@mflatt I just tried to reproduce the problem and again got the empty form. On the other hand, I was able to add the package with a workaround, see https://github.com/racket/racket-pkg-website/issues/76#issuecomment-856615872 . So the failures seem to depend on the entered data.
When you successfully added a package, was your data similar to the data in my screenshots, i. e. filling all of description, tags and branch (and of course using the Git suffix checkbox)? Maybe you can reproduce my problem with data more similar to the data from my screenshots.
Here are a couple old write-ups I did related to my commercial uses of Racket:
• https://defn.io/2019/08/20/racket-ecommerce/ • https://defn.io/2020/01/04/remember-internals/ I always feel a little bit embarrassed reading these old posts :sweat_smile:, but hopefully they’re helpful to you.
I’ve pushed a change to add the “same library?” check. Does that allow libcairo2.dylib to be installed as it should be?
Thanks! I will try later tonight
@popa.bogdanp would you change anything if you were to write them now? I’m curious.
@sarna.dev I feel like the point in the first article about error reporting is less relevant to me right now because I’ve learned to navigate around the limitations, but that just boils down to me knowing more now than I did then. I never released a version of Remember for Windows or Linux so I don’t know if I’m as convinced now that that approach leads to less work overall compared to just using racket/gui. I do still think you end up with apps that feel more natural in their respective environments, though.
@popa.bogdanp thanks for explaining! hunting Racket errors definitely gets easier with time, I have no idea why I can pinpoint and fix them way faster than before haha
Out of curiosity — how do you obtain the exported data from Slack in the first place? Given that (a) this slack is on the free tier, yet (b) you have records for #general stretching back to 2015, it seems like Something Clever is being done…
(I have a few other Slacks that I’d very much like to archive…)
adding a separate make-temporary-directory function would make this situation clearer, IMO. does that sound like a good idea to you too? I could send a pull request.
That also seems reasonable.
ah, thanks @popa.bogdanp. I read through them, nice work!
The archives are obtained from Slack by @samth ; importantly, we took the first archives before a free trial expired, so we had the entire data set
We only ended up on the free trial because someone accidentally linked this Slack with their corporate slack, which automatically put us on a free trial of the Plus plan.
Before I go reinventing the wheel, if I had a list of minutes and seconds, e.g., 0:15 0:25 0:20 1:10 3:40 0:02 …, is there anything already in Racket that would let me produce the sum (5:52)? In PostgreSQL there’s an interval type, and I was wondering if there’s anything like that in Racket.
@gknauth you probably want to use the gregor library by @jaz
@samth Thanks, I do use gregor in a number of my projects, I just forgot all the good stuff that’s in there. Much appreciated.
@gknauth I’d say there’s still not a super-convenient way to do that in gregor. Time-of-day arithmetic wraps around midnight, so unless you’re sure the sum isn’t more than a day, it’s not really appropriate for this. You can do it, however, using periods and period-between: #lang racket/base
(require gregor
gregor/time
gregor/period)
(define periods
(list (seconds 15)
(seconds 25)
(seconds 20)
(period [minutes 1] [seconds 10])
(period [minutes 3] [seconds 40])
(seconds 2)))
(define period-sum
(foldl +period (seconds 0) periods))
(period-between
(datetime 0)
(+period (datetime 0) period-sum)
'(minutes seconds))