#jackythat should fix the issues with bridgy backfeed
#jackythen I'm going to look into fixing my test suite and refactoring out the authentication logic
#jackywanna get to a point where I don't have to define it in koype but I can rely on something else to do it
#jacky(pre-cursor to AutoAuth support for me as well as the separation of concerns for my site)
#[Michael_Beckwiti need to do some review of the spec and whatnot for atom, cause i think i went about editing the xml tags when i shouldn't have, when trying to implement a custom feed for some data.
#[Michael_Beckwitprobably a case of i should have just been amending what went in the tags themselves more than anything
#[jgmac1106]!tell snarfed any guides to using Bridgy but including media or audio? Currently I POSSE another note pointing to posts or notes with audio and video
swentel, gxt and mblaney joined the channel
#Zegnat[jgmac1106]: what are you trying to bridge?
mblaney and superjen96 joined the channel
#jamietanna[m]jgmac1106 Bridgy to Twitter works with photos (not yet tried with videos) if that helps?
#jamietanna[m]Just looking into allowing my Micropub server to tell me if I've already created a piece of content - I don't see anything in https://www.w3.org/TR/micropub/ about what to return in the case it already exists, so I guess it would be a 201 with a `Location` header to the existing post, rather than a https://http.cat/409 ?
#Zegnatjamietanna[m]: what is the definition of already existing?
#ZegnatE.g. submitting the exact same payload twice? Defining an mp-slug that already exists without being an update?
#jamietanna[m]Good question. Submitting the same payload twice - I'm planning to stop me from re-publishing i.e. a `like-of` to the same post
#ZegnatI think for the exact same payload, I would also just return the Location header of the previously created post
#jamietanna[m]Perfect - that's what I was thinking per the spec, but from my experience with RESTful APIs I'd expect a `Conflict` to say "you've already done that". All good though, as this makes it easier :D
#ZegnatThough I wonder if some sort of cool-down period would be needed. Submitting a like within a day seems like an accidantal repeat, but what if I like an article again after a year? Should that just become last year’s post, or should it create a new post so the thing resurfaces for people who follow me.
#ZegnatI am unsure if you want the spec to define this behaviour, as there isn’t anything wrong with wanting to create duplicates imo
#jamietanna[m]That's fair, hadn't thought of that usecase
#ZegnatOBviously you only need to support the use-case that you are likely to run into yourself on your personal homempage :) I just like to overthink
[jgmac1106] joined the channel
#[jgmac1106]audio clips and video clips, I am sure I am hitting the file size limits, maybe I just need to link to the videos rather than embed in notes
#[jgmac1106]but they look purty inside of my notes
#[jgmac1106]So I publish a note with a video and then I publish another note saying "hey I just published a video and I link to the first note. I use bridgy to POSSE the second note
#jamietanna[m]Through the stuff I'm doing I've actually found a fair few duplicates I didn't know about - woops!
#ZegnatInteresting. I would expect that if Bridgy couldn’t POSSE the actual video or audio, it would leave something else in its place on Twitter
#Zegnat[jgmac1106]: have you filed an issue on what you would expect Bridgy to do with your posts?
#ZegnatSo I guess what you are expecting is that, with POSSE to Twitter, the tweet includes either the embedded video (if possible) or some indication that there is video on the original post.
jenelizabeth joined the channel
#ZegnatBTW, I think the u-video class is in the wrong place, needs to be on the <source> element for the video URL to be picked up
#[jgmac1106]thx, I set it up to add captions just out of time now, will do when I get home, I have been putting u-video in wrong place in every video post then
#Zegnat`u-` instructs the mf2 parser to get a URL from the element you put it on. If you are using source elements for the video URLs, that is where you need to put it.
#ZegnatI do not know of any parsers that treat nested elements like source-within-video special
#ZegnatMaybe it turns out Bridgy does the right thing if you move the u-video :D I do not actually know
[snarfed] joined the channel; mblaney left the channel
#Loqi[snarfed]: [jgmac1106] left you a message 3 hours, 41 minutes ago: any guides to using Bridgy but including media or audio? Currently I POSSE another note pointing to posts or notes with audio and video
#[snarfed]audio no, since afaik none of its silos support it
[tantek] and [LewisCowles] joined the channel
#[LewisCowles]adding S3 idno plugin to see if it can work on heroku (5 years old)
#[LewisCowles]half tempted to see if it works with regular S3 before all the heroku nonsense providers
#[LewisCowles]other-half tempted to make/find sftp file & filesystem driver for it
#[LewisCowles]so far had to patch to get around load balancer, had to upload the config which prohibits public repo storage of code. It's looking like it was designed for shared hosting
#jeremycherfas!tell jamietanna[m],Zegnat I like the way Pinboard unobtrusively tells me I have saved something before and asks whether I want to edit or replace.
