๐ It's 11 PM. I'm trying to explain my sister in law why I'm so excited:
I managed to run a little R shiny app INSIDE a quarto report rendered _statically_ using Github:
https://t.co/okFHrCwEC9
๐
Now she believes I'm a weirdo ๐
@JosiahParry Fwiw my use cases for promises have been allowing users to execute a long running SQL query without clogging up the session for other users. I've been eyeing the new ExtendedTasks features in Shiny to see if that can supplement what I need to do and not use promises
@JosiahParry I use promises in my role.... And I still get confused whenever I use it. The docs are pretty rough tbh, idk if this is helpful but This was the last article I read that I remember having a simple example that was helpful
https://t.co/QUUHXTQQBZ
@JosiahParry Essentially what my boss did a few years back where I work now. Now we have many teams using R, have internal R teaching sessions, and build really cool things really quickly. He solved problems, now we have 40+ R users from beginning to advanced.
@tiddlydump @JosiahParry @tjmahr@RyanPlain fwiw though, i was using duckdb + dbplyr :) and slacking my colleague every little nugget of duckdb info i found on this journey haha i was loving it more and more regardless the rstudio crashes after hitting 25gbs of ram usage along the way ๐
@tiddlydump @JosiahParry @tjmahr@RyanPlain and here is where i read more about getting it to spill to disk - https://t.co/c9ZB3yiPOd
but I felt like i had to do a lot of docs reading (which is fine!) to get to that point and get my process to work.
@tjmahr @JosiahParry @RyanPlain@tiddlydump When I updated the connection to the screenshot it ran fine and didn't crash which I was learning the hard way lol
@tjmahr @JosiahParry @RyanPlain@tiddlydump Something I wish was more clear in docs - the default dbcConnect(duckdb()) will still utilize RAM when running expensive calls... I found out this week using this connection to force it to spill onto disk and cap the memory. I needed a distinct over 500million rows
@beeonaposy Warwick NY is nice - in Orange County, has a nice town center (coffee shops, restaurants, some boutique shops) schools aren't bad, but close enough to the city that there are daily commuters. Chester ny is nice as well but not as walkable
@CFlensburg@rstatstweet This posit/rstudio article might be of some help. I've done a similar setup for when I share an RProject / common parent folder. (though we generally turn the rdata option off to limit that) https://t.co/qqAWWtNpAI
@zevross@rstats_tweets I've written something like that for myself.... Might not be the most elegant of code but hey it works lol I have it mapped to my Ctrl + alt +v shortcut. https://t.co/qlvT5rnHvI
@sharoz I see this all the time now that I've been using Windows more, but (in general) RStudio doesn't become unresponsive for me when I start to move my mouse. Never saw it on my MacBook. I thought I was the only one seeing this because it'd disappear before I got screenshots
@dgkeyes update original to have it say "done" or something like that? Depending on how long it takes report to render, could set CRON schedule to check every 10 min, 30, etc...
@dgkeyes Might be clunky - but use a CRON schedule to run 1/min and check a googlesheet. Have them sub a google form with the name to that sheet. When script runs, if it sees a new entry, have it render/source the other script that emails them w/ report? then update the original sheet
@tanya_shapiro as interviewee: If I were to start today, what are the short and medium (~6ish month) term goals I'd be working on. Gives me a sense of the type of work i'd be engaging in and their priorities of that moment.