ok, but feel free to have a break :) | 12:07 |
| Merio | something odd happens here... it's funny :) I've done nothing and now it's working... hmmmm! | 12:07 |
| dhans | heh :) | 12:08 |
| dhans | and please just tell me which parts of your JS code you want to join in our demo instance? | 12:09 |
| Merio | ok so just wait for me 10 minutes, I'm going to revert some changes and then I'm going to help you on the visualizations :) | 12:09 |
| Merio | bbs | 12:09 |
| dhans | ok, now visualize works again, but be prepared, this is certainly not what we want :-P | 12:14 |
| Merio | dhans: back :) | 12:19 |
| *** thrain42 has joined #melange | 12:19 | |
| *** ChanServ sets mode: +v thrain42 | 12:19 | |
| Merio | dhans: ok, so basically now we have always the same table/JSON returned for every statistics, but the statistics are collected well, is it true? | 12:21 |
| dhans | I think so | 12:21 |
| dhans | and the function that actually draws the table is defined in base.html (drawTable) and has nothing in common with the JS code rules that you proposed :) | 12:25 |
| Merio | dhans: no problems, I can do it :) | 12:26 |
| Merio | ok I've pushed another change, what I need to do to push it in the live instance? Is it enough to follow "deploying to google app engine" thing in the gettingstarted wiki page? | 12:28 |
| Merio | dhans: just to be sure I'm not doing something odd :) | 12:28 |
| dhans | let me see it | 12:28 |
| Merio | ok, I've pushed some changes and merged with your changes | 12:29 |
| Merio | Now I would like to push it in the live instance and I don't want to break things :) | 12:29 |
| dhans | so can you make another simple change :), because I did not push my last update that made visualize worked :P | 12:30 |
| *** Erant has quit IRC | 12:31 | |
| dhans | you should have something like: (r'^%(url_name)s/(?P<access_type>visualize_stats)/%(scope)s$', in line 96 in views/models/statistic.py so please change scope for key_fields | 12:31 |
| dhans | and about deploying to google app engine - yes, thirdparty/google_appengine/appcfg.py update build is enough | 12:32 |
| *** Erant has joined #melange | 12:33 | |
| Merio | Push it in the repository then :) | 12:33 |
| Merio | I'm committing a little change | 12:33 |
| Merio | dhans: committed | 12:34 |
| *** durin42 has quit IRC | 12:34 | |
| dhans | but I started to rewrite this some other functions and I bet that if I push now, visualization won't work again :P | 12:34 |
| Merio | eheh ok, I'll work on something else in the meantime | 12:34 |
| Merio | dhans: I think it's better if you commit stuff more granularly, I see from Melange's commit log that it's that the way to go | 12:35 |
| *** solydzajs has joined #melange | 12:35 | |
| Merio | (but we can just ask during the conference call ;)) | 12:35 |
| Merio | (or we can just ask solydzajs now :P) | 12:36 |
| *** penyaskito has joined #melange | 12:36 | |
| dhans | I agree, but just forgot and started to do something new | 12:40 |
| Merio | ok :) Just ping me when you're done :) | 12:41 |
| *** r0bby has joined #melange | 12:45 | |
| dhans | Merio: ok, committed, merged and pushed:) | 12:47 |
| Merio | dhans: great :) | 12:47 |
| Merio | do you already sent the stuff into the live instance? I would like to try ^__^ | 12:48 |
| dhans | so I can say no:) | 12:48 |
| dhans | I mean 'no I have not' :) | 12:48 |
| Merio | ehhe ok :) | 12:48 |
| Merio | so I'm trying | 12:48 |
| *** ChanServ sets mode: +o solydzajs | 12:49 | |
| solydzajs | yo what's up ? | 12:49 |
| solydzajs | Merio: ? | 12:49 |
| dhans | solydzajs: hello paweł :) | 12:50 |
| Merio | solydzajs: we're doing our IRC meeting ... so we're pushing a working demo into the live instance and going to start joining our code | 12:50 |
