Development Discussion Board

yes, integrating with other packages. The first hook is not as important as the last hook, but both are nice to have. For the early hook, it is nice to be able to set defines to override error reporting settings, etc. as they might not be compatible with other packages. If the hook system is DB driv...

I use Amazon Cloudfront Content Distribution Network to host cacheable assets such as images, scripts, etc -- it helps with site performance. Uploading the assets (or using origin push, depending on the CDN), can be left to the user (might be a 'nice to have' in v4), but it should be easy to overrid...

Yep, I know that -- that's not going to cut it for me though. "contrib" hardly reflects what will be over 50% of the downloaded package. Even just renaming it to "non-phpbb/" would be better, but "contrib/" is not a good name -- kindof the entire point of the request.

1. An ability to package the "MODX packagage" as a single installation step in a larger process. The current packaging guidelines are too strict to allow for this. For example, step 1 could be "install software A" (or even "install phpBB"), then step 2 would be to depl...

I agree on getting rid of all the redundant routes through to essentially the same thing. In addition, I have plenty of users who also overlook the administrator/global moderator/user permissions drop-down at the top of the screen on the assign permissions page. Also you have a naming collision (in ...

Excellent idea. Many mods use the same areas and template edits.

Some off the top of my head I would find useful in overall_header and overall_footer:

VERY_TOP
META
EARLY_SCRIPTS
HEAD_END
BODY_BEGIN
EXTRA_NAVIGATION
BEFORE_COPYRIGHT
AFTER_COPYRIGHT
BODY_END

I agree that UMIL should be in the core. I don't think it is very natural for users installing MODs to have to potentially overwrite the UMIL directory with each MOD -- it is not consistent with any other file copy action they will have to perform. Just another layer of complexity/opacity for the en...

The first problem I have with the current hook system is that whether the core function is subsequently executed or not seems to be controlled by the return value of the hooked-in user function. i.e. returning an empty string vs returning null seem to do different things (it's been a while since I w...

Hi, I posted a fairly detailed response on the forum in the initial Q&A thread -- upshot was basically that manual code edits should go the way of the dinosaurs and as such MODX wasn't so important in the grand scheme of things. (energy better spent on new hooks and filters). However, in the int...

I see nothng stupid in starting this discussion.. the original point is well-made and sensible. Unlike the "they'll do what they want to do" non-responses.