diff options
| author | Heikki Linnakangas <heikki.linnakangas@iki.fi> | 2012-07-05 22:16:29 +0300 |
|---|---|---|
| committer | Heikki Linnakangas <heikki.linnakangas@iki.fi> | 2012-07-05 22:32:04 +0300 |
| commit | 138313ebaa985030d75d429c3b3cb7138e62b10f (patch) | |
| tree | 0a894fd991890fee743a7099355b157c527c59c7 /src | |
| parent | b8aca12d77612007ebde22db14e3971ef664d553 (diff) | |
| download | postgresql-138313ebaa985030d75d429c3b3cb7138e62b10f.tar.gz postgresql-138313ebaa985030d75d429c3b3cb7138e62b10f.zip | |
Fix mapping of PostgreSQL encodings to Python encodings.
Windows encodings, "win1252" and so forth, are named differently in Python,
like "cp1252". Also, if the PyUnicode_AsEncodedString() function call fails
for some reason, use a plain ereport(), not a PLy_elog(), to report that
error. That avoids recursion and crash, if PLy_elog() tries to call
PLyUnicode_Bytes() again.
This fixes bug reported by Asif Naeem. Backpatch down to 9.0, before that
plpython didn't even try these conversions.
Jan Urbański, with minor comment improvements by me.
Diffstat (limited to 'src')
| -rw-r--r-- | src/pl/plpython/plpython.c | 69 |
1 files changed, 62 insertions, 7 deletions
diff --git a/src/pl/plpython/plpython.c b/src/pl/plpython/plpython.c index 95efd0b42e6..2f67caba6db 100644 --- a/src/pl/plpython/plpython.c +++ b/src/pl/plpython/plpython.c @@ -4873,16 +4873,71 @@ PLyUnicode_Bytes(PyObject *unicode) const char *serverenc; /* - * Python understands almost all PostgreSQL encoding names, but it doesn't - * know SQL_ASCII. + * Map PostgreSQL encoding to a Python encoding name. */ - if (GetDatabaseEncoding() == PG_SQL_ASCII) - serverenc = "ascii"; - else - serverenc = GetDatabaseEncodingName(); + switch (GetDatabaseEncoding()) + { + case PG_SQL_ASCII: + /* + * Mapping SQL_ASCII to Python's 'ascii' is a bit bogus. Python's + * 'ascii' means true 7-bit only ASCII, while PostgreSQL's + * SQL_ASCII means that anything is allowed, and the system doesn't + * try to interpret the bytes in any way. But not sure what else + * to do, and we haven't heard any complaints... + */ + serverenc = "ascii"; + break; + case PG_WIN1250: + serverenc = "cp1250"; + break; + case PG_WIN1251: + serverenc = "cp1251"; + break; + case PG_WIN1252: + serverenc = "cp1252"; + break; + case PG_WIN1253: + serverenc = "cp1253"; + break; + case PG_WIN1254: + serverenc = "cp1254"; + break; + case PG_WIN1255: + serverenc = "cp1255"; + break; + case PG_WIN1256: + serverenc = "cp1256"; + break; + case PG_WIN1257: + serverenc = "cp1257"; + break; + case PG_WIN1258: + serverenc = "cp1258"; + break; + case PG_WIN866: + serverenc = "cp866"; + break; + case PG_WIN874: + serverenc = "cp874"; + break; + default: + /* Other encodings have the same name in Python. */ + serverenc = GetDatabaseEncodingName(); + break; + } + rv = PyUnicode_AsEncodedString(unicode, serverenc, "strict"); if (rv == NULL) - PLy_elog(ERROR, "could not convert Python Unicode object to PostgreSQL server encoding"); + { + /* + * Use a plain ereport instead of PLy_elog to avoid recursion, if + * the traceback formatting functions try to do unicode to bytes + * conversion again. + */ + ereport(ERROR, + (errcode(ERRCODE_INTERNAL_ERROR), + errmsg("could not convert Python Unicode object to PostgreSQL server encoding"))); + } return rv; } |