307605 - Eliminate method nsIMAPGenericParser::at_end_of_line()

Status: VERIFIED FIXED

(MailNews Core :: Networking: IMAP, defect)

Description

[Hans-Andreas Engel]

The method |nsIMAPGenericParser::at_end_of_line()| can be eliminated and be
replaced by using the field |fAtEndOfLine|.

Comment 1

[Hans-Andreas Engel]

There are three possibilities for the value of fNextToken:
1) |fNextToken| is set to |nsnull|
2) |fNextToken| is obtained via |strtok|.  Then it cannot be CRLF (in fact, it 
   does not contain any whitespaces)
3) |fNextToken| is set explicitly to |CRLF|.  In this case, also |fAtEndOfLine| 
   is set to PR_TRUE (this must added in one place).

So, if |nsCRT::strcmp(fNextToken, CRLF) == 0|, then |fAtEndOfLine| is true. Thus,
    fAtEndOfLine || (nsCRT::strcmp(fNextToken, CRLF) == 0)
evaluates to |fAtEndOfLine|.  Since the above statement is the return value of
|at_end_of_line()|, we can replace all calls to this function by |fAtEndOfLine|.

Comment 2

[Hans-Andreas Engel]

Attachment: eliminate nsIMAPGenericParser::at_end_of_line()

Comment 3

[David :Bienvenu]

Comment on attachment 195353 [details] [diff] [review]
eliminate nsIMAPGenericParser::at_end_of_line()

seems ok - I'll apply the patch and run with it.

Comment 4

[David :Bienvenu]

Comment on attachment 195353 [details] [diff] [review]
eliminate nsIMAPGenericParser::at_end_of_line()

I've been running with this for a bit and haven't seen any problems.  I'll run
with it for a day or two more. Please ping me then to get me to check it in -
thx!

Comment 5

[David :Bienvenu]

fix checked in, thx, Hans.

Comment 6

[Stephen Donner [:stephend] Not actively reading bugmail]

Both David and I have been using this on different IMAP servers, and I haven't
seen any fallout.

Verified FIXED using LXR and my source tree for code verification, and build
2005-10-01-05 for feature testing on Windows XP SeaMonkey trunk.