#[eli_oat]Does anyone know of a quick and easy way to see an example of the JSON output form a micropub client? I’m trying to work off of the w3.org docs, but I’m not having much luck mapping the incoming data
#[eli_oat]word, sorry — I wasn’t clear. This is in different context, using micropub.rock I’m able to get the first scenario to work sans HTML, but the second one, which I think is the correct format, just displays “ARRAY”
#[eli_oat]when I try to test it against receiving HTML
#[eli_oat]I am able to successfully post plain JSON h-entries, and HTML JSON h-entries to my site using micropub.rocks and I am able to post to my site using Quill’s Editor interface. I am not, however, able to post to my site using Quill’s note interface. I’ve tried enabling and disabling Quill’s “Send HTML Content” toggle.
#[eli_oat]Because it is so young, or is there something architectural that seems fragile? What I'm doing is storing posts as json, constructing a combine view of those posts as a json feed, and then keying UI off of the feed
#Zegnat[eli_oat], are you storing any extra meta data, like post kind et al? How are you doing that in JSON Feed?
#[eli_oat]It supports arbitrary keys as long as they have a leading underscore.
#[eli_oat]And the json of each post isn't necessarily json feed compliant, since it can get reformatted when it is added to the feed view, so I can add extra info no probleeemo
#myfreewebi store json of each post too, but mf2 formatted (each property is an array). i actually store feeds as mf2 objects as well so i can reconfigure them on the fly from micro-panel
#ZegnatAh, alright [eli_oat]. So it is more JSON storage then JSON Feed storage ;) I think that was the only thing KartikPrabhu would have found fragile
#[eli_oat]That makes sense. Storing as mf2 formatted json is also a really great idea. Hmmm ?
#[eli_oat]Right now I store all my data as plaintext (because I'm stupid obsessed with flat files cmss), but am backing everything up to couchdb. I could reformat on the way to couch to mf2
#sknebelmyfreeweb I gotta try wire up micro-panel to my site, it seems surprisingly similar to something I wanted to build ;)
#sknebel[eli_oat]: I think you'll find flat-files are very popular around here
#ZegnatThat update to /jsonfeed I just did needs more work. There is now a _indieweb property extension to JSON Feed committed to Known and definitely worth documenting.
#myfreewebsome people prefer jf2 to avoid dealing with the arrays everywhere. but i designed micro-panel around arrays :D
#ZegnatIf JSON Feed does gain traction we might as well make sure to document IndieWeb-data-in-JSON-Feed from the start
#myfreewebre "obsessed with flat files cmss"… sweetroll started out as json files in git. switching to postgres felt so good
#myfreeweblike i could have built some indexing on top of flat files, maybe use a db in addition to files, but i never did any of that
#[eli_oat]I spend most of my day managing a couple postgres dbs, so am trying to avoid mostly because of that ?
#[eli_oat]"most of my day" is an exaggeration, but a chunk of it, at least.
#myfreewebhttps://github.com/myfreeweb/mf2sql is my postgres schema. more functions than schema actually :D it can extract nested mf2 objects into separate records when writing, and inline them back when reading. which leads to hilarious bugs sometimes: once i had a micropub client id that was equal to my site's url, it inlined the home feed into the client id property :D
#Loqi[myfreeweb] mf2sql: PostgreSQL schema for storing microformats2 JSON objects
#myfreeweb[eli_oat]: hmm my site didn't like your h-entries https://unrelenting.technology/replies/2017-05-28-11-32-03 no author, no url even. i guess it's probably a good idea for me to add the fetched URL if the url property is missing… but i think everyone publishes permalinks as u-url
KevinMarks, [kevinmarks] and [eli_oat] joined the channel
#[eli_oat]@myfreeweb I fixed the missing URL issue. As for the author — should that just be on the main page of my domain, or should I add that to each post within the h-entry?
#ben_thatmustbemeThis all shows more of why I want to unify jf2 and jsonfeed. They are quite similar. But to be able to easily and programmatically change mf2 Json to jsonfeed/jf2 I think would make offloading the conversion super simple and it would gain even more traction
#myfreeweb[eli_oat]: since you don't have your h-card as the h-entry's author, adding rel=author to your home link (a href="https://eli.li" title="back") will instruct authorship parsers to go there to find the h-card
#ben_thatmustbemeIn other words I want to just write it once or just use an external service and as long as my mf2 is correct, I know the Json feed /jf2 is correct
KevinMarks, jeremycherfas and [cleverdevil] joined the channel
#[cleverdevil]@Zegnat I'm the one who added JSON Feed to Known, and added those _indieweb extensions.
#[cleverdevil]I'd love to see someone hash out the details and all agree on something that we can use.
#[cleverdevil]If a bunch of IndieWeb sites adopt JSON Feed with the extensions, perhaps we can get the extensions "standardized" and baked into the spec itself.
#ZegnatI think I credited you correctly on the wiki for that, [cleverdevil]
#[cleverdevil]Ah, cool, not so worried about credit, more interested in getting feedback on what I did with the extensions.
#ZegnatWe’ll see what happens with it. I just started a section for documenting it and was then called away. I might be able to document more of it later
#[eddie]Obviously the data comes from compass, but do you store the gps file in your post directory too? Or do you just Store the start/end time, and when the static map image is being created, do you just grab the stuff from compass at that moment and only leave the gps in compass?
KevinMarks joined the channel
#aaronpkCompass actually posts a file with geojson data of the trip, and my site generates the map from that
#aaronpkSo I have the raw data from each trip along with the map image stored with each post
KevinMarks and KevinMarks_ joined the channel
#myfreewebhttps://unrelenting.technology/notes/2017-05-28-23-33-32 ← in case anyone else thought that AWS Lambda is a great fit for an extended micropub media-endpoint thing that does fancy processing (e.g. WebP images in addition to PNG/JPEG) D:
#Loqi[Greg] So Amazon Lambda has a 6 MB limit on request (and response) size. Binary files have to be Base64 encoded (LOL) which makes the limit even SMALLER! So my micropub media endpoint chokes on full DSLR resolution photos. Yeah the "right way" is to have th...