| *** r0bby has quit IRC | 12:51 | |
| *** r0bby has joined #melange | 12:51 | |
| solydzajs | Merio: ok sounds good | 12:51 |
| solydzajs | Merio: tty some more during conference call | 12:51 |
| solydzajs | Merio: I probably gonna take a nap, still jetlagged a little bit | 12:51 |
| Merio | solydzajs: ok great :) | 12:51 |
| Merio | dhans: I have to run build.sh first? | 12:51 |
| dhans | Merio: you should | 12:52 |
| Merio | solydzajs: ok, sleep tight so ^__^ ttyl :) | 12:52 |
| Merio | dhans: it's giving me this => google/appengine/ext/remote_api/remote_api_stub.py:232: SyntaxWarning: assertion is always true, perhaps remove parentheses? | 12:53 |
| Merio | assert (txdata.thread_id == thread.get_ident(), | 12:53 |
| Merio | dhans: is it normal or need I to change something to work with python 2.5 instead of 2.6? | 12:54 |
| dhans | I am working with 2.5, but I also get a lot of errors/ warnings while executing build.sh | 12:55 |
| *** r0bby has quit IRC | 12:56 | |
| *** r0bby has joined #melange | 12:56 | |
| Merio | dhans: it says I don't have permissions :P | 12:57 |
| dhans | to upload? | 12:57 |
| Merio | yep | 12:57 |
| Merio | You do not have permission to modify this app (app_id=u'statistic'). | 12:57 |
| dhans | strange | 12:58 |
| Merio | let me see, perhaps I mistyped something | 12:58 |
| dhans | hmm | 12:58 |
| dhans | what about your app.yaml file? | 12:59 |
| Merio | it has application: statistic | 12:59 |
| dhans | please try to change it to dhansmelange | 12:59 |
| Merio | need I to change the name? | 12:59 |
| Merio | Oh crap :P | 12:59 |
| Merio | eheheh it's working now, shame on me :P | 13:00 |
| dhans | I know this name is a little bit selfish, but when I created this instance I didn't know someone else would work on it :P | 13:01 |
| Merio | dhans: I know you're going to change all my commits to your name! :P | 13:02 |
| dhans | :P | 13:02 |
| Merio | eheh ok, so let's see what's happening then ^__^ | 13:02 |
| Merio | ok it's working, but it's giving me 500 server errors when visualizing. Is it a known problem or something happened with my build? | 13:04 |
| dhans | nope | 13:04 |
| dhans | I was just explaining | 13:04 |
| Merio | ah ok perfect then | 13:05 |
| dhans | but try for Students Per Degree :) | 13:05 |
| Merio | yep tried... it's working :) | 13:05 |
| dhans | ok, so now it works like this: when you click 'visualize', the function visualizeSample is called and and it calls logic.getDataTableObject | 13:05 |
| * Merio going to the source code :) | 13:05 | |
| dhans | this function calls function getDataTable + statistic.id | 13:06 |
| dhans | but, for now, we have only getDataTableStudentsPerDegree | 13:06 |
| Merio | ok great | 13:07 |
| Merio | got it | 13:07 |
| Merio | it's not possible for us to have those initializations in the statistic.py data model instead of defining a function for every single statistic? | 13:08 |
| Merio | Otherwise I think it's not flexible enough for the other types of stats | 13:09 |
| dhans | so, got lost, by initializations you mean? | 13:10 |
| Merio | I mean: description = {"degree": ("string", "Degree"), | 13:11 |
| Merio | "number": ("number", "Number")} | 13:11 |
| Merio | if we've that description inside statistic.py, then we just need to retrieve it for every stat and just have a single function for all | 13:12 |
| dhans | ok, so that's what I thought:) | 13:12 |
| Merio | eheh perfect then :) | 13:12 |
| Merio | something like this | 13:12 |
| Merio | description = simplejson.loads(statistic.json_description) | 13:13 |
| Merio | and then changing the iteration to | 13:13 |
| dhans | it's a good idea | 13:13 |
| dhans | I was not sure if one generic function would be enough for creating data tables, but maybe it is | 13:14 |
| Merio | for key in json_string.keys(): | 13:14 |
| Merio | for key_desc in description.keys(): | 13:14 |
| Merio | <<build the data >> | 13:14 |
| Merio | data.append(data_built); | 13:14 |
| dhans | ok | 13:15 |
| Merio | I can try to do something like that, so I can just start up with a little bit of Python, but I don't know if it would mean stopping you at the moment | 13:15 |
| dhans | no | 13:15 |
| Merio | "no"... meaning? :p | 13:16 |
| dhans | no, it would not stop me :) | 13:16 |
| Merio | nothing could stop you! ;) eheh I mean it's going to be a great change (also for the model and the data collection), just wanted to be sure it's all ok :) | 13:17 |
| Merio | dhans: what do you do to test the data? Is seed_db enough? | 13:18 |
| *** durin42 has joined #melange | 13:18 | |
| *** ChanServ sets mode: +v durin42 | 13:18 | |
| dhans | to be honest, I have not used it:) | 13:19 |
| Merio | so just did all stuff manually? | 13:19 |
| dhans | I have always created one program, added a few (let's say 5) students and it was quite ok:) | 13:20 |
| Merio | (I'm going to change the wiki page in the meantime and sending a mail in the dev list about the live demo instance) | 13:20 |
| Merio | I think it could be an opportunity to enhance seed_db, I'll see what to do about it | 13:21 |
| *** penyaskito has quit IRC | 13:24 | |
| dhans | just one question about google visualization api: the DataTable object is a generic type of object and when we have it, we can create all types of visualizations? | 13:24 |
| *** penyaskito has joined #melange | 13:24 | |
| Merio | yes, that kind of abstraction is what I'm doing in the JS layer | 13:25 |
| dhans | because I am just a little bit concerned if one generic function that generates DataTable will be sufficient for some more sophisticated stats | 13:28 |
| dhans | like for StudentsPerDegreePerAge that you mentioned | 13:28 |
| *** thrain42 has quit IRC | 13:29 | |
| Merio | (wiki page updated and mail sent ^__^) | 13:30 |
| dhans | thanks | 13:30 |
| Merio | Well, perhaps it's not going to be sufficient | 13:30 |
| Merio | Well.. it's not sufficient at all, because the "half-baked" JSONs are not covered at all | 13:31 |
| dhans | yes, I know, I am just talking about data_tables | 13:31 |
| Merio | But it's sufficient for the basic data to work, then we will do another function for the other stuff | 13:31 |
| Merio | And if we're lucky enough to achieve more sophisticated stats during the GSoC we'll see how to generalize that function. I think it's not good aiming too far away, just get basics done with a reasonable amount of flexibility and an eye to the future | 13:32 |
| dhans | it sound good... | 13:32 |
| dhans | *sounds | 13:32 |
| Merio | ^_^ | 13:32 |
| Merio | Ok so I'll try to get my first hands on Python ^__^ What will you do in the meantime? (just to be in sync, it's perfectly fine to say nothing :P) | 13:33 |
| dhans | I have to leave for a while:-) | 13:33 |
| dhans | so will be back in like 25 minutes | 13:34 |
| Merio | that's ok, see you later during conference call then :) | 13:34 |
| Merio | no probs c ya :) | 13:34 |
| *** MatthewWilkes has quit IRC | 13:35 | |
| *** dhaun has joined #melange | 13:43 | |
| *** jamtoday has joined #melange | 14:12 | |
| *** jamtoday has quit IRC | 14:15 | |
| Merio | dhans: ping | 14:17 |
| *** jamtoday has joined #melange | 14:18 | |
| Merio | /who freenode/staff/* | 14:32 |
| Merio | Hmm | 14:32 |
| Merio | :P | 14:32 |
| *** jamtoday has quit IRC | 14:32 | |
| *** Merio has quit IRC | 14:42 | |
| *** Merio has joined #melange | 14:42 | |
| *** Lennie has joined #melange | 14:50 | |
| *** ChanServ sets mode: +o Lennie | 14:50 | |
| *** Merio was kicked by Lennie (Rejoin :D) | 14:52 | |
