Closed Bug 479104 Opened 16 years ago Closed 16 years ago

move windows plugin code to unicode

Categories

(Core Graveyard :: Plug-ins, defect)

ARM
Windows CE
defect
Not set
normal

Tracking

(Not tracked)

RESOLVED FIXED

People

(Reporter: vlad, Assigned: vlad)

References

Details

Attachments

(1 file, 3 obsolete files)

The current windows code is all non-unicode; for Windows CE, we need it to be unicode. Here's a patch that uses all the magic TSTR TCHAR etc. stuff.
Attachment #362955 - Flags: review?(jst)
Vlad, Crowder, you all need to talk more :( Vlad, see bug 422784 (which is waiting for my review since forever, need to get to that now that we're past 1.9.1 on the trunk). Question is, which patch do we prefer. I haven't looked in enough detail yet to know the answer...
This patch is probably better, and certainly fresher, though I haven't tried it in a WinCE build, yet.
Ack, my fault, sorry -- I was just working through compile problems as a I ran into them =/ This patch is simpler, but crowder's approach of actually doing the conversion to unicode-everywhere might actually be better for consistency... we won't have to guess which mode we're in if problems come up on the desktop.
So should I review Crowders patch then, and ignore this one, or should we take this one first and work towards Crowders patch? Crowder, do you think your patch is still good to go, or does it need significant work before it's worth reviewing?
After talking with crowder, it might make more sense to review this one -- we probably do want unicode everywhere at some point, but this patch is simpler for review purposes and will solve the immediate problem.
This is the bug we want reviewed, so I've duped bug 422784 here.
Attached patch updated patch (obsolete) — Splinter Review
Updated patch; fixed some things that the try server complained about. (Mainly missing tchar.h include on the desktop.)
Assignee: nobody → vladimir
Attachment #362955 - Attachment is obsolete: true
Attachment #364359 - Flags: review?(jst)
Attachment #362955 - Flags: review?(jst)
Here's an additional patch to the previous, just placed here for ease of review -- it fixes sizeof() -> NS_ARRAY_LENGTH, and also fixes RegSetValueEx.. apparently REG_SZ means either char or wchar depending on which version you call, but the length is still in bytes. It's all very horrible!
Attachment #364359 - Attachment is obsolete: true
Attachment #364359 - Flags: review?(jst)
Attached patch updated patchSplinter Review
Combined patch with the above two.
Attachment #364576 - Attachment is obsolete: true
Attachment #364577 - Flags: review?(jst)
Attachment #364577 - Flags: superreview+
Attachment #364577 - Flags: review?(jst)
Attachment #364577 - Flags: review+
Comment on attachment 364577 [details] [diff] [review] updated patch Looks good.
I had to make a few small build changes, and did a tryserver build here -- https://build.mozilla.org/tryserver-builds/2009-03-01_14:48-vladimir@mozilla.com-479104unicode-plugins/ I've confirmed that all the plugins I have installed are recognized the same with this patch as without, so gonna land this shortly.
Checked in.
Status: NEW → RESOLVED
Closed: 16 years ago
Resolution: --- → FIXED
Comment on attachment 364577 [details] [diff] [review] updated patch >+t_NS_NewNativeLocalFile(wchar_t *path, PRBool b, nsILocalFile **retval) >+{ >+ return NS_NewNativeLocalFile(NS_ConvertUTF16toUTF8(path), b, retval); >+} Native != UTF8 so this will fail for plugins in non-ASCII paths...
Hm, indeed. Though that code path is only with UNICODE, so it won't affect our current win32 builds. Will fix in a followup.
Product: Core → Core Graveyard
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: