Re: [htdig-dev] Re: Importing cookies

[Gilles D. <gr...@sc...>]

According to Gabriele Bartolini:
> >been set to a new HtCookieInFileJar object, which doesn't get deleted.
> >Shouldn't the delete cookie_file statement be moved outside of the
> >innermost if clause, and past the end of the else clause?
> 
> It is deleted later by the main function, through the base class 
> HtCookieJar pointer.

OK, but that still doesn't explain the inconsistency.  The
HtCookieInFileJar object is allocated whether or not the result is
non-zero, so why do you delete the object after using it when the result
is zero, but leave it up to a completely different function otherwise?
It just doesn't make sense to me, and it's clearly not "defensive
programming".  If you don't need the object anymore, you should delete
it unconditionally.  If you do need the HtCookieInFileJar object for
later when the result code is non-zero, that hasn't been made clear.

> >expert on every piece of code added to 3.2.  If you add a description to
> >defaults.cc yourself, doing your best to describe it, I'll gladly fix any
> >grammatical errors or ask you about ambiguities I find in the description,
> >but I don't want to have to document things I don't understand or use.
> >Ditto for testing - I can't test cookie support in htdig, because I
> >don't use them on my system.
> 
> Sorry Gilles, I didn't explain myself correctly. I just wanted you to 
> eventually correct my entry in the defaults file, as you know my english 
> has a marked spaghetti accent.

Well, on Feb. 1 you wrote: "Then, as always, Gilles (I know I am terrible
but you should know me already!) please find me a suitable description
for the 'cookies_input_file' attribute to be put in the defaults.cc file
(and defaults.xml too!)."  The words "find me a suitable description"
certainly seem to imply you were expecting me to write it, not just
proofread it, especially since you never gave me anything to proofread.
Anyway, let me know when you've written something, and I'll have a look.
(Maybe post to htdig-dev for others to comment too.)

> I did not mean to ask you to test the cookies code, sorry about it! I was 
> referring just to the description stuff (or maybe the distributed 
> cookies.txt file).

I misinterpreted what you wrote on Feb. 1:  "Please test it and let me
know if I can commit in the next week."  I thought you were asking me,
but that e-mail was addressed to all the developers.

> I have switched to Debian now (my friends at my local LUG convinced my 
> after years of RedHat) and hopefully tomorrow I'll be able to commit the code.

I have to ask, was it for practical or ideological reasons that they
convinced you to switch?  ;-)

-- 
Gilles R. Detillieux              E-mail: <gr...@sc...>
Spinal Cord Research Centre       WWW:    http://www.scrc.umanitoba.ca/
Dept. Physiology, U. of Manitoba  Winnipeg, MB  R3E 3J7  (Canada)