This removes about 2/3 of the occurrences of nsXPIDLString in the tree. The
places where nsXPIDLStrings are null-checked are replaced with |rv| checks.
The patch also removes a couple of unused declarations from
nsIStringBundle.idl.
Note that nsStringBundle::GetStringFromNameHelper() was merged into
GetStringFromName(), because they both would have had the same signature.
We have a minimum requirement of VS 2015 for Windows builds, which supports
the z length modifier for format specifiers. So we don't need SizePrintfMacros.h
any more, and can just use %zu and friends directly everywhere.
MozReview-Commit-ID: 6s78RvPFMzv
Most of the names passed to nsIStringBundle::{Get,Format}StringFromUTF8Name
have one of the two following forms:
- a 16-bit C string literal, which is then converted to an 8-bit string in
order for the lookup to occur;
- an 8-bit C string literal converted to a 16-bit string, which is then
converted back to an 8-bit string in order for the lookup to occur.
This patch introduces and uses alternative methods that can take an 8-bit C
string literal, which requires changing some signatures in other methods and
functions. It replaces all C++ uses of the old methods.
The patch also changes the existing {Get,Format}StringFromName() methods so
they take an AUTF8String argument for the name instead of a wstring, because
that's nicer for JS code.
Even though there is a method for C++ code and a different one for JS code,
|binaryname| is used so that the existing method names can be used for the
common case in both languages.
The change reduces the number of NS_ConvertUTF8toUTF16 and
NS_ConvertUTF16toUTF8 conversions while running Speedometer v2 from ~270,000 to
~160,000. (Most of these conversions involved the string
"deprecatedReferrerDirective" in nsCSPParser.cpp.)
Makes initing Places services cheaper, by delaying the connection creation to the first time
it's actually needed.
Same way, delays reading the bookmark roots at the first time they are requested.
Deprecates the concept of lazy observers, since they are no more needed, we can just use addObserver.
Simplifies the startup path: always sends "places-init-complete" (both as a category and a topic) when
the connection starts and adds a "locked" database state when we can't get a working connection.
Makes PlacesCategoriesStarter register for the new category, since it's cheaper than being a bookmarks
observer.
Fixes a couple race conditions in keywords and expiration due to new startup timings.
Removes a test in test_keywords.js that is no more easily feasible, since it'd requires a pre-build
places.sqlite that should be kept up-to-date at every version.
MozReview-Commit-ID: 6ccPUZ651m0
Root domain icons should be retained as far as possible, and only expired when
a domain is removed from the database. We do that through the moz_hosts trigger.
Additionally, since we return root domain icons without the need for a direct
association, we can save a lot of database space and performance by not storing
associations at all for root domain icons.
This has some downsides, since we don't exactly know if the page was really
associated with that root icon or not, but the perf gains are far superior.
Note that we still create associations during migration, since it would be too
expensive to examine every single url to figure out if it's a root domain icon.
MozReview-Commit-ID: 3mlfcOV8ixC
Updates consumers to the new behavior.
Some consumers are changed to use the "page-icon:" protocol, since it's not
trivial to join the icons table and get a single result out of it. In most cases
the join would return multiple results since a page can have multiple icon payloads.
These consumers for now will return the biggest payload, bug 1347532 will fix
some of them to properly pass a #size=NN fragment.
Note that, even before, these were just "moz-anno:favicon:" uris, and the
payload had to be fetched from the database.
Some other consumers for now just fallback to the largest payload, by passing 0
to GetFaviconURLForPage.
The favicon optimization still happens on the main-thread, bug 1346139 will
handle that problem.
Most of the changes involve handling the modified IconData objects, that now
retain an array of payloads, rather than just one. But note that .ico files are
not yet split into single frames, due to imagelib missing APIs that will be handled
in bug 1337402.
The other changes involve fixing queries to properly join with the new tables.
Finally, note that thanks to the FOREIGN KEYS support, removing from moz_icons or
moz_pages_w_icons will also remove relations from moz_icons_to_pages.
The system only supports square icons, so icons are resized based on their larger side.
This doesn't include new tests, those will be in a following changeset.
MozReview-Commit-ID: JUkpquhpS8y
Giving '0' (literal zero) to nsCOMPtr is now ambiguous, as both
nsCOMPtr(decltype(nullptr)) and nsCOMPtr(T*) could be used.
In any case, our coding standards mandate the use of 'nullptr' for pointers.
So I'm changing all zeroes into nullptr's where necessary.
MozReview-Commit-ID: LXiZTu87Ck6
The new name makes the sense of the condition much clearer. E.g. compare:
NS_WARN_IF_FALSE(!rv.Failed());
with:
NS_WARNING_ASSERTION(!rv.Failed());
The new name also makes it clearer that it only has effect in debug builds,
because that's standard for assertions.
This patch makes most Run() declarations in subclasses of nsIRunnable have the
same form: |NS_IMETHOD Run() override|.
As a result of these changes, I had to add |override| to a couple of other
functions to satisfy clang's -Winconsistent-missing-override warning.
By using nsCOMArray::Forget, we can pass the references we hold in an
nsCOMArray directly to the XPCOM caller, without having to go through an
extra cycle of AddRef/Release calls. This way should be slightly faster
and it also results in smaller code. Notice that we don't have to worry
about removing the OOM check in the places that use Forget(), because we
were already using moz_xmalloc to allocate the outparam array, and
moz_xmalloc aborts the process on OOM, just as Forget() does.
The nsNavHistoryFolderResultNode::GetQueries change is thrown in for fun.
In all of the places touched by this patch, the smart pointer we're
appending is about to become unused, so simply .forget()'ing its
reference into the appropriate nsCOMArray works just fine.
