Minor feedback - Githubissues

joaotavora / snooze

Common Lisp RESTful web development

207 stars 22 forks source link

Minor feedback #2

Closed PuercoPop closed 4 years ago

PuercoPop commented 8 years ago

Hi Joao, after some cursory reading about snooze.

There started-p and resource-p don't follow convention. It is bad enough that CL has the archaic p/-p convention instead of ? to pile on irregularity. single words have no hyphen, otherwise -p at the end. I'm guessing you are aware of the convention but have some objection to it?
There is no patch http is missing
breaking up common.lisp in files instead of sections with delimited ^L would make code more approachable for new comers. This is not an issue yet for snooze but can become an issue as the codebase grows like in swank.

joaotavora commented 8 years ago

I'm guessing you are aware of the convention but have some objection to it?

I am aware of it, but I stick to -p. I do follow the defthingy define-new-thingy convention, cause that doesn't hamper legibility as much.

There is no patch http is missing

patches welcome! pun intended.

breaking up common.lisp in files instead of sections with delimited ^L would make code more approachable for new comers.

Yes. Also remembering why I came up with common.lisp and the snooze-common package might help :-). What was I thinking, what the heck is it "common" with?

joaotavora commented 8 years ago

I am aware of it, but I stick to -p. I do follow the defthingy define-new-thingy convention, cause that doesn't hamper legibility as much.

I just notice that I do follow it in some places like stringp and errorp... so much for coherence. Well let me try to come up with something not totally ad-hoc at all... Let's say I follow it when the (short) word is a CL concept, and not something from my problem domain. So integerp, conditionp but frobnicator-p.

PuercoPop commented 8 years ago

Why go halfway causing confusion on newcomers and oldtimers a like? Why not just use ? for predicates like the scheemers.

Also ensure-uri could be replaced with yuri

mdbergmann commented 4 years ago

It is bad enough that CL has the archaic p/-p convention instead of ? to pile on irregularity.

What a nonsense. p/-p is Common Lisp. ? is Scheme.

joaotavora commented 4 years ago

Since I had this discussion, I was pointed to this relevant section of CLtL2:

https://www.cs.cmu.edu/Groups/AI/html/cltl/clm/node69.html

joaotavora commented 4 years ago

Also, I think this issue can be closed. But discussion can continue, of course.

mdbergmann commented 4 years ago

Sure, close it.