| Commit message (Collapse) | Author | Age | Files | Lines |
| |
|
| |
|
| |
|
|
|
|
|
|
|
| |
This removes the old style extension support, you can read the
rationale in the bug report.
https://bugzilla.gnome.org/show_bug.cgi?id=685630
|
|
|
|
|
|
|
| |
org.gnome.epiphany.web.enable-adblock, controls whether adblock is
enabled (globally).
https://bugzilla.gnome.org/show_bug.cgi?id=681657
|
|
|
|
|
|
|
|
| |
ephy-profile-migrator
Instead, just append the new variable to the existing environment.
https://bugzilla.gnome.org/show_bug.cgi?id=683538
|
|
|
|
| |
https://bugzilla.gnome.org/show_bug.cgi?id=683929
|
|
|
|
|
|
| |
To convert any possible directory separator into an underscore.
https://bugzilla.gnome.org/show_bug.cgi?id=683711
|
|
|
|
|
|
|
|
|
|
| |
Since both the frecent store and the history window don't need to be
updated immediately after a url is visited, we can use ::urls-visited
instead. The advantage of this is that we reduce considerably the load
when updating both models when loading many pages at the time,
specially during startup.
https://bugzilla.gnome.org/show_bug.cgi?id=683550
|
|
|
|
|
|
|
|
| |
This signal is emitted in bulks and is intended for clients that are
not interested in every single URL that is visited but only to know
that visits have happened.
https://bugzilla.gnome.org/show_bug.cgi?id=683550
|
| |
|
|
|
|
| |
The default dot dir is now in ~/.config, not ~/.gnome2.
|
| |
|
|
|
|
|
|
|
|
| |
ephy-profile-migrator"
This reverts commit adfecc2bb03ed280543dc5ad4ba6fa350ff00d8c.
This breaks the migrator tests.
|
|
|
|
|
|
|
|
| |
ephy-profile-migrator
Instead, just append the new variable to the existing environment.
https://bugzilla.gnome.org/show_bug.cgi?id=683538
|
|
|
|
|
|
|
|
|
|
| |
All code paths use int, and this breaks on architechtures where
long is not the same as int (such as x86_64).
https://bugzilla.gnome.org/show_bug.cgi?id=683029
[Do the same for the visit time column]
Signed-off-by: Claudio Saavedra <csaavedra@igalia.com>
|
|
|
|
|
|
|
| |
If there is an error building a statement, the returned value is
always NULL. Calling g_object_unref() on them will lead to trouble.
https://bugzilla.gnome.org/show_bug.cgi?id=683475
|
|
|
|
|
|
|
| |
If there is an error processing the statement, then unref it before
returning.
https://bugzilla.gnome.org/show_bug.cgi?id=683475
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
ephy_snapshot_service_get_snapshot_async() receives an option web view
parameter, that it's only used in case the snapshot is not the in the
thumbnails cache. We can split the method into
ephy_snapshot_service_get_snapshot_async() to get a snapshot from a web
view and ephy_snapshot_service_get_snapshot_for_url_async() to get a
snapshot from the cache. The former uses the latter to try first if the
web view URI is in the cache.
Patch includes other cleanups and fixes:
- Add missing ephy_snapshot_service_save_snapshot_finish()
- Add EphySnapshotServiceError to handle errors
- Use GSimpleAsyncResult API instead of using GIOScheduler directly
- Use different async data structs for every async operation
https://bugzilla.gnome.org/show_bug.cgi?id=683327
|
|
|
|
|
| |
We need this to actually activate the current item, in order to
support ctrl+click to open in a new tab.
|
|
|
|
|
|
|
|
|
|
| |
Needs a fairly recent gnome-common. To use do:
- Run configure with --enable-code-coverage
- Type 'make check-code-coverage'
- Open the HTML results
https://bugzilla.gnome.org/show_bug.cgi?id=683297
|
|
|
|
| |
recursive delete function is leaking a reference.
|
|
|
|
|
|
| |
To do this, add the render policy as a column to the model and use it
as a cellrenderer attribute. The default value is already PRELIT. When
removing the cell, flip it to NEVER.
|
|
|
|
|
| |
This way the default item is PRELIT. This will be useful when using
this type as a cell renderer attribute later on.
|
|
|
|
|
| |
It makes no sense to emit the delete signal if the delete button is
not rendered at all.
