Commit Graph

101 Commits

Author SHA1 Message Date
Giovanni Campagna
048a14f1f3 Search: use the same settings object for loading search providers
In recent glib, change notifications don't actually happen unless all
keys have been read, in an effort to reduce unnecessary dbus
traffic for shortlived GSettings object and avoid AddMatch calls.
But we care about changes here, so we need to make sure we're
subscribed, and an easy way to do so is to reuse the same object
to watch for changes and to load the active providers at startup.

https://bugzilla.gnome.org/show_bug.cgi?id=746509
2015-03-27 13:11:22 -07:00
Giovanni Campagna
fd6ef482f0 Search: watch for changes to enabled as well
When introducing support for "default disabled" search providers
this part was overlooked, so enabling a default disabled provider
would be ignored until the next login.

https://bugzilla.gnome.org/show_bug.cgi?id=746509
2015-03-27 13:11:22 -07:00
Giovanni Campagna
8b5a44e119 Search: be resilient against buggy search providers
If a search provider returns a meta without a name, don't crash
constructing the actor.

https://bugzilla.gnome.org/show_bug.cgi?id=745861
2015-03-11 13:47:48 -07:00
Cosimo Cecchi
8345fe006e search: remove redundant check
We check for (metasNeeded.length == 0) at the beginning of the function,
which is only ever called when when a non-zero number of results is
received back from the provider. Effectively, this means that
(metas.length != metasNeeded.length) will also catch (metas.length == 0)
and print a nicer message to the log.
2015-01-07 15:26:21 -08:00
Cosimo Cecchi
67a807e849 search: improve log message when metas length mismatch
Log also the name of the provider and the expected/received metas.
2015-01-07 15:26:21 -08:00
Cosimo Cecchi
cea946e745 search: don't forget to notify caller when failing to ensure actors
The updateSearch() function is called in SearchResults every time new
search hits are available from a search provider; SearchResults will
wait for updateSearch() to complete in a callaback, to update the
overall progress of the search operation.

updateSearch() will call _ensureResultActors(), which will in turn call
getResultMetas() on the search provider, which is an operation that can
fail arbitrarily or return inconsistent data, as it's entirely in the
hands of the search provider.

In case _ensureResultActors() returns a failure, updateSearch() is
currently failing to notify the passed-in callback, which might leave
SearchResults in an inconsistent state: make sure the asynchronous flow
always ends up with a notification to the updateSearch() callback.
2015-01-07 15:26:21 -08:00
Patrick Ward
72b0a3d78f search: Change order of duplicate search check
Checks for a duplicate search before setting the current search
to true and before cancelling the current search. This ensures that
if a duplicate search occurs while the previous search is still active,
the duplicate search will not incorrectly cancel or change the state
of the previous search.
2014-10-01 07:32:06 -06:00
Jasper St. Pierre
e6cd112379 search: Reset the search display when there's no search terms
Destroy the previous search view when there's no search terms, to make
sure old search items don't show up.
2014-09-30 00:55:01 -06:00
Jasper St. Pierre
2cefc8be27 search: Match an empty array, not a null one
When the terms are missing, we pass an empty array, not a null one.
2014-09-30 00:55:01 -06:00
Jasper St. Pierre
991179835e search: Factor out a function 2014-09-30 00:55:01 -06:00
Cosimo Cecchi
778ef365d9 search: hide provider actors when they're created
When the overview is created, search is populated with one actor for
each provider. As they're not hidden though, they will contribute to the
overall size request of the search page, which will shift upwards the
overview grid.

Reviewed-By: Jasper St. Pierre <jstpierre@mecheye.net>
2014-09-29 22:42:25 -07:00
Florian Müllner
aa7b0a285b search: Do not use deprecated GSettings:schema 2014-09-12 23:56:56 +02:00
Jasper St. Pierre
c49f2e1384 search: Only queue an immediate search if we don't have one queued 2014-09-11 18:01:23 -06:00
Jasper St. Pierre
9208473e59 search: Be a bit more careful in _doSearch about timeouts 2014-09-11 17:51:12 -06:00
Jasper St. Pierre
08ece4c186 search: Fix timeout ordering 2014-09-11 17:50:56 -06:00
Jasper St. Pierre
99c29366ee search: Make sure to clear the timeout 2014-09-11 17:40:00 -06:00
Jasper St. Pierre
7a11964dfb search: Fix indentation 2014-09-11 17:14:54 -06:00
Jasper St. Pierre
c4922f6624 search: Move the timeout for searching into setTerms
This lets us considerably clean up the event flow here and change how
things are structured. It also makes sure that we never show "No
Results" -- search.js not being aware of the timeout means that it might
not think that any work was being done when we show the page.
2014-09-11 17:10:26 -06:00
Jasper St. Pierre
4f2070a7c6 search: Track search status more accurately
Keep a flag depending on whether a provider is in-flight, and use that
to determine what status label to show.

