[Bug 25587] Be more clear that progress events are about bytes being transferred

https://www.w3.org/Bugs/Public/show_bug.cgi?id=25587

--- Comment #10 from Takeshi Yoshino <tyoshino@google.com> ---
(In reply to Anne from comment #9)
> Well, that is the reason why I set them all to 0 at that point. But leaving
> them all the same and freezing seems okay too (if the UI code only looks at
> progress). I don't feel strongly. 
> 
> It's a little uglier to keep state around I think, but I don't mind it.

I see. CC-ed Glenn to hear thoughts.

FYI, here's what each browser is doing so far:

I used a script which sends a response using chunked Transfer-Encoding. It
flushes 1024 byte every 1 sec. After several chunks, it sends an invalid chunk
header (starts with asterisks instead of size of chunk in hex) to induce a
network error.

All of IE, Chrome and Gecko keep the "loaded" value.


IE 11.0.9600.17239

readystatechange1 * 2
loadstart(false, 0, 18446744073709552000) * 1
readystatechange2 * 1
readystatechange3 * 1
progress(false, 1024, 18446744073709552000) * 1
readystatechange3 * 1
progress(false, 2048, 18446744073709552000) * 1
readystatechange3 * 1
progress(false, 3072, 18446744073709552000) * 1
readystatechange4 * 1
error(true, 3072, 3072) * 1
loadend(true, 3072, 3072) * 1


39.0.2149.0 canary (64-bit)

Please ignore "(x, y, z)" for readystatechange events. It's due to a bug in
Chrome.

readystatechange1(false, 0, 0) * 1
loadstart(false, 0, 0) * 1
readystatechange2(false, 0, 0) * 1
readystatechange3(false, 0, 0) * 1
progress(false, 1024, 0) * 1
readystatechange3(false, 0, 0) * 1
progress(false, 2048, 0) * 1
readystatechange3(false, 0, 0) * 1
progress(false, 3072, 0) * 1
readystatechange4(false, 0, 0) * 1
progress(false, 3072, 0) * 1
error(false, 3072, 0) * 1
loadend(false, 3072, 0) * 1


Firefox Aurora 34.0a2 (2014-09-07)

readystatechange1 * 1
loadstart(false, 0, 0) * 1
readystatechange2 * 1
readystatechange3 * 1
progress(false, 1024, 0) * 1
readystatechange3 * 1
progress(false, 2048, 0) * 1
readystatechange3 * 1
progress(false, 3072, 0) * 1
readystatechange4 * 1
error(false, 3072, 0) * 1
loadend(false, 3072, 0) * 1

-- 
You are receiving this mail because:
You are the QA Contact for the bug.

Received on Monday, 8 September 2014 09:05:33 UTC