#dev 2021-08-04
2021-08-04 UTC
jacky, Seirdy, KartikPrabhu, Stasimon and capjamesg joined the channel
# 
[snarfed] #hugops [jacky], hope https://jacky.wtf/ is back up soon!
# 
capjamesg[d] Do full text search engines typically index whole pieces of content?
# capjamesg[d] I am only indexing titles, headings, meta descriptions, and keywords on my blog search engine.
# capjamesg[d] This came to mind last night but I haven't made a decision because I wondered how this would impact performance.
# capjamesg[d] (with sqlite, that is)
neceve joined the channel
hendursa1 and petermolnar joined the channel
# 
vikanezrimaya I knew it! I knew streams in Rust had a try_collect() method, they must've had one!
# vikanezrimaya Before that I was trying to panic inside a stream and then catch unwinding the stack outside to see if an error occured
# vikanezrimaya I couldn't even know which error it was
# vikanezrimaya And I'm pretty sure that a database connection used inside the stream might not've been properly returned into the pool, causing pool exhaustion and subsequent database connection errors!
# vikanezrimaya The code even type-checks, so it must compile. I have to try it in production to see if it's effective
# vikanezrimaya And people call me smart. No, I do things in ways people are not supposed to do things, which is clearly indicated in the documentation for the methods I use to do things
# vikanezrimaya catch_unwind() literally says "do not use this to catch recoverable application errors"
# vikanezrimaya And I still used it and successfully pierced my foot with it, resulting in my application needing restarts roughly every other day
# vikanezrimaya And people call me smart
KartikPrabhu joined the channel
# vikanezrimaya I really wanted to do some frontend work today but I fixed a backend bug instead. Accidentally.
# vikanezrimaya I just wanted to see if post["properties"]["author"] gets autoexpanded for presentation
# vikanezrimaya (it does not, but I could probably fix it)
chee joined the channel
# vikanezrimaya https://fireburn.ru/posts/kittybox-db-connection-leak - here's a somewhat technical explanation of what I've done wrong
# vikanezrimaya if someone's interested
shoesNsocks and rockorager joined the channel
# hala-bala[m] Is there good tutorial how to obtain token with help of indieweb ?
# 
rockorager I think I (finally) have the two main things I wanted to get done for a "minimal" config indieweb starter site: self-hosted webmentions and micropub. If anyone wants to "alpha" test it, here's the site: https://11ndieweb.netlify.app/
# hala-bala[m] I am trying to create extention and I would like to authenticate to be able to use my micropub instance
# rockorager The main thing is you'll need a Github Personal Access Token once you click the Deploy to Netlify button
# hala-bala[m] that will do, thank you aaronpk
# rockorager It has self-hosted webmentions with configurable "accepted domains", so once I get around to building the API, it could also serve as a single-click self hosted (well, Netlify hosted) webmention receiver for any domain
# hala-bala[m] rockorager: I will check it out for sure, I am running 11ty myself so I might take some parts.
# hala-bala[m] Has anybody developed private blog channels and has written about it ? Kinda like, post you want to share only with family/friends etc.
# rockorager I wrote the endpoints as an 11ty plugin (eleventy-plugin-11ndieweb), which basically just gives you the lambda function for the webmention and micropub endpoints
# rockorager The catch is that there is basically no config for a different folder structure (yet?) for the way micropub saves your files
# hala-bala[m] rockorager: I can modify that if anything. I have been pretty happy with https://github.com/getindiekit/indiekit for that.
# rockorager Indiekit is definitely more polished than this! :)
hendursaga, rockorager, hendursa1, [grantcodes], [jeremycherfas] and shoesNsocks1 joined the channel
# hala-bala[m] aaronpk: how do you deal with cors ? how do you fetch links without hitting cors issues ?
# hala-bala[m] when doing link discovery on website
# hala-bala[m] do I really need to create server to get html to get links to be able to use them ?
# hala-bala[m] is there microservice which takes website url and returns array of micropub, indieauth endpoint etc ?
# hala-bala[m] Basically I am trying to authenticate user in webextention (basically js website) to be able to post content as him/her. Based on the blog you shared, I am supposed to parse html to get links for indieauth and micropub, which I feel can be done on server but not in client :/. So I might have to make server to do that part for me.
# hala-bala[m] and while I can allow no cors on my own website, if I want to publish the extension, I don't expect people to fiddle cors options.
# 
aaronpk check this out https://grantcodes.github.io/micropub/#/
# hala-bala[m] yea, same issue ```Client side useage
# hala-bala[m] Although this library is intended to be usable client side you will likely run into CORS issues so be careful about that.```
# hala-bala[m] but I think I can take auth endpoint and micropub as settings value and jump over this proccess.
hendursaga joined the channel
# 
[aciccarello] Nice work on 11ndieweb rockorager++
jacky joined the channel
KartikPrabhu, neceve and capjamesg joined the channel
petermolnar, jacky, KartikPrabhu, chee and rockorager joined the channel
BinarySavior joined the channel