Age | Commit message (Collapse) | Author |
|
Was a regression from ddbf2ff.
This commit also introduce a "#define snprintf _snprintf", so the rest of the file was simplified to make use of this
instead of distinguishing between snprintf and _snprintf in the various functions.
|
|
|
|
|
|
parser to break.
Documentation for HTTPClient.
Added a query_string_from_dict method to HTTPClient to create a x-www-form-urlencoded valid query string for GET and POST messages.
String now has http_escape() and http_unescape() methods to help facilitate the above query_string_from_dict method.
|
|
The issue was that world_wrap would skip over newlines, without adding them to the output.
|
|
|
|
Before > referred to < and < to >, which is incorrect...
|
|
|
|
This works if the property has been documented (about half are at this point)
|
|
|
|
and resources.
A general speedup should be apparent, with even more peformance increase when compiling optimized.
WARNING: Tested and it seems to work, but if something breaks, please report.
|
|
|
|
|
|
|
|
|
|
Fix sprintf errors
|
|
|
|
|
|
|
|
|
|
|
|
|
|
-added h_offset and v_offset to 3D Camera, should allow to do the same as in #1102
|
|
-Collada exporter supports Blend Shapes (even on actions via set driven keys)
|
|
|
|
-=-=-=-=-=-=-=-=-=-=-=-=-=
-Begin work on Navigation Meshes (simple pathfinding for now, will improve soon)
-More doc on theme overriding
-Upgraded OpenSSL to version without bugs
-Misc bugfixes
|
|
-=-=-=-=-=-=
-Support for lightmap baker, have fun figuring out how it works before tutorial is published.
|
|
|
|
Conflicts:
modules/multiscript/register_types.cpp
platform/android/java/src/com/android/godot/GodotLib.java
|
|
-fix some menus
-fixed bug in out transition curves
-detect and remove file:/// in collada
-remove multiscript for now
-remove dependencies on mouse in OS, moved to Input
-avoid fscache from screwing up (fix might make it slower, but it works)
-funcref was missing, it's there now
|
|
Signed-off-by: Vinzenz Feenstra <evilissimo@gmail.com>
|
|
Entering a somethign in the 'Path field' triggers the
`_path_changed` signal being triggered.
This in turn calls Globals::localize_path(const String& p_path) with the
currently entered string.
localize_path then is replacing backslashes with slashes and calls
afterwards `String::simplify_path`
String::simplify_path is checking wheter a string starts with:
- res://
- local://
- user://
If any of those is true it removes this section.
However, if any of the first letters of those are matching begins_with
returns true, which is wrong. It should only return true if the whole
string is matched at the beginning.
This caused the whole desaster and lead localize_path into an endless loop
because out of `u` suddenly became user:// which it then tried again to
localize and so on.
This fix, fixes the root of the problem which is begins_with which should
not return true if not the whole search string was matched.
Signed-off-by: Vinzenz Feenstra <evilissimo@gmail.com>
|
|
|