|
|
|
|
|
|
|
|
|
| |
ephy_profile_utils_do_migration()
This is passed further on to the migrator process. Right now this is
unused.
https://bugzilla.gnome.org/show_bug.cgi?id=681679
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
directories
This is necessary since we add new columns to the history database in
step 8 of migration. However, non-default profile directories don't
have a .migrated file right now but, since up to step 7 it was
possible to use these profiles without migration, it is safe to assume
that a non-default profile without a .migrated file can be migrated
starting from step 8.
https://bugzilla.gnome.org/show_bug.cgi?id=681679
|
|
|
|
|
|
|
| |
This method will let us know when the current dot directory is the
default one and not a user-specified one nor a web application one.
https://bugzilla.gnome.org/show_bug.cgi?id=681679
|
|
|
|
| |
This allows us to extend this further cleanly.
|
| |
|
|
|
|
|
| |
Since this is also used for persistency of temporary directories, but
also for --profile.
|
| |
|
| |
|
|
|
|
|
|
|
|
| |
Fixes a crash when epy is started with the bookmarks window open. The old
code was trying to set the favicon in an already released GValue. Instead of
doing that, we now force a repaint of the row once we get the favicon.
https://bugzilla.gnome.org/show_bug.cgi?id=673795
|
|
|
|
|
|
|
|
|
|
|
| |
The EphyNodeView is based on the following stack of models:
Sort model -> Filter model -> EphyTreeModelNode
This function returns a GtkTreeIter pointing to a row in the sort model from
a EphyNode that belongs to the EphyTreeModelNode.
https://bugzilla.gnome.org/show_bug.cgi?id=673795
|
| |
|
|
|
|
| |
This is needed for distcheck, which runs tests before installing.
|
|
|
|
|
|
|
| |
This should help catch the mistake of not installing the new migrator
when the profile version changes.
https://bugzilla.gnome.org/show_bug.cgi?id=683134
|
|
|
|
| |
This will allow us to catch mismatches between browser and migrator binary.
|
|
|
|
|
| |
Otherwise they look empty until the icon is loaded. This is specially
annoying when opening the overview and all the icons are scheduled for load.
|
|
|
|
|
|
|
|
| |
We were naively emitting a few signals from the history service thread
while doing this is not thread safe. Add a few helper methods to run
this in a main loop idle.
https://bugzilla.gnome.org/show_bug.cgi?id=683040
|
|
|
|
|
|
| |
Add a ephy_overview_store_animated_remove() that shrinks the thumbnail
in a timeout until it's small enough and then removes it from the
model.
|
| |
|
| |
|
|
|
|
|
|
|
| |
Instead of using a hardcoded cairo frame, add a property to define the
frame around thumbnails. Also rework the code around the
default-thumbnail property to make it possible to use this there as
well.
|
| |
|
|
|
|
| |
position of the close x
|
|
|
|
| |
Squared snapshots are not that great, so use a 4:3 aspect ratio.
|
|
|
|
| |
This way there's no flashing of thumbnails between the query and the reply.
|
|
|
|
| |
Just to spare some code repetition.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
is needed
Currently, we were not retrieving thumbnails from the service if they
were outdated. This would cause some pages in the overview not to
display a thumbnail at all even if one is available (but old).
Fix this by always retrieving a thumbnail but storing its mtime in the
model, and making ephy_snapshot_store_needs_snapshot() check the age
of snapshots when deciding whether a snapshot update is needed or not.
|
|
|
|
| |
update is needed
|
| |
|
| |
|
|
|
|
|
|
|
| |
This subclass of EphyOverviewStore displays the most frecently visited
pages.
https://bugzilla.gnome.org/show_bug.cgi?id=455173
|
| |
|
| |
|
|
|
|
|
|
| |
These rows store information relevant to the overview: the update time
of the url's thumbnail and wheter the url should be hidden from the
overview's frecent view.
|
| |
|
| |
|
| |
|
|
|
|
| |
This adds a couple of new required columns to the urls table.
|
|
|
|
|
|
|
| |
This should be subclassed by the models for the history and/or
currently open views.
https://bugzilla.gnome.org/show_bug.cgi?id=455173
|
|
|
|
|
|
|
|
|
|
|
| |
deletions
GdMainView connects to GdMainViewGeneric:delete-item-clicked and emits
its :item-deleted signal which, if unhandled, simply removes the item
in question from the underlying model.
Users of GdMainView can handle this signal and stop the default
handler from being invoked, thus avoiding removal of the item.
|
|
|
|
|
|
| |
Also, emit the GdMainViewGeneric:delete-item-clicked when the button
is clicked. The signal is still unhandled, so clicking in the button
will not make any difference yet.
|
|
|
|
|
|
| |
Implementors of this interface may connect emit this signal to notify
the GdMainView when a user is requesting to delete one item from the
view.
|
|
|
|
|
| |
This is a renderer that draws a "close" button on top of its contents
and that emits a signal when the button is activated.
|
|
|
|
|
|
| |
This widget courtesy of gnome-documents
https://bugzilla.gnome.org/show_bug.cgi?id=455173
|
|
|
|
|
|
|
| |
This service provides pixbufs for URLs while caching these locally as
thumbnails. gnome-desktop-thumbnail is used to handle caching.
https://bugzilla.gnome.org/show_bug.cgi?id=668578
|
| |
|
|
|
|
|
| |
We were only emitting it if we happened to have a callback method
associated with the history message, that does not make sense.
|
| |
|
|
|
|
|
|
|
|
| |
If there's no profile dir there should not be anything to do for the
profile migrator (since in theory epiphany has never been
executed). This might backfire for people that manually mess with
their profile dir (removing it), but those people hopefully know what
they are doing.
|
|
|
|
|
| |
Otherwise we'll try to migrate old data in new ephy installs, which
does not make sense.
|
|
|
|
|
|
|
|
| |
Let's make EphyWindow the one in charge of deciding whether a URI is
actually shown or not in the location entry. This allows to remove
some code to that effect in EphyLocationController (and perhaps some
more in EphyWebView in the future), and makes this feature more
extensible for the future.
|
| |
|
|
|
|
| |
And add unit tests for it.
|
|
|
|
|
|
|
|
| |
Use webkit_web_view_save() API for HTML pages that saves the web page
into a MHTML file. For any other MIME types supported by the web view,
save the main resource data to a file.
https://bugzilla.gnome.org/show_bug.cgi?id=679368
|
|
|
|
| |
https://bugzilla.gnome.org/show_bug.cgi?id=679764
|
|
|
|
|
|
|
| |
If libgcr is available it's used to show the information about the TLS
certificate in the dialog too.
https://bugzilla.gnome.org/show_bug.cgi?id=681506
|
| |
|
|
|
|
|
|
| |
WebAudio support in WebKitGTK is not stable yet but it would be good
to allow Epiphany users to test this feature nonetheless. The
enable-webaudio gsetting is set to false by default.
|
|
|
|
|
|
|
| |
execute_quit() is called already when the QUIT message is received,
there is no need to call it after the thread loop quits.
https://bugzilla.gnome.org/show_bug.cgi?id=680529
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
GTK+ has changed the UX of the GtkFileChooser. This has (pragmatically)
deprecated "last save dir" stored by applications themselves.
For Epiphany this means we no longer need a "persist-key" in our file
chooser, and that we have to clean up some GtkFileChooser API use in
followup patches.
This commit removes the property and uses of it.
Bug #655508
|
|
|
|
|
|
| |
to avoid hard coding the package name.
https://bugzilla.gnome.org/show_bug.cgi?id=672023
|
|
|
|
|
|
| |
All the properties are write-construct-only.
https://bugzilla.gnome.org/show_bug.cgi?id=671595
|
|
|
|
|
|
|
| |
This is a hack. One that we should remove when we decide what to do with
extensions.
https://bugzilla.gnome.org/show_bug.cgi?id=680907
|
|
|
|
| |
https://bugzilla.gnome.org/show_bug.cgi?id=679369
|
| |
|
|
|
|
|
|
|
| |
This fixes the cases where sites like www.foo.com save cookies for
domain .foo.com, so that when an app is created for www.foo.com, cookies
in the current jar for the domain .foo.com are not copied to the app
cookie jar.
|
|
|
|
| |
https://bugzilla.gnome.org/show_bug.cgi?id=678612
|
|
|
|
|
| |
Leaving in ephy-request-about only the code specific to the soup feature
implementation.
|
|
|
|
| |
function
|
|
|
|
|
| |
It makes easier to check whether download finished, and it's compatible
with WebKit2.
|
|
|
|
| |
It builds and basic functionality works.
|
|
|
|
| |
https://bugzilla.gnome.org/show_bug.cgi?id=678405
|
| |
|
|
|
|
| |
https://bugzilla.gnome.org/show_bug.cgi?id=673271
|
|
|
|
|
|
|
| |
It creates a data URI for the given filename. Use the new function when
building error and applications pages.
https://bugzilla.gnome.org/show_bug.cgi?id=677025
|
| |
|
| |
|
| |
|
|
|
|
| |
https://bugzilla.gnome.org/show_bug.cgi?id=677720
|
|
|
|
|
|
| |
So that they are bindable.
https://bugzilla.gnome.org/show_bug.cgi?id=675804
|
|
|
|
| |
https://bugzilla.gnome.org/show_bug.cgi?id=673270
|
|
|
|
|
|
|
| |
Better explain the logic of the function and reorder the conditions.
This makes ~/Downloads the fallback instead of ~/Desktop.
https://bugzilla.gnome.org/show_bug.cgi?id=673337
|
|
|
|
|
|
|
| |
Unsigned integer can not be negative, therefore it is redundant to
check whether it is greater than or equal to zero.
https://bugzilla.gnome.org/show_bug.cgi?id=675888
|
|
|
|
|
|
|
| |
If the org.gnome.Epiphany.web.do-not-track setting is enabled, we'll
send the DNT: 1 header with every outbound request we make.
See http://donottrack.us/ for more details.
|
|
|
|
| |
That way we can run the tests from any directory in debug mode.
|
|
|
|
|
|
| |
The path to the just built migrator is known at compile time, so no
need to create strings at runtime. Also, share the name of the binary
itself to avoid silly bugs like last time.
|
|
|
|
| |
Mistakenly changed it to use underscores.
|
| |
|
| |
|
|
|
|
| |
Use --do-step/-d and an index.
|
|
|
|
| |
So we can properly test it.
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
| |
We don't use XML files to create our toolbars anymore, so this is not
needed.
|
|
|
|
|
|
|
|
|
|
| |
We need to update their desktop files (they had references to the
profile directory) and the symlink of the .desktop file in the Shell
applications directory.
It would be much easier to just delete and re-add the applications,
but unfortunately that would wipe out the existing profile data in the
apps (like cookies).
|
|
|
|
|
| |
It's quite cumbersome to re-construct from its name and URL, so just
add it here. We'll use it to migrate the profile data.
|
|
|
|
|
|
| |
We are going to use it in the profile migrator, so it needs to be
there. Besides, this code just deals with plain data in the profile
and application dir, so it makes sense for it to be here.
|
|
|
|
| |
Do nothing if the destination history file already exists.
|
|
|
|
|
| |
Otherwise we'll read the .migrated file *before* we move the old
profile. Not OK, it will make us start from scratch all the migration
|
| |
|
|
|
|
|
|
|
| |
Sqlite limits the length of a LIKE pattern to 50000 bytes, therefore
we need to make sure that longer strings are not used as queries.
https://bugzilla.gnome.org/show_bug.cgi?id=674848
|
|
|
|
|
|
| |
Migrate profile directory to XDG config dir
https://bugzilla.gnome.org/show_bug.cgi?id=522810
|
|
|
|
|
|
|
|
|
|
| |
Since we are about to migrate our profile dir, allow file helpers init
to not ensure the profile dir exists (it was hardcoded until now). For
this we get rid of the ugly boolean parameters and add a flags
parameter, which preserves the old behaviors and allows for this new
option.
We update all the callers in the tree.
|
|
|
|
|
| |
Call it directly from main, since we want it to happen before the file
helpers initialization.
|
|
|
|
| |
https://bugzilla.gnome.org/show_bug.cgi?id=674870
|
|
|
|
| |
https://bugzilla.gnome.org/show_bug.cgi?id=673337
|
|
|
|
| |
https://bugzilla.gnome.org/show_bug.cgi?id=673337
|
|
|
|
| |
We'll use this when we land the Overview.
|
|
|
|
|
|
| |
We need this to be an enum, since we'll a third option in the future
for the Overview (to never show the tabs bar). For now just add the
two values we have now and migrate the code and the user data.
|
| |
|
|
|
|
|
|
|
|
|
| |
With this policy the session will only be restored if the application
has exited unexpectedly, but not if the user manually closes it.
There are no code changes needed for this to work, having a (valid)
different value than always/never in the setting makes things just
work.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
We add a new gsettings key, 'restore-session-policy', with two valid
values: 'always' and 'never'. A brief explanation of our session state
mechanism follows.
There are three ways to exit Epiphany:
1) Activate 'Quit' in the application menu
2) Close the last application window
3) Kill the process manually, SIGSEGV, or other similar unexpected
event.
For 1) and 2), we'll now do the same thing:
a) Call ephy_session_close
b) Exit the application manually
ephy_session_close will check the new restore-session-policy setting,
and only save the session state if it's set to 'always'. Before it
used to manually destroy all present windows. We now let EphyShell or
EphyWindow do this, EphySession only manages the session state saving.
For 3), the process will die with the state saved up to that point,
there's nothing we can do. For that reason, on startup also check the
new setting; if it's set to 'never' ignore the session state, open a
window in the homepage, and delete the old state file.
https://bugzilla.gnome.org/show_bug.cgi?id=673453
|
|
|
|
|
|
|
| |
The GFile API provides useful error messages, print them when any
operation fails to aid debugging.
https://bugzilla.gnome.org/show_bug.cgi?id=673666
|
|
|
|
| |
https://bugzilla.gnome.org/show_bug.cgi?id=673665
|
|
|
|
|
|
|
|
|
|
| |
Invalid enumerators can be returned even when no GError is set.
Check if the enumerator is non-NULL before proceeding, and adjust the
default return value so it is not TRUE when g_file_enumerate_children
fails.
https://bugzilla.gnome.org/show_bug.cgi?id=673337
|
|
|
|
|
|
| |
It's a two-line save, and there are no users of this API.
https://bugzilla.gnome.org/show_bug.cgi?id=673337
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
Align the elements of the GtkEntryCompletion popup with those in the
location entry. The code comes with a detailed comment and a scheme of
how the aligment is done now.
Because of the unhandled pixels of GtkEntryCompletion, this code might
need an update if anything in GTK+ or Adwaita changes.
https://bugzilla.gnome.org/show_bug.cgi?id=672927
|
|
|
|
| |
https://bugzilla.gnome.org/show_bug.cgi?id=672927
|
|
|
|
| |
https://bugzilla.gnome.org/show_bug.cgi?id=672927
|
|
|
|
|
|
|
| |
Makes the symbolic-icon of the completion rows change its color
accordingly to the selection state. With Adwaita: black -> white.
https://bugzilla.gnome.org/show_bug.cgi?id=672927
|
|
|
|
| |
Since we don't have one anymore.
|
|
|
|
| |
Since we don't have one anymore.
|
|
|
|
| |
https://bugzilla.gnome.org/show_bug.cgi?id=672926
|
|
|
|
| |
https://bugzilla.gnome.org/show_bug.cgi?id=672926
|
|
|
|
| |
The brave new world is to use the session support in GtkApplication.
|
|
|
|
| |
https://bugzilla.gnome.org/show_bug.cgi?id=672907
|
|
|
|
| |
https://bugzilla.gnome.org/show_bug.cgi?id=672525
|
|
|
|
|
|
|
| |
Link the glow and the menu buttons together by using a "linked" style
class on the box containing them.
https://bugzilla.gnome.org/show_bug.cgi?id=672712
|
|
|
|
|
|
|
| |
If the old history file exists but has no actual visits we'll enter
into an infinite loop.
https://bugzilla.gnome.org/show_bug.cgi?id=672547
|
| |
|
|
|
|
|
|
|
|
| |
Use the new webkit favicon database for favicons here. This
might not work perfectly, but that's a webkitgtk issue, see
https://bugs.webkit.org/show_bug.cgi?id=81665
https://bugzilla.gnome.org/show_bug.cgi?id=672480
|
|
|
|
|
|
|
|
|
|
| |
Most data files are listed through EXTRA_DIST, so they are left in the
top srcdir during VPATH builds. The new EphyWebView unit test needs
data/pages/error.html, so distcheck was broken because ephy_file
("error.html") was failing since it looked for them in $pwd/data.
To fix this, look for files in the top srcdir instead of $pwd/data,
which will lead to the builddir data data dir during distcheck.
|
|
|
|
|
|
|
|
| |
We don't want the floating bar to extend without boundaries in the
available size, since some hyperlinks can be very long (and make the bar
cover the whole screen).
https://bugzilla.gnome.org/show_bug.cgi?id=671520
|
|
|
|
|
|
|
| |
Instead of hardcoding all visits as 'TYPED' properly distinguish
between bookmarks, following links and typing URIs in the entry.
