Skip to content

gh-130273: Fix incorrectly colored error locations when wide unicode characters exists#130277

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

Draft
HarryLHW wants to merge 1 commit into
base:main
Choose a base branch
from
Draft

gh-130273: Fix incorrectly colored error locations when wide unicode characters exists #130277

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
46 changes: 35 additions & 11 deletions Lib/traceback.py
View file
Open in desktop
Original file line numberDiff line numberDiff line change
Expand Up@@ -656,18 +656,29 @@ def output_line(lineno):
line = result[-1]
colorized_line_parts = []
colorized_carets_parts = []

for color, group in itertools.groupby(itertools.zip_longest(line, carets, fillvalue=""), key=lambda x: x[1]):
caret_group = list(group)
if color == "^":
colorized_line_parts.append(ANSIColors.BOLD_RED + "".join(char for char, _ in caret_group) + ANSIColors.RESET)
colorized_carets_parts.append(ANSIColors.BOLD_RED + "".join(caret for _, caret in caret_group) + ANSIColors.RESET)
elif color == "~":
colorized_line_parts.append(ANSIColors.RED + "".join(char for char, _ in caret_group) + ANSIColors.RESET)
colorized_carets_parts.append(ANSIColors.RED + "".join(caret for _, caret in caret_group) + ANSIColors.RESET)
line_part_col = 0
for char, group in itertools.groupby(carets):
carets_part = "".join(group)
width = len(carets_part)
if char == "^":
color = ANSIColors.BOLD_RED
elif char == "~":
color = ANSIColors.RED
else:
colorized_line_parts.append("".join(char for char, _ in caret_group))
colorized_carets_parts.append("".join(caret for _, caret in caret_group))
colorized_carets_parts.append(carets_part)
line_part_col, line_part = _offset_at_width(line, width, start=line_part_col)
colorized_line_parts.append(line_part)
continue

colorized_carets_parts.append(color)
colorized_carets_parts.append(carets_part)
colorized_carets_parts.append(ANSIColors.RESET)

colorized_line_parts.append(color)
line_part_col, line_part = _offset_at_width(line, width, start=line_part_col)
colorized_line_parts.append(line_part)
colorized_line_parts.append(ANSIColors.RESET)
colorized_line_parts.append(line[line_part_col:])

colorized_line = "".join(colorized_line_parts)
colorized_carets = "".join(colorized_carets_parts)
Expand DownExpand Up@@ -965,6 +976,19 @@ def _display_width(line, offset=None):
for char in line[:offset]
)

def _offset_at_width(line, width, start=0):
col = start + width
part = line[start:col]
if part.isascii():
return col, part

from unicodedata import east_asian_width

for i in range(start, len(line)):
if width <= 0:
return i, line[start:i]
width -= 2 if east_asian_width(line[i]) in _WIDE_CHAR_SPECIFIERS else 1
Copy link
Contributor

@grayjkgrayjkFeb 18, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

does this correctly handle zero-width characters? e.g. the '\u0301' from 'cafe\u0301'

Copy link
ContributorAuthor

@HarryLHWHarryLHWFeb 19, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

No.
However, _display_width() does not correctly handle zero-width characters for carets positions either.

>>> café = 1/0 Traceback (most recent call last): File "<python-input-0>", line 1, in <module> café =1/0~^~ZeroDivisionError: division by zero >>>

They can be fixed in the same way. I wonder how I can find all the zero-width characters.

Copy link
Contributor

@grayjkgrayjkFeb 19, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It looks like there are many zero-width characters: https://wcwidth.readthedocs.io/en/latest/specs.html#width-of-0

Copy link
ContributorAuthor

@HarryLHWHarryLHWFeb 19, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This is how wcwidth implements wcswidth() and wcwidth(): https://wcwidth.readthedocs.io/en/latest/_modules/wcwidth/wcwidth.html#wcswidth
Maybe it is too complicated for our traceback? Could we have a simple way that can handle most of the cases?

return len(line), line[start:]


class _ExceptionPrintContext:
Expand Down
Loading