#[snarfed]should work in a python shell in the bridgy virtualenv after you've imported meetup
gxt, JKing and gRegorLove joined the channel
#jamietanna[m]I'm now getting `google.cloud.ndb.exceptions.ContextError: No current context. NDB calls must be made in context established by google.cloud.ndb.Client.context.` ?
#[snarfed]ah yeah, sorry, this is annoying. so you actually need something like:
#[snarfed]ah that's just from a cookie. ok then, features isn't getting set to publish during signup, but your index.html line seems right. time to debug!
#jamietanna[m]Thanks - let me know if there's anything I can help with
#[snarfed]hah sorry i didn't mean i was starting to debug
#GWGSo, what other custom integrations that don't require an API key can I integrate into Parse This?
leg and nickodd joined the channel
#jamietanna[m]Oh gotcha snarfed! I may need a hand if you have some time, as you'll know more what you're looking for
#[snarfed]jamietanna ok! btw it's probably time for me to grab the meetup API creds so i can test. evidently your emails don't get to me, maybe try twitter DM?
#jamietanna[m]That's odd I wonder what's happened. But yeah I'll try that now 👍🏽
#[snarfed]oh i guess i need to make a meetup account now 😆
#[snarfed]debugging now. btw jamietanna a couple minor nits: you'll need to add a meetup_icon.png file to static. also mind changing the user-visible app name in meetup's API to Bridgy, without a period? (very nitpicky, apologies!)
#[snarfed]oh also i think i see the bug, commenting in github now
#jamietanna[m]not to worry! Re meetup_icon.png is that to Bridgy itself, or granary/oauth-dropins?
#jamietanna[m]And re the username, do you mean change it to `Meetup Jamie Tanna` or `Meetupcom Jamie Tanna`? Or?
#[snarfed]in bridgy. you'll see it after you get the user showing up. and it's the app name, right now it's Brid.gy with a dot
#[jgmac1106]good to see IndieWeb building blocks spreading, watch out for Lord Businessman and his Kragle trying to keep you trapped in the social media silos
#[tantek]or if I'm feeling particularly dry, 1 is too many 🙂
#jamietanna[m]!tell snarfed I've raised a PR for oauth-dropins, so once that and granary are merged we're almost there! I've got it RSVPing via `curl`, but not yet via the UI - if you're able to see if you can spot why that'd be appreciated!
#GWG[tantek]: On a standard page, I have an h-feed and the alternate rss/json feeds for old school. This isn't dry..these are links to the tag archive, the author archive, etc
sivy joined the channel
#[tantek]nah, there should either be zero (encourage old feed readers to parse h-feed), or 1 valid Atom 1.0 feed (since it's an IETF RFC that can be updated by an open process) that represents literally the same set of content that is on the page
#[tantek]any other feeds for any other sets of content belong as links on the HTML versions of those pages
#[tantek]e.g. for a tag archive, have a tag archive HTML page, and then have that HTML page link to its feed for a tag archive
#[tantek]e.g. for an author archive, have an actual author archive HTML page, and then have that HTML page link to its feed
#aaronpkit's no fun clicking on an rss link just to get a pile of XML in your face
#[tantek]I've yet to see a good example / reason to have more than one feed link. Unless you're experimenting with a new format like JSONFeed, I'd allow that 🙂
#[tantek]GWG you are always asking about design advice. You just got some ^^^
#[tantek]If you ever want to or get requests for different feeds of stuff, first build the easily human viewable h-feed of that different "feed of stuff" and link to it visibly
#GWG[tantek]: Right now, I don't mark up links to my archives with the indication is is a feed.
#[tantek]even if that's just a list of hyperlinks in a navigation box
#[jgmac1106]I have many a reason to have multiple feeds on my homepage, when people want to add feeds into their reader they drop the homepage in and then want to selectthe feeds to subrcribe to
#[tantek]and then on each of those *HTML* pages, you can provide a non-HTML feed file link if you want
#Loqi[Greg McVerry] Step Two of @IndieWebCamp Online complete: I layed out my grid on paper and time to write the rules
#[jgmac1106]people want to subscribe to the audio version, the teaching tips, and then books I am reading, So I will have multiple h-feeds on the page and then use granary to convert those h-feeds into rss
#aaronpkthe point about showing html first still applies, and is probably even more important in those casers
#[jgmac1106]funny I wrote about HTML first in that project update....but I am starting with CSS first...oops
#[jgmac1106]but on my homepage I include feeds that aren't even on my website but on different websites if I think itis a feed interesting to my audience
#sknebelI think GWG is asking about the links to those HTML versions
#aaronpkI suppose it's just my non-note/photo things that are not syndicated, like events or trips or food posts
[Christina_Hendr joined the channel
#GWGI just want, when I do feed discovery, to present different feeds.
#GWGSince most people when subscribing to a site in Microsub would use the home page URL, I'd like that to be able to show my most popular feeds and have a secondary page that tells you how to find the rest.
[jeremycherfas] joined the channel
#[jgmac1106]What is the difference between grid-column and grid-template-columns?
#LoqiIt looks like we don't have a page for "difference between grid-column and grid-template-columns" yet. Would you like to create it? (Or just say "difference between grid-column and grid-template-columns is ____", a sentence describing the term)
#aaronpkwhy am I having such a hard time dealing with storing photo responses for meetable
#KartikPrabhu[jgmac1106]: grid-column assigns a column to the child of the grid element. grid-template-column declares the columns on the grid on the parent element containing the grid
#aaronpki'm trying to store pre-resized versions of photos on disk to avoid doing it on the fly
#aaronpkphotos can be added either via webmention or uploaded directly
#aaronpkphotos can be stored either on disk or with one of Laravel's storage providers like S3
#aaronpkwebmention responses can contain multiple photos
#aaronpkphotos are displayed in a grid, and can be reordered arbitrarily, including rearranging photos across posts so they are not necessarily grouped by the post they came from
#aaronpkthese are a lot of constraints to keep in mind as I am doing this
#aaronpkI think my current plan is that when a photo is added, I will also store resized versions by prefixing the filename with the size (so a file like public/responses/10/abcdefg.jpg gets stored also as public/responses/10/1600x-abcdefg.jpg public/responses/10/230x230-abcdefg.jpg)
#aaronpkand i'm going to only store the original name in the database, and when I display the photo I will add the size prefix on the fly assuming the file is already there
#aaronpkit feels a little fragile, but I think any other solution is going to be a lot more involved
#aaronpkthe "right" way I suppose would be to make a new one-to-many table for the photos that belong to a response, and then I could store each size's URL in that table and do the overall ordering there
#jamietanna[m]!tell snarfed looks like, as we can only use one `redirect_uri` I've updated it to http://localhost:8080/publish/meetup/finish - but need a quick hand getting it to go through the publish flow from the home page, as it currently goes through the listen flow
[snarfed] joined the channel
#[snarfed]ugh, only one redirect url, so antiquated
#Loqi[snarfed]: jamietanna[m] left you a message 1 hour, 46 minutes ago: I've raised a PR for oauth-dropins, so once that and granary are merged we're almost there! I've got it RSVPing via `curl`, but not yet via the UI - if you're able to see if you can spot why that'd be appreciated!
#Loqi[snarfed]: jamietanna[m] left you a message 1 minute ago: looks like, as we can only use one `redirect_uri` I've updated it to http://localhost:8080/publish/meetup/finish - but need a quick hand getting it to go through the publish flow from the home page, as it currently goes through the listen flow
#jamietanna[m]sorry, by listen flow I mean the backfeed / listen functionality. Because we've only implemented publish for Meetup, I'd prefer to only have the option to set up publish so the UX isn't confusing
#[snarfed]oh and re the icon file, it will be rendered at 24x24. so most of the silos are 48x48, that's ideal, but can be any size as long as it's square and looks good small
#jamietanna[m]!tell snarfed I think we're pretty much done now 🙌 I've got it working from no-authZ to publishing, the main thing left is the doc updates and the ability to disable it - unless there's anything else you think that needs to be done
#[snarfed]1jamietanna awesome! sounds like you can demo today then, if you're hoping to
#jamietanna[m]Snarfed ah for IWC Online? Unfortunately it's late here in the UK so it's bed time for me but that's a good idea - I should've recorded it for demos
nickodd left the channel
#[jgmac1106]1jaimetanna I will make a place online for demos and you can add one later, if people make short videos I will splice them all together
#[jgmac1106]1haven't figured out the online demo yet, might have to be asynchronous