XHR with .responseType=blob returns 0 size blob

• Component: XML
• Status: RESOLVED FIXED

[Taiju Tsuiki]

XHR with .responseType = 'blob' returns Blob, but its .size is set to 0.

[Taiju Tsuiki]

Created attachment 146779 [details]
failcase

[Taiju Tsuiki]

Created attachment 146780 [details]
+reading blob with FileReader

Note that, we can read this blob with FileReader, even if its size is 0.

[Kinuko Yasuda]

Created attachment 146854 [details]
Patch

[Alexey Proskuryakov]

Comment on attachment 146854 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=146854&action=review

> Source/WebCore/xml/XMLHttpRequest.cpp:287
>          if (m_binaryResponseBuilder.get()) {

This is old code, but this ".get()" is unnecessary, and should be removed sooner or later.

[Alexey Proskuryakov]

<rdar://problem/11640154>

[Kinuko Yasuda]

(In reply to comment #4)
> (From update of attachment 146854 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=146854&action=review
> 
> > Source/WebCore/xml/XMLHttpRequest.cpp:287
> >          if (m_binaryResponseBuilder.get()) {
> 
> This is old code, but this ".get()" is unnecessary, and should be removed sooner or later.

Thanks, will delete .get() before landing.

[Kinuko Yasuda]

Committed r120042: <http://trac.webkit.org/changeset/120042>

[János Badics]

Unfortunately this new test fails on every bots since it was introduced in 120042. See https://bugs.webkit.org/show_bug.cgi?id=88844

[Kinuko Yasuda]

(In reply to comment #8)
> Unfortunately this new test fails on every bots since it was introduced in 120042. See https://bugs.webkit.org/show_bug.cgi?id=88844

It looks it's passing at least on chromium.  (In reply to comment #8)
> Unfortunately this new test fails on every bots since it was introduced in 120042. See https://bugs.webkit.org/show_bug.cgi?id=88844

Sorry for the breakage, I'll take a look. (At least it's passing on chromium?)

[János Badics]

> It looks it's passing at least on chromium.  (In reply to comment #8)
> Sorry for the breakage, I'll take a look. (At least it's passing on chromium?)
I'm sorry, it passes only on chromium : )

[Kinuko Yasuda]

(In reply to comment #10)
> > It looks it's passing at least on chromium.  (In reply to comment #8)
> > Sorry for the breakage, I'll take a look. (At least it's passing on chromium?)
> I'm sorry, it passes only on chromium : )

It looks like XHR_RESPONSE_BLOB is not enabled on those platforms and I needed to add the test to Skipped for them.  I'm going to make Skipped changes while I'll take another look.

[Mario Sanchez Prada]

(In reply to comment #11)
> [...]
> It looks like XHR_RESPONSE_BLOB is not enabled on those platforms and I needed 
> to add the test to Skipped for them. 
> I'm going to make Skipped changes while I'll take another look.

I think it would have been better to have filed a bug and use BUGWK###### instead of BUGWKGTK SKIP in the expectations file, so we have a clue were to look at when unskipping it.

Already did it: http://trac.webkit.org/changeset/120059 (new bug reported: 88859)

[Kinuko Yasuda]

(In reply to comment #12)
> (In reply to comment #11)
> > [...]
> > It looks like XHR_RESPONSE_BLOB is not enabled on those platforms and I needed 
> > to add the test to Skipped for them. 
> > I'm going to make Skipped changes while I'll take another look.
> 
> I think it would have been better to have filed a bug and use BUGWK###### instead of BUGWKGTK SKIP in the expectations file, so we have a clue were to look at when unskipping it.
> 
> Already did it: http://trac.webkit.org/changeset/120059 (new bug reported: 88859)

Thanks for doing this!!   I was going to file a new bug but must have done so when I update Skipped/TestExpectations.