126231 - several font download dialogs appear upon entering the unicode page

Status: VERIFIED FIXED

(Core :: Internationalization, defect, P2)

Description

[Tracy Walker [:tracy]]

seen on windows commercial build 2002-02-18-06-trunk

-goto the above URL
the page loads fine, but four download dialogs appear suggesting that these 
languages need to be intstalled in order for the page to view correctly. But, 
the page is correctly rendering the various fonts named in the dialogs. (Korean, 
Japanese, Traditional Chinese and Simplified Chinese)

Comment 1

[Teruko Kobayashi]

I can reproduce this with http://www.unicode.org/iuc/iuc10/x-ncr.html

If I go to the Japanese, Korean, Simplified Chinese and Traditional Chinese page
one by one, each font download dialog appears.  This is correct.

Comment 2

[Yuying Long]

See bug 126083 for more detatil of regression date.

Comment 3

[Roy Yokoyama]

*** Bug 126083 has been marked as a duplicate of this bug. ***

Comment 4

[Roy Yokoyama]

Not sure why my check-in for 119927 causes this problem.
Oh well, I'll accept this and see what's going on.......

Comment 5

[Roy Yokoyama]

Ok, i think I know what's going on
we used to pass 'ja' to 
rv = fontEnum->HaveFontFor(aFontPackID, &have);
but now we pass 'lang:ja' to it.

I will past a patch to trip out the 'lang:'.

Comment 6

[Roy Yokoyama]

Attachment: strip "lang:" and get the actual langID

This is because we didn't want to change
the contract between gfx and FontPackageProxy.
but we moved the call fontEnum->HaveFontFor(aFontPackID, &have);
from /gfx to /intl/locale

We need to strip the "lang:" from aFontPackID.
ftang: can you review?

Comment 7

[Roy Yokoyama]

ftang: can you veiw?

cc'ing ftnag

Comment 8

[Frank Tang]

Comment on attachment 70157 [details] [diff] [review]
strip "lang:" and get the actual langID

r=ftang

Comment 9

[Roy Yokoyama]

brendan: can you /sr this small patch?

Comment 10

[Brendan Eich [:brendan]]

Comment on attachment 70157 [details] [diff] [review]
strip "lang:" and get the actual langID

Sure -- we know there's always a lang: at the front, and that aFontPackID[5]
won't index beyond the end of a short string passed in by an errant caller?

/be

Comment 11

[Brendan Eich [:brendan]]

In other words, should we assert that (strncmp(aFontPackID, "lang:", 5) == 0)?

/be

Comment 12

[Roy Yokoyama]

Attachment: Adding Assertion

I think it's better to test for validity

Comment 13

[Roy Yokoyama]

checked in

Comment 14

[Roy Yokoyama]

*** Bug 125971 has been marked as a duplicate of this bug. ***

Comment 15

[Tracy Walker [:tracy]]

verified fixed on commercial windows build 2002-02-20-06-trunk

Comment 16

[Frank Tang]

*** Bug 126130 has been marked as a duplicate of this bug. ***