If the migration phase has to show a dialog the idle callback for
intializing mail stores will run too soon. Instead, hook it onto
the EShellBackend start() method.
Migration code can initialize mail stores early if it needs to.
To reduce GConf usage in em-composer-utils.c:
- Relevant functions in em-composer-utils.c now take arguments for
reply and forward styles.
- Redundant forwarding functions were removed:
em_utils_forward_attached()
em_utils_forward_inline()
em_utils_forward_quoted()
- EMailReader now has "forward-style" and "reply-style" properties,
which get bound to the appropriate EShellSettings properties in
modules/mail/e-mail-config-reader.c. These same EShellSettings
properties are bound to the combo boxes in Composer Preferences.
This marks the end of unintrusive error dialogs, which were too
unintrusive. We now show errors directly in the main window using
the EAlert / EAlertSink framework.
Global variables in shared libraries are a bad idea. EMailBackend now
owns the MailSession instance, which is actually now EMailSession.
Move the blocking utility functions in mail-tools.c to e-mail-session.c
and add asynchronous variants. Same approach as Camel.
Replace EMailReader.get_shell_backend() with EMailReader.get_backend(),
which returns an EMailBackend. Easier access to the EMailSession.
Given the way the autosave feature was awkwardly bolted on to the
composer, an EExtension seemed like a natural fit. And it helped
clean up some object lifecycle hacks (and bugs).
What we have now is a new module consisting of two EExtensions:
EComposerAutosave extends EMsgComposer and determines when to
kick off an asynchronous autosave operation.
EComposerRegistry extends EShell and offers to restore orphaned
autosave files on startup (which is also asynchronous now).
e-autosave-utils.c holds the actual asynchronous functions and a few
other miscellaneous utility functions.
Source code for the new module lives in /modules/composer-autosave.
e_preferences_window to take factory callbacks and store a reference
to the shell. - This makes start-up substantially faster, particularly
on Atom (eg.).
Remove a number of idle handlers used to create these UIs in the
first instance, cleaning the code.
EMFormatHTML now holds a sealed EWebView instead of a public GtkHTML,
accessible through em_format_html_get_web_view().
Rename e_mail_reader_get_html_display() to e_mail_reader_get_formatter()
and have it return an EMFormatHTML instead of an EMFormatHTMLDisplay,
since that's usually the type you want (or else an EMFormat, but never
an EMFormatHTMLDisplay).
Composer and signature editor always default to HTML in Express mode.
Hide the corresponding composer preference. This will not affect the
user preference in normal mode.
After analyzing this again I'm confident we really don't need it.
The only state change is from FALSE to TRUE at startup, and that
one-time event happens while the mail shell backend is starting up
(see: e_shell_backend_start()).
If a need arises to query for this in the future I'll extend the
EShellBackend API with an e_shell_backend_started() function, but
for now there's no need.
EMailBackend is an abstract subclass of EShellBackend that handles
online and offline modes and application shutdown. Placing this in
the shared mail library allows Anjal to reuse it. Evolution's mail
module further extends this class as EMailShellBackend.
Yes, this signal is kind of an ugly monster. I'm not sure how to improve this
significantly. But this commit removes the last EMFolderTreeModel and EShell
dependencies from MailFolderCache, which is a big step towards splitting off
the backend.
https://bugzilla.gnome.org/show_bug.cgi?id=604627
Instead of pushing the updates to the right places, the folder cache simply
emits the appropriate signals and other objects are responsible for listening
and handling them appropriately. This allows us to cut down the dependencies of
MailFolderCache significantly, which is a huge step towards allowing us to split
it off for the backend.
Another nice thing about this is that it allows us to trim a lot of 'public' api
from the filter, vfolder, and config classes that were only used by the cache.
Now that stuff can all be internal since they're pulling changes rather than
having the changes pushed.
The last remaining problematic dependency in MailFolderCache is
EmFolderTreeModel. That is next on the chopping block.
https://bugzilla.gnome.org/show_bug.cgi?id=604627
The changes are mainly including the e-alert-header.h header instead of just
e-alert.h. This allows us to include e-alert.h in non-UI situations when
necessary.
The EError mechanism is used both for error dialogs as well as basic alerts or
user prompts, so we should give it a more general name which matches this use.
This patch also cleans up a few includes of e-alert.h (formerly e-error.h) that
were not actually being used.
https://bugzilla.gnome.org/show_bug.cgi?id=602963
There were a number of problems:
- Walk the GType tree to find EPluginHook subclasses, instead of
just registering the immediate children.
- Some EConfig and EEvent subclasses were not being configured
properly (particularly the mail and calendar subclasses).
- Add preference window pages after the main loop starts to make
sure all plugins and plugin hooks are installed first.
- Require all EPlugin and EPluginHook subtypes be registered before
loading plugins. This drastically simplifies the EPlugin/EPluginHook
negotiation.
- Turn most EPluginHook subtypes into GTypeModules and register their
types from an e_module_load() function (does not include shell hooks).
- Convert EPluginLib and the Mono and Python bindings to GTypeModules
and register their types from an e_module_load() function, and kill
EPluginTypeHook.
Milan Crha