"question" label here, because as usual I can see why this is happening, it is just another surprise (note I differentiate between "surprise" and "expected" :) )
In SQLAlchemy, we have an outputtypehandler that in one particular case, is essentially this:
def handler(cursor, name, default_type, size, precision, scale):
if precision and scale == 0:
return cursor.var(
cx_Oracle.NATIVE_INT,
255,
arraysize=cursor.arraysize,
outconverter=None,
)
else:
return None
it seems well intentioned enough until one of my users decided to stick 9999999999999999999 into it, producing the error:
cx_Oracle.DatabaseError: ORA-01455: converting column overflows integer datatype
Apparently the max int size for NATIVE_INT is 32 bits, so...maybe if you can confirm that, and what kind of datatype is coming back if I dont do that. What's the purpose of NATIVE_INT exactly, and do I have to worry about this on the input side as well (though this user hasn't tried that yet :) )
I can either not return a var at all, or I can do this:
return cursor.var(
int, 255, arraysize=cursor.arraysize, outconverter=None
)
and both of those seem to work great for everything I'm testing plus these new tests. In #208 it seemed like I had decided to use NATIVE_INT but that was on the input side. On the output side can you confirm I should set "int" on the var and/or not return the var?
RetroSearch is an open source project built by @garambo | Open a GitHub Issue
Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo
HTML:
3.2
| Encoding:
UTF-8
| Version:
0.7.4