[Pypgsql-devel] [ pypgsql-Bugs-1055180 ] __unicodeConvert doesn't recognise subclasses of UnicodeType

[SourceForge.net <no...@so...>]

Bugs item #1055180, was opened at 2004-10-27 12:02
Message generated for change (Comment added) made by ghaering
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=116528&aid=1055180&group_id=16528

Category: PgSQL
Group: None
Status: Open
>Resolution: Fixed
Priority: 5
Submitted By: Mike McGavin (jesterzog)
Assigned to: Nobody/Anonymous (nobody)
Summary: __unicodeConvert doesn't recognise subclasses of UnicodeType

Initial Comment:
Cursor.__unicodeConvert() (in PgSQL.py) doesn't encode
unicode strings all the time.

If the object that is passed is of a type that is a
subclass of a UnicodeType, rather than of the
UnicodeType itself, the __unicodeConvert() method
doesn't recognise it as unicode, even though it is.

This appears to be because most of the comparisons
within the __unicodeConvert() method use the "is"
operator to check the object type, which only compares
exact types and doesn't take sub-classing into account.

This bug may be able to be fixed quite easily, by
changing the comparisons (about 7 of them) to use
'isinstance()' instead of the 'is' operator.  ie. 
Instead of:

    if type(obj) is UnicodeType:
        doSomething()

the code should read:

    if isinstance(obj, UnicodeType):
        doSomething()


----------------------------------------------------------------------

>Comment By: Gerhard Häring (ghaering)
Date: 2005-02-27 19:40

Message:
Logged In: YES 
user_id=163326

Fixed in CVS.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=116528&aid=1055180&group_id=16528