diff options
| author | Serhiy Storchaka <storchaka@gmail.com> | 2018-04-19 06:23:03 (GMT) |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2018-04-19 06:23:03 (GMT) |
| commit | 69524821a87251b7aee966f6e46b3810ff5aaa64 (patch) | |
| tree | c846e23df670d6fa22ecba626d2280efd222cc86 /Tools | |
| parent | b7e1eff8436f6e0c4aac440036092fcf96f82960 (diff) | |
| download | cpython-69524821a87251b7aee966f6e46b3810ff5aaa64.zip cpython-69524821a87251b7aee966f6e46b3810ff5aaa64.tar.gz cpython-69524821a87251b7aee966f6e46b3810ff5aaa64.tar.bz2 | |
bpo-33189: pygettext.py now accepts only literal strings (GH-6364)
as docstrings and translatable strings, and rejects
bytes literals and f-string expressions.
Diffstat (limited to 'Tools')
| -rwxr-xr-x | Tools/i18n/pygettext.py | 14 |
1 files changed, 9 insertions, 5 deletions
diff --git a/Tools/i18n/pygettext.py b/Tools/i18n/pygettext.py index 13d7a64..b46dd33 100755 --- a/Tools/i18n/pygettext.py +++ b/Tools/i18n/pygettext.py @@ -232,6 +232,10 @@ def escape_nonascii(s, encoding): return ''.join(escapes[b] for b in s.encode(encoding)) +def is_literal_string(s): + return s[0] in '\'"' or (s[0] in 'rRuU' and s[1] in '\'"') + + def safe_eval(s): # unwrap quotes, safely return eval(s, {'__builtins__':{}}, {}) @@ -317,8 +321,8 @@ class TokenEater: def __call__(self, ttype, tstring, stup, etup, line): # dispatch ## import token -## print >> sys.stderr, 'ttype:', token.tok_name[ttype], \ -## 'tstring:', tstring +## print('ttype:', token.tok_name[ttype], 'tstring:', tstring, +## file=sys.stderr) self.__state(ttype, tstring, stup[0]) def __waiting(self, ttype, tstring, lineno): @@ -327,7 +331,7 @@ class TokenEater: if opts.docstrings and not opts.nodocstrings.get(self.__curfile): # module docstring? if self.__freshmodule: - if ttype == tokenize.STRING: + if ttype == tokenize.STRING and is_literal_string(tstring): self.__addentry(safe_eval(tstring), lineno, isdocstring=1) self.__freshmodule = 0 elif ttype not in (tokenize.COMMENT, tokenize.NL): @@ -353,7 +357,7 @@ class TokenEater: def __suitedocstring(self, ttype, tstring, lineno): # ignore any intervening noise - if ttype == tokenize.STRING: + if ttype == tokenize.STRING and is_literal_string(tstring): self.__addentry(safe_eval(tstring), lineno, isdocstring=1) self.__state = self.__waiting elif ttype not in (tokenize.NEWLINE, tokenize.INDENT, @@ -378,7 +382,7 @@ class TokenEater: if self.__data: self.__addentry(EMPTYSTRING.join(self.__data)) self.__state = self.__waiting - elif ttype == tokenize.STRING: + elif ttype == tokenize.STRING and is_literal_string(tstring): self.__data.append(safe_eval(tstring)) elif ttype not in [tokenize.COMMENT, token.INDENT, token.DEDENT, token.NEWLINE, tokenize.NL]: |