Web lists-archives.com

RE: [PHP] BC break in 7.2 caused by undocumented and unauthorised change

That first link of yours merely asked us both to calm down, which I did. It certainly was not a warning from a moderator that if I persisted that I would be banned.

That second link of yours contains this quote from Stanislav Malyshev: “you don't have the rights to be in the list if the community does not want it”. This means that even if you do not break any published rules that you can be banned just because someone does not like what you have to say. So if someone proposes a BC break and I argue against it he can complain and have me banned. Does this sound fair to you?

Your third link indicates that you do not understand how session_name() is supposed to work. It can either be used to get the current session name, set a new one, or both at the same time. This makes it perfectly clear that it is valid to run session_name() while a session is active. The only point to recognise is that if you set the session name this will not take effect until the next session_start(), and it is ONLY session_start() which will fail is a session is active.

When programmers use session_name() they use it in either the get or set modes and rarely for both at the same time. That is why few people spotted the bug that when you try to do both it actually does neither. This is the only bug is session_name(), and this was never fixed.

In 7.2 the behaviour was changed so that session_name(‘newname’) will now fail if a session is active. So instead of fixing a genuine bug a new artificial bug was introduced. I say “artificial” simply because it was changed because someone thought that it was not logical to set a session name while another session was already active. The manual clearly states that it should be possible to run session_name() while a session is active. It is only session_start() that will fail if a session has already been started. Thus the following sequence was perfectly valid until 7.2:





I repeat, the change in 7.2 did not fix the real bug but introduced a new and artificial bug.

Tony Marston

From: Christoph M. Becker <cmbecker69@xxxxxx>
Sent: Thursday, May 24, 2018 2:02:38 PM
To: Tony Marston; php-general@xxxxxxxxxxxxx
Subject: Re: [PHP] BC break in 7.2 caused by undocumented and unauthorised change

On 23.05.2018 at 17:29, Tony Marston wrote:

> If you read that post you will see that it was mostly complaining about
> the person known as "rhsoft". All I did was argue AGAINST his viewpoint,
> yet I was still banned in what I would call an arbitrary manner. Note
> that I say "banned" and not "suspended" as I was not informed of what
> action was to be taken against me, nor was I given a chance to explain
> myself.

FTR: you have been repeately admonished for offending the mailing list
rules and netiquette by several people, for instance in
<https://nam04.safelinks.protection.outlook.com/?url=http%3A%2F%2Fnews.php.net%2Fphp.internals%2F100765&data=02%7C01%7C%7Cfe08c799e18549a00d6b08d5c176a211%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C636627637619303667&sdata=RLSQJDIk7x94Tt7HCLv34xn0fsxD49T2yFk6lEYPwXU%3D&reserved=0>.  Even after it has suggested
to suspend your mailing list access, you have not shown insight or
regret, see <https://nam04.safelinks.protection.outlook.com/?url=https%3A%2F%2Fexternals.io%2Fmessage%2F101479%23101501&data=02%7C01%7C%7Cfe08c799e18549a00d6b08d5c176a211%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C636627637619303667&sdata=0S%2BBiLk%2BD2TCXx19opRNizyzBoy40MN%2BEp%2BV0ASDK3k%3D&reserved=0>, for instance.

Wrt. the actual issue see my comment in <https://nam04.safelinks.protection.outlook.com/?url=https%3A%2F%2Fbugs.php.net%2F76358&data=02%7C01%7C%7Cfe08c799e18549a00d6b08d5c176a211%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C636627637619303667&sdata=WgFXFURmwH5NmRwyC0sP1I3ehXRMHxmQf4pIKT6wqww%3D&reserved=0>.

Christoph M. Becker