sqlobject-cvs — CVS notification

[SQL-CVS] [ sqlobject-Bugs-1463974 ] Column name collides with internal method

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

Bugs item #1463974, was opened at 2006-04-04 12:20
Message generated for change (Tracker Item Submitted) made by Item Submitter
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540672&aid=1463974&group_id=74338

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: General
Group: SQLObject release (specify)
Status: Open
Resolution: None
Priority: 5
Submitted By: Patrick Coleman (ptrck)
Assigned to: Nobody/Anonymous (nobody)
Summary: Column name collides with internal method

Initial Comment:
Hi,

My database has a column 'expire' in a certain table.
It appears that when SQLObject calls expire() (line
778, dbconnection.py) on a cache object from that
table, rather than calling the expire method to expire
the object from cache it actually returns the value
from the 'expire' column, which being a Long of course
isn't callable, resulting in the attached exception.

I propose that internal methods used by SQLObject use
names that are unlikely to collide with column names
(perhaps by prepending __SQLObject__ or similar), and
that these internal names be documented so database
designers can avoid using names that may collide.

SQLObject is SQLObject-0.7.0-py2.4, as shipped with
Turbogears 0.8a3-py2.4.

Thanks,
Patrick
--
http://www.labyrinthdata.net.au

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

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540672&aid=1463974&group_id=74338

[SQL-CVS] [ sqlobject-Bugs-1461361 ] orderBy: datetime.datetime cannot be compared with None

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

Bugs item #1461361, was opened at 2006-03-30 04:42
Message generated for change (Tracker Item Submitted) made by Item Submitter
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540672&aid=1461361&group_id=74338

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: Postgres
Group: None
Status: Open
Resolution: None
Priority: 5
Submitted By: Nobody/Anonymous (nobody)
Assigned to: Nobody/Anonymous (nobody)
Summary: orderBy: datetime.datetime cannot be compared with None

Initial Comment:
When datetime.datetime is used for DateTimeCols:
Using defaultOrder with a DateTimeCol fails when there
is an empty DateTimeCol (None). This does not happen
with the mx modules.

"can't compare datetime.time to NoneType"

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

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540672&aid=1461361&group_id=74338

[SQL-CVS] r1675 - SQLObject/branches/0.7-bugfix/sqlobject/manager

[<sub...@co...>]

Author: ianb
Date: 2006-03-29 11:17:33 -0700 (Wed, 29 Mar 2006)
New Revision: 1675

Modified:
   SQLObject/branches/0.7-bugfix/sqlobject/manager/command.py
Log:
Fix problem with sqlobject-admin and the changing return value of createTableSQL

Modified: SQLObject/branches/0.7-bugfix/sqlobject/manager/command.py
===================================================================
--- SQLObject/branches/0.7-bugfix/sqlobject/manager/command.py	2006-03-28 12:23:39 UTC (rev 1674)
+++ SQLObject/branches/0.7-bugfix/sqlobject/manager/command.py	2006-03-29 18:17:33 UTC (rev 1675)
@@ -836,7 +836,15 @@
                               + '_' + dbName + '.sql')
             if sim:
                 continue
-            create, constraints = cls.createTableSQL()
+            # @@: This is a little hacky, but sometimes I'm getting string
+            # results here, and sometimes tuples (apparently, if the code is 
+            # meant to work?), so we'll allow for both
+            result = cls.createTableSQL()
+            if isinstance(result, basestring):
+                create = result
+                constraints = []
+            else:
+                create, constraints = result
             if constraints:
                 constraints = '\n-- Constraints:\n%s\n' % (
                     '\n'.join(constraints))

[SQL-CVS] [ sqlobject-Bugs-1460100 ] orderBy: datetime.datetime cannot be compared with None

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

Bugs item #1460100, was opened at 2006-03-28 07:57
Message generated for change (Tracker Item Submitted) made by Item Submitter
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540672&aid=1460100&group_id=74338

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: Postgres
Group: None
Status: Open
Resolution: None
Priority: 5
Submitted By: Nobody/Anonymous (nobody)
Assigned to: Nobody/Anonymous (nobody)
Summary: orderBy: datetime.datetime cannot be compared with None

Initial Comment:
When datetime.datetime is used for DateTimeCols:
Using defaultOrder with a DateTimeCol fails when there
is an empty DateTimeCol (None). This does not happen
with the mx modules.

