Re: lynx-dev cookie file observation/question

[brian j. pardy]

[First off, apologies for not being around to answer some of the 
cookie questions that have come up, I'm subscribed to the list but 
recently lost my ability to ssh in to the box my mail is on, and I
haven't set up IMAP yet, and -- short version, feel free to personally
CC me on any questions, anybody]

Kim wrote:

> Ever since the (now fixed) snafu a few releases ago, caused the
> .cookie file to get wiped out, I've been dup'ing it when testing
> a new release.
> 
> A curiousity that I've noticed, is that on each lynx invocation,
> the new .cookie file that is written upon exiting lynx, is written
> in the reverse order of the pre-invocation file (ie, invoke/exit
> lynx twice in a row, and the .cookie file is back to the original
> order).
> 
> Functionally, I'm sure the ordering of the entries probably doesn't
> matter, but I find the behavior somewhat "unusual".
> 
> Without slogging thru the code, I wonder if this is being done for
> a reason (performance, I'd guess), or is just an oversight of some
> sort ...?

Indeed that does happen, and led me down quite a few incorrect paths 
when I was attempting to debug the trunc problems.  It's an annoyance
more than anything, AFAICT it is not being done for a reason -- it's
just an artifact of how they're stored in cookie_list (in store_cookie())
when read from the file.  Each new cookie is inserted at the head of
the list, not at the end.