Allow duplicate and key-only query-string parameters about htty HOT 8 CLOSED

My patch in #56 resolves the first issue (every key no longer needs a value). While replacing the existing entry is not always what you want, it is a more common use case, so I would propose:

• query-set continues to replace if the specified name already exists
• query-add always adds a new element

This is much more consistent with the common meanings of 'set' and 'add' in most programatic environments. As an aside, with the internal refactoring of query string management included in the patch, implementing query-add is pretty trivial.

What is the difference between query-unset NAME and query-remove NAME?

from htty.

query-remove NAME removes the last value keyed by NAME, whereas query-unset NAME would remove all values keyed by NAME.

from htty.

Ah, makes sense. May I propose query-remove NAME [VALUE] where if value is not specified it does as you suggest, but if value is specified it removes that specific instance?

from htty.

Excellent idea. I’ll apply it to ‘cookie-remove’, too.

from htty.

The patch in #63 should close this issue.

from htty.

query-add, query-remove, and query-set are all working as desired now.

I like the enhancement you made to the concept of query-remove that looks for matches by both name and value. It would be good if query-unset worked this way, too:

• query-add NAME [VALUE [NAME [VALUE ...]]]
• query-remove NAME [VALUE]
• query-set NAME [VALUE [NAME [VALUE ...]]]
• query-unset NAME [VALUE]
• query-unset-all

from htty.

So something like query-unset foo would remove all instances of foo in the query string regardless of value and query-unset foo bar would only remove instances that were foo=bar ?

from htty.

That’s what I was thinking, yes.

from htty.