The Styles API section of the Mapbox API Documentation site now lists Traffic Day v2 and Traffic Night v2, so this change adds those styles to all the places where styles are listed.
Also switched iosapp and macosapp to unversioned style factory methods since MGLStyleDefaultVersion is no longer applicable for all styles.
Undeprecated the unversioned style URL factory methods in MGLStyle for consistency with the Android and Qt SDKs. Added warnings about using them with the runtime styling API.
Refactored mbgl::util::default_styles to track different versions for different styles.
Upgraded from v9 default styles to v10 wherever the developer expects to get the latest and greatest, as well as in a couple tests where it may be beneficial to ensure that we can handle a two-digit version number in the style URL.
* [ios] Use map view frame to calculate annotation view reuse adjustments
The value for determining the visible viewport buffer and also the distance to move offscreen an annotation view outside of that buffer was based on the annotation view width and height. This changes that to use the map viewport width and height as constants and avoids a class of bugs where the annotation view would become detached from the tracking system when it did not have a size or was animating from a small to large size.
Pulled various translations from Transifex.
Although Transifex supports the stringsdict format, it doesn’t allow translators to customize the format. This change splits out a separate “level” key for the same format string and moves any prefixes and suffixes into the “level” and “count” keys. For now, each localization has been updated manually, but future updates will come from Transifex.
This adds a `convertJSON` template, to be used like:
Error error
optional<Foo> foo = convertJSON<Foo>(string, error);
Internally, it parses the string with RapidJSON and then calls `convert<Foo>(parsed, error)`.
While here, rationalize GeoJSON converters and fix error handling for Tileset conversion in OfflineDownload.
This reverses #3095. Rationale:
* We're now exposing source attributes as a public API. Making those attributes unavailable at certain times complicates that API.
* We're preparing to split RenderSource out of Source. Removing this removes a point of coupling between the two.
