debug whole file

[pfitzseb]

Outstanding issues:

• REPL mode switching is broken.
• We should get rid of unnecessary return statements in top-level code.

Related issues:

• fixes [FR] Add @debug alias for @enter Juno.jl#451
• fixes Debugging without running from REPL Juno.jl#425
• fixes [FR] "Start debugging" command  Juno.jl#315

Also finally fixes JunoLab/Juno.jl#118.

[codecov]

Codecov Report

Merging #269 into master will increase coverage by 3.12%.
The diff coverage is n/a.

@@            Coverage Diff             @@
##           master     #269      +/-   ##
==========================================
+ Coverage   50.83%   53.95%   +3.12%     
==========================================
  Files          41       42       +1     
  Lines        2093     2311     +218     
==========================================
+ Hits         1064     1247     +183     
- Misses       1029     1064      +35     

Impacted Files	Coverage Δ
src/completions.jl	85.48% <0.00%> (ø)	⬆️
src/datatip.jl	78.12% <0.00%> (ø)	⬆️
src/static/toplevel.jl	97.14% <0.00%> (ø)	⬆️
src/workspace.jl	78.37% <0.00%> (ø)	⬆️
src/goto.jl	86.89% <0.00%> (ø)	⬆️
src/display/showdisplay.jl	24.75% <0.00%> (ø)	⬆️
src/utils.jl	89.77% <0.00%> (ø)	⬆️
src/display/errors.jl	38.29% <0.00%> (ø)	⬆️
src/eval.jl	69.86% <0.00%> (ø)	⬆️
src/debugger/workspace.jl	0.00% <0.00%> (ø)	⬆️
... and 7 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update ea89450...2d9cce1. Read the comment docs.

[aviatesk]

Will give a review later.

[aviatesk]

Why is there a problem in REPL mode switching ? I still can't see why only toplevel debugging has he issue ...

[pfitzseb]

Because it's not started from the REPL, and asynchronously switching REPL modes doesn't work well.

[aviatesk]

Do we still need this ? Aren't breakpoints added in startdebugging anyway for now ?

[pfitzseb]

We do, yeah. Not entirely sure why though :P

[aviatesk]

Because it's not started from the REPL, and asynchronously switching REPL modes doesn't work well.

Isn't possible to "pretend" that it's started from REPL when we call debug_file ?

[aviatesk]

@pfitzseb d4d36ac and JunoLab/atom-julia-client@a14acef works nicely as far as I've tested, great work ! Code certainly looks really hacky though.

[aviatesk]

I occasionally get the following error (when e.g. after use some debugging command):

RROR: syntax: invisible character \u200b near column 1
Stacktrace:
 [1] top-level scope at none:0
 [2] eval at .\boot.jl:331 [inlined]
 [3] eval_code(::JuliaInterpreter.Frame, ::Expr) at ~\julia\packages\JuliaInterpreter\src\utils.jl:481
 [4] eval_code(::JuliaInterpreter.Frame, ::String) at ~\julia\packages\JuliaInterpreter\src\utils.jl:462
 [5] interpret at ~\julia\packages\Atom\src\debugger\repl.jl:94 [inlined]     
 [6] interpret at ~\julia\packages\Atom\src\debugger\repl.jl:93 [inlined]
 [7] (::Atom.JunoDebugger.var"#65#66")(::REPL.LineEdit.MIState, ::Base.GenericIOBuffer{Array{UInt8,1}}, ::Bool) at ~\julia\packages\Atom\src\debugger\repl.jl:40
 [8] #invokelatest#1 at .\essentials.jl:710 [inlined]
 [9] invokelatest at .\essentials.jl:709 [inlined]
 [10] run_interface(::REPL.Terminals.TextTerminal, ::REPL.LineEdit.ModalInterface, ::REPL.LineEdit.MIState) at ~\julia\julia\usr\share\julia\stdlib\v1.5\REPL\src\LineEdit.jl:2354
 [11] run_interface(::REPL.Terminals.TextTerminal, ::REPL.LineEdit.ModalInterface) at ~\julia\julia\usr\share\julia\stdlib\v1.5\REPL\src\LineEdit.jl:2348
 [12] debugprompt() at ~\julia\packages\Atom\src\debugger\repl.jl:56
 [13] (::Atom.JunoDebugger.var"#53#55")() at .\task.jl:358

Do you think you can fix this within the current approach ?

[pfitzseb]

That happens when hideprompt is called while the debugger (or another repl mode that's not closed with backspace) is active. I thought I'd stopped that from happening with this line, but maybe we should whitelistthe base REPL modes instead.

[aviatesk]

Ah I see. Not sure why isdebugging filtering doesn't work for me.

[aviatesk]

Sorry @pfitzseb, I cluttered with branches and just found I've not included 0614e71 -- will try this again and if the error doesn't happen I guess we can move to release 👍

[aviatesk]

Okay, so I found I only see the error when I start debugging within tree-view (which I added in JunoLab/atom-julia-client#684)
Will try to look at why that causes a problem, but I think you can merge for now if you want.

[aminya]

Yay. 🚀