As I was writing some tags in auto complete, I can't help to notice that each time I type another letter, the suggest list disappears and then show up again. Usually this is not a problem but in the situation below, it can be some what annoying.
1. suppose there are one tag. "aaaaa"
2. I typed "aaa" and "aaaaa" and w/e other suggestion pops up
3. I typed another "a", the suggestion disappears. I typed the "a" because I'm a fast typer so my brain falls behind my hand.
4. Now I have to wait a second or so depend on the server's speed.
A good autocomplete shouldn't let that happen, so in #3, the situation below should happen
I typed another "a", the javascript found one of the suggestion matches what I typed, so it doesn't disappear. It sends a query to the server,. a second later, the server query come back with results and populate the rest of the auto complete.
A small client side script modification can make this module more usable.
I'm happy about the JS cache the result from each query.
maybe it's possible to use the cache on other suggestions, for example, I have
"cba","acb","aaa"
1.I first typed c.
2."cba", "acb" come out
3. now I remove c and put a
4. JS will search though it's list of candidate, and return "cba", "acb", and send a request to server and fetch a new list to populate the rest. so user get a faster response time
also, it might be better for autocomplete to sort the returns with the query string first. Currently, if I type "c", it returns
"acb"
"cba"
because usually humans tend to type the first letter of the word first, auto complete should sort it so it returns
"cba"
"acb"
Some vocabulary should allow synonyms in the auto complete too, sometimes a person might only know the synonym but not the term.
Comments
Comment #1
dpearcefl commentedIs this still a need? If so, it should be tested against D8 and moved there if the issue still occurs.