224254 - make CSS parser use nsresult for |aErrorCode|

Status: RESOLVED FIXED

(Core :: CSS Parsing and Computation, defect, P3)

Description

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

The CSS parser has long had a TODO comment that it should use nsresult for its
|aErrorCode| values.

The following patch is largely the result of:

s/PRInt32& aErrorCode/nsresult& aErrorCode/
s/PRInt32 errorCode/nsresult errorCode/

but also:
 * removes three checks for -1 error code meaning EOF and replaces them with
assertions, since I can't find any place that sets the error code to -1 on EOF
 * removes the TODO comment

Comment 1

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

Attachment: patch

Comment 2

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

I tested that the "test styles" bookmarklet at
http://www.squarefree.com/bookmarklets/webdevel.html still works as I type.

Comment 3

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

(I should really do that in a debug build, though.)

Comment 4

[Boris Zbarsky [:bzbarsky]]

Are the callers who used to check for -1 likely to throw exceptions on EOF when
we don't want them to?  (I guess no, since the scanner does not set rv to -1 on
EOF....)

Comment 5

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

No assertions in the same test in a debug build, so I don't think anything is
returning -1 in aErrorCode for EOF.

Comment 6

[Boris Zbarsky [:bzbarsky]]

I'll try to get to this sometime this week...

Comment 7

[Boris Zbarsky [:bzbarsky]]

Comment on attachment 134542 [details] [diff] [review]
patch

r+sr=bzbarsky; I'm sorry this took so long...  The EOF thing is indeed not a -1
anymore, since we just use rv from the stream for that error code, and streams
return NS_OK and 0 bytes for a read at EOF.

Comment 8

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

Fix checked in to trunk, 2003-11-13 14:48 -0800.