| Laney | yeah I gave up and went back to a lucid container :P | 09:38 |
|---|---|---|
| lifeless | :P | 09:40 |
| wgrant | Things seem to be a lot smoother with a fresh late-Precise-generated Lucid container | 09:44 |
| alexh | hi | 16:02 |
| jelmer | hi alexh | 16:03 |
| alexh | any hints on why the bug activity API returns an empty entries set but a non-zero total_size? | 16:03 |
| alexh | e.g. https://api.launchpad.net/1.0/bugs/123920/activity | 16:03 |
| _mup_ | Bug #123920: Bluetooth Logitech Dinovo Keyboard/Mouse don't work <cft-2.6.27> <iso-testing> <linux-source-2.6.22> <metabug> <patch> <Bluez-utils:New> <bluez (Ubuntu):Triaged> <linux (Ubuntu):Invalid> <linux-source-2.6.22 (Ubuntu):Won't Fix> <udev (Ubuntu):Invalid> < https://launchpad.net/bugs/123920 > | 16:03 |
| alexh | happens with every bug I've tried, not just that one, fwiw | 16:05 |
| Laney | Is there a make harness that lets me do doctest stuff? | 18:53 |
| Laney | (or, what do I import to get that stuff?) | 18:57 |
| === almaisan-away is now known as al-maisan | ||
| === al-maisan is now known as almaisan-away | ||
| lifeless | Laney: what do you mean? | 19:22 |
| Laney | I'm trying to make a doctest and it'd be good to be able to test this interactively | 19:26 |
| Laney | but I don't know what to import into the harness repl to get those functions | 19:26 |
| Laney | it doesn't really matter; I'm feeling it out without this. | 19:26 |
| lifeless | ah | 19:29 |
| lifeless | so we consider doctests one step up from the spawn of evil | 19:30 |
| lifeless | you might like to not make one | 19:30 |
| Laney | yeah? | 19:30 |
| Laney | I thought they were the way to test UI changes | 19:30 |
| lifeless | yah | 19:30 |
| lifeless | very much deprecated | 19:30 |
| lifeless | having testable *documentation* is still good, and if you are writing docs you expect folk to read, feel free to make them a doctest | 19:30 |
| lifeless | but such are not replacements for unit tests or integration tests | 19:31 |
| cjwatson | Laney: unfortunately the codebase is not always a good indicator of modern practice :-/ | 19:48 |
| cjwatson | another approach is: make your test fast enough that instead of running things interactively in a harness, you can just repeatedly run the test | 19:49 |
| Laney | cjwatson: I don't know how to test the UI then :P (unless I am misusing the terminology and interactive tests are something different from doctests) | 19:50 |
| Laney | the reason I want to use the harness is because I don't really have a view of the test interfaces yet so it's most efficient to just be able to bash on stuff | 19:51 |
| Laney | fyi, the file I am proposing to edit is lib/lp/soyuz/stories/soyuz/xx-distributionsourcepackagerelease-pages.txt | 19:51 |
| cjwatson | doctests are noninteractive | 19:51 |
| Laney | the repl-style tests I mean | 19:52 |
| cjwatson | well, anything you can do in that can be recast as a unit test fairly straightforwardly | 19:52 |
| cjwatson | just a matter of learning the various self.assert* helpers | 19:53 |
| Laney | indeed. I just thought these were the way that UI tests were done. | 19:53 |
| lifeless | so a UI test could mean: | 20:05 |
| lifeless | - JS tests | 20:05 |
| lifeless | - html content checking | 20:05 |
| lifeless | - non-ajax form checking | 20:05 |
| lifeless | - ajax integration test | 20:05 |
| lifeless | perhaps a better question is, what change are you making ? | 20:06 |
| lifeless | also, morning all | 20:06 |
| lifeless | bbiab, shopping run | 20:06 |
| mwhudson | lifeless: did you see http://cortesi.github.com/pathod/ | 22:22 |
| lifeless | no, looking | 22:23 |
| lifeless | nice | 22:23 |
| lifeless | fault injection test server. woot. | 22:23 |
| lifeless | nice - 200@100k,ascii_letters | 22:25 |
| wgrant | alexh: You're unauthenticated. | 22:34 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!