[RESOLVED] running scripts sometimes prints a NOTICE:

ATS16805

Accoding to phpinfo()
I'm using PHP 4.3.10 on Apache 1.3.33

I'm storing my files in my local Apache webroot, and running the php files on http://localhost/phpfiles.php

I'm working through a Sams teach yourself PHP book. Assuming that the code examples in the book are correct, I wonder why when i copy the book's code into a php file in EditPlus, then save that file and view it in the browser, the output will sometimes show a Notice:

is this some type of debugging feature? is there a way to turn it off? should i be concerned?

this reminds me of similar warnings when using ColdFusion. if i had CF's debugging on, it would give me a clue as to where there was something wrong w/ my code. that's why i thought that perhaps it's a PHP or Apache debugging mode.

My concern is this-- what if i published similar code to my web site-- would my visitors see the "Notice" text?

could someone please explain, or point me to an explanation?

thank you!

drew010

its good to debug scripts using the error level, obviously on production you dont want this. you can turn it off in your php.ini by changing error_reporting to the desired value. you can also set this from within your script.
see http://php.net/error_reporting for examples and error constants.

ATS16805

very cool. yes, i agree about debugging the scripts during development... i was simply concerned about the very action of the error reporting-- i didn't know if it was in fact error reporting (debugging as i said above), or if in fact there was something critically wrong w/ my code.

thank you. you definitely gave me the info i'm looking for.

actually... one more question about this.

considering that i am getting a "Notice", does this mean that my code is faulty, and i should change it, or is it simply indicating a characteristic of the code, and not that it's necessarily an "error".

i realize that when i get the same type of printout, indicating a "Parse Error" that my code is in fact erroneous.

so, then, i guess what i'm asking is:
what is the text-book definition of a "Notice"?

drew010

notices are definately not critical errors of any sort. mostly they are there to help you catch errors or potential errors in your scripts.

for example if i try to do
if ($x == 5)
but $x does not exist, it will print a warning telling me the variable doesnt exist, which is helpful if im expecting that to execute true and it never does, php will tell me x doesnt exist.

similar for arrays, if you try to access an index that doesnt exist, they tell you undefined index.

you may also get warnings from passing invalid arguments to functions, for example mysql_fetch_array requires a valid link id, if you pass an invalid one, it will tell you that the resource id passed to it was invalid, which can also help if your queries arent successfully executing in mysql and you are wondering why.

ATS16805

thanks, drew!

the level of detail in your reply definitely helped a lot!
(phew!... wipes brow...)

what about the "undefined constant... assumes" notice. i've seen that one a lot as i've been trying some code examples from this book. all along i thought it was at least something to be a little worried about... but, from what you've said, that's apparently not necessarily the case.

if i read you correctly, the notices i've seen or cited here are more just to get me to take a closer look at what's going on, assuming i might back myself into a corner on down the line, if i were to continue w/ the code... it's possible that something might go wrong. does that sound about right?

Drakla

The undefined constant usually occurs when you use a word as a key to an associative array and forget to put quotes around it, e.g. $POST[name] instead of $POST['name']. Using old tutorials with php is a nightmare, and it's really disheartning seeing people trying their best but learning from an outdated source.

ATS16805

actually, this book (an e-book in the form of a chm file) was published in 2003-- at least that's what is written.

i thought perhaps it was a PHP revision issue. is that what you're suggesting?