[Webkit-unassigned] [Bug 93166] Web Inspector: Select the shortest match in the FilteredItemSelectionDialog

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=93166





--- Comment #20 from johnjbarton <johnjbarton at chromium.org>  2012-09-24 13:55:38 PST ---
(In reply to comment #19)
> 1. I think selecting a match other than the first one is misleading.

The order of the items in the match list has no significance. The first one is not special in any way. Thus selecting the first one is effectively random.

> 2. I don't think that the shortest match is necessarily the best one.

The shortest match is not always the one the user seeks, but it is much more likely to be the one the user seeks than any other item and in particular it is more likely than the random first item.

The shortest match algorithm allows users to filter the list by words they recall. For example "script" will bring up all files related to "script". This allows users to use partial knowledge to get a browse-able list they can refine or click on.

> 3. This patch adds some complexity to the code which will be redundant once we change the regular expression to match from the beginning of the string (I consider this as the main reason of false positives).

Unfortunately Bug 97332 is not a good direction. The entries in the item list are *not* false positives. They are potential selections filtered by the bits of knowledge the user recalls. 

Compare the Chrome Omnibox to the location bar from olden days. In the olden days we had to type "http://blah.blah.etc" to navigate. Now we type things we recall and get a completion dialog prompting our memory. Bug 97322 takes from the future into the past.

I hope you will reconsider.

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.