135771 - filename keeps spinning; should stop instead and have red X

Status: VERIFIED FIXED

(SeaMonkey :: Composer, defect)

Description

[sujay]

using 4/4 build of netscape

1) launch netscape
2) jump to yahoo.com
3) File | Edit Page
4) make a change/edit
5) Publish
6) leave username and passwd. blank
but give it a filename of "yah.html"
7) click on Publish

Notice in the progress dialog you get red X's for all the unsuccesful
files..this is fine..however, the filename yah.html keeps spinning/publishing.
It goes on for several minutes....it should stop immediately and also
have a red X next to that filename.

Comment 1

[Charles Manske]

This problem is in either HTTP or nsIWebBrowserPersist code. We don't get the 
final nsIWebProgressListener::onStateChange message to tell us we should fail.
Here's the dubug output for this case:
***** onStateChange request: undefined
      state flags:  STATE_START,
 * requestSpec=undefined, pubSpec=http://www.yahoo.com/foo.html, aStatus=0
*****        status is 0
***** onStateChange; NO REQUEST CHANNEL
***** onStateChange request: undefined
      state flags:  STATE_STOP,
 * requestSpec=undefined, pubSpec=http://www.yahoo.com/foo.html, aStatus=0

Kathy: The fixes we worked on this morning don't help this case.

Comment 2

[Kathleen :Brade]

Sujay, DSL Michael--do you still see this problem?

Comment 3

[michael wendell]

I am still seeing this problem on the 04-09 trunk.

Comment 4

[Kathleen :Brade]

Attachment: patch

note that this patch has some other fixes (like for bug 135527)

The problem isn't in the persist code but in the handling of cancellation.  We
won't get notification for all files if we cancel.  The fix is to set the
"busy" state to a non-busy state (failure).

Comment 5

[Akkana Peck]

Comment on attachment 78562 [details] [diff] [review]
patch

r=akkana, but please remove the extra dump() or make it conditional on a
debugging variable.

Comment 6

[kinmoz]

Comment on attachment 78562 [details] [diff] [review]
patch

sr=[e-mail]

Comment 7

[Kathleen :Brade]

users are very confused by the continuing animation despite the publishing
action being canceled; nominate this for adt approval on branch

Comment 8

[Jaime Rodriguez, Jr.]

adt1.0.0+ (on ADT's behalf) approval for checkin into 1.0. Pls check this in to
the trunk and 1.0 branch today.

Funny thing the patch we see in this bug, seems to include a patch for another
bug we just approved ... hmmmmmm.

Comment 9

[Asa Dotzler [:asa]]

Comment on attachment 78562 [details] [diff] [review]
patch

a=asa (on behalf of drivers) for checkin to the 1.0 branch

Comment 10

[Kathleen :Brade]

This fix was checked into both the trunk and the 1.0 branch

Please verify in both locations.

Comment 11

[sujay]

verified on both 4/15 trunk and 4/13 branch.