- Moves `Bytes` from `util` to `eval` module
- Accepts bytes in `str` function for bytes -> str conversion
- Adds `bytes` function for str | array -> bytes conversion
- Adds `array` function for bytes -> array conversion
- Adds `len`, `at`, and `slice` methods for bytes
- Adds `encoding` parameter to `read` function
Sometimes, optimized line breaking outputs lines that are wider
than the allotted width so that spacing between words can be
narrowed to fit. However, the width calculation for the fit-to-width
case does not account for this, causing the resulting width to exceed
the width of the region. This commit takes the minimum of the
calculated width with the width of the region. Fixes#1825.
This positions placed elements relative to the real container size instead of relative to the base size of the region. This makes its usage more versatile.
Fixes#82Fixes#685Fixes#1705
Fixes#713. They were previously classified by Unicode as relations, which caused a '+' sign after one of them to be treated as a unary operator rather than a binary operator. This PR overrides the Unicode classification and classifies them as normal instead, which causes a '+' sign after one of them to be properly treated as a binary operator.
Co-authored-by: Ryan Pitasky <111201305+rpitasky@users.noreply.github.com>
Co-authored-by: Daniel Csillag <15617291+dccsillag@users.noreply.github.com>
This ensures that the ranges of the shaped glyphs are monotonically
decreasing in right-to-left-text, thus avoiding nonsensical results
in find_safe_to_break that later causes a panic (see #1373).
Additionally, debug assertions have been added to catch non-monotonic
glyph ranges.
Matt Fellenz
Beiri22
damaxwell
Laurenz
bluebear94
Pg Biel
lolstork
Sébastien d'Herbais de Thun
Myriad-Dreamin
Liam Murphy
KG
MALO
sitandr
Eric Biedert
Mathias Fischler
Lynn