Simon Montagu wrote:

> I've appointed myself as clean-up and documentation policeman for the
> Bidi code. I'm working on the code and I will be posting some
> documentation here.
> 
> Steve, if you, Chris Waterson and Kevin McCluskey could send more
> detailed comments directly to me at [EMAIL PROTECTED] it would be
> very helpful.
> 
> Best regards,
> Simon

 From Simon Fraser:

========================================================




-------- Original Message --------
Subject: Re: [Fwd: Re: Bug 62777 (BiDi feature added by IBM) needs 
extensive review]
Date: Thu, 04 Jan 2001 11:44:41 -0800
From: [EMAIL PROTECTED] (Simon Fraser)
Organization: Netscape Communications Corp.
To: Steve Clark <[EMAIL PROTECTED]>
CC: [EMAIL PROTECTED]
References: <[EMAIL PROTECTED]>



Steve Clark wrote:
> 
> I had already included Simon on the reviewers list for BiDi support,
> but I also need someone to review the selection and caret portions of
> the layout changes, along with related classes (like
> nsFrameIterator).  Simon, I'm guessing you're the man for the caret
> code.  How about selection?  Is that Mike?  Anyone else.
> 
> Since this request is late, please don't feel pressure to do this by
> the 1:00 mtg today.

Comments here (since I'm too lazy to go to the meeting):

nsICaret.h:
  They added non-virual methods and member variables to an XPCOM
  interface -- a complete no-no.

  In addition, would prefer use of PRPackedBool rather than PRBool
  for member variables (esp. when > 1).

nsCaret.h/cpp
  Is frame->GetStyleData() guaranteed to return a non-null style
  info pointer? Thye never check it.


nsHTMLEditRules.cpp
  Would prefer to see
    if (NS_SUCCEEDED(res))
  instead of 
    if NS_SUCCEEDED(res)

  Also, need to be more careful about methods that can return NS_OK
  but a null frame/content pointer. So after the call to
    primaryFrame->GetChildFrameContainingOffset
  should check frameBefore before dereferencing it. Same with frameAfter
  lower down.

  Should use a nsCOMPtr here and for baseLevel, because of an early return:
    nsIAtom* embeddingLevel = NS_NewAtom("EmbeddingLevel");

nsTextEditRules.cpp

  Similar comments apply as to nsHTMLEditRules.cpp.

Otherwise, things look fine.

Simon


Reply via email to