Closed Bug 1431400 Opened 7 years ago Closed 5 years ago

[meta] Very hard to debug gmail with the debugger

Categories

(DevTools :: Debugger, defect, P5)

58 Branch
defect

Tracking

(Not tracked)

RESOLVED WORKSFORME

People

(Reporter: Gijs, Unassigned)

References

Details

(Keywords: meta)

Attachments

(1 file)

When looking into bug 1429831, I ran into a number of issues:


1) I can't seem to pretty-print JS in HTML files.
2) the list of event listeners that shows up when clicking the [ev] bubble thing is hard to navigate when it's long, and seems to be randomly ordered (try clicking the [ev] on <body> when logged into gmail and viewing a message)
3) performance is really sluggish. Stepping took multiple seconds. For any watched variables, It always displayed "(unavailable)" for a second or 5, and then updated with the actual value.
4) "find in files" doesn't ever return any results for me
5) using 'find' on a file, then scrolling, then clicking the gutter to set breakpoints, somehow scrolls the viewport (maybe back to where 'find' tried to go, I'm not sure).
6) state updates of the debugger seem... confused. When hitting a breakpoint, some views say "not paused" while other views have updated to show the debugger is paused, etc.
7) after closing and reopening the debugger, the debugger pane was blank
8) reloading the page takes many seconds, if not minutes, and the debugger is unusable during this time.

I'm happy to file individual bugs for these and/or other issues, but I don't know which of these are already on file or not...
Keywords: meta
Product: Firefox → DevTools

Hey :Gijs, would you mind checking and seeing if this is still relevant. Also, do you think we still need the meta keyword?

(In reply to :Gijs (he/him) from comment #0)

When looking into bug 1429831, I ran into a number of issues:

  1. I can't seem to pretty-print JS in HTML files.

This is still an issue.

  1. the list of event listeners that shows up when clicking the [ev] bubble
    thing is hard to navigate when it's long, and seems to be randomly ordered
    (try clicking the [ev] on <body> when logged into gmail and viewing a
    message)

This seems to have been fixed.

  1. performance is really sluggish. Stepping took multiple seconds. For any
    watched variables, It always displayed "(unavailable)" for a second or 5,
    and then updated with the actual value.

Performance is better.

  1. "find in files" doesn't ever return any results for me

This seems to now do something if I hit enter, but not before. Not sure if that's intentionally different from the cmd-p / "find files" stuff, but it's confusing...

  1. using 'find' on a file, then scrolling, then clicking the gutter to set
    breakpoints, somehow scrolls the viewport (maybe back to where 'find' tried
    to go, I'm not sure).

I don't see this at the moment.

  1. state updates of the debugger seem... confused. When hitting a
    breakpoint, some views say "not paused" while other views have updated to
    show the debugger is paused, etc.

It still gets confused when stepping - I ended up with a debugger that showed with a green "debugging" icon (so thought it was paused) while the rest of the UI gave no indication it was paused (no stack, no highlighted line, pause/continue/stepping buttons were as if not stopped). Also, line numbers for breakpoints / function definitions for pretty printed code seem wrong. I'll add a screenshot.

So yeah, keeping this open seems right. As for the meta keyword, you tell me. I can try getting individual bugs on file here if that's still useful. Jason?

Flags: needinfo?(jlaster)

Thanks for looking into it. I agree that it would be nice to file some more specific bugs:

  1. I can't seem to pretty-print JS in HTML files.
  2. state updates of the debugger seem confused

with 6, it would be great if you have some clear steps. This is a clear P1 if it is reproducible.

Flags: needinfo?(jlaster)
Depends on: 1494153
Depends on: 1532237
Depends on: 1532240
Depends on: 1532244

The pretty print issue was already on file, I've filed bugs for everything else I ran into.

Priority: -- → P2
Priority: P2 → --
Type: enhancement → defect
Priority: -- → P5

Is it alright if we close this bug?

(In reply to Jason Laster [:jlast] from comment #6)

Is it alright if we close this bug?

Sure.

Status: NEW → RESOLVED
Closed: 5 years ago
Resolution: --- → WORKSFORME
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Creator:
Created:
Updated:
Size: