Re: [phpslash-users] comment section

[Ajay S. <ss...@od...>]

tobozo <to...@io...>, let me have access to his site which had the
same problem and I fixed it.  Checkout the latest CVS for the fix (and
several other important fixes.

The problem is that the template variable COMMENT_ID was getting carried
over from the parent comment to the form.  Even though we are assign
COMMENT_ID the value in $ary[comment_id] (which is undefined in a new
comment), it seems that PHP 3.0.16 doesn't update COMMENT_ID.  Thus,
when you hit Submit on the new comment form, PHPSlash see's the
parent_id in the COMMENT_ID fild and performs an update.

This is also a security problem because anyone can overwrite a comment
by screwing with the form and then sending it back to the code.  This
has been fixed also with a check for $seclev.

later,
ajay


On Sun, 11 Mar 2001, Tigran wrote:

> Some update on this problem.
> I was running rc4 had the same problem then deleted everything and reloaded
> 0.6 still have the problem.
> Running on phplib 7.2c
> Yes it does happen on all articles
> I run the commands you gave in MySQL and got the same number returned for both.
>
>
> Something interesting I did not see before.  A comment on the article will
> render a new comment.  However a reply to a comment erases the original
> comment.  it seems to write over the same comment id.  And it displays this
> message.
>
> found the comment_id, '61', so now we update.
> Updated comment 61, on reply on forth by tigo, <>
>
> so it just updates the existing comment id.
>
> any thoughts
>
> thank you so very much for helping me with this.
>
> The site is www.usanogh.com and the article with all the test I have is
> "Many Thanks"
> http://www.usanogh.com/articles/html/article.php3?story_id=26
>
>
> At 09:18 PM 3/8/01 -0500, you wrote:
> >Hmmmm.. I assume phplib is 7.2c?  I really don't know what wrong.. and
> >without the ability to turn on debuging and watch the queries get passed I
> >couldn't even make a guess.  I assume this happens for every article?  Did
> >you check the db_sequence table?  This is where we store the most recently
> >used id, perhaps it's out of synch with the site (ie: maybe there's a way to
> >force a id into phpslash from the GET params and dodge a key lookup.. This
> >could cause serious havoc.
> >
> >Try doing something like this in MySQL:
> >select nextid from db_sequence where seq_name='psl_comment_seq';
> >select comment_id from psl_comment order by comment_id desc limit 1;
> >
> >The two resulting values should be the same.
> >
> >
> >
> >Each query should produce two numbers, they should be the same.  If they
> >aren't you have a problem inthe db, if they are you problem lies in the core
> >and / or templates...  Did you change / omit any of the {} placeholders in
> >the templates?  Perhaps something in a form is getting crossed?
> >
> >-n
>
> _________________________________________________
> "Science is like sex: sometimes something useful comes out,
> but that is not the reason we are doing it" -- Richard Feynman
>
>
> _______________________________________________
> phpslash-users mailing list
> php...@li...
> http://lists.sourceforge.net/lists/listinfo/phpslash-users
>

-- 
----------------------------------------------------------------
Satyajot (Ajay) Sharma
ss...@od...
Digital Odyssey System Administrator
Voting for the lesser of two evils is still voting for evil.
----------------------------------------------------------------