char(8) vs char8

D'Arcy J.M. Cain wrote:

This was an unexpected difference between these two types and I wonder
if it was meant to be this way. Previously, a char8 field with the
string 'abc' would return 'abc' as expected. Now, with char(8), I get
back 'abc ' instead. You can see this with my PygreSQL module
or the C interface (which my module uses, of course.) This causes a
lot of my programs to break.

char(x) is the datatype 'bpchar' (blank padded char). Thus it is padded
with spaces to the field width.

Couldn't you use something like "select rtrim(column) from table". This
will trim the spaces off.

Byron

This was an unexpected difference between these two types and I wonder
if it was meant to be this way. Previously, a char8 field with the
string 'abc' would return 'abc' as expected. Now, with char(8), I get
back 'abc ' instead. You can see this with my PygreSQL module
or the C interface (which my module uses, of course.) This causes a
lot of my programs to break.

This is the expected behavior for the char() datatype. If you have other variable
length fields in the table, you can simply use varchar() which does not blank pad
to the specified length. If you don't have other variable length fields then you
loose some performance if you switch to varchar().

I have made a quick change to my Python module to handle this. Should
I clean it up or can I expect the behaviour to go back the way it was?

No, it will stay.

Andreas

This was an unexpected difference between these two types and I wonder
if it was meant to be this way. Previously, a char8 field with the
string 'abc' would return 'abc' as expected. Now, with char(8), I get
back 'abc ' instead. You can see this with my PygreSQL module
or the C interface (which my module uses, of course.) This causes a
lot of my programs to break.

I have made a quick change to my Python module to handle this. Should
I clean it up or can I expect the behaviour to go back the way it was?

The behavior you want is varchar() rather than char().

- Tom

Thus spake Byron Nikolaidis

string 'abc' would return 'abc' as expected. Now, with char(8), I get
back 'abc ' instead. You can see this with my PygreSQL module

Couldn't you use something like "select rtrim(column) from table". This
will trim the spaces off.

It wouldn't be as convenient as this example.

import pg
for d in pg.connect('database').query('select * from table').dictresult():
print "Num: %(field1)3d, '%(field2)s'" % d

-- 
D'Arcy J.M. Cain <darcy@{druid|vex}.net>   |  Democracy is three wolves
http://www.druid.net/darcy/                |  and a sheep voting on
+1 416 424 2871     (DoD#0082)    (eNTP)   |  what's for dinner.

Thus spake Thomas G. Lockhart

string 'abc' would return 'abc' as expected. Now, with char(8), I get
back 'abc ' instead. You can see this with my PygreSQL module

The behavior you want is varchar() rather than char().

Right. Dump and reload time I guess. >:-/

-- 
D'Arcy J.M. Cain <darcy@{druid|vex}.net>   |  Democracy is three wolves
http://www.druid.net/darcy/                |  and a sheep voting on
+1 416 424 2871     (DoD#0082)    (eNTP)   |  what's for dinner.