"can't compare datetime.time to NoneType"

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

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540672&aid=1460100&group_id=74338

[SQL-CVS] [ sqlobject-Patches-1458925 ] Fix for bug 1458595 (destroySelf in Transaction = deadlock)

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

Patches item #1458925, was opened at 2006-03-26 14:35
Message generated for change (Comment added) made by cpisto
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540674&aid=1458925&group_id=74338

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
Status: Open
Resolution: Invalid
Priority: 5
Submitted By: Cody Pisto (cpisto)
Assigned to: Oleg Broytmann (phd)
Summary: Fix for bug 1458595 (destroySelf in Transaction = deadlock)

Initial Comment:
This patch fixes bug 1458595, in which actions that
resulted in Transaction._SO_delete being called
resulted in an additional database connection being
opened, causing a deadlock while multiple connections
await on eachothers commit.

This patch is against svn trunk,
r1668. (dbconnection.py)

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

>Comment By: Cody Pisto (cpisto)
Date: 2006-03-28 07:32

Message:
Logged In: YES 
user_id=118227

sorry, lost track of time, patch will follow today ;)

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

Comment By: Oleg Broytmann (phd)
Date: 2006-03-28 05:21

Message:
Logged In: YES 
user_id=4799

Problems? :)

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

Comment By: Cody Pisto (cpisto)
Date: 2006-03-27 08:45

Message:
Logged In: YES 
user_id=118227

Sure, ill refactor-
Revised patch will follow this evening.

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

Comment By: Oleg Broytmann (phd)
Date: 2006-03-27 07:46

Message:
Logged In: YES 
user_id=4799

Your patch duplicates the query from DBAPI._SO_delete() to
Transaction._SO_delete(). Would you mind to refactor your
patch - split DBAPI._SO_delete() into two methods - one to
generate a query string, and another to execute the query;
then use the first method to generate a query in the
Transaction._SO_delete()?

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

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540674&aid=1458925&group_id=74338

[SQL-CVS] r1674 - in home/phd/SQLObject/paramstyles/sqlobject: . tests

[<sub...@co...>]

Author: phd
Date: 2006-03-28 05:23:39 -0700 (Tue, 28 Mar 2006)
New Revision: 1674

Modified:
   home/phd/SQLObject/paramstyles/sqlobject/sqlbuilder.py
   home/phd/SQLObject/paramstyles/sqlobject/tests/test_converters.py
Log:
Merged patches from the revisions 1671:1673 from the trunkthe patch 1450568: faster sqlbuilder.Insert.

Modified: home/phd/SQLObject/paramstyles/sqlobject/sqlbuilder.py
===================================================================
--- home/phd/SQLObject/paramstyles/sqlobject/sqlbuilder.py	2006-03-28 12:22:48 UTC (rev 1673)
+++ home/phd/SQLObject/paramstyles/sqlobject/sqlbuilder.py	2006-03-28 12:23:39 UTC (rev 1674)
@@ -552,20 +552,18 @@
             allowNonDict = False
         if template is not NoDefault:
             insert += " (%s)" % ", ".join(template)
-        first = True
         insert += " VALUES "
+        listToJoin = []
+        listToJoin_app = listToJoin.append
         for value in self.valueList:
-            if first:
-                first = False
-            else:
-                insert += ", "
             if type(value) is type({}):
                 if template is NoDefault:
                     raise TypeError, "You can't mix non-dictionaries with dictionaries in an INSERT if you don't provide a template (%s)" % repr(value)
                 value = dictToList(template, value)
             elif not allowNonDict:
                 raise TypeError, "You can't mix non-dictionaries with dictionaries in an INSERT if you don't provide a template (%s)" % repr(value)
-            insert += "(%s)" % ", ".join([sqlrepr(v, db) for v in value])
+            listToJoin_app("(%s)" % ", ".join([sqlrepr(v, db) for v in value]))
+        insert = "%s%s" % (insert, ", ".join(listToJoin))
         return insert
 
     def __sqllist__(self):

Modified: home/phd/SQLObject/paramstyles/sqlobject/tests/test_converters.py
===================================================================
--- home/phd/SQLObject/paramstyles/sqlobject/tests/test_converters.py	2006-03-28 12:22:48 UTC (rev 1673)
+++ home/phd/SQLObject/paramstyles/sqlobject/tests/test_converters.py	2006-03-28 12:23:39 UTC (rev 1674)
@@ -112,9 +112,39 @@
     assert sqlrepr(instance, 'mysql') == "SELECT 'test'"
 
 def test_insert():
+    # Single column, no keyword arguments.
     instance = Insert('test', [('test',)])
     assert sqlrepr(instance, 'mysql') == "INSERT INTO test VALUES ('test')"
 
+    # Multiple columns, no keyword arguments.
+    instance2 = Insert('test', [('1st', '2nd', '3th', '4th')])
+    assert sqlrepr(instance2, 'postgres') == "INSERT INTO test VALUES ('1st', '2nd', '3th', '4th')"
+
+    # Multiple rows, multiple columns, "valueList" keyword argument.
+    instance3 = Insert('test', valueList=[('a1', 'b1'), ('a2', 'b2'), ('a3', 'b3')])
+    assert sqlrepr(instance3, 'sqlite') == "INSERT INTO test VALUES ('a1', 'b1'), ('a2', 'b2'), ('a3', 'b3')"
+
+    # Multiple columns, "values" keyword argument.
+    instance4 = Insert('test', values=('v1', 'v2', 'v3'))
+    assert sqlrepr(instance4, 'mysql') == "INSERT INTO test VALUES ('v1', 'v2', 'v3')"
+
+    # Single column, "valueList" keyword argument.
+    instance5 = Insert('test', valueList=[('v1',)])
+    assert sqlrepr(instance5, 'mysql') == "INSERT INTO test VALUES ('v1')"
+
+    # Multiple rows, Multiple columns, template.
+    instance6 = Insert('test', valueList=[('a1', 'b1'), ('a2', 'b2')], template=['col1', 'col2'])
+    assert sqlrepr(instance6, 'mysql') == "INSERT INTO test (col1, col2) VALUES ('a1', 'b1'), ('a2', 'b2')"
+
+    # Multiple columns, implicit template (dictionary value).
+    instance7 = Insert('test', valueList=[{'col1': 'a1', 'col2': 'b1'}])
+    assert sqlrepr(instance7, 'mysql') == "INSERT INTO test (col2, col1) VALUES ('b1', 'a1')"
+
+    # Multiple rows, Multiple columns, implicit template.
+    instance8 = Insert('test', valueList=[{'col1': 'a1', 'col2': 'b1'},
+                                        {'col1': 'a2', 'col2': 'b2'}])
+    assert sqlrepr(instance8, 'mysql') == "INSERT INTO test (col2, col1) VALUES ('b1', 'a1'), ('b2', 'a2')"
+
 def test_update():
     instance = Update('test', {'test':'test'})
     assert sqlrepr(instance, 'mysql') == "UPDATE test SET test='test'"

[SQL-CVS] r1673 - in SQLObject/branches/0.7-bugfix/sqlobject: . tests

[<sub...@co...>]

Author: phd
Date: 2006-03-28 05:22:48 -0700 (Tue, 28 Mar 2006)
New Revision: 1673

Modified:
   SQLObject/branches/0.7-bugfix/sqlobject/sqlbuilder.py
   SQLObject/branches/0.7-bugfix/sqlobject/tests/test_converters.py
Log:
Applied the patch 1450568: faster sqlbuilder.Insert.


Modified: SQLObject/branches/0.7-bugfix/sqlobject/sqlbuilder.py
===================================================================
--- SQLObject/branches/0.7-bugfix/sqlobject/sqlbuilder.py	2006-03-28 12:17:30 UTC (rev 1672)
+++ SQLObject/branches/0.7-bugfix/sqlobject/sqlbuilder.py	2006-03-28 12:22:48 UTC (rev 1673)
@@ -483,20 +483,18 @@
             allowNonDict = False
         if template is not NoDefault:
             insert += " (%s)" % ", ".join(template)
-        first = True
         insert += " VALUES "
+        listToJoin = []
+        listToJoin_app = listToJoin.append
         for value in self.valueList:
-            if first:
-                first = False
-            else:
-                insert += ", "
             if type(value) is type({}):
                 if template is NoDefault:
                     raise TypeError, "You can't mix non-dictionaries with dictionaries in an INSERT if you don't provide a template (%s)" % repr(value)
                 value = dictToList(template, value)
             elif not allowNonDict:
                 raise TypeError, "You can't mix non-dictionaries with dictionaries in an INSERT if you don't provide a template (%s)" % repr(value)
-            insert += "(%s)" % ", ".join([sqlrepr(v, db) for v in value])
+            listToJoin_app("(%s)" % ", ".join([sqlrepr(v, db) for v in value]))
+        insert = "%s%s" % (insert, ", ".join(listToJoin))
         return insert
 
 registerConverter(Insert, SQLExprConverter)

Modified: SQLObject/branches/0.7-bugfix/sqlobject/tests/test_converters.py
===================================================================
--- SQLObject/branches/0.7-bugfix/sqlobject/tests/test_converters.py	2006-03-28 12:17:30 UTC (rev 1672)
+++ SQLObject/branches/0.7-bugfix/sqlobject/tests/test_converters.py	2006-03-28 12:22:48 UTC (rev 1673)
@@ -112,9 +112,39 @@
     assert sqlrepr(instance, 'mysql') == "SELECT 'test'"
 
 def test_insert():
+    # Single column, no keyword arguments.
     instance = Insert('test', [('test',)])
     assert sqlrepr(instance, 'mysql') == "INSERT INTO test VALUES ('test')"
 
+    # Multiple columns, no keyword arguments.
+    instance2 = Insert('test', [('1st', '2nd', '3th', '4th')])
+    assert sqlrepr(instance2, 'postgres') == "INSERT INTO test VALUES ('1st', '2nd', '3th', '4th')"
+
+    # Multiple rows, multiple columns, "valueList" keyword argument.
+    instance3 = Insert('test', valueList=[('a1', 'b1'), ('a2', 'b2'), ('a3', 'b3')])
+    assert sqlrepr(instance3, 'sqlite') == "INSERT INTO test VALUES ('a1', 'b1'), ('a2', 'b2'), ('a3', 'b3')"
+
+    # Multiple columns, "values" keyword argument.
+    instance4 = Insert('test', values=('v1', 'v2', 'v3'))
+    assert sqlrepr(instance4, 'mysql') == "INSERT INTO test VALUES ('v1', 'v2', 'v3')"
+
+    # Single column, "valueList" keyword argument.
+    instance5 = Insert('test', valueList=[('v1',)])
+    assert sqlrepr(instance5, 'mysql') == "INSERT INTO test VALUES ('v1')"
+
+    # Multiple rows, Multiple columns, template.
+    instance6 = Insert('test', valueList=[('a1', 'b1'), ('a2', 'b2')], template=['col1', 'col2'])
+    assert sqlrepr(instance6, 'mysql') == "INSERT INTO test (col1, col2) VALUES ('a1', 'b1'), ('a2', 'b2')"
+
+    # Multiple columns, implicit template (dictionary value).
+    instance7 = Insert('test', valueList=[{'col1': 'a1', 'col2': 'b1'}])
+    assert sqlrepr(instance7, 'mysql') == "INSERT INTO test (col2, col1) VALUES ('b1', 'a1')"
+
+    # Multiple rows, Multiple columns, implicit template.
+    instance8 = Insert('test', valueList=[{'col1': 'a1', 'col2': 'b1'},
+                                        {'col1': 'a2', 'col2': 'b2'}])
+    assert sqlrepr(instance8, 'mysql') == "INSERT INTO test (col2, col1) VALUES ('b1', 'a1'), ('b2', 'a2')"
+
 def test_update():
     instance = Update('test', {'test':'test'})
     assert sqlrepr(instance, 'mysql') == "UPDATE test SET test='test'"

[SQL-CVS] [ sqlobject-Patches-1458925 ] Fix for bug 1458595 (destroySelf in Transaction = deadlock)

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

Patches item #1458925, was opened at 2006-03-27 01:35
Message generated for change (Comment added) made by phd
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540674&aid=1458925&group_id=74338

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
Status: Open
>Resolution: Invalid
Priority: 5
Submitted By: Cody Pisto (cpisto)
>Assigned to: Oleg Broytmann (phd)
Summary: Fix for bug 1458595 (destroySelf in Transaction = deadlock)

Initial Comment:
This patch fixes bug 1458595, in which actions that
resulted in Transaction._SO_delete being called
resulted in an additional database connection being
opened, causing a deadlock while multiple connections
await on eachothers commit.

This patch is against svn trunk,
r1668. (dbconnection.py)

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

>Comment By: Oleg Broytmann (phd)
Date: 2006-03-28 16:21

Message:
Logged In: YES 
user_id=4799

Problems? :)

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

Comment By: Cody Pisto (cpisto)
Date: 2006-03-27 19:45

Message:
Logged In: YES 
user_id=118227

Sure, ill refactor-
Revised patch will follow this evening.

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

Comment By: Oleg Broytmann (phd)
Date: 2006-03-27 18:46

Message:
Logged In: YES 
user_id=4799

Your patch duplicates the query from DBAPI._SO_delete() to
Transaction._SO_delete(). Would you mind to refactor your
patch - split DBAPI._SO_delete() into two methods - one to
generate a query string, and another to execute the query;
then use the first method to generate a query in the
Transaction._SO_delete()?

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

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540674&aid=1458925&group_id=74338

[SQL-CVS] [ sqlobject-Patches-1458380 ] Let admin create tables with no foreign keys first

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

Patches item #1458380, was opened at 2006-03-25 21:08
Message generated for change (Comment added) made by phd
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540674&aid=1458380&group_id=74338

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
>Status: Closed
>Resolution: Fixed
Priority: 5
Submitted By: Dennis Brakhane (dennis)
Assigned to: Nobody/Anonymous (nobody)
Summary: Let admin create tables with no foreign keys first

Initial Comment:
Suppose I have the following model:

class A(SQLObject):
    foo = StringCol()
    b = ForeignKey("B")

class B(SQLObject):
    bar = StringCol()

If I now let sqlobject-admin create the tables for me,
the order in with these are created is not defined.
If table A is created before table B, PostgreSQL will fail:

CREATE TABLE a (
    id SERIAL PRIMARY KEY,
    foo TEXT,
    b_id INT, CONSTRAINT b_id_exists FOREIGN KEY (b_id)
REFERENCES b (id)
)

Fails with "Relation >b< does not exist"

So, I've patched manager/command.py to create tables in
the order of classes with fewest foreign keys created
first.

This might not catch all cases, but it "works for me(TM)"

Patch is attached


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

>Comment By: Oleg Broytmann (phd)
Date: 2006-03-28 16:19

Message:
Logged In: YES 
user_id=4799

Ok, then.

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

Comment By: Dennis Brakhane (dennis)
Date: 2006-03-27 19:47

Message:
Logged In: YES 
user_id=26932

> I think I'd be better to use this feature instead of
> unreliable "fewest foreign keys". Would you mind
> recreating your patch?

Erm... What exactly should I recreate? It seems like
manager.py already has this functionality built-in.

I'm using TurboGears svn and didn't realize the external
reference to SQLObject's SVN was to the "0.7-bugfix" branch
and not the trunk. So I think the bug is already fixed.


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

Comment By: Oleg Broytmann (phd)
Date: 2006-03-27 18:32

Message:
Logged In: YES 
user_id=4799

.createTable() in the trunk has got a feature - it can delay
creating foreign keys and instead return SQL query to create
them later. So one can create all tables, collects these
delayed SQL queries and run them later, when all tables are
in place. Look in the FAQ:
http://svn.colorstudy.com/SQLObject/docs/FAQ.txt , question
"Mutually referencing tables".

I think I'd be better to use this feature instead of
unreliable "fewest foreign keys". Would you mind recreating
your patch?

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

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540674&aid=1458380&group_id=74338

[SQL-CVS] [ sqlobject-Patches-1450568 ] faster sqlbuilder.Insert

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

Patches item #1450568, was opened at 2006-03-15 20:58
Message generated for change (Comment added) made by phd
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540674&aid=1450568&group_id=74338

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
>Status: Closed
>Resolution: Accepted
Priority: 5
Submitted By: Davide Alberani (alberanid)
>Assigned to: Oleg Broytmann (phd)
Summary: faster sqlbuilder.Insert

Initial Comment:
In the attachment, a very simple patch to improve
performances (about 12 times faster, with a list of
~20000 items, tested with python2.3) of the
sqlbuilder.Insert
class, when a very long valueList list is provided.

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

>Comment By: Oleg Broytmann (phd)
Date: 2006-03-28 16:18

Message:
Logged In: YES 
user_id=4799

Applied in the revision 1672 in the trunk. Thank you!

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

Comment By: Davide Alberani (alberanid)
Date: 2006-03-28 00:50

Message:
Logged In: YES 
user_id=170840

I've extended the test_insert() function of the
test_converters.py file; you can find the new function in
the attached "insert_tests.py" file.

Now some common conditions are tested.

It seems to works correctly, and the patch was really
simple, but better safe than sorry. :-)

If you need something else, feel free to ask.

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

Comment By: Oleg Broytmann (phd)
Date: 2006-03-27 18:37

Message:
Logged In: YES 
user_id=4799

Please add some tests so I can run them before and after
applying the patch and verify that the patch doesn't break
Insert().

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

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540674&aid=1450568&group_id=74338

[SQL-CVS] r1672 - in SQLObject/trunk/sqlobject: . tests

[<sub...@co...>]

Author: phd
Date: 2006-03-28 05:17:30 -0700 (Tue, 28 Mar 2006)
New Revision: 1672

Modified:
   SQLObject/trunk/sqlobject/sqlbuilder.py
   SQLObject/trunk/sqlobject/tests/test_converters.py
Log:
Applied the patch 1450568: faster sqlbuilder.Insert.


Modified: SQLObject/trunk/sqlobject/sