TODO
TODO
TODO
Text may render differently when starting at fractional coordinates, so
:py:meth:`.FreeTypeFont.getmask` and :py:meth:`.FreeTypeFont.getmask2` now
support a start
argument. This tuple of horizontal and vertical offset
will be used internally by :py:meth:`.ImageDraw.text` to more accurately place
text at the xy
coordinates.
The exact
encoding option for WebP is now supported. The WebP encoder
removes the hidden RGB values for better compression by default in libwebp 0.5
or later. By setting this option to True
, the encoder will keep the hidden
RGB values.
XMP data can now be
decoded for WEBP images through getxmp()
.
When saving a JPEG image, a comment can now be written from :py:attr:`~PIL.Image.Image.info`, or by using an argument when saving:
im.save(out, comment="Test comment")
TODO
TODO