124864 - nsISelectionPrivate::ToStringWithFormat needs constants for flags param

Status: VERIFIED DUPLICATE of bug 102029

(Core :: DOM: Editor, defect)

Description

[Mike Pinkerton (not reading bugmail)]

/* wstring toStringWithFormat (in string formatType, in unsigned long flags,
in PRInt32 wrapColumn); */
  NS_IMETHOD ToStringWithFormat(const char *formatType, PRUint32 flags, PRInt32
wrapColumn, PRUnichar **_retval) = 0;

what are the flags supposed to be? i see several places in the code where people
are doing:

        htmlstring = privateSelection.toStringWithFormat("text/html", 128+256, 0);

Magic, i tell ya. It's all magic. Add some enum constants to the IDL. Your
friends will thank you.

Comment 1

[Akkana Peck]

The flags are defined in nsIDocumentEncoder.h.  There's a bug to IDL-ize that
file, but it's on hold because the document encoder may become obsolete with the
new document persist classes.  Arguably the flags should move there and we
should make this part of bug 99624.

Comment 2

[Kathleen :Brade]

add dependency for idl-izing of nsIDocumentEncoder (bug 102029)

cc jst who didn't understand the need for idl-izing the docEncoderjst:  it'd be
preferable to many people if we could use
nsIDocumentEncoder.OutputWithAbsoluteLinks (for example), instead of using
"magic numbers"

Comment 3

[mjudge]

this is a problem with the document encoder not selection. I dont know what to 
do with this. i am marking it future.   I assume we will change the document 
encoder long before we fix this.

Comment 4

[Mike Pinkerton (not reading bugmail)]

if the issue is with the document encoder, who owns it? shouldn't the bug be
reassigned to the correct owner instead of futuring it?

Comment 5

[Akkana Peck]

It's somewhat academic who owns this bug or whether it's futured since it's
really a dup of bug 102029 (fixing that bug would automatically fix this one)
and that bug has had a patch in it for 5 months.  The issue is getting sr/approval.

Comment 6

[mjudge]

then i am marking this a dup

*** This bug has been marked as a duplicate of 102029 ***

Comment 7

[sujay]

verified.