We'll use this to compute the frecency of history items.
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
The main purpose of the cancellable API in ephy-history-service
is to let the user notify when the results of the operation
are no longer needed and the callback call can be omitted. Since
performing a read operation, in such cases, makes no sense, we
cancel them altogether. However, given the nature of the service,
we don't cancel write operations.
https://bugzilla.gnome.org/show_bug.cgi?id=671901
|
|
|
|
| |
https://bugzilla.gnome.org/show_bug.cgi?id=671635
|
|
|
|
|
| |
Signed-off-by: Priit Laes <plaes@plaes.org>
Reviewed-by: Xan Lopez <xan@igalia.com>
|
| |
|
| |
|
|
|
|
|
| |
By now, the public API only supports filtering by time range, but
filtering by string matching is also possible.
|
|
|
|
| |
If there are no hosts, that's not a failure, so return TRUE instead.
|
|
|
|
|
|
| |
If no host can be selected, always select the "all hosts" row.
https://bugzilla.gnome.org/show_bug.cgi?id=671575
|
|
|
|
|
|
|
| |
This adds always a default item "All sites" which, when selected, will
show all urls in all hosts.
https://bugzilla.gnome.org/show_bug.cgi?id=671575
|
|
|
|
| |
https://bugzilla.gnome.org/show_bug.cgi?id=671594
|
|
|
|
|
| |
This breaks selection in _BROWSE mode, and we don't really need it
anyway.
|
|
|
|
| |
https://bugzilla.gnome.org/show_bug.cgi?id=671559
|
|
|
|
|
|
|
| |
After removing all the urls in a host, we don't need that host
anymore.
https://bugzilla.gnome.org/show_bug.cgi?id=671559
|
| |
|
|
|
|
| |
Simply bail out and don't crash the browser.
|
|
|
|
|
| |
This puts "Others" to the beginning of the list and the rest
alphanumerically sorted.
|
|
|
|
|
| |
So that it doesn't trigger a double-click in the first press. Use
a handler instead of connecting to the signal.
|
| |
|
|
|
|
|
| |
EVER needs to be at the end, not in the second position, since that's
the order in the ephy history window.
|
|
|
|
|
| |
Since that's their natural place. Also create a base class to hold the
common code they share.
|
|
|
|
| |
So that AFTER handlers are executed as well
|
|
|
|
| |
This method will remove all the history related to that host.
|
| |
|
|
|
|
| |
To retrieve all hosts in the history
|
| |
|
| |
|
| |
|
|
|
|
|
|
| |
Otherwise the default code path is enough. Also, the host is always
going to be NULL for newly created URLs, so double check this is the
case and don't try to handle anything else.
|
| |
|
|
|
|
|
|
| |
We need to consider the case of an "invalid" host with a non default
zoom level, since the profile migrator will send us that kind of
thing.
|
|
|
|
| |
And use it in EphyEmbed instead of the legacy signal.
|
| |
|
|
|
|
|
|
|
|
|
| |
Similar to the old EphyHistory 'add-page' signal. It allows blocking
the visit by handling the signal, since the actual logic to add the
visit is in the class' default handler.
The add_visit(s) methods won't emit this signal, but for now we'll
assume that whoever is using those knows what he's doing.
|
|
|
|
| |
Fits much better with the naming model in that class.
|
|
|
|
|
|
| |
It was a very thin wrapper on top of the history service, it does not
seem to make much sense. For now move the two useful helper methods
down to the service without further changes.
|
|
|
|
|
| |
This is what the old history code did. Looks better than an empty
string.
|
| |
|
|
|
|
| |
It's not really that useful.
|
|
|
|
| |
It's not used at all. Again, format change.
|
|
|
|
|
|
|
|
|
|
| |
Zoom levels have always been stored per host in Epiphany, not per URL
(otherwise you'd have to re-apply the zoom level again and again when
visiting, say, every news entry in your favorite newspaper). Change
things to work like that.
Note: this changes the SQL table format for the history, so you'll
need to re-migrate your history.
|
|
|
|
|
| |
We now fetch the history from the SQL backend using a service thread,
and merge the results with our old bookmark data.
|
|
|
|
|
|
|
|
|
|
| |
When we retrieve a url from the backend from a user given
EphyHistoryURL, it's possible for this to contain updated information
for the page. In such case, don't overwrite this with the one stored
in the backend.
