#dev 2021-12-09

2021-12-09 UTC
KartikPrabhu joined the channel
#
[KevinMarks]
I have some proportion of that in unmung
#
[KevinMarks]
then convert to jf2
KartikPrabhu and campegg joined the channel
#
Loqi
unmung has 1 karma over the last year
#
[snarfed]
unmung++
[jeremycherfas] joined the channel
#
Loqi
unmung has 2 karma over the last year
klez, lagash, rrix_, kogepan, [manton] and [Sean] joined the channel
#
capjamesg
I saw it on your unmung home page.
#
[KevinMarks]
I have a few very rough versions on unmung - indiecards, story cards, and the hovercards I was playing with. I should tidy them up
tetov-irc and akevinhuang2 joined the channel
#
Loqi
ok, I added "https://americanexpress.io/yak-shaving/" to the "See Also" section of /issue_tracker https://indieweb.org/wiki/index.php?diff=78249&oldid=77981
Seb[d] and jeremycherfas joined the channel
#
[snarfed]
[Sean] may be easier to talk here than on https://github.com/snarfed/bridgy/issues/1094
#
Loqi
[jalsco] #1094 Probably is "user-error" but still I need some assistance ✨please🙏🏻
#
[snarfed]
I'm still not seeing h-entry (or h-feed) anywhere on https://grok.starkraving.asia/
#
[Sean]
hey
#
[Sean]
wait
#
[Sean]
seeing h-entry (or h-feed) - on the home page is being done via that <link>
#
[snarfed]
what? I don't follow
#
[snarfed]
oh, sorry, ok, looking
#
[Sean]
link `
#
[Sean]
```<link rel="feed" type="text/html" href="https://grok.sta..... blahblah... hfeed ">```
#
[snarfed]
that's an XML file, served with content-type: text/xml; charset=utf-8. h-entry etc need to be in an HTML file
#
[Sean]
it was html yesterday -
#
[Sean]
i changed to xml so easier to click thru from github article
#
[Sean]
hahahaha
#
Loqi
hehe
#
[snarfed]
it seems like microformats are giving you a lot trouble. at this point I'd strongly recommend either switching to a theme that has them built in, or take a break and take some time to learn microformats more deeply
#
[Sean]
u think we should stop now - when just changing that back from `xml` to `html` is all that is needed ?
#
[Sean]
not really trouble
#
[snarfed]
you've hit a lot of problems so far, largely around microformats. I expect this probably isn't the last 😁
#
[Sean]
better to work through it and understand each item - for future sites
#
[snarfed]
up to you!
#
[snarfed]
in that case, again, I'd recommend slowing down on this and take a day or two to really learn and understand https://microformats.org/wiki/microformats2 from the beginning
#
[Sean]
ok - it is changed back to HTML and the cdn cache cleared ...
#
[Sean]
↩️ yes we have spent many hours on that page over last few weeks.
#
[snarfed]
not seeing any change on the home page. you might want to turn off the CDN entirely while you're developing, and only turn it back on for launch
#
[Sean]
maybe clear yr browser cache
#
[snarfed]
I'm using curl
#
[snarfed]
gotta run, good luck! back later
#
[Sean]
`curl https://grok.starkraving.asia | grep hfeed`
#
[Sean]
Is a mystery to me ... will try bit later.. thanks again for all your time.
#
[Sean]
shows
#
[Sean]
`<link rel="feed" type="text/html" href="https://grok.starkraving.asia/.well-known/hfeed.html">`
#
[Sean]
I am not able to turn off the CDN , but I cannot explain why it show `hfeed.xml` to some place in the world. Maybe it takes time for CDN to do its thing. I really don't know
gRegor joined the channel
#
[snarfed]
[Sean] looks like it worked! you just needed to click the _Crawl now_ button, or wait for the scheduled crawl. I clicked that button and Bridgy found webmention targets
#
[snarfed]
(it didn't _send_ them, since https://grok.starkraving.asia/2021/12/08/qldpolicevax/ doesn't advertise a webmention endpoint, but that's easy to fix)
sp1ff joined the channel
#
petermolnar
https://github.com/solid/community-server/ - haiyaa... this is the whole let's throw random standards out there "for people who want to use it". Long live XHTML 2 Strict!
#
[Sean]
yes i see now 👍👍😁 the message has gone to "No webmention support:" whereas before it was " No webmention targets"
#
[Sean]
you say ... "*thats easy to fix*" ?
#
[Sean]
```you mean like adding `<link rel="webmention" href="https://webmention.io/grok...etc ` ```
#
[Sean]
to the article page
#
[Sean]
???
#
capjamesg[d]
You can read about how to set up webmention on the webmention.io site.
#
[Sean]
yeah it has laready been working - including the responses in backfeed - thats not the problem
#
[snarfed]
[Sean] yes
#
capjamesg[d]
The “Use it on your site” section on the homepage provides the meta tag(s) you can use to set up webmention and ping backs.
#
[Sean]
the problem is - taking out the EXPLICIT LINK from the tweet/posting ... means all this extra
#
[Sean]
↩️ all articles - and all reply tweets etc - that do not explicitly tweet the LINK -- ... becasue before (when backfeed was working ) the only difference --- when none of this extra was needed -- was the actual tweet contained the link... All this is just to get past that ... Anyway, at least we are there now !
