| *** brinzhang_ has joined #openstack-doc | 00:51 | |
| *** brinzhang has quit IRC | 00:54 | |
| *** goldyfruit has quit IRC | 03:01 | |
| *** KeithMnemonic1 has quit IRC | 03:06 | |
| *** KeithMnemonic has joined #openstack-doc | 03:12 | |
| openstackgerrit | Ian Wienand proposed openstack/openstack-manuals master: Move proposal job into project-config https://review.opendev.org/678133 | 03:26 |
|---|---|---|
| openstackgerrit | Ian Wienand proposed openstack/openstack-manuals master: Move proposal job into project-config https://review.opendev.org/678133 | 03:34 |
| openstackgerrit | Merged openstack/openstack-manuals master: Move proposal job into project-config https://review.opendev.org/678133 | 04:25 |
| *** kopecmartin|off is now known as kopecmartin | 05:55 | |
| openstackgerrit | Andreas Jaeger proposed openstack/training-guides master: WIP: Update to normal docs jobs https://review.opendev.org/678155 | 06:08 |
| AJaeger | amotoki, asettle, stephenfin, I'm now happy with the changes we have for PDF building, see: | 06:38 |
| AJaeger | https://review.opendev.org/#/c/678077 https://review.opendev.org/678142 and https://review.opendev.org/664555 | 06:38 |
| AJaeger | I'm testing this in https://review.opendev.org/#/c/678080 | 06:39 |
| AJaeger | thanks amotoki to getting us there! | 06:39 |
| amotoki | yay! I hope the zuul-job patch lands soon :) | 07:03 |
| amotoki | We also need to send a mail to announce our current status and the basic instructions | 07:04 |
| *** jawad_axd has joined #openstack-doc | 07:07 | |
| AJaeger | I'll ping for those changes to be merged... | 07:08 |
| *** rcernin has quit IRC | 07:15 | |
| *** miloa has joined #openstack-doc | 07:42 | |
| AJaeger | amotoki: first +2s coming in ;) | 08:29 |
| asettle | AJaeger, ohhhhhhh :o | 08:57 |
| asettle | This is exciting! | 08:57 |
| * AJaeger just gave a -1 on https://review.opendev.org/#/c/664555/ - this will not work in general as is yet. | 08:58 | |
| asettle | Sure sure | 08:58 |
| AJaeger | no time to dig deeper - asettle if you want to discuss what to do ("design"), it would be nice. | 08:59 |
| asettle | AJaeger, talking about the template design of the PDF? | 08:59 |
| AJaeger | amotoki, asettle : https://review.opendev.org/#/c/664555/ is merged ;) | 08:59 |
| amotoki | \o/ | 09:00 |
| * AJaeger needs to step out for a bit and leaves 664555 for you ;) | 09:00 | |
| asettle | Uhhh | 09:02 |
| asettle | stephenfin, you got any clarity here cause I don't? | 09:03 |
| stephenfin | whuh? | 09:04 |
| asettle | Andreas was asking me to look into the design? I'm confused? | 09:05 |
| AJaeger | asettle: read my last comment on 664555, please | 09:07 |
| AJaeger | design : What should happen if api-ref job is run and the tox.ini has a pdf environment? | 09:07 |
| asettle | I mean, I have | 09:07 |
| asettle | Oh. Thank you | 09:07 |
| asettle | :) | 09:07 |
| AJaeger | Figure out what we need - and once that's clear, we can define what to change... | 09:08 |
| AJaeger | will be out for a bit... | 09:08 |
| asettle | o/ | 09:09 |
| -amotoki- is a bit tired by working on common jobs with nested dependencies :) | 09:12 | |
| amotoki | AJaeger: good catch on build-openstack-api-ref/guide | 09:16 |
| amotoki | as of now, we assume pdfdocs tox env is a counter part of docs tox env | 09:17 |
| amotoki | we can add similar variants for api-ref and api-guide but I am not sure this is what stephenfin wants. | 09:18 |
