Why so complicated? extGeshi will stay canonically lowercase and $ext will be lower-cased?

Off topic: I wonder then this will return true:

in_array($filename, $extensions)

The code fror content type detection does not probe for the entire filename.

Line in question:

./filedetails.php:$extn = strtolower(strrchr($path, '.'));

Off topic: I wonder then this will return true:

in_array($filename, $extensions)

Hm... Just a theory: MAKE files don't contain a classical filename extension. But they can be identifiered using their name "Makefile", so I guess in this case, we would need to add "Makefile" to the GeShi extensions.

Why so complicated? extGeshi will stay canonically lowercase and $ext will be lower-cased?

Yes, that is indeed easier. But personally I feel safer when both sides are lower-cased before comparing them. Back to my example, it's possible that people add "Makefile" to extGeshi, and they might be very frustrated because it doesn't work anymore (or is WebSvn converting extGeshi to all-lower-case somewhere in the program flow?).

in regards filedetails.php it looks like $extn is forced lower-case and then used at two instances:

• in_array($extn, $zipped)
• $contentType[$extn]

In this case I also want to become independent of case-sensitivety, so I would prefer to change:

• Don't force $extn to be lower-case
• in_array($extn, $zipped) => i would also like to compare case-insensitively
• $contentType[$extn] => search for the (first) element in the array, indepdenent of the case

What do you think?

Off topic: I wonder then this will return true:

in_array($filename, $extensions)

Hm... Just a theory: MAKE files don't contain a classical filename extension. But they can be identifiered using their name "Makefile", so I guess in this case, we would need to add "Makefile" to the GeShi extensions.

Why so complicated? extGeshi will stay canonically lowercase and $ext will be lower-cased?

Yes, that is indeed easier. But personally I feel safer when both sides are lower-cased before comparing them. Back to my example, it's possible that people add "Makefile" to extGeshi, and they might be very frustrated because it doesn't work anymore (or is WebSvn converting extGeshi to all-lower-case somewhere in the program flow?).

I agree on Makefiles, I have just tried. There is no mapping and it works exactly that way. I am revoking my comment for the entire file. More than that, I would not lowercase the whole filename because of cases like Makefile and CMakeLists.txt.

in regards filedetails.php it looks like $extn is forced lower-case and then used at two instances:

* in_array($extn, $zipped)

* $contentType[$extn]

In this case I also want to become independent of case-sensitivety, so I would prefer to change:

* Don't force `$extn` to be lower-case

* in_array($extn, $zipped) => i would also like to compare case-insensitively

* $contentType[$extn] => search for the (first) element in the array, indepdenent of the case

What do you think?

Hmm, lowercasing the entire filename looks wrong to me. Makefile != makefile in terms of any make implementation. I wouldn't touch it.

So for the same of simplicity I still prefer:

if (in_array($filename, $extensions) || in_array(strtolower($ext), $extensions)) {

Ok, I have applied the changes. But I think we really need to add a hint to the documentation that $contentType,
$zipped, and $extGeshi must only contain lower-case entries.

Ok, I have applied the changes. But I think we really need to add a hint to the documentation that $contentType, $zipped, and $extGeshi must only contain lower-case entries.

I agree, please make a separate PR for the doc change.