[3.11] gh-108590: Fix sqlite3.iterdump for invalid Unicode in TEXT columns (GH-108657) (#108674)

(cherry picked from commit 400a1cebc7)
This commit is contained in:
Corvin 2023-08-30 06:29:33 -04:00 committed by GitHub
parent a2c05a414d
commit 5a6d1238bb
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
3 changed files with 41 additions and 2 deletions

View file

@ -7,6 +7,28 @@
# future enhancements, you should normally quote any identifier that
# is an English language word, even if you do not have to."
from contextlib import contextmanager
def _force_decode(bs, *args, **kwargs):
# gh-108590: Don't fail if the database contains invalid Unicode data.
try:
return bs.decode(*args, **kwargs)
except UnicodeDecodeError:
return "".join([chr(c) for c in bs])
@contextmanager
def _text_factory(con, factory):
saved_factory = con.text_factory
con.text_factory = factory
try:
yield
finally:
con.text_factory = saved_factory
def _iterdump(connection):
"""
Returns an iterator to the dump of the database in an SQL text format.
@ -63,8 +85,9 @@ def _iterdump(connection):
table_name_ident,
",".join("""'||quote("{0}")||'""".format(col.replace('"', '""')) for col in column_names))
query_res = cu.execute(q)
for row in query_res:
yield("{0};".format(row[0]))
with _text_factory(connection, bytes):
for row in query_res:
yield("{0};".format(_force_decode(row[0])))
# Now when the type is 'index', 'trigger', or 'view'
q = """