#
[Sean]
↩️ but thats all setup and working for days/weeks already
#
[snarfed]
Bridgy has to know which post on your site goes with which tweet somehow. Right now it looks in either the tweet itself or your post
#
[Sean]
↩️ these are to regain the functionality that we already had working after - taking out the explicit link - and - unfortunately not being able to let bridgy handle the shortening automatically (due to the network arch here ) ---
#
[Sean]
↩️ Thanks again - I will now sleep - already 3:30am here .. I will make the changes tomorrow . Actually have some bigger probs here (unrelated to this). Really not too difficult. There are some gaps in the doc. But I am amazed that Bridgy ONLY needs the webmention endpoint in the posting-article when the URL of that posting is not contained in the tweet. ...
#
[Sean]
↩️ when the URL-of-article is in the tweet text, how is Bridgy able to work out the webmention endpoint ?? I thought it would need to be told what it is is that isntance also ... I can only guess it gets it from the twitter-profile-linking-t--home-page which contains the endpoint
#
[Sean]
↩️ if that guess is correct - then it should be able to get it from there in either case - whether the URL is contained in tweet text or not .
#
[Sean]
↩️ but it cannot - so i will add it to all articles
#
[Sean]
↩️ no problem ! 😁👍🙏✨
#
[snarfed]
↩️ Bridgy fetches your article's URL and looks for the webmention endpoint there. that's how webmention discovery always works!
#
[Sean]
↩️ and , even not there, the replies/reactions DID work when only difference was the <article-link> in original tweet ... not any endpoint - except on the homepage.
#
[Sean]
↩️ where it gets the endpoint when <kink contained in tweet text> ?
#
[Sean]
↩️ only can be - from the homepage - becasue thats the only place that has it
#
[Sean]
↩️ anyway... all good - just curiosity now --- good noght
#
[Sean]
↩️ 😉
#
[snarfed]
↩️ webmention discovery generally still needs the endpoint in every page. Bridgy is a bit more complicated in that it briefly caches your endpoint and uses it for your whole site, but that's a minor Bridgy detail that's not true more broadly
KartikPrabhu joined the channel
#
capjamesg[d]
I am struggling to get the sphynx autodoc function feature to work with read the docs.
#
capjamesg[d]
[snarfed] or anyone else have you encountered this issue?
#
capjamesg[d]
I tried adding my library into the docs/requirements.txt file but that didn’t help.
#
capjamesg[d]
I also tried the path change that I saw in the granary conf.py file but that didn’t fix the issue (but it did help).
#
capjamesg[d]
I get a warning on build that indieweb_utils cannot be found.
#
[snarfed]
yeah it can take work to get right. are you testing locally? try getting it working locally first
#
capjamesg[d]
How do you build the docs locally?
#
[snarfed]
basically just sphinx-apidoc and then sphinx-build
#
capjamesg[d]
That is so helpful!
#
capjamesg[d]
[snarfed]++
#
Loqi
[snarfed] has 35 karma in this channel over the last year (66 in all channels)
#
capjamesg[d]
As always, I appreciate your help!
#
capjamesg[d]
I’ll try to get it working locally. That will make debugging much easier.
#
[snarfed]
welcome! good luck!
EvanBoehs[m], jamietanna[m], benatkin, micahrl[m], LaBcasse[m], hala-bala[m], Lohn, unrelentingtech, npd[m], Abhas[m], HarryCoburn[m], vikanezrimaya, PeterMolnar[m], samwilson, doubleloop[m], akevinhuang, diegov, rommudoh[m], reed, mlncn and mambang[m] joined the channel
#
mlncn
sheepishly waves, long time no see
rommudoh[m], reed, Abhas[m], benatkin, samwilson, doubleloop[m], hala-bala[m], vikanezrimaya, Lohn and diegov joined the channel
mambang[m], jamietanna[m], unrelentingtech, nekr0z, LaBcasse[m], EvanBoehs[m], HarryCoburn[m], micahrl[m], npd[m], PeterMolnar[m] and tetov-irc joined the channel
#
mlncn
POSSE to Facebook, for Pages and Groups only: At Agaric we've been supporting https://portside.org in posting to its Facebook page and in upgrading to Drupal 9 and to 3.x versions of Social API / Social Post Facebook modules, we want to try to make this as easy as possible for as many people as possible
rommudoh[m], reed, Abhas[m], benatkin, doubleloop[m], hala-bala[m], samwilson, vikanezrimaya, Lohn, diegov, mambang[m] and unrelentingtech joined the channel
#
mlncn
Known is no longer going to Facebook, right? Hasn't been set up to be for pages/groups?
#
mlncn
Anyone happen to know how WordPress JetPack does it? Haven't deciphered yet, https://github.com/Automattic/jetpack/tree/master/projects/plugins/jetpack/modules/publicize
#
mlncn
we're forging ahead building on https://www.drupal.org/project/social_api but trying to see how the wider world does it
#
mlncn
Also! Anyone POSSEing to *specific* Mastodon instances? Approach in Drupal of course has been to make your website federate directly, which is awesome, but we've seen a need to POSSE post from website to pre-existing account on a Mastodon instance.