How would this work with multi? A multi has a built in share, so presumably one 
could stop an easy in a multi if that share had an appropriate locking setup. 
But what would be the expected behaviour then? Should it be (1) a stop call 
causes application code to regain control from curl_multi_perform promptly, but 
without damaging any of the other easies, or (2) the stop call aborts one easy, 
and only aborting *all* the easies guarantees control returned to the 
application promptly? Or something else entirely?

If the expected behaviour is (1), it seems that curl_multi_stop should be a 
thing, and then really it ought to be possible to stop a curl_multi_perform 
without damaging *any* of its easies, and perhaps then resume it (and all of 
its easies) later.
-- 
Christopher Head

-------------------------------------------------------------------
Unsubscribe: https://cool.haxx.se/list/listinfo/curl-library
Etiquette:   https://curl.haxx.se/mail/etiquette.html

Reply via email to