This makes possible to later update this information when saving a new
visit.
|
|
|
|
|
| |
If there is no visit time range there is no need to perform a inner
join between these two tables, so spare the hassle.
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
|
|
| |
Xan, talk to me after class.
|
| |
|
| |
|
| |
|
|
|
|
| |
They were leftovers from old APIs
|
|
|
|
|
|
|
|
|
| |
EphyHistoryService provides a high-level API to store history
information. It will processed by a worker thread using SQLite to
provide a fast, responsive service to the main UI.
Based on the code by Martin Robinson (mrobinson@igalia.com) and
Claudio Saavedra (csaavedra@igalia.com).
|
|
|
|
|
|
|
| |
We'll use them to implement the new history/bookmarks storage backend.
Code by Martin Robinson (mrobinson@igalia.com) and Claudio Saavedra
(csaavedra@igalia.com)
|
|
|
|
|
|
| |
Free GFiles created during argv parsing.
https://bugzilla.gnome.org/show_bug.cgi?id=671095
|
|
|
|
| |
We can just use it as SOUP_URI_SCHEME_HTTPS is an intern static string.
|
|
|
|
|
|
|
|
| |
It was actually more than a warning. It was preventing epy to recover
passwords previously stored in the keyring, as all of them were stored using
"/" as the path of the URI.
https://bugzilla.gnome.org/show_bug.cgi?id=670431
|
|
|
|
|
|
| |
Instead of gtk+ stock icon, use the symbolic version. Respects LTR/RTL.
https://bugzilla.gnome.org/show_bug.cgi?id=669131
|
|
|
|
|
|
| |
If WebGL is enabled in WebKitGTK+, this new GSettings switch
disables and enables WebGL settings at runtime. WebGL is disabled
by default currently because Epiphany may use Clutter in the future.
|
|
|
|
|
|
| |
In 99% of the cases we were using it for gtk_window_set_icon_name,
which does nothing in GNOME 3. In the couple of valid use cases, just
use "web-browser" directly.
|
|
|
|
| |
We can kill data/icons now.
|
| |
|
|
|
|
|
| |
Use the symbolic icon for the completion entry, I'm sure Lapo will
approve.
|
| |
|
|
|
|
|
|
| |
At this point we were only using the #defines in EphyLocationEntry to
decide which symbolic icon to use, so keep that and delete everything
else.
|
|
|
|
|
|
|
| |
Both the icons themselves and the names in the header file. This just
removes those without absolutely any reference to them in the entire
source tree, but I suspect we can remove most of them since they are
not actually used.
|
|
|
|
|
|
|
|
| |
Instead of a GtkToolItem. A future commit will move the code in
EphyLocationAction not to be a GtkAction, which is why this was a
GtkToolItem in the first place.
https://bugzilla.gnome.org/show_bug.cgi?id=668206
|
|
|
|
| |
And fix a couple of warnings.
|
| |
|
|
|
|
| |
And remove them from the existing menubar.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Factor the logic that fakes clicks from a middle click in
EphyMiddleClick(Tool)Button by forwarding a left click to GTK+ when we
receive a middle click. Since ephy_gui_is_middle_click stops working
in this case, add the minimal logic in EphyLinkAction to make it work
again (basically, cache the button that activated the action inside
the action itself).
The EphyMiddleClickable(Tool)Button classes were written by Alexandre
Mazari.
https://bugzilla.gnome.org/show_bug.cgi?id=628364
|
| |
|
| |
|
| |
|
| |
|
|
|
|
| |
Now we can remove our internal implementation.
|
|
|
|
|
|
|
|
|
|
|
| |
In some cases trying to get the content-type of a download right at the
start of the transfer won't give us useful information.
To handle this case we query for the content-type and a proper icon
every 10%, hoping that GIO/Soup will make a smarter guess. This is an
arbitrary number that hopes to work with most cases.
Bug #662059
|
|
|
|
|
|
|
| |
GtkImage works fine with GIcon, this saves us some lines of code and
allocating a GdkPixbuf.
Bug #662059
|
|
|
|
|
|
|
|
|
|
| |
This new API gets the content-type of a download using GIO, or Soup if
GIO is still not available for the download.
GIO is our first option since its local guesses are more reliable. Soup
can be cheated by servers or confused by still too incomplete downloads.
Bug #662059
|
| |
|