Baffling: code works/doesn't work

Anon

Hi,

If you can help me with this, I'd appreciate it. This page: http://www.fcPremier.com/francais/ is a new part of my site.

It's basically one page that calls presentation and content data depending on user clicks.

But case "article": is the one portion that is not working right.

For some browsers, seemingly all MS, browsers, this type of page: http://www.fcpremier.com/francais/?do=article&id=000254 works fine. For others, mostly Netscape, the content does not appear.

The weird thing is that the code to be executed is all serverside.

I can tell by checking source when it doesn't work that the page has gone out, done a function on an included page, and returned with some data. The data that I see returned is hardcoded into the function.

But all the database data - the article content - is not included!

And yet, on other browsers, it is!!

How could this switch depending on browser - it's all serverside code.

How could the source code send from the server be different depending on which browser is requesting info?

I don't get it.

The code is too complex, with includes etc. to post here, but I hope someone can give me a clue.
-john

Anon

I've just tried your link

http://www.fcpremier.com/francais/?do=article&id=000254

and it shows up fine, in both NN4 and IE5.

Even more baffling?

Anon

Thanks for trying it. Glad it works for you.

http://www.fcpremier.com/francais/?do=article&id=000254

My Mac PowerBook G3 (lombard edition) does not display that page in either IE5 or NN4.6.

My Caldera box running NN does display that page.

Most people in my office seem to be able to get it, but some don't. All PCs seem to be able to get it.

Wierder and wierder.

Thanks anyways!

Anon

According to the validator at www.w3.org, the only real error with the page is an extra close font tag. It should work, unless, sometimes the page returns broken parts and other times returns proper html.

Note that netscape is usually a bit more finicky about tables and such.

the validator can be reached at this address:

http://validator.w3.org/

Anon

Seems strange. When it doesn't work, is the page ended correctly, or is it interrupted after a precise point ? Are only the db-data missing ?
Actually, all I can tell you right now is that there are several mistakes in the french introducing text; the correct one is :

"Bienvenue sur le site PREMIER

Nous développons des programmes et des produits de qualité afin d'aider les professeurs et les élèves à avoir plus de succès personnellement, socialement ainsi que dans leur milieu scolaire, et ce, de la maternelle jusqu'à l'université."

The text in the article 254 is correct.
I don't know if you care, but I tell you anyway ;-)

Anon

I'm now wondering if the problem with accessing the page is simply an IP-based thing.

I have my Mac PowerBook set for our USA HO network, and was in our Canadian HO. It usually is no problem, because both assign IPs via DHCP servers.

Since I don't get the error on my Mac at home .. I'm wondering if this could be the issue.

Grasping at straws here ...
:-)

Anon

Without knowing what you're getting in the browsers where it "doesn't work," I would think that the problem lies in your url:
http://www.fcpremier.com/francais/?do=article&id=000254

Some browsers may have a problem with using a query string without specifying a page. I would suggest that you try specifying the filename of whatever the file being accessed is (e.g. http://www.fcpremier.com/francais/default.php?do=article&id=000254).

-Keegan

P.S. Check out Tim Perdue's article on this site entitled something like "Building Dynamic Pages with Search Engines in Mind." I use a somewhat similar system on a site of mine, where there is one page which gets the content from a specified file, and I have managed to avoid having the '?' in my urls. It's a very interesting article, anyway.

Anon

I noticed something interesting. In your main page, the URL is /francais?do=article&id=000254, i.e. with no slash before ?. When you do this, Apache sends you a redirect toward /francais/?do=article&id=000254. Maybe some browsers do not handle this redirect correctly.

Anon

Well, redirection is a serverside thing, not browser (at least as far as I know) so I don't think that could be it.

I've just had to build a different way of getting that content out, since I was not managing to solve the problem. The IP address thing was not the problem, I connected with computers that didn't work in our office while at home, and they still didn't work.

Now it's working (with the workaround) but I'm still baffled. How can one browser get a certain html source file sent to it and another browser get a different html source file sent to it?

Display differences I can understand, but the html source itself was different.

Since the vars being passed to the application were being passed in the URL, and since the browsers that did not work had html that was hardcoded into the function in the middle of the page (just not the article/db data that the function was supposed to call), I can only assume that different browsers somehow differ in how they issue calls to the server in URLs like this:

http://www.fcPremier.com/francais/?do=article?id=000255

Very, very odd.

Thanks for the help you've given, everyone.
-john

Anon

Hmmm ... no dice.

Tried that, and it did not fix the problem. Still baffled, but have found a workaround.

Thanks for the info.
-john

Anon

Thanks for the français help!
-john