GCC has some optimization for the inclusion guard, but they only work if
the check is the outermost one.
We're fairly inconsistent because of historical reasons, so we should
ensure that we follow the same pattern in every public header.
ClutterScript should be able to automatically call gettext() and friends
on strings loaded from a UI definition, prior to passing the string to
the object it is constructing.
The basic implementation is trivial:
- set a translation domain on the ClutterScript instance
- mark the translatable strings inside the JSON data, like:
"property" : {
"translatable" : true,
"string" : "a translatable string"
}
The hard part is now getting the tools we use to extract the
translatable strings to understand the JSON format we use inside
ClutterScript.
GLib has gained support for compiling ancillary data files into the same
binary blob as a library or as an executable.
We should add this feature to ClutterScript, so that it's possible to
bundle UI definitions with an application.
Currently, defining states for object signals can only be done by
defining a ClutterState inside the ClutterScript definition. We should
allow creating a (named) ClutterState in code, and associating it to a
ClutterScript instance — and have the Script resolve the "state" field
of a signal definition correctly.
The G_CONST_RETURN define in GLib is, and has always been, a bit fuzzy.
We always used it to conform to the platform, at least for public-facing
API.
At first I assumed it has something to do with brain-damaged compilers
or with weird platforms where const was not really supported; sadly,
it's something much, much worse: it's a define that can be toggled at
compile-time to remove const from the signature of public API. This is a
truly terrifying feature that I assume was added in the past century,
and whose inception clearly had something to do with massive doses of
absynthe and opium — because any other explanation would make the
existence of such a feature even worse than assuming drugs had anything
to do with it.
Anyway, and pleasing the gods, this dubious feature is being
removed/deprecated in GLib; see bug:
https://bugzilla.gnome.org/show_bug.cgi?id=644611
Before deprecation, though, we should just remove its usage from the
whole API. We should especially remove its usage from Cally's internals,
since there it never made sense in the first place.
Bug 1099 - No ClutterScript API to get a list of IDs in a given file
* clutter/clutter-script.[ch]: Add clutter_script_list_objects(),
a function for retrieving all the objects built by a ClutterScript
instance. (Based on a patch by Noah Gibbs)
Bug 989 - Add a search path for clutter script assets
* clutter/clutter-script.h:
* clutter/clutter-script.c:
(clutter_script_finalize),
(clutter_script_add_search_paths),
(clutter_script_lookup_filename): Add the ability to define multiple
search paths inside ClutterScript and to look up a specific filename
inside those search paths. This is useful to define a set of
directories where the assets for a UI definition are and still
reference those assets by their name instead of the full path. (989,
based on a patch by Matthew Allum)
* clutter/clutter-texture.c:
(clutter_texture_set_custom_property): Use the newly added
clutter_script_lookup_filename() function.
* clutter/json/json-parser.[ch]: Use gssize, size we allow -1
as a length (meaning "use the whole string").
(json_parser_load_from_data): Use the passed length instead of
using strlen() all the time.
* clutter/clutter-script.[ch]: Ditto as above.
* clutter.symbols: Update with the new public symbols
* clutter/clutter-script.h:
* clutter/clutter-script-private.h:
* clutter/clutter-script.c:
(parse_signals), (json_object_end),
(signal_info_free), (object_info_free): Parse the "signals"
member for GObjects.
(clutter_script_connect_signals),
(clutter_script_connect_signals_full): Add new API for autoconnecting
signal handlers using the UI definition files.
* tests/test-script.c:
* tests/test-script.json: Test signal autoconnection.
* clutter/clutter-scriptable.[ch]: Rename ::set_name and ::get_name
to ::set_id and ::get_id, to avoid potential confusion with the
ClutterActor:name property.
* clutter/clutter-script.h:
* clutter/clutter-script.c (clutter_script_construct_object): Use
clutter_scriptable_set_id().
(clutter_get_script_id): Add a public function to retrieve the ID
used in the UI definition files from an object.
* clutter/clutter-actor.c: Do not set the name of the actor with
the ID set in the UI definition files.
* tests/test-script.c: Test clutter_get_script_id().
* clutter.symbols: Update with the new symbols.
* clutter/clutter-script.[ch]: Slight API change in the
clutter_script_get_objects() function: now it takes
object name/object return location pairs and returns the
number of objects found and returned.
* tests/test-script.c: Exercise the clutter_script_get_objects()
function.
* clutter/clutter-script.h:
* clutter/clutter-script.c: Add a ::get_type_from_name() virtual
function for bindings to override. The current implementation
calls g_type_from_name() and our lazy class resolver.
* clutter.symbols: Update.
* doc/reference/clutter-sections.txt: Update.
* clutter/clutter-script-private.h:
* clutter/clutter-script.h:
* clutter/clutter-script.c: Allow id-less objects: as long
as they have a "type" member, a unique id will be provided.
(json_object_end): Add merge id to the object information
structure.
(apply_behaviours), (add_children): Keep the unresolved
objects around.
(construct_stage), (clutter_script_construct_object): If an
object has unresolved children or behaviours try resolving
them when we ask for it.
(json_parse_end), (clutter_script_ensure_objects): Ensure
that the objects are fully constructed as best as we can when
finished parsing.
(object_info_free), (remove_by_merge_id):
(clutter_script_unmerge_objects): Remove objects under the
same merge id returned by the loading functions. (Fixes
bug #558)
* clutter/clutter-script-private.h:
* clutter/clutter-script.h:
* clutter/clutter-script.c: Add licensing information to
the newly added files.
* clutter/clutter-script.c: Support creating behaviours with
ClutterScript. ClutterAlpha objects are implicit, but
timelines can be both explicit objects using their id or
implicit objects. Make the property resolution and translation
more robust. Support the pixbuf property.
* tests/test-script.c: Test the newly added features.
* docs/reference/clutter-docs.sgml:
* docs/reference/clutter-sections.txt: Add ClutterScript.
Initial implementation of the UI definition files. (#424)
* clutter/json/Makefile.am:
* clutter/json/*.[ch]: In-tree copy of JSON-GLib, a GLib-based
JSON parser/generator library. We use it in-tree because we might
need to change the API. Ideally, we'd depend on it.
* clutter/clutter.h:
* clutter/clutter-script-private.h:
* clutter/clutter-script.[ch]: ClutterScript, the scenegraph
generator class. It parses JSON streams in form of buffers and
files and builds the scene.
* clutter/clutter-debug.h:
* clutter/clutter-main.c: Add a "script" debug flag
* clutter/Makefile.am: Build glue.
* tests/Makefile.am:
* tests/test-script.c: Add a test case for the ClutterScript.
* configure.ac: Depend on GLib 2.14, so we can use the
g_hash_table_get_key() and g_hash_table_get_values() functions
for the time being; we can probably reimplement those, but we
are going to need 2.14 anyway if we are going to implement a
list model using GSequence.