This ensures that we only show "No Results" when we're done searching
and we're sure that all providers have returned results back to us.
2014-09-11 17:10:26 -06:00
Jasper St. Pierre
03177dc474 search: Merge SearchSystem with SearchResults
The complexities of tracking these two things separately, with the
display on one half, and the searching on the other half, is difficult
to manage. Squash it all together.
2014-09-11 17:10:26 -06:00
Jasper St. Pierre
b9b4886a6f search: Don't bother clearing the result display twice
It's just extra work, and breaks our hide/show hack to prevent the
flashing on search results.
2014-09-11 17:10:25 -06:00
Florian Müllner
3981b27d8f search: Remove DND code from GridSearchResult
Providers that need drag-and-drop behavior can implement this via the
createResultObject() hook (as the app search provider already does), no
need to duplicate that code in the generic result objects
(ListSearchResult already does not implement DND).

https://bugzilla.gnome.org/show_bug.cgi?id=734726
2014-09-02 00:07:32 +02:00
Carlos Soriano
62786c09a8 appDisplay: Animate appIcon for new window of apps
Following design mockups, animate the icons on AllView, FrequentView,
Dash and Search to zoom out when opening a new window of the app or when
the app is not running and the user execute it.

https://bugzilla.gnome.org/show_bug.cgi?id=734726
2014-09-02 00:07:32 +02:00
Florian Müllner
67c216a6fe search: Allow providers to return the complete result object
This makes the existing createResultObject() match its name better
 and avoids oddities like nested StButtons in app search results.

https://bugzilla.gnome.org/show_bug.cgi?id=734726
2014-09-02 00:07:32 +02:00
Carlos Soriano
1c0c38574d search: Remove unused function 2014-08-20 18:41:18 +02:00
Florian Müllner
62bfde45aa search: Fix default drag actor source
Another unnoticed fallout from commit 792b963bda ...
2014-08-19 20:34:08 +02:00
Florian Müllner
d5aa276c41 search: Fix custom drag actor source
The existing code broke when commit 792b963bda changed the custom
result actor hook to return an object instead of an actor - stop
trying to go through a _delegate to make it work again.
2014-08-19 19:12:02 +02:00
Florian Müllner
d0fe1211f2 search: Fix visibility of "more" icon
Since commit 19749bb37f, the icon is always visible - ooops.
2014-08-19 18:13:40 +02:00
Giovanni Campagna
926177785d search: add support for default disabled search providers
Search providers that should be disabled by default come with
a DefaultDisabled=true key in their keyfile, and are enabled
with the "enabled" whitelist, not with the "disabled" blacklist.

https://bugzilla.gnome.org/show_bug.cgi?id=734110
2014-08-19 15:29:22 +02:00
Carlos Soriano
9970671bb1 Search: Don't use IconGrid width for maxDisplayedResults
Currently to know how many results we could show for GridResults
we use the width of the bin containing those results. Since it's
expanding it shouldn't be a problem. But it becomes a problem when
no results are displayed, thus the container becomes hidden and
it losts its allocation.
In the next introduction of terms in search we call again
maxDisplayedResults but it doesn't have allocation yet, and therefore no
results are displayed (currently a bug on IconGrid makes the min size =
one icon, so actually we show one and only one icon in this case).

To solve that use the parent container which contains the search results
of all providers or the text label with not displayed results, so it
always have the real available width to calculate maxDisplayedResults.

Thanks Alban Browaeys for the debugging footwork.

https://bugzilla.gnome.org/show_bug.cgi?id=732416
2014-07-19 18:12:43 +02:00
Jasper St. Pierre
83cb26d70e js: Adapt to GSettings API change
The 'schema' property has been deprecated for a long time. Even though
this will likely be reverted in glib, let's stop using it.
2014-06-24 15:17:09 -04:00
Jasper St. Pierre
a83b9ed6d7 search: Make sure to destroy old provider displays when we unregister
When we unregistered providers, like when we refreshed the list of
active remote providers, we would forget to destroy the old provider
display after the fact. This left an empty "skeleton" provider display
still in the search results that would never be filled in. Make sure
to destroy it properly.

https://bugzilla.gnome.org/show_bug.cgi?id=728597
2014-04-21 23:36:52 -04:00
Rui Matos
e339e2658d search: Ensure that the default result is visible in the scroll view
The default result is set to selected when key focus enters the search
entry. We must also scroll the view if needed.

https://bugzilla.gnome.org/show_bug.cgi?id=726759
2014-03-26 11:42:32 +01:00
Giovanni Campagna
492558a2d2 search: survive the provider reporting the wrong number of metas
If a provider crashes during GetResultMetas, or returns the wrong
number of metas, report it as failing but don't cause a JS exception.

https://bugzilla.gnome.org/show_bug.cgi?id=725020
2014-03-12 19:26:41 +01:00
Giovanni Campagna
b78e00f372 search: destroy result actors before forgetting about them
We can't let live (ie, never destroyed) actors undergo GC, because
they will emit :destroy signals during finalization and assert/crash
libmozjs. Properly destroy all actors before letting the GC
free them.

https://bugzilla.gnome.org/show_bug.cgi?id=724798
2014-03-12 19:26:41 +01:00
Jasper St. Pierre
842c792868 search: Only do a subsearch if the previous results have returned from DBus
There's a potential race condition in the search code: if we have an
outstanding search call to a provider for search "A", and if before it comes
back we do a subsearch for "AB", we won't have any results to pass along.

Previously, we used an empty list when storing the provider results, so we
effectively told the remote search app to filter through this empty list for
any search results that meet the new query, meaning we showed the user 0
results for the provider in this case.

Now that we don't store an empty list, but instead store `undefined`, this race
raises a warning. Solve it by doing an initial search query in this case
instead.

The search code isn't too smart about chained subsearches: now, if we hit this
race while already on a subsearch, we'll do an initial search for the subsearch
query instead, but that is much better than showing the user nothing. This
could be fixed in the future for a performance improvement.

Reviewed-by: Florian Müllner <fmuellner@gnome.org>
2013-11-04 14:50:45 -05:00
Jasper St. Pierre
4965b1ca7b search: Fix previous commit
It was pushed by accident before it was tested...
2013-11-02 20:11:13 -04:00
Jasper St. Pierre
9cd7ea9371 search: Make the internal search interface callback-based
Long ago, the search system worked in a synchronous manner: providers
were given a query, and results were collected in a single array of
[provider, results] pairs, and then the search display was updated
from that.

We introduced an asynchronous search system when we wanted to potentially
add a Zeitgeist search provider to the Shell in 3.2. For a while, search
providers were either async or sync, which worked by storing a dummy array
in the results, and adding a method for search providers to add results
later.

Later, we removed the search system entirely and ported the remaining
search providers to simply use the API to modify the empty array, but the
remains of the synchronous search system with its silly array still
lingered.

Finally, it's time to modernize. Promises^WCallbacks are the future.

Port the one remaining in-shell search engine (app search) to the new
callback based system, and simplify the remote search system in the
process.
2013-11-02 20:07:06 -04:00
Jasper St. Pierre
dc2468b27b remoteSearch: Do filtering here of providers here
We already do ordering at load time, so why not filtering?
2013-11-02 20:07:06 -04:00
Jasper St. Pierre
252617bd70 search: Remove unnecessary import and full reference
I'm surprised this worked, actually...
2013-11-02 18:01:26 -04:00
Florian Müllner
f12378cf7b search: Hide overview when activating result
This was (accidentally?) dropped by commit 3749b09366.

https://bugzilla.gnome.org/show_bug.cgi?id=711205
2013-10-31 16:19:25 +00:00
Jasper St. Pierre
d401b493a4 search: Fix Return for searching in the overview 2013-10-30 17:43:12 -04:00
Jasper St. Pierre
e630fec63a search: Fix adding items to icon grids 2013-10-30 13:06:56 -04:00
Jasper St. Pierre
af06b78605 searchDisplay: Cache result display actors
When we create a result actor, cache it, so it can be used for
subsearches of the same initial. For now, to keep memory usage
and the stage graph relatively clean, don't persist the actors
across searches, but maybe we should do this in the future.

This also means that we don't query getResultMetas for items
that we've seen in the same initial search.

https://bugzilla.gnome.org/show_bug.cgi?id=704912
2013-10-30 13:01:20 -04:00
Jasper St. Pierre
3749b09366 searchDisplay: Make the search result actors stateless, by removing terms
We want to cache result actors between searches, so we shouldn't
instantiate them with search-specific info.

https://bugzilla.gnome.org/show_bug.cgi?id=704912
2013-10-30 13:01:20 -04:00
Jasper St. Pierre
0590962d36 viewSelector: Move all of the provider-loading logic to SearchSystem
The existing provider system is split between a confusing mess of
RemoteSearch, SearchSystem, SearchDisplay, and ViewSelector, partly
because of the vestigal in-shell search system. Move most of the
logic to search.js so it's easier to read.
2013-10-30 12:59:41 -04:00
Jasper St. Pierre
c0c20d49a5 search: Always fetch the list of search providers
We fetch and store the list of providers from the search system when we
construct SearchResults, but we never update this list when providers are
changed at runtime, causing various bugs making the search not seem as
snappy as it should be. Make sure to always fetch the list of providers
from the search system.
2013-10-30 12:59:31 -04:00
Jasper St. Pierre
dd44219aa5 Merge searchDisplay.js and search.js
search.js used to do a lot more, but now that most of the
functionality has been moved to the remote search system,
it doesn't do a lot. Merge searchDisplay.js into it.
2013-10-29 16:02:32 -04:00
Jasper St. Pierre
8cc1fe007d search: Actually crash when seeing errors from a native search provider
We don't implement many of these, and not catching the error lets us
see stack traces and other such information a lot faster.

https://bugzilla.gnome.org/show_bug.cgi?id=704912
2013-10-29 16:02:32 -04:00
Jasper St. Pierre
4efd363134 search: Ensure that we correctly remove remote search providers
When we reload the remote search providers, we currently try to remove
all remote providers, and then re-scan. It turns out that we sometimes
remove the wrong providers from the remote provider list, causing us to
have some providers not correctly unloaded.

https://bugzilla.gnome.org/show_bug.cgi?id=700283
2013-07-26 17:39:05 -04:00