`GLib.TestSuite` is a special type of object, as it doesn't actually
have an associated copy function or a reference/unreference function.
In practice, that actually means there's no way to actually have 2
strong references to such an object, or to copy it somehow. Due to the
way GObject properties in Vala work, that means you can't really use
them either.
Ever since vala commit 5f0a146f65, this got reflected properly in the
internally maintained GLib VAPI (as people were experiencing double
frees otherwise), but it was added only conditionally for GLib 2.70 or
later. In practice, that means you only get vala compiler issues if you
(impliclty) set `--target-glib=2.70` (or a later version).
To fix this for vala-unit and our own Geary tests, this commit changes
the `ValaUnit.TestCase` class to only allow "stealing" the strong
reference to the `GLib.TestSuite`, rather than somehowing allowing
people to hold a weak reference, and by also making it a protected field
rather than a property.
Since this changes the API of ValaUnit, we also bump the version. In
general though, I hope people aren't using ValaUnit externally, since
these types of API/ABI breaks can happen with Vala libraries.
Link: 5f0a146f65
Currently, a lot of our build variables that are defined at
configuration time, are spread out across the code base, often declared
as `extern` which can break when moving around sections of the code
across files.
This commit introduces a "Config" namespace which basically maps to the
definitions in `config.h`, but allows us to properly access them too
from the Vala source code.
By doing so, it helps us to more explicitly see where we rely on a
build variable from this file (which should be obvious from the `Config'
namespace).
To make it ourselves a bit easier in Meson too, we can declare an
internal dependency, which helps ensure that we pull in the dependency
where needed.
Move SQL generation for FTS search from ImapDb.Account to SearchQuery.
Convert to use Geary.SearchQuery.Term instances to generate SQL, rather
than parsing the expression. Simplify the generated SQL substantially
and generate MATCH values that work with SQLite FTS5.
Some servers (e.g. Dovecot) use placeholder strings instead of the empty
string (e.g. "MISSING_DOMAIN") in FETCH Envelope responses when a
required address part (e.g. local part, domain) are empty.
This adds a quirk that can be enabled for such servers to check for the
placeholders and if found, replace them with empty strings.
Defining an interface for the composer to access application objects
and services decouples it from Application.Controller, allowing it to be more
easily unit tested.
Replace use of Application.Client and Application.Controller in the
composer and add some basic unit tests.
Engine mocks don't need to be in the `Geary` namespace, and including
them there makes it difficult to use them in client tests, so put them
all in their own name-space and corresponding directory.
Break out the generic testing code into something easily re-used, and
improve the API substantially:
* Use generics to reduce the number of equality tests to effectively
a single one
* Make all assert args consistent in that the actual value is always
listed first.
* Add convenience API for common string/array/collection assertions
Revert merge request GNOME/geary!374 for now since the shared process
model breaks old-style WebProcess message handler IPC.
This can be un-reverted when out JS is ported to the new Messages API
that is landing in WebKitGTK 2.28.
This reverts commit e4a5b85698, reversing
changes made to 66f6525480.
Support manual validation where needed, add unit tests.
This also slightly changes the behaviour of non-required field, since
an empty non-required field should be valid.
The GVariant type "*" only matches a single data type, not many, and
the sense of the test to check serialised ids was wrong anyway. As a
reuslt, this method probably never worked. Add a unit test.
The only reason it was in the engine was so it could be used by both
the client and the web extension, without worrying about the
webkit2gtk and webkit2gtk_web_extension packages conflicting. However
it didn't really belong there, and added a dependency for the engine
on javascriptcoregtk which doesn't belong. So this fixes all that.
Add a ContactHarvester arg to ImapDB.Folder.create_or_merge_email and
call it on the resulting email. Add and update a new harvester
property to MinimalFolder and pass that in as needed.
Replaces harvesting code in ImapDB code so as to be stand-alone and
reusable. Implement harvesting policy to only harvest from a set of
whitelisted special folder types (so junk and trash is not harvested)
and only harvest valid, non-spoofed, addresses.
If a FillWindowOperation didn't load a full amount of messages, it just
assumed that there were no more to load. This is not true however when
loading locally, the folder's vector isn't fully expanded, and it gets
to the end of the vector.
This patch fixes the operation to also queue another fill if the
monitor's folder message window is smaller than the folder's total
message count.
Fixes#289
Some software like the above will mangle From mailbox names by appending
"via Some Service" to the From mailbox name. This messes up generating
default avatars for the actual people sending these messages, so
attempt to de-mangle the names.
This involves moving primary originator determination from the engine
to the client, since it's now a policy thing. Add unit tests.
Don't try to dray null initials, actually return null per API contract
when no initials can be found, don't include non-alphanumerics in the
initials. Add test case.
Instead of each top-level IMAP folder being a FolderRoot object, then
children of that being FolderPath objects, this makes FolderRoot an
"empty" FolderPath, so that both top-level and descendant folders are
plain FolderPath objects. Aside from being more technically correct,
this means that empty namespace roots can now be used interchangably
with non-empty namespace roots (addressing issue #181), and custom
folder implementations no longer need to provide their own trivial,
custom FolderRoot.
To support this, a notion of an IMAP root and a local root have been
added from which all remote and local folder paths are now derived,
existing places that assume top-level == root have been fixed, and
unit tests have been added.
