PostgreSQL

Hi,

I'm resending this because it seems that it did not reach the mailing list:




Hi all again,

I noticed two small issues with PostgreSQL package:

1. PGConnection class >> fieldConverterClass returns PGFieldConverter
class but the file is not in the package.xml.
2. in the definition of PGFieldConverter >> writeQuotedTime, the
variable aDateTime is used; may it should be aTime?

Anyway, how do I make PackageLoader >> fileInPackage to fail if a class
definition is not found?

Thanks for your time,
Stefan


_______________________________________________
help-smalltalk mailing list
[hidden email]
http://lists.gnu.org/mailman/listinfo/help-smalltalk

On 11/25/2009 10:02 AM, Stefan Izota wrote:
>
>
> 1. PGConnection class >> fieldConverterClass returns PGFieldConverter
> class but the file is not in the package.xml.
> 2. in the definition of PGFieldConverter >> writeQuotedTime, the
> variable aDateTime is used; may it should be aTime?

Does the attached patch fix it?

> Anyway, how do I make PackageLoader >> fileInPackage to fail if a class
> definition is not found?

This is currently not done because some packages have undefined class
references for Squeak compatibility.  The right way to do it would be to
wrap the file in with a check that "Undeclared keys" does not change.

It should warn, not fail anyway.

Paolo

_______________________________________________
help-smalltalk mailing list
[hidden email]
http://lists.gnu.org/mailman/listinfo/help-smalltalk

Paolo Bonzini wrote:
> On 11/25/2009 10:02 AM, Stefan Izota wrote:
>>
>>
>> 1. PGConnection class >> fieldConverterClass returns PGFieldConverter
>> class but the file is not in the package.xml.
>> 2. in the definition of PGFieldConverter >> writeQuotedTime, the
>> variable aDateTime is used; may it should be aTime?
>
> Does the attached patch fix it?

Yes, it does. Thanks.


_______________________________________________
help-smalltalk mailing list
[hidden email]
http://lists.gnu.org/mailman/listinfo/help-smalltalk

diff --git a/packages/dbi/ResultSet.st b/packages/dbi/ResultSet.st
index e20e824..7dd2af8 100644
--- a/packages/dbi/ResultSet.st
+++ b/packages/dbi/ResultSet.st
@@ -147,11 +147,13 @@ case I only hold the number of rows affected.'>
     printOn: aStream [
  "Print a representation of the receiver on aStream."
  <category: 'printing'>
- self rows do:
- [:row |
- row printOn: aStream.
- aStream nl]
-    ]
+ self isSelect ifTrue: [
+ self rows do:
+ [:row |
+ row printOn: aStream.
+ aStream nl]
+ ]
+ ]
 
     statement [
  "Return the Statement, if any, that generated the result set."

_______________________________________________
help-smalltalk mailing list
[hidden email]
http://lists.gnu.org/mailman/listinfo/help-smalltalk

On 11/25/2009 01:04 PM, Stefan Izota wrote:
>
>
> I attached a small patch that fixes printing a dml resultset to a stream
> (like executing an UPDATE in REPL).

I committed a variant that prints something also for a DML resultset.

Paolo


_______________________________________________
help-smalltalk mailing list
[hidden email]
http://lists.gnu.org/mailman/listinfo/help-smalltalk

Sorry to bother again, but shouldn't it be

self isSelect ifFalse: [ ^super printOn: aStream ].

instead of

self isSelect ifFalse: [ ^super printOn: aString ].

Paolo Bonzini wrote:


_______________________________________________
help-smalltalk mailing list
[hidden email]
http://lists.gnu.org/mailman/listinfo/help-smalltalk

On 11/25/2009 02:02 PM, Stefan Izota wrote:
> Sorry to bother again, but shouldn't it be

Should have been "sorry to point out a stupid mistake". :-)

> self isSelect ifFalse: [ ^super printOn: aStream ].
>
> instead of
>
> self isSelect ifFalse: [ ^super printOn: aString ].

Thanks.

Paolo


_______________________________________________
help-smalltalk mailing list
[hidden email]
http://lists.gnu.org/mailman/listinfo/help-smalltalk