spaceone | jay, finally i found the time to fix the performance leak in circuits.http which was caused by using self.wait() | 00:03 |
---|---|---|
spaceone | serving a HTML page with dynamic content (4986 bytes) takes real 0m0.017s now on localhost. | 00:05 |
spaceone | this is fine I think | 00:05 |
pdurbin | spaceone: I bet they're keeping prologic busy. :) | 00:19 |
spaceone | ^^ | 00:19 |
*** Coldblackice has joined #circuits | 00:36 | |
prologic | spaceone: good work :) | 01:18 |
prologic | pdurbin: yes :) | 01:18 |
spaceone | ^^ | 01:19 |
spaceone | prologic: release of next circuits version still planned in dec. ? | 01:19 |
spaceone | i had too much work so i could not do much | 01:20 |
spaceone | I implemented a SAML service provider in our server components | 01:20 |
spaceone | prologic: i wrote a blog article about it: https://www.univention.com/2015/11/single-sign-on-for-ucs-4-1/ | 01:21 |
prologic | spaceone: yes I think so | 01:44 |
prologic | my stuff arrives tomorrow | 01:44 |
prologic | so hopefully I'll be able to spend a bit more free time on things | 01:44 |
prologic | cool (re blog) I'll have a read :) | 01:44 |
pdurbin | spaceone: do you like SAML? | 01:46 |
spaceone | no, i hate it, pdurbin | 01:46 |
pdurbin | lol | 01:46 |
pdurbin | spaceone: you can come commiserate with me in ##shibboleth some day :) | 01:47 |
spaceone | :D i have java even more | 01:47 |
spaceone | i am planning to create circuits.http.server.saml.sp | 01:48 |
spaceone | there is a really painful and bad designed but working python saml lib called pysaml2 | 01:49 |
pdurbin | spaceone: there's also https://github.com/onelogin/python-saml | 01:56 |
pdurbin | and https://github.com/pitbulk/python3-saml | 01:57 |
pdurbin | both mentioned at https://github.com/omab/python-social-auth/issues/388 | 01:58 |
spaceone | arg, you told me this 3 month too late :D | 02:00 |
pdurbin | spaceone: put it in circuits.http.server.saml.sp :) | 02:01 |
spaceone | this looks a lot more clean | 02:02 |
pdurbin | python3-saml? | 02:04 |
spaceone | no, https://github.com/onelogin/python-saml | 02:05 |
spaceone | but hm | 02:05 |
spaceone | it doesn't seem to be secure | 02:05 |
pdurbin | no? uh oh | 02:05 |
spaceone | they are using string interpolation to build up XML | 02:05 |
spaceone | without escaping anything | 02:05 |
spaceone | and they are raising Exception directly instead of a nice subclass | 02:06 |
spaceone | but in general it looks nice | 02:09 |
spaceone | way less code than pysaml2 | 02:09 |
spaceone | only one code path to get into specific state | 02:09 |
spaceone | s | 02:09 |
spaceone | better API | 02:09 |
spaceone | i'll have a closer look soon | 02:10 |
pdurbin | spaceone: maybe you'll love SAML! | 02:10 |
spaceone | thanks for telling me about this | 02:10 |
spaceone | pdurbin: lol, never... it's not working with REST | 02:10 |
pdurbin | spaceone: do you like OAuth better? | 02:10 |
spaceone | probably, but OAuth gets implemented in some month | 02:11 |
spaceone | so i have no experience with it yet | 02:11 |
pdurbin | spaceone: you'll be adding it to UCS? | 02:11 |
spaceone | yes | 02:11 |
pdurbin | spaceone: cool. after you do I'd like to hear your comparison of OAuth and SAML, please | 02:12 |
spaceone | okay | 02:12 |
pdurbin | :) | 02:12 |
*** linkedinyou has quit IRC | 05:55 | |
*** linkedinyou has joined #circuits | 07:57 | |
*** Coldblackice has quit IRC | 09:22 | |
*** Coldblackice has joined #circuits | 10:22 | |
*** Coldblackice has quit IRC | 12:39 | |
*** owrgmeucwcpcytpc has joined #circuits | 16:22 | |
*** Workster has quit IRC | 16:22 | |
*** Workster has joined #circuits | 16:22 | |
*** oneandzero has joined #circuits | 18:32 | |
oneandzero | test | 18:34 |
oneandzero | am i radiating? | 18:34 |
*** oneandzero has quit IRC | 18:40 | |
*** Coldblackice has joined #circuits | 19:48 | |
*** linkedinyou has quit IRC | 21:01 | |
*** Coldblackice has quit IRC | 21:33 | |
*** Coldblackice has joined #circuits | 23:48 |
Generated by irclog2html.py 2.11.0 by Marius Gedminas - find it at mg.pov.lt!