Stream: project-error-handling

Topic: Meeting 2021-08-02


view this post on Zulip Jane Lusby (Aug 02 2021 at 17:51):

Meeting in a little over an hour

view this post on Zulip Jane Lusby (Aug 02 2021 at 17:59):

Agenda: https://hackmd.io/jZMyx6YhQ1upUDB4w8iR5g

view this post on Zulip Jakub Duchniewicz (Aug 02 2021 at 18:02):

I will update myself now as I am on the road in several minutes:
Working on the RFC for Backtrace in core. Have to look through the review and iterate over it again

view this post on Zulip Jane Lusby (Aug 02 2021 at 18:03):

:+1:

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:01):

Meeting time!

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:01):

Sean you got the notes today?

view this post on Zulip Sean Chen (he/him) (Aug 02 2021 at 19:01):

Sure

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:01):

alright

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:02):

Okay let's start with action items

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:02):

and we can do this by person

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:03):

so Sean, you're working on the trial-and-error crate and std::error::Report RFC

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:04):

did you want to give a status update on those?

view this post on Zulip Sean Chen (he/him) (Aug 02 2021 at 19:05):

Sure, I'm working right now on documenting the two modules in the crate, which consist of the Report type and DynError type.

view this post on Zulip Sean Chen (he/him) (Aug 02 2021 at 19:05):

Once those docs get accepted by Jane, I'll be moving on to adding examples and tests for them.

view this post on Zulip Sean Chen (he/him) (Aug 02 2021 at 19:05):

After that, using the crate as a proof of concept for the Report RFC.

view this post on Zulip Sean Chen (he/him) (Aug 02 2021 at 19:06):

That's about it.

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:06):

:+1:

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:06):

ill review the newest changes after this meeting

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:06):

and @Jakub Duchniewicz 's RFC, though I'm not sure if I should review it now or wait until hes had a chance to respond to @Nick Cameron's comments

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:09):

@Charles Ellis O'Riley Jr. do you want to give your status update next?

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:09):

For your action items I currently have:

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:09):

view this post on Zulip Charles Ellis O'Riley Jr. (Aug 02 2021 at 19:10):

ok. I did a PR on thde GH action

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:10):

that's just waiting on my review as well right?

view this post on Zulip Charles Ellis O'Riley Jr. (Aug 02 2021 at 19:11):

Right. I realized after I did it that I should have b ranced it. Right now it's from main.

view this post on Zulip Charles Ellis O'Riley Jr. (Aug 02 2021 at 19:12):

If it's necessary, If I can backout, I can resubmit it from a branch.

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:13):

as long as its from a fork I don't think it matters if its on your main branch or not

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:13):

but I'll let you know if there are issues

view this post on Zulip Charles Ellis O'Riley Jr. (Aug 02 2021 at 19:13):

Cool. Going forward, I'll branch.

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:14):

For my update I'm still working on the negative trait coherence change. Haven't had a chance to touch it in the last week or two

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:15):

And this afternoon I'll be having a meeting with @Nick Cameron to go over the dyno RFC and splitting it out from the generic member access RFC so he can take over pushing that and we can move that RFC forward

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:16):

Next up, issue triage

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:17):

so last week I updated the libs team agenda generator so it can work for error handling project group agendas

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:17):

so lets start by going over the PG-Error-Handling issues

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:17):

first one is:

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:17):

"Fix the error trait"

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:17):

so this is blocked on the backtrace stabilization questions

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:18):

which is currently in progress with the work @Jakub Duchniewicz is doing on backtrace in core

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:18):

so no actions needed there

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:18):

Next "Tracking issue for error source iterators"

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:19):

I think this is more or less stalled for lack of a champion pushing it

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:19):

I'm going to add this to the list of unassigned action items incase anyone is looking for more work to pick up

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:23):

Next: "Formatting for std::backtrace::Backtrace"

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:23):

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:23):

I think this is done?

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:24):

or at least all the main changes suggested have been implemented

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:24):

i think there were some clever suggestions using the precision flag that we didn't do

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:25):

but I expect we will want to close this issue and maybe open the leftover changes as a separate issue

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:27):

the stabilize the backtrace feature PR is associated with the fix the error trait tracking issue so not gonna go over that one

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:27):

and https://github.com/rust-lang/rust/issues/77474 "add track_caller support to ?" is just looking for an implementor

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:28):

though I expect it will run into performance issues and may be difficult to get

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:30):

next up is https://github.com/rust-lang/rust/issues/32837, pluggable panic implementations

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:31):

I'm guessing that's related to #[panic_handler] but I'm not really sure

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:31):

taking a look

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:32):

hmm no

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:32):

seems to be related to the panic runtime?

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:32):

i guess even lower level

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:32):

I'm gonna assume this isn't something we need to care about atm

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:32):

moving on

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:32):

likewise, optimizations for Results is out of scope for this project group atm imo

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:33):

nicer assert messages is being worked on by @Maarten de Vries iirc

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:34):

https://github.com/rust-lang/rust/issues/54889

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:34):

implementing Try for ExitStatus

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:34):

I'm not certain when we should / would do this

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:34):

I think if we did do this now it would be insta-stable with respect to ? usage

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:34):

but that might be totally okay

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:34):

worth looking into

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:36):

next up, tracking issue for stabilizing Error::type_id

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:36):

https://github.com/rust-lang/rust/issues/60784

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:36):

I think this one is also relatively low priority so probably not worth focusing on just yet

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:36):

but a good idea for us to work on eventually

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:37):

Tracking issue for PanicInfo::message

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:37):

https://github.com/rust-lang/rust/issues/66745

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:38):

I think this one just needs someone to figure out where it left of and what the next steps are

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:38):

not sure how important it is

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:38):

probably less important than error in core + integrating error with panics

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:38):

but it seems related

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:38):

vaguely

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:38):

https://github.com/rust-lang/rust/pull/79085 is blocked on the backtrace in core work

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:39):

https://github.com/rust-lang/rust/issues/79676

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:39):

we added this already right?

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:39):

this is just chilling for now I think

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:40):

yay sean

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:41):

https://github.com/rust-lang/rust/issues/80190 is a very interesting API addition for Option

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:41):

though I think its probably too niche

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:42):

probably worth bringing up in the next libs-api meeting

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:43):

which i just did

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:43):

should get added to the agenda

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:43):

https://github.com/rust-lang/rust/issues/84908 was just recently added

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:43):

no need for us to do anything there

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:43):

and last

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:43):

https://github.com/rust-lang/rust/pull/86826

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:43):

a very cool PR by @Mara

view this post on Zulip Mara (Aug 02 2021 at 19:44):

hello

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:44):

which we should probably follow up on to see if we can get it merged

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:44):

hello

view this post on Zulip Mara (Aug 02 2021 at 19:44):

yeah it's not entirely clear what to do with it

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:44):

we're just reviewing all of the A-Error-Handling issues on the rust repo and got to your io::Error change to include the path

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:45):

Mara said:

yeah it's not entirely clear what to do with it

is there something blocking it?

view this post on Zulip Mara (Aug 02 2021 at 19:45):

it's nice for debugging. but if you can't rely on the information being there, it's not of much value. especially if it might disappear in the future or if it isn't available on another platform.

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:45):

i think you might underestimate the value of it just being in the debug info

view this post on Zulip Mara (Aug 02 2021 at 19:45):

maybe

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:45):

at least for now

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:45):

long term it will probably be less useful

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:45):

once we get it so unwrap actually reports Errors

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:45):

or at least can actually report them

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:45):

we may still default to reporting via Debug with the std error handler

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:45):

open questions

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:46):

my vote would be to try to just merge it

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:48):

Anyways, I think that's all for the meeting

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:48):

Does anyone have anything else they'd like to go over or shall we adjourn for the day?

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:49):

gonna take that as an okay to call it

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:50):

thank you everyone ^_^

view this post on Zulip Jane Lusby (Aug 02 2021 at 19:50):

:wave:

view this post on Zulip Charles Ellis O'Riley Jr. (Aug 02 2021 at 19:50):

:+1:


Last updated: Jan 29 2022 at 09:51 UTC