53895 - [WINDOWS] mimetype.description should not contain (*.suffix)

Status: VERIFIED FIXED

(Core Graveyard :: Plug-ins, defect, P4)

Description

[Henrik Gemal]

If you look at the about:plugins page the mimetype.description incorrectly 
contains *.suffix information.
Fx the "Netscape Navigator Plug-in for Adobe Acrobat v4.05"
mimetype.description is "Acrobat (*.pdf)"

in Netscape 4.x the description is just "Acrobat" which is correct. The suffix 
info can be retrived from the mimetype.suffixes

Expected:
the "mimetype.description" returns just the description not any info about the 
suffixes!

Comment 1

[ekrock's old account (dead)]

Not a Netscape 6 RTM blocker. FUTURE. This bug has been marked Future because
the Netscape engineer it is assigned to is overburdened.

Comment 2

[Henrik Gemal]

the "Description" field contains reduntant info, since suffixes can be obtained 
via the "Suffixes" object.

Netscape 4.x didn't include suffixes in "Description".

Comment 3

[Henrik Gemal]

I'm not sure how Netscape 4.x did this but Mozilla seems to just read the
FileOpenName property of the plugin file. This is done in:
http://lxr.mozilla.org/mozilla/source/modules/plugin/base/src/nsPluginsDirWin.cpp#405

perhaps someone could look at how Netscape 4.x did this?

Comment 4

[Jason Summers]

Nav4 also gets the description from the FileOpenName property, but it strips off 
the suffix information. A little experimentation suggests that it strips off 
everything starting with the last open parenthesis found in the string. (In 
other words, it's a hack.)

Comment 5

[rubydoo123]

the suffix comes from the plug-in

Comment 6

[Jason Summers]

The FileOpenName field contains both the description and the suffix, as is
required by the plug-in specification. When a script queries the description,
Mozilla returns the entire FileOpenName field instead of correctly parsing out
the description. Granted that this is a very minor problem, but marking it
INVALID in this way implies that there is something wrong with what the plug-ins
are doing, which is not true.

Comment 7

[Peter Lubczynski]

Yeah, if 4.x did this, we really should emulate. There may be JS out there
depending on this quirk.

Here's how 4.x did it:
http://lxr.netscape.com/nova/source/lib/plugin/npwplat.cpp#655

Comment 8

[Peter Lubczynski]

--->taking bug

Comment 9

[av (gone)]

I have a patch.

Comment 10

[av (gone)]

Attachment: patch v.1

I did not make it platform specific. Not sure if this should be Windows only.
Please review.

Comment 11

[Peter Lubczynski]

Comment on attachment 91144 [details] [diff] [review]
patch v.1

r=peterl

Comment 12

[Alec Flett]

Comment on attachment 91144 [details] [diff] [review]
patch v.1

sr=alecf

Comment 13

[av (gone)]

In the trunk.

Comment 14

[shrirang khanzode]

verified fixed on trunk 0812.