[itecz26]

CSS escapes are decoded more than once

unescape_ident() and the string branch of tokenize() decode escapes in separate re.sub() passes (unicode escapes, then simple escapes, plus escaped newlines first for strings). Each pass rescans what the previous one produced, so escape boundaries are lost:

>>> GenericTranslator().css_to_xpath(r"*[x='\\31 ']")
"descendant-or-self::*[@x = '1']"        # expected '\31 '
>>> GenericTranslator().css_to_xpath(r"*[x='\5c 41']")
"descendant-or-self::*[@x = '41']"       # expected '\41'
>>> GenericTranslator().css_to_xpath(r"a\\31")
'descendant-or-self::a1'                 # expected name() = 'a\31'

CSS Syntax §4.3.7 consumes each escape exactly once, left to right, so \\ is a literal backslash and the result of \5c is never read as the start of another escape. Here a selector can end up matching a different value or element name than the one it spells, which is the usual double-decoding problem for anything that inspects or builds selectors before translating them. Decoding every escape with one combined pattern in a single pass fixes both the identifier and the string path; valid input without a backslash next to an escape is unaffected and the existing tests pass unchanged.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 99.79%. Comparing base (91f39df) to head (c3c52ac).

Additional details and impacted files

@@           Coverage Diff           @@
##           master     #203   +/-   ##
=======================================
  Coverage   99.79%   99.79%           
=======================================
  Files           3        3           
  Lines         987      987           
  Branches      161      162    +1     
=======================================
  Hits          985      985           
  Misses          1        1           
  Partials        1        1           

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.