User-themable stacktraces

[tecosaur]

Back in #49586, I claimed that it gave us the ability to start systematically solving annoyances like #40228 by making the styling used user-customisable.

This is one of the first PRs following through on that, by using a set of new faces (to be added in JuliaLang/StyledStrings.jl#4) and hence actually resolves #40228.

Along the way, we also make the file paths clickable links 🙂.

To illustrate the user-customisability, I've put a garish customisation in my faces.toml

[julia.stacktrace.location]
foreground = "bright_blue"

[julia.stacktrace.filename]
foreground = "bright_green"

[julia.stacktrace.fileline]
foreground = "bright_yellow"

[julia.stacktrace.frameindex]
foreground = "bright_magenta"

[julia.stacktrace.basemodule]
foreground = "bright_red"

and here's a resulting screenshot

[tecosaur]

It would be great if somebody could review this PR. Friendly bump 🙂

[tecosaur]

As before, if someone could spare the time to take a look at this it would be appreciated.

[vtjnash]

LGTM

Related future question: is printstyled still usable, with face = :julia_stacktrace_location instead of color=, rather than needing to explicitly name the AnnotatedString type name?

[tecosaur]

Thanks for giving this a look Jameson!

[tecosaur]

Related future question: is printstyled still usable, with face = :julia_stacktrace_location instead of color=, rather than needing to explicitly name the AnnotatedString type name?

Currently, we haven't touched printstyled as part of the work with the AnnotatedString type and StyledStrings (other than JuliaLang/StyledStrings.jl#12), so there is no face keyword.

I'd be inclined to not bother with printstyled("hey", face = :julia_stacktrace_location) since you can just use print(styled"{julia_stacktrace_location:hey}").

[tecosaur]

You may or may not want to squash merge this given the first commit is just introducing uripath, but it's only used in the context of the second content.

[tecosaur]

Since this PR is approved now, I'll go ahead and merge JuliaLang/StyledStrings.jl#32.

[tecosaur]

Hmmm, the tests seem to have a lot of Unhandled Task ERROR: EOFError: read end of file and Unhandled Task ERROR: IOError: read: connection reset by peer (ECONNRESET).

It does look like some stack trace tests also need to be updated. I'll do so shortly.

[vtjnash]

Yeah, notably looks like a StackOverflowError in some place, preventing accurately getting a stacktrace though

[vtjnash]

Removing merge-me because the error seems related:

MethodError(f=Base.stacktrace_path, args=(".\\error.jl", 35), world=0x000066c1)
2024-02-15 05:36:21 UTC	jl_method_error_bare at C:/workdir/src\gf.c:2251
2024-02-15 05:36:21 UTC	jl_method_error at C:/workdir/src\gf.c:2269
2024-02-15 05:36:21 UTC	jl_lookup_generic_ at C:/workdir/src\gf.c:3099
2024-02-15 05:36:21 UTC	ijl_apply_generic at C:/workdir/src\gf.c:3114
2024-02-15 05:36:21 UTC	#print_module_path_file#1036 at .\errorshow.jl:811
2024-02-15 05:36:21 UTC	print_module_path_file at .\errorshow.jl:805

[tecosaur]

Ah, seems like I had an ::Int64 where I should have had ::Int. Let's see if that makes CI happy.

I've also rebased to the current HEAD for fun, ignore the dramatic-looking push that GH shows above, it's just the two commits you've seen before with two lines tweaked.

[fingolfin]

Lots of CI errors -- but perhaps unrelated to this PR?!

[staticfloat]

MethodError(f=Core.kwcall, args=((modulecolor=:default, digit_align_width=3), typeof(Base.print_module_path_file)(), Base.IOContext{Base.PipeEndpoint}(io=Base.PipeEndpoint(handle=0x07baf8b0, status=3, buffer=Base.GenericIOBuffer{GenericMemory{:not_atomic, UInt8, Core.AddrSpace{Core}(0x00)}}(data=Memory{UInt8}(0, 62d389d0)[], reinit=false, readable=true, writable=true, seekable=false, append=true, size=0, maxsize=2147483647, ptr=1, offset=0, mark=-1), cond=Base.GenericCondition{Base.Threads.SpinLock}(waitq=Base.IntrusiveLinkedList{Task}(head=nothing, tail=nothing), lock=Base.Threads.SpinLock(owned=0)), readerror=nothing, sendbuf=nothing, lock=Base.ReentrantLock(locked_by=nothing, reentrancy_cnt=0x00000000, havelock=0x00, cond_wait=Base.GenericCondition{Base.Threads.SpinLock}(waitq=Base.IntrusiveLinkedList{Task}(head=nothing, tail=nothing), lock=Base.Threads.SpinLock(owned=0)), _=(0, 0)), throttle=10485760), dict=Base.ImmutableDict{Symbol, Any}(parent=Base.ImmutableDict{Symbol, Any}(parent=#<null>, key=#<null>, value=#<null>), key=:limit, value=true)), nothing, "none", 1), world=0x00006727)

This does look related.

[tecosaur]

Yea, I need to get to going through these CI errors, I'll see if I can do so this weekend.

[tecosaur]

Considering this is a straightforward new feature, I'm inclined to think at this point in the 1.11 release process we might as well give up on having it in 1.11 and just have it in 1.12 once CI is made happy.