99633 - printing regression tests ask for a printer

Status: VERIFIED FIXED

(Core :: Layout: Tables, defect)

Description

[Bernd]

with a current CVS under win98 if I run rtest baseline in the table/printing
directory it creates only rgd files for the first two files and ask's for a printer
at file:///s|/mozilla/layout/html/tests/table/printing/bug54829.html. If I
dismiss the dialog it stalles.

Comment 1

[Bernd]

Attachment: hack to ease the pain

Comment 2

[Bernd]

This is a hack as usual, the problem is that we start a print job even if we are
in regression tests. Further in DoPrint we check for the regression output, but
this is already to late as we started already a print job. 

Comment 3

[Bernd]

taking the bug

Comment 4

[rods (gone)]

*** Bug 102763 has been marked as a duplicate of this bug. ***

Comment 5

[rods (gone)]

Attachment: should use NS_OK instead of '0' and it includes a small debug fix for me.

Comment 6

[rods (gone)]

r=rods for the patch

Comment 7

[rods (gone)]

Actually now that I think of it, shouldn't we also skip the EndDocument call 
and there is a file pointer?

Comment 8

[Bernd]

 OnEndPrinting(NS_OK); // removes listener
 >>>> NS_ASSERTION(0!=0, "gotcha");
  if (mPrintPS) {
    mPrintDC->EndDocument();
#ifdef DEBUG_PRINTING
    fprintf(mDebugFD, "****************** End Document ************************\n");
#endif
  }


I added an assertion its never called during regression tests so it looks that
the patch is sufficient

Comment 9

[Marc Attinasi]

Comment on attachment 51854 [details] [diff] [review]
should use NS_OK instead of '0' and it includes a small debug fix for me.

sr=attinasi

Comment 10

[karnaze (gone)]

Comment on attachment 51854 [details] [diff] [review]
should use NS_OK instead of '0' and it includes a small debug fix for me.

r=karnaze

Comment 11

[karnaze (gone)]

Adding nsbranch keyword, since printing regression testing is mandatory prior to 
layout checkins.

Comment 12

[Kevin McCluskey (gone)]

Marking nsbranch+. Needed to make printing regression tests work. 

Comment 13

[Jaime Rodriguez, Jr.]

Marking [PDT] for tracking.

Comment 14

[Bernd]

fix checked in into trunk, reassigning the bug to rods, as I do not build branch

Comment 15

[Jaime Rodriguez, Jr.]

This looks like a candidate for the branch. Can we talk about it in the PDT today?

Comment 16

[Jaime Rodriguez, Jr.]

pls check this into the branch - PDT+ 

Comment 17

[rods (gone)]

checked into the branch

Comment 18

[Amarendra Hanumanula]

can someone verify this bug.. I do not have a debug build on win98..
Thanks

Comment 19

[lchiang]

[e-mail] (Bernd Mielke) - can you help to verify this? Thank you.

Comment 20

[Amarendra Hanumanula]

I got a debug build and ran the regression tests(rtests) in the directory
 /mozilla/layout/html/tests/table/printing.. It did create all the baselines for 
all the testcases in that directory but finally it gave an asserstion 
"NS_ENSURE_TRUE() in file \mozilla\docshell\base\nsWebShell.cpp
But I think this is different form this bug
Marking verified