| *** Merio has joined #melange | 14:52 | |
| *** ChanServ sets mode: +v Merio | 14:52 | |
| *** florinciu has joined #melange | 14:58 | |
| solydzajs | dhans: ping | 15:02 |
| Lennie | ajaksu | 15:07 |
| Lennie | ping | 15:07 |
| ajaksu | Lennie: pong | 15:07 |
| Lennie | skype! | 15:07 |
| Lennie | if you can | 15:07 |
| ajaksu | I need about 3 minutes, installing skype on a windows laptop :) | 15:08 |
| * ajaksu keeps getting the mice mixed :) | 15:08 | |
| *** jamtoday has joined #melange | 15:16 | |
| ajaksu | jamtoday: not in the call? | 15:18 |
| *** penyaskito has quit IRC | 15:37 | |
| Lennie | ajaksu, thanks for joining in :D | 15:37 |
| Lennie | unfortunetly we discussed most of the important stuff yesterday | 15:37 |
| Lennie | bbs | 15:37 |
| *** Lennie is now known as Lennie|Food | 15:37 | |
| *** Lennie|Food is now known as Lennie|Food|Gone | 15:37 | |
| *** Lennie|Food|Gone is now known as Lennie|Gone | 15:38 | |
| *** Lennie|Gone is now known as Lennie | 15:48 | |
| ajaksu | got a long phone call right after the conference call ended, so after a long while: thanks for not laughing out loud everytime I tried to speak English :D | 16:13 |
| *** jamtoday has quit IRC | 16:20 | |
| Lennie | oh hell no :D | 16:22 |
| Lennie | ajaksu, you should join next week when madhusudan is there | 16:22 |
| Lennie | indian accents are the best :D | 16:22 |
| Lennie | I'm dutch so I suck as well :) | 16:23 |
| ajaksu | I couldn't have discussed and understood as much in a call as I did yesterday here on IRC :) | 16:23 |
| Lennie | hehe | 16:28 |
| *** penyaskito has joined #melange | 16:48 | |
| *** N4L|Lennie has joined #melange | 16:54 | |
| *** Lennie has quit IRC | 16:55 | |
| Erant | N4L? | 17:16 |
| *** dhaun has quit IRC | 17:17 | |
| *** ajuonline has joined #melange | 17:36 | |
| ajuonline | seen Lennie | 17:36 |
| ajuonline | :P | 17:36 |
| * ajuonline whacks r0bby | 17:36 | |
| *** ChanServ sets mode: +o N4L|Lennie | 17:37 | |
| *** N4L|Lennie is now known as Lennie | 17:37 | |
| Lennie | Nuubs4Life | 17:37 |
| ajuonline | lol | 17:37 |
| Lennie | Lennie was taken when the internet came back ^^ | 17:37 |
| ajuonline | what happened to the new TBBT Episodes? | 17:37 |
| ajuonline | S02 over? | 17:37 |
| ajuonline | ..dont see the Heroes ones as well :/ | 17:38 |
| Lennie | yes | 17:38 |
| Lennie | it's summer break | 17:38 |
| Lennie | back in winter :) | 17:38 |
| Lennie | s2e23 was the last one | 17:39 |
| ajuonline | oh dear :/ | 17:39 |
| ajuonline | how long? in months? | 17:39 |
| Lennie | no clue | 17:40 |
| Lennie | lemme check | 17:40 |
| Lennie | september | 17:41 |
| Lennie | sao | 17:44 |
| Lennie | so | 17:44 |
| Lennie | how are you? | 17:44 |
| Lennie | ajaksu | 17:44 |
| Lennie | ajuonline * | 17:44 |
| Lennie | how's your GSoC going? | 17:44 |
| ajuonline | ah not started :/ | 17:44 |
| ajuonline | i have college issues. | 17:44 |
| ajuonline | should get done with them this week | 17:45 |
| Lennie | that sucks :( | 17:45 |
| ajuonline | yeah my mentor tells me he will fail me :/ | 17:45 |
| Lennie | I can't blame him | 17:46 |
| ajuonline | bbut thats okay :) cant help it. i need to graduate first. | 17:46 |
| Lennie | since you havent started yet | 17:46 |
| ajuonline | i know | 17:47 |
| *** jamtoday has joined #melange | 17:54 | |
| Lennie | ajuonline, nn :) | 18:10 |
| *** Lennie has quit IRC | 18:11 | |
| jamtoday | ajaksu: the reason for the surveys not restoring after a failed POST is the JS has been commented out | 18:28 |
| jamtoday | I was using a hidden field hack | 18:28 |
| ajaksu | jamtoday: yup, I mentioned that in the issue | 18:28 |
| jamtoday | were you planning on using a different method? | 18:28 |
| jamtoday | ajaksu: good thinking on testing in other browsers | 18:30 |
| jamtoday | ajaksu: I tested everything on all the browsers but never surveys | 18:30 |
| jamtoday | I moved some logic to the surveys logic module, and just now fixed some strange behavior for empty surveys | 18:32 |
| ajaksu | jamtoday: aha, got it: I got rid of EditSurvey and didn't investigate what should be stored in that hidden charfield instead... | 18:32 |
| jamtoday | the field is called survey_html and it's just a jQuery.html() dump | 18:33 |
| ajaksu | oops, then I commented one thing in JS and another one in python (survey_content) | 18:33 |
| jamtoday | that's what it looks like...not hard to fix | 18:34 |
| jamtoday | What do you think about eliminating the link_id field for survey questions | 18:34 |
| ajaksu | you can revert it if it doesn't take too much time, otherwise I'm finishing a security fix for another project and will start melange hacking in about one hour :) | 18:35 |
| jamtoday | not eliminating the property, just the field. I think it can safely be auto-generated | 18:35 |
| jamtoday | ok thats fine | 18:35 |
| ajaksu | It's OK to get rid of it, we might also want to use the question instead of the link_id-like identifier for CSV and result lists | 18:36 |
| ajaksu | Or make the question actually a Name or Title, with textual question living inside the widget? | 18:37 |
| *** jamtoday has quit IRC | 18:41 | |
| *** jamtoday has joined #melange | 18:46 | |
| *** Merio has quit IRC | 18:55 | |
| *** florinciu has quit IRC | 19:05 | |
| jamtoday | ajaksu: new commit - field deletion bug fixed, css adjusted...selection fields aren't saving | 19:27 |
| jamtoday | and some useful new additions to seed_db | 19:28 |
| ajaksu | jamtoday: let me try to fix the selection fields :) | 19:28 |
| ajaksu | jamtoday: my guess is that it's the content-saving JS, testing... | 19:33 |
| jamtoday | you're right it's in the form submit function | 19:40 |
| jamtoday | or the form submit function is far too specific | 19:40 |
| jamtoday | we want to make sure these bugs don't keep re-emerging...i can fix this one | 19:40 |
| ajaksu | I actually wasn't able to reproduce it... let me check the merge | 19:44 |
| jamtoday | figured it out | 19:47 |
| jamtoday | it's only when you don't submit an option | 19:47 |
| jamtoday | That's actually the behavior we want, but I'll add a prompt if the survey contains selection fields with no options | 19:47 |
| jamtoday | also thinking that jquerytools would probably be a more efficient choice...it's about a third of the size for dialogs | 19:48 |
| jamtoday | but anyways, jQuery UI is fine | 19:48 |
| jamtoday | I fixed some issues with the delete option and sortable not working for selection options | 20:00 |
| ajaksu | jamtoday: div #survey_widget input {width:320px;} makes checkboxes look weird in Opera -> http://github.com/ajaksu/Melange/commit/c03cb00f885ec6e341d61754ce428c70e2a8d7f7 | 20:01 |
| tpb | <http://ln-s.net/3Kph> (at github.com) | 20:01 |
| *** dhans has quit IRC | 20:04 | |
| jamtoday | i was just editing css as well | 20:12 |
| jamtoday | conflicted merge | 20:12 |
| jamtoday | ....i'll fix this in a moment...after you can merge upstream with me | 20:13 |
| ajaksu | ok, I'm working on the deadline as interval now :) | 20:13 |
| jamtoday | merged and pushed | 20:22 |
| ajaksu | pushed the deadline as interval feature: we now have an 'opening' attribute that is the starting of the period in which a survey can be taken | 20:42 |
| *** jamtoday has quit IRC | 20:52 | |
| *** jamtoday has joined #melange | 20:59 | |
| *** jamtoday has left #melange | 20:59 | |
| *** mithro has joined #melange | 21:15 | |
| *** ChanServ sets mode: +v mithro | 21:15 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!