Re: Re: note 69872 rejected and deleted from language.types.array by notes editor bjori

On 25 Sep 2006, at 17:50, Hannes Magnusson wrote:

I'll speak even slower: r e a d t h e g u i d e l i n e s o n h o w t o r e p o r t a d o c b u g.

Last time I looked doc and code bugs were all in one system. My impression of the problem was that the behaviour was wrong and the docs were right, because what it does is certainly quite unexpected. The simple problem that led me to track this down was that I had an object property that seemed to vanish after casting to an array. Calling the bug bogus didn't make any sense at all - either PHP or the docs had to be wrong, and marking it as bogus just says "There is no problem". It seems like I guessed wrong on the bug category, but that doesn't make the report bogus. A more appropriate course of action would have been to change the category to a documentation problem. Requiring a reporter to know WHY a bug occurs is just obtuse - a bug report is typically only an observation that a problem exists, hopefully with some accuracy regarding when.

If the docs are _wrong_ don't you think they should be changed? It would help way more people to fix the docs rather then submitting note arguing with the manual. That will only confuse people.

I wasn't arguing with the manual! Quite the opposite in fact - The function I posted acts like the docs said PHP should! Given that my bug report was bogused, posting a workaround for PHP's apparently incorrect behaviour seemed like a perfectly reasonable course of action, especially given that there was no other source for this information. What's more, now I hear that the documentation has been updated (thank you), the function is still useful as it saves having to figure out how to parse out the additional information it includes.

And if you can, we would _love_ a patch against phpdoc (since this is community driven effort after all) correcting this issue.

I would have been more than happy to do that. The only problem in this case is that the cause of the problem was not made apparent until after the report was bogused, so I was not in any position to suggest a fix.

(As a side note: how do you like the comments currently on that page? I at least am not going to read them all)

One minor idea that might help with this - implement digg-style thumbs-up/down, and possibly 'report as obsolete/inaccurate/miscategorised/inappropriate' so that comments can get rated and editors can find out which ones are of lesser interest without having to read them all. If digg is anything to go by, people just like clicking stuff. Thank you for your attention, and for fixing the docs. I've amended my original bug report to reflect this information. Marcus PS: someone with appropriate karma, please feel free to trash this comment of mine: https://master.php.net/manage/user-notes.php?action=delete+68828 as it's got errors in. -- Marcus Bointon Synchromedia Limited: Creators of http://www.smartmessages.net/ marcus@synchromedia.co.uk | http://www.synchromedia.co.uk/