| amotoki | if we go to this route, '-pdf' suffix or 'pdf-' prefix would make more sense | 09:20 |
| -amotoki- will be offline next one or two hours | 09:21 | |
| asettle | amotoki, I don't see the need at this point to gen the api-ref and api-guide into PDFs. Perhaps that could almost be a subsequent ideal if someone so wants. But at this point in time, the main priority is the user facing content. | 09:44 |
| *** brinzhang_ has quit IRC | 09:46 | |
| stephenfin | We can't generate PDFs for the api-guide anyway since the plugin only supports HTML output | 10:36 |
| AJaeger | ok, I'll tweak the job late today... | 10:55 |
| asettle | Good to know. Thanks AJaeger | 11:02 |
| *** tesseract has joined #openstack-doc | 11:12 | |
| *** goldyfruit has joined #openstack-doc | 11:14 | |
| *** hemna has joined #openstack-doc | 11:40 | |
| amotoki | asettle: I agree our primary goal is a user-facing document (though whether the contributor doc is a user-facing document is a question for a long time since doc migation). The problem is if "pdfdocs" tox env exists PDF build will be run for API ref or API guide too (unintentionally). | 12:55 |
| amotoki | asettle: A question is how to disable PDF build for them. | 12:55 |
| amotoki | asettle: I see two ways; the one is to have a flag to disable PDF build. | 12:56 |
| amotoki | asettle: the other way is to define a prefix or a suffix for PDF build. this means if a tox env for HTML build is 'docs' and a prefix 'pdf-' is for PDF build, when 'pdf-docs' exists PDF build will run. | 12:57 |
| amotoki | by doing so, for API ref, there is no 'pdf-api-ref' by default so PDF build will not be run. | 12:58 |
| asettle | Thanks for the explanation amotoki | 12:58 |
| AJaeger | amotoki: I'll fix the change so that it will not be run for API* jobs anymore. | 12:59 |
| asettle | I commented on the PR asking whether or not it would be easier to disable the PDF jobs? AJaeger 's suggestion was to have another tox env | 12:59 |
| AJaeger | amotoki: One way: Create a new base job, have api-ref, api-guide and openstack-tox-docs inherit from the new base job... | 12:59 |
| AJaeger | But adding a flag would work as well | 13:00 |
| asettle | The flag is an interesting options - provides choice. But then we're still building huge PDFs with both docs and api docs | 13:00 |
| asettle | Which, thinking out loud, would that not be incredibly messy and enormous? | 13:00 |
| amotoki | AJaeger: or we can update openstack-tox-docs which look for a tox env whose name is 'pdf-<tox-env-for-docs>' | 13:01 |
| AJaeger | amotoki: yes, that works as well... | 13:01 |
| AJaeger | and is one way to disable it ;) | 13:01 |
| AJaeger | name it "this-environment-should-not-exist-otherwise-you-are-screwed" ;) | 13:02 |
| amotoki | A simplest way at the moment seems to introduce a flag in openstack-tox-docs and specify the flag to false in api-ref/api-guide job. | 13:02 |
| amotoki | hmm... when I search "parent: openstack-tox-docs" in codesaerch.o.o, there are more jobs who inherit openstack-tox-docs. | 13:04 |
| AJaeger | those all build the same docs, so no problem for these | 13:08 |
| amotoki | AJaeger: btw, build-openstack-api-(ref|guide) inherit opendev-tox-docs, so changes in openstack-tox-docs does not affect them. | 13:08 |
| AJaeger | You only need to care about those thta change teh tox environemn | 13:08 |
| AJaeger | amotoki: I need to update my tree ;) | 13:08 |
| amotoki | Talking about jobs in openstack-zuul-jobs, only build-openstack-deploy-guide might be affected. | 13:11 |
| AJaeger | amotoki: let me rebase the change to master... | 13:33 |
| AJaeger | amotoki: I need to rebase the build-api- jobs on openstack-tox-docs for requiremetns files | 13:35 |
| amotoki | the work on the common job for PDF build is much more complicated that what I thought first..... I wonder folks who want to support PDF build imagined it :( | 13:38 |
| hemna | morning | 13:40 |
| hemna | how is it all goin? | 13:40 |
| amotoki | hemna: we are still working on the common job. we've found another problem on a way to support/detect 'pdfdocs' in the common job definition.... I am not sure the common job lands today. | 13:41 |
| * AJaeger is updating it right now | 13:44 | |
| hemna | ok | 13:46 |
| hemna | I've been away the last 2 days at the cinder midcycle | 13:46 |
| asettle | How was it? | 13:47 |
| hemna | it was good | 13:47 |
| hemna | small team this time | 13:47 |
| hemna | I still haven't found a solution to the memory issue with TeX | 13:47 |
| *** jawad_axd has quit IRC | 13:49 | |
| amotoki | hemna: I think it happens for a sample config file, right? | 13:51 |
| * AJaeger just updated https://review.opendev.org/664555 - what do you think? | 13:53 | |
| AJaeger | amotoki: this renames the tox environment to "pdf-docs" | 13:53 |
| *** jawad_axd has joined #openstack-doc | 13:53 | |
| AJaeger | and offers a way to disable it. | 13:53 |
| AJaeger | Did I cover everything? ;) | 13:54 |
| amotoki | AJaeger: looking at it.. | 13:54 |
| AJaeger | thanks | 13:55 |
| * asettle goes to review | 13:56 | |
| *** jawad_axd has quit IRC | 13:58 | |
| asettle | Looks good, other than the one question I had inline | 14:01 |
| amotoki | I think it is good for now. In future, we can calculate tox_pdf_docs from tox_envlist and we can run PDF build automatically when tox_pdf_docs exists. | 14:02 |
| amotoki | so we can drop tox_pdf_docs and tox_skip_pdf | 14:03 |
| AJaeger | asettle: I'm sure stephenfin got it working - but it might be a different package list. | 14:08 |
| AJaeger | So, unless he confirms that the list as is works on Fedora, let's leave it. | 14:08 |
| AJaeger | thanks, amotoki | 14:08 |
| asettle | AJaeger, that's fair. It would be nice to confirm at least one other distro. hemna - by chance, are you testing with SUSE? | 14:14 |
| asettle | Just referencing my comment here: https://review.opendev.org/#/c/664555/21/roles/prepare-build-pdf-docs/README.rst | 14:14 |
| amotoki | AJaeger: asettle: one thing I know is Fedora has no xindy package, but latex_use_xindy = False is our current practice, so it might not be a problem. | 14:17 |
| * asettle thinks | 14:17 | |
| asettle | Hmm okay | 14:17 |
| asettle | stephenfin, would be good if you could confirm when you have a mo | 14:17 |
| amotoki | AJaeger: asettle: other than that, bindep.txt in several projects covers ubuntu and rpm based distros. | 14:18 |
| asettle | Good to know | 14:21 |
| AJaeger | something is wrong, https://zuul.opendev.org/t/openstack/build/fe819c27f21740f2b52f8bdd4f3c6fed only has PDF, no HTML ;( | 14:22 |
| asettle | Haha oh no | 14:22 |
| AJaeger | amotoki: yes, I know that it covers that - and I'm happy to change the text or code | 14:22 |
| * AJaeger needs coffee and will the ncheck... | 14:22 | |
| asettle | AJaeger, this looks good though! https://openstack.fortnebula.com:13808/v1/AUTH_e8fd161dc34c421a979a9e6421f823e9/logs_80/678080/4/check/openstack-tox-docs/fe819c2/pdf/horizon.pdf | 14:24 |
| amotoki | but no html artifacts.... | 14:24 |
| asettle | Nope | 14:25 |
| amotoki | AJaeger: looking at ara report, both tox runs were called with tox -epdf-docs..... our expectation is tox -e docs and tox -e pdf-docs though.. | 14:26 |
| amotoki | is confused with what happens..... | 14:29 |
| AJaeger | amotoki: yes, that's the problem - now figuring out why... | 14:31 |
| *** jawad_axd has joined #openstack-doc | 14:46 | |
| hemna | back, my memory issue happens when building the cinder pdf | 14:49 |
| *** kopecmartin is now known as kopecmartin|off | 14:51 | |
| *** jawad_axd has quit IRC | 14:51 | |
| amotoki | hemna: are you still hitting this? | 14:57 |
| AJaeger | amotoki: didn't you had a way with latex_elements and memory increase? | 14:58 |
| amotoki | AJaeger: I didn't. in neutron and nova doc, I avoid this by skipping inline sample config files IIRC. | 14:59 |
| amotoki | maxlistdepth=10 helps us when we hit deeper list issue | 15:00 |
| AJaeger | ah | 15:00 |
| AJaeger | import_role is bad, include_role did the trick - https://zuul.opendev.org/t/openstack/build/386ee418bde04aabafafbc2967cfa430 looks fine, doesn' it? | 15:01 |
| amotoki | AJaeger: ah.... I should check the difference between include_role and import_role, but I didn't know improt_role affects other same role in a playbook.. | 15:04 |
| AJaeger | amotoki: import is at initial setup, include evaluates when used. I didn't know that either ;( | 15:06 |
| amotoki | AJaeger: thanks. I now seem to understand the description of import_role in the ansible doc. | 15:08 |
| *** jawad_axd has joined #openstack-doc | 15:11 | |
| hemna | I just tried to run my cinder patch against Leap | 15:12 |
| hemna | and it failed | 15:12 |
| hemna | it failed to install the requirements | 15:13 |
| hemna | amotoki: but yes, I am still hitting the memory issue | 15:13 |
| *** jawad_axd has quit IRC | 15:16 | |
| hemna | Could not find a version that satisfies the requirement Sphinx===2.2.0 | 15:16 |
| hemna | (on Leap 15.1 | 15:16 |
| AJaeger | I can install sphinx on Leap 15.1 wihtout problems, question is what is missing for it... | 15:18 |
| hemna | hrmm | 15:19 |
| hemna | I'm on master now too and trying to run tox -edocs | 15:19 |
| hemna | fails on sphinx | 15:19 |
| amotoki | Leap is openSUSE? | 15:20 |
| amotoki | pip installs sphinx from PyPI, so I don't see any reason which depends on distros.... | 15:20 |
| AJaeger | yes, openSUSE hsa two flavors - openSUSE Leap and openSUSE tumbleweed (the rolling release) | 15:21 |
| AJaeger | amotoki: yes, I installed via pip as well | 15:21 |
| AJaeger | hemna: can you paste the output? | 15:21 |
| AJaeger | asettle: could you +2A https://review.opendev.org/677802, please? | 15:23 |
| AJaeger | hemna: do you have cinder changs with pdf building up that I can look at with 664555 ? | 15:31 |
| *** jamesmcarthur has joined #openstack-doc | 15:47 | |
| hemna | sure just a sec | 15:50 |
| hemna | https://review.opendev.org/#/c/677301/ | 15:53 |
| hemna | that's my cinder patch to build pdf | 15:53 |
| hemna | http://paste.openstack.org/show/762691/ | 15:58 |
| hemna | that's the cinder tox -edocs failure on master in Leap 15.1 | 15:58 |
| hemna | not sure why it can't install sphinx 2.2.0 | 15:59 |
| AJaeger | hemna: want to make 677301 depends-on 664555 ? | 16:01 |
| hemna | I suppose we can | 16:03 |
| hemna | I'm just trying to run it locally | 16:03 |
| hemna | think that patch will fix the memory issue? | 16:03 |
| hemna | ok so you are naming the job 'pdf-docs' not 'pdfdocs' | 16:03 |
| AJaeger | hemna: mmh, does cinder not use python3 for base-python? | 16:03 |
| AJaeger | hemna, yes, pdf-docs | 16:03 |
| AJaeger | so, we can use pdf-api-ref later | 16:04 |
| hemna | ok I'll have to update my 3 patches to rename it to pdf-docs | 16:04 |
| hemna | the docs job in cinder tox.ini is python3 | 16:04 |
| AJaeger | hemna, do you have python3 installed? | 16:05 |
| hemna | https://github.com/openstack/cinder/blob/master/tox.ini#L134 | 16:05 |
| AJaeger | hemna: indeed - my local copy was on pike... | 16:05 |
| hemna | yup | 16:06 |
| hemna | python3 rungs | 16:06 |
| hemna | runs | 16:06 |
| AJaeger | hemna: tox -e docs works fine for me - and that paste is really strange, I'm out of ideas | 16:08 |
| hemna | yah I'm confused as to why it fails | 16:09 |
| AJaeger | hemna: it looks like sphinx is not available. Are you using a local pypi mirror? | 16:11 |
| *** tesseract has quit IRC | 16:11 | |
| AJaeger | line 154 says that version 1.8.5 is the latest it could find - so, where do you get sphinx from? | 16:11 |
| hemna | http://paste.openstack.org/show/762693/ | 16:17 |
| hemna | I tried to pip install it manually into the same venv | 16:17 |
| hemna | it shows the list of versions it knows about for sphinx | 16:17 |
| hemna | 1.8.5 is the latest | 16:17 |
| hemna | ?@! | 16:18 |
| hemna | no custom ~/.pip either | 16:19 |
| hemna | https://pypi.org/project/Sphinx/#files | 16:19 |
| hemna | 2.2.0 is the only version available on pypi | 16:19 |
| hemna | so I'm confused | 16:19 |
| AJaeger | 1.8.5 is the latest python2 version. Is your venv really using python3? | 16:20 |
| AJaeger | hemna: doc/requirements.txt has: | 16:20 |
| AJaeger | sphinx!=1.6.6,!=1.6.7,>=1.6.3,<2.0.0;python_version=='2.7' # BSD | 16:20 |
| AJaeger | So, if your venv by accident uses python2 instead of python3, that might be the reaons | 16:21 |
| *** AJaeger_ has joined #openstack-doc | 16:31 | |
| *** AJaeger has quit IRC | 16:35 | |
| *** AJaeger_ is now known as AJaeger | 16:38 | |
| *** AJaeger_ has joined #openstack-doc | 16:47 | |
| *** AJaeger has quit IRC | 16:51 | |
| *** miloa has quit IRC | 17:02 | |
| hemna | ok | 17:03 |
| hemna | python2 is uninstalled | 17:04 |
| hemna | python3.4 is what's on the system | 17:04 |
| AJaeger_ | 3.4? I have 3.6.5 installed | 17:05 |
| AJaeger_ | hemna, re-install python3 package, I think something is strange with your system if you still have 3.4 | 17:08 |
| *** jamesmcarthur has quit IRC | 17:11 | |
| *** jamesmcarthur has joined #openstack-doc | 17:11 | |
| amotoki | hemna: I succeeded to build PDF file for cinder doc. Is it okay to update your cinder change? | 17:12 |
| amotoki | hemna: As OpenStack developers, 3.6 and 3.7 are the expected version for python3 now :) | 17:13 |
| hemna | yah, this is Leap 15.1 | 17:14 |
| hemna | trying to do an update now to see if it helps | 17:14 |
| *** jamesmcarthur has quit IRC | 17:22 | |
| amotoki | hemna: it looks like that literalinclude of cinder.conf sample file causes memory issue. Commenting out that literalinclude solves the memory issue. | 17:30 |
| amotoki | hemna: In addition, I need to set maxlistdepth: 10 in latex_elements. | 17:30 |
| amotoki | hemna: by these two changes I can generate the PDF file for cinder. | 17:30 |
| amotoki | good night | 17:33 |
| hemna | ok that sounds good | 17:51 |
| *** cp has quit IRC | 17:58 | |
| AJaeger_ | amotoki, good night! | 18:14 |
| AJaeger_ | asettle, stephenfin, hemna , amotoki : All changes are in ;) Happy PDF Building :) | 18:14 |
| *** AJaeger_ is now known as AJaeger | 18:14 | |
| openstackgerrit | Merged openstack/contributor-guide master: use promote job https://review.opendev.org/677802 | 18:56 |
| openstackgerrit | Merged openstack/openstack-manuals stable/newton: Update for openstack-doc-tools 2.0 https://review.opendev.org/678031 | 18:59 |
| AJaeger | stephenfin, asettle, are you going to send an email to the mailing list with updated instructions about PDF goal? you should be unblocked now ;) | 19:00 |
| openstackgerrit | Merged openstack/openstack-manuals stable/ocata: Update for openstack-doc-tools 2.0 https://review.opendev.org/678036 | 19:02 |
| hemna | oh well, I can't get python to upgrade to 3.7 on leap | 19:04 |
| * hemna is giving up | 19:04 | |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-manuals stable/newton: Update URL of contributor guide https://review.opendev.org/678289 | 19:14 |
| *** jamesmcarthur has joined #openstack-doc | 19:28 | |
| *** jamesmcarthur has quit IRC | 19:32 | |
| *** jamesmcarthur has joined #openstack-doc | 19:32 | |
| *** jamesmcarthur has quit IRC | 19:43 | |
| openstackgerrit | Merged openstack/openstack-manuals stable/newton: Update URL of contributor guide https://review.opendev.org/678289 | 19:44 |
| *** jamesmcarthur has joined #openstack-doc | 19:45 | |
| *** jamesmcarthur has quit IRC | 19:50 | |
| *** jamesmcarthur has joined #openstack-doc | 20:14 | |
| AJaeger | asettle, amotoki, stephenfin, have a look at the nova PDF at https://review.opendev.org/#/c/676730 and read my comment ;( Hope either of you can look into this at some time... | 20:17 |
| * AJaeger joins hemna in giving up | 20:17 | |
| *** jamesmcarthur has quit IRC | 20:18 | |
| *** jamesmcarthur has joined #openstack-doc | 20:19 | |
| AJaeger | we could probably enforce to use the projectname (that's a zuul variable) | 20:23 |
| *** jamesmcarthur has quit IRC | 21:46 | |
| *** jamesmcarthur has joined #openstack-doc | 21:48 | |
| *** jamesmcarthur has quit IRC | 21:50 | |
| *** jamesmcarthur has joined #openstack-doc | 21:51 | |
| *** jamesmcarthur has quit IRC | 22:07 | |
| *** jamesmcarthur has joined #openstack-doc | 22:07 | |
| *** rcernin has joined #openstack-doc | 22:11 | |
| *** KeithMnemonic has quit IRC | 22:13 | |
| *** jamesmcarthur has quit IRC | 22:13 | |
| *** jamesmcarthur has joined #openstack-doc | 22:31 | |
| *** jamesmcarthur has quit IRC | 22:37 | |
| *** jamesmcarthur has joined #openstack-doc | 22:39 | |
| *** jamesmcarthur has quit IRC | 22:46 | |
| *** jamesmcarthur has joined #openstack-doc | 22:50 | |
| *** jamesmcarthur has quit IRC | 23:01 | |
| *** rcernin has quit IRC | 23:11 | |
| *** rcernin has joined #openstack-doc | 23:12 | |
| *** jamesmcarthur has joined #openstack-doc | 23:15 | |
| *** jamesmcarthur has quit IRC | 23:17 | |
| *** jamesmcarthur has joined #openstack-doc | 23:19 | |
| *** jamesmcarthur has quit IRC | 23:22 | |
| *** jamesmcarthur has joined #openstack-doc | 23:22 | |
| *** jamesmcarthur has quit IRC | 23:32 | |
| *** jamesmcarthur has joined #openstack-doc | 23:38 | |
| *** jamesmcarthur has quit IRC | 23:42 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!