mhall119 | jose: hmm, I thought we got summit moved to https | 02:07 |
---|---|---|
jose | nope, still using http for me | 02:07 |
mhall119 | jose: you should file an RT to get it using https, there's nothing to be done in the code | 02:16 |
jose | will do | 02:16 |
mhall119 | thanks | 02:16 |
dholbach | good morning | 07:30 |
dholbach | davidcalle: I'm looking at a very strange phenomenon with using django.test.Client.get() to test local links | 08:51 |
dholbach | the following used during the test: | 08:51 |
dholbach | print((url, type(res), len(res.content))) | 08:51 |
dholbach | gives: | 08:51 |
dholbach | ('/file2/', <class 'django.http.response.HttpResponseRedirect'>, 0) | 08:51 |
dholbach | ('/en/file2/', <class 'django.http.response.HttpResponseNotFound'>, 12641) | 08:51 |
dholbach | HttpResponseNotFound doesn't seem to be what we think it is :) | 08:52 |
dholbach | I think I'll just test for len(django.test.Client.get(url).content)>0 and that's it :) | 08:53 |
davidcalle | dholbach, indeed, I was actually wondering the same thing about httpresponsenotfound when I tried your test last week, maybe Django doesn't do exactly the same thing when it's being tested VS in browser. | 09:02 |
dholbach | davidcalle: yep, looks like it | 09:09 |
dholbach | hum... len(res.content)>0 also doesn't work - the 404 page also has content :) | 09:17 |
dholbach | davidcalle: finally fixed it by using cms.test_utils.testcases.CMSTestCase.get_request() | 14:01 |
dholbach | (and fixed a couple of other bits along the way) | 14:01 |
davidcalle | dholbach: Nice! | 14:01 |
dholbach | davidcalle: links are working now and are part of the test suite | 14:21 |
dholbach | I think we're getting close to having this usable | 14:22 |
* davidcalle tests | 14:25 | |
dholbach | davidcalle: do you remember what we did to fix things like https://bugs.launchpad.net/developer-ubuntu-com/+bug/1531200? | 14:51 |
ubottu | Launchpad bug 1531200 in Ubuntu Developer Portal "Snappy/Snapcraft docs don't show sublists" [Undecided,New] | 14:51 |
davidcalle | dholbach: I don't, let me try | 15:14 |
mhall119 | davidcalle: it looks like the new JS scopes API doc package still hasn't hit their PPA, is something blocking it? | 15:15 |
davidcalle | dholbach: bzr messed up while importing the pip cache of your branch, setting it up again, hold on :) | 15:17 |
dholbach | ah yes... I might need to add this new module to pip | 15:20 |
dholbach | I mean adding the markdown extension to the cache | 15:22 |
dholbach | brb, need to restart my session - my mouse pointer just vanished | 15:22 |
davidcalle | For this specific file, with nested lists, this works: html=markdown.markdown(md_text, tab_length=3) | 15:28 |
davidcalle | dholbach: I guess (and hope) that matches github way of doing nested lists, so this would be consistent. | 15:29 |
davidcalle | (And not mess up other files) | 15:29 |
davidcalle | dholbach: do we have nested lists somewhere else? (/me looks at snappy/security) | 15:29 |
dholbach | davidcalle: hum, do we know that that's consistent with what github does? | 15:30 |
dholbach | snappy - docs/meta.md | 15:31 |
dholbach | snappy - docs/rest.md | 15:32 |
dholbach | snappy - docs/security.md | 15:32 |
dholbach | that's all files with nested lists AFAICS right now | 15:32 |
davidcalle | dholbach: so, other docs (that imports fine with nested ul), follow two rules: no spaces before a bullet point "* ", and 4 spaces before a nested one " * " | 15:33 |
davidcalle | dholbach: snapcraft doc writer is clearly evil. | 15:33 |
davidcalle | (and I'm just going to disregard the fact that we should probably write guidelines for imported md :) ) | 15:35 |
davidcalle | dholbach: are you used to do github mps? Maybe you can teach me? :) | 15:37 |
dholbach | I failed a couple of times already ;-) | 15:37 |
dholbach | I'm happy to help | 15:37 |
davidcalle | dholbach: https://github.com/ubuntu-core/snapcraft/pull/218 \o/ Thanks :) | 16:02 |
dholbach | nice work | 16:03 |
dholbach | davidcalle: so the one image we currently have (on https://github.com/ubuntu-core/snapcraft/blob/master/README.md) works because it points to a remote URL - should the importer block if it finds a local image (for now... until we have a better idea how to fix it)? | 16:22 |
davidcalle | dholbach: tha's probably the best idea, yeah | 16:22 |
dholbach | ok | 16:22 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!