| Commit message (Collapse) | Author | Files | Lines |
|---|
|
This PR fixes the code to avoid saving default environment every time
the project is run whitin the editor.
Should fix #17727. Sorry for the troubles!
|
|
This adds more scaling options, in addition to a custom scaling option
which allows any scale between 0.75 and 3.0 to be used.
|
|
This reverts commit e42576548f2c0ae2c6cb24ce2b0437ffb8978d65.
|
|
|
|
|
|
|
|
|
|
|
|
Fixes #16923. I'm not a fan of the special case for scripts in editor_node.cpp, but in any case,
I made it so it wouldn't make the external editor to re-open just because we switched scenes.
|
|
|
|
|
|
|
|
Works both for the editor and games.
Projects can still use "debug/settings/stdout/print_fps" to enable it
permanently. The --print-fps option takes precedence (so works even if
the project setting is disabled). That setting is also no longer redefined
on the fly based on the verbose flag, that was a mess.
|
|
|
|
|
|
Fix #17644
|
|
|
|
When `p_points.size() > p_colors.size()`, it crashed with invalid
array access to `p_colors`. Also, when `p_colors` was an empty
`Vector` it crashed due a missing `else` checking the `size`
condition, as the code handling that special case exists.
This PR fixes the missing `else` for `p_colors.size == 0` and,
following the `canvas_item_add_multiline` spirit, it only uses the
first color for the whole polyline if points and colors differ in
size.
Fix #17621.
|
|
Fixes the mistake I made in #17603 to make it require Python > 3.6.
|
|
Font update after resize relies on the viewport size which was updated
after the font was already refreshed, which resulted in artifacts when
it was rendered into the actual/new viewport size.
Fixes #15173.
|
|
|
|
|
|
|
|
Aside from the colors themselves, the rejection was adding somes issues later in the mesh import process.
|
|
- Setup runtime main args during initialization. This must be done manually by embedders who do not call mono_runtime_run_main. Fixes NullReferenceException in System.Environment.
- Continue to search the assembly in the rest of the search locations if loading it from one of them failed.
|
|
|
|
The messages generated by some assertions can be confusing to users.
|
|
Text overflowed canvas as tables didn't calculate correctly the width
of their columns. They used the whole table width available for each
column. Also, the `cell` parameter was wrongly parsed if used with its
optional argument (expand ratio).
This PR fixs the parsing of `cell` parameter (i.e. `cell=e`) and the
distribution of the full table width between columns, but it overrides
automatically the `expand` flag if the column is smaller than it could
be, to allow a better UX out-of-the-box. It keeps the `expand_ratio`
flag to let the user customize how every column grows in relation to
the rest.
Partially fix #6289.
|
|
When adding a directory path to the inventory of the pack, an empty file name was being added to the file list. That made `Directory.get_ntext()` signal end-of-list too early so that files in a subdirectory were missed.
Fixes #15801.
Helps with #16798.
|
|
More clarification.
|
|
When `_save_all_scenes` or `save_resource_in_path` was called, they
always saved all the scenes and the resource no matter if they were
modified or not. For example, when `saving before run` option was
checked, it always overwrote the current scene and the default
environment simply by opening and runing the project.
This PR adds checks for unsaved scenes (using the same `unsave` check
others method used) and modified resources (comparing last modified
time and last import time).
Fix #6025.
|
|
A subprocess call wasn't specifying an encoding,
so this gave a TypeError in Python 3.
|
|
|
|
|
|
Add a new function to check action names, `_validate_action_name`, in
the spirit of `_valprop`. Offending characters include non-printable
ascii, and `\/=:"`. Also set only one text for the UI message.
|
|
|
|
|
|
As `KEY_F3` was used both for changing to script editor window and, in
the script editor, for finding the next result in the last search, and
the key event is **not** consumed, the resulting behaviour was similar
to press `F3` twice, first to change to script editor and second to
find the next result of a previous search.
This PR sets the `key_pressed` status of `InputEvent` to `false` if
this event is responsible of an editor change, simulating the
consumption of the event.
Fix #17334
|
|
Open mapping files in textmode as they are textfiles.
|
|
|
|
|
|
|
|
Fixup to #17296.
|
|
|
|
|
|
Disallow reserved keywords as class names and prefix base class with the Godot
namespace if it's the same as the class name.
Fixes #12483
|
|
|
|
|
|
|
|
|
