Profile avatar
lcas.dev
Software person. Building @deno.land. Creator of Fresh. tc39.es delegate. he/him šŸ³ļøā€šŸŒˆšŸŒšŸŒ»šŸ’š lcas.dev
182 posts 3,586 followers 169 following
Regular Contributor
Active Commenter
comment in response to post
Hey, that's not nice - what are we faking?
comment in response to post
This - ƖBB knows about some Spanish trains that Renfe fails to display in their own planner
comment in response to post
I don’t know who wrote this, but I know exactly who wrote this
comment in response to post
How'd you imagine those to be synced back down to your local folder?
comment in response to post
or did the trains turn the humans gay and we just never knew
comment in response to post
Great!
comment in response to post
Also San Sebastian->Madrid was not bookable through Interrail for them, they had to fall back to getting a full price ticket Talked to multiple people at various ticket offices, they all have the same "bug in the system" excuse
comment in response to post
Renfe has not been selling Interrail reservations for some lines "due to a bug in the booking system" for 2 weeks now. Unable to book online from anyone, or in person at the ticket office A friend of mine is unable to book Ourense->Madrid on Thu for example, even though seats are still available
comment in response to post
No it’s not dead. But we’re going to look at some other ways to do state before committing that this is the right approachā„¢ļø and slapping a GA label on it
comment in response to post
WinterTC is officially called TC55 already because Ecma TCs are numbered. But it is colloquially referred to as WinterTC because that's more fun
comment in response to post
www.digitalocean.com/community/tu...
comment in response to post
And for everyone always complaining about Deutsche Bahn: todays trip was ~10 hours total, and we arrived with a grand delay of 2 minutes
comment in response to post
Hahahhahaha
comment in response to post
chatgpt impression of three balls of anxity in a trenchcoat. no resemblance IMO :D
comment in response to post
There is literally no assignment happening in my code sample.
comment in response to post
`print(type(eval("500")))` returns `<class 'int'>`
comment in response to post
I'm sorry, but what you are saying is just wrong. The thing on the left hand side of the `is` evalutes not to a string, but a number. `eval` evaluates the string literal "500" into a number 500.
comment in response to post
`a=500; a is 500;` does not always return `False`. If both of the numbers are in the same compilation unit (ie same file, same eval, etc), CPython may dedupe them into a single heap object.
comment in response to post
No, they are both numbers. I am evaluating the string "500" (ie a running Python code file containing just the characters `500`). The reason that `is` doesn't compare to `True` is because `is` does pointer equality, not value equality.
comment in response to post
Yes, that’s the reasoning. Numbers are heap objects, heap objects have identity, and `is` does identity comparison (not value comparison). CPython does interning for small numbers, which is why you can’t ever observe this for the number 0 for instance.
comment in response to post
We did quite the maneuver honestly. The train driver had to get out of the train 1/5 of the way between Frankfurt Flughafen and Mannheim and walk 400m to the other side of the train to reverse back the other direction.
comment in response to post
Because of a burning train on the Riedbahn or something like that? Very difficult to find information
comment in response to post
I’m so sorry Adam. This is incredibly fucked up
comment in response to post
Check the ā€œParticipatingā€ section of this page: github.com/wintercg/admin
comment in response to post
Please do this through the agent option - we shouldn't be adding a bunch of random unstandardized options to the fetch options bag directly. Fetch agents and such are actually a topic of discussion at WinterTC right now, so maybe you want to join that :)
comment in response to post
Yes
comment in response to post
You have to specify that you rely on them statically (but we’re also pretty smart if you do thinks like `import(ā€œ./foo/ā€œ + dynamic)`
comment in response to post
This dependencies object contains only the dependencies that are actually used from your entrypoints. So if you have a dependency on `lodash` in your original package, but you don't ever import from lodash in any file referenced from your entrypoint, it won't be in the final package.json.
comment in response to post
2. When building the tarball, we again walk all modules from the entrypoint and create the relevant .js and .d.ts files, now using bare specifiers again so Node can resolve them. We also synthesize a package.json containing the entrypoints + a dependencies obj.
comment in response to post
Sorta ignore it. It works like this: 1. On publish, we resolve all imports in all modules reachable from your entrypoints and replace the bare specifiers in the source with `npm:` / `jsr:` imports (with the version ranges from your package.json / deno.json / import map etc)
comment in response to post
Figure out the dependencies that need to be in package.json
comment in response to post
JSR does this all automatically indeed
comment in response to post
That seems very atypical for Milan though. In my experience Italy leads Europe on network speed and cost. Nico had 2.5Gbit for 20 €/m in the middle of Turin
comment in response to post
Oh that’s great! Maybe you could add that to the ā€œGetting thereā€ section? I doesn’t mention the organized buses right now