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.
Our test wasn't using the appropriate callback `.end()` function, which
was failing the test with the latest Vala nightly releases, breaking the
Flatpak build in our CI already too. This should fix that issue.
You can test that this should be the correct order by adding a line for
each on a file and then running
```
$ cat "$TEST_FILE" | LC_ALL=C sort -r
```
Fixes: 43c23270a6 ("migrate locale from en_US.UTF-8 to C.UTF-8")
When marking a message, a race condition can happen:
- A remote folders update is already running
- We mark the message locally and replay it remotely
- The previous remote update result restores invalid values locally
The parser would think the flag was a response code if the flag started
with a '[' character. Flags like these are sometimes returned by Gmail
which broke fetching emails from Gmail.
This commit checks if the parser is already parsing flags and if so
treats any starting '[' characters as part of the flag string.
Without that the client tests fail like
** (test-client:74671): CRITICAL **: 11:32:40.613: void webkit_user_content_manager_add_style_sheet(WebKitUserContentManager*, WebKitUserStyleSheet*): assertion 'styleSheet' failed
This was so far masked since the composer-page-state-test JS tests did
that but test suites should be independent.
Remove "\n " sequences from RFC822 text before decoding so that any
long mailboxes/mailbox lists folded before the 80 char limit are
unfolded.
Fixes#1009
Ensure OR is actually used to separate disjuncts, don't prefix match
non-stemmed terms when a stemmed version exists or when EXACT is
specified. Ensure column is correctly specified per disjunct.
SQLite FTS5 doesn't allow all negated terms in matches, and since NOT
is a binary operator (rather than unary) negated terms should be listed
after any positive terms.
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.
Adds classes that allow building arbitrary query expressions and require
an instance to be provided to Geary.SearchQuery, to be set as a
property.
This enables query expressions to be parsed by clients instead of the
engine, in which ever whay they choose.
Add SQL migration that drops the old FTS4 MessageSearchTable table,
re-create as a FTS5 table, clean up the column names a bit, and adds a
flags column so unread/starred queries can be made fast.
Define a SQLite FTS5 extension function `geary_matches()` to replace
the FTS3 `offsets()` function which no longer exists in FTS5, based on
Tracker's implementation.
Update code to FTS5 conventions (docid -> rowid, etc), use new column
names, populate and update the flags column as the email's flags
change, and use new match function for getting matching tokens.
Advanced searches are probably currently broken, these will be fixed
by subsequent commits.
Using `whitespace: pre-wrap` to format plain text email sometimes
causes additional width to be allocated by the plain text blocks that
then does not get used due to the constraints on the HTML element.
The allocated space remains however and hence an un-needed horizontal
scrollbar appears.
Using `break-spaces` instead seems to help since it allows breaks after
a space character, leading to the additional space not otherwise being
allocated.
Now that the `:focus-within` pseudoclass is supported, use this rather
than some custom JS to update custom HTML classes. This also prevents
spurious mutation events from firing.
Stop needlessly wrapping object members and array elements in
variant variants.
Don't wrap object values in variants since the code is already using
vardicts for these. Return a variant array if a JS array contains values
of all the same type and don't wrap these in variants, else return
a tuple, which don't need to be wrapped either.
Update web extension to check for errors when invoking page state
methods and pass a message back if found. Check for this, decode and
throw a vala error in the WebView if found.
Use WebKitGTK UserMessage objects for invoking JS methods rather than
serialising to JS strings and running those. This is possibly slightly
less efficient, but removes the onus on serialising to and parsing from
JS and once switched over from message handlers to UserMessage objects
will be using a single uniform IPC interface for both.
Convert `get_protocol_state` to an automatic property, so that rather
than requiring explcit signals for lifecycle events, a GObject notify
signal can be used instead.
Convert disconnect signal to a property so it can be accessed if needed.
Convert code listening to the disconnect signal to listen to notify
signals for `protocol_state` instead, and hence also treat the session
as disconnected when a logout is in progress.
Fixes#986
Since both submitting a command no longer requires a cancellable, and it
is desirable to avoid sending a queued command that has already been
cancelled beforehand, add a new `Command.should_send` Cancellable
property to specify if a command should (still) be sent or not, and stop
passing a cancellable to ClientSession when submitting commands.
Allow call sites to pass in existing cancellable objects, and thus also
add it it as a ctor property to the Command class and all subclasses.
Lastly, throw a cancelled exception in `wait_until_complete` if send
was cancelled so that the caller knows what happened.
Remove redundant cancellable argument from
`Imap.Client.command_transaction_async` and rename it to
`submit_command` to make it more obvious about what it does.
