Skip to content

Ensure that canonically equivalent strings have the same width #37

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Apr 22, 2024
Merged

Ensure that canonically equivalent strings have the same width #37

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 3 additions & 0 deletions Cargo.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -22,6 +22,9 @@ std = { version = "1.0", package = "rustc-std-workspace-std", optional = true }
core = { version = "1.0", package = "rustc-std-workspace-core", optional = true }
compiler_builtins = { version = "0.1", optional = true }

[dev-dependencies]
unicode-normalization = "0.1.23"

[features]
default = []
bench = []
Expand Down
30 changes: 25 additions & 5 deletions scripts/unicode.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -150,14 +150,15 @@ def load_zero_widths() -> "list[bool]":
"""Returns a list `l` where `l[c]` is true if codepoint `c` is considered a zero-width
character. `c` is considered a zero-width character if

- it is in general categories `Cc`, `Mn`, or `Me` (determined from `UnicodeData.txt`),
- it is in general category `Cc`,
- or if it has the `Grapheme_Extend` property (determined from `DerivedCoreProperties.txt`),
- or if it has the `Default_Ignorable_Code_Point` property (determined from `DerivedCoreProperties.txt`),
- or if it has a `Hangul_Syllable_Type` of `Vowel_Jamo` or `Trailing_Jamo` (determined from `HangulSyllableType.txt`).
"""

zw_map = []

# Characters with general category `Cc`, `Mn`, or `Me` have 0 width...
# Characters with general category `Cc` have 0 width
with fetch_open("UnicodeData.txt") as categories:
current = 0
for line in categories.readlines():
Expand All @@ -168,7 +169,7 @@ def load_zero_widths() -> "list[bool]":
raw_data[1],
raw_data[2],
]
zero_width = cat_code in ["Cc", "Mn", "Me"]
zero_width = cat_code == "Cc"

assert current <= codepoint
while current <= codepoint:
Expand All @@ -188,10 +189,16 @@ def load_zero_widths() -> "list[bool]":
# `Default_Ignorable_Code_Point`s also have 0 width:
# https://www.unicode.org/faq/unsup_char.html#3
# https://www.unicode.org/versions/Unicode15.1.0/ch05.pdf#G40095
#
# `Grapheme_Extend` includes characters with general category `Mn` or `Me`,
# as well as a few `Mc` characters that need to be included so that
# canonically equivalent sequences have the same width.
with fetch_open("DerivedCoreProperties.txt") as properties:
single = re.compile(r"^([0-9A-F]+)\s+;\s+Default_Ignorable_Code_Point\s+")
single = re.compile(
r"^([0-9A-F]+)\s+;\s+(?:Default_Ignorable_Code_Point|Grapheme_Extend)\s+"
)
multiple = re.compile(
r"^([0-9A-F]+)\.\.([0-9A-F]+)\s+;\s+Default_Ignorable_Code_Point\s+"
r"^([0-9A-F]+)\.\.([0-9A-F]+)\s+;\s+(?:Default_Ignorable_Code_Point|Grapheme_Extend)\s+"
)

for line in properties.readlines():
Expand Down Expand Up @@ -240,6 +247,19 @@ def load_zero_widths() -> "list[bool]":
# (which are considered 0-width on their own) to form a composed Hangul syllable with
# width 2. Therefore, we treat it as having width 2.
zw_map[0x115F] = False

# Unicode spec bug: these should be `Grapheme_Cluster_Break=Extend`,
# as they canonically decompose to two characters with this property,
# but they aren't.
zw_map[0x0CC0] = True
zw_map[0x0CC7] = True
zw_map[0x0CC8] = True
zw_map[0x0CCA] = True
zw_map[0x0CCB] = True
zw_map[0x1B3B] = True
zw_map[0x1B3D] = True
zw_map[0x1B43] = True

return zw_map


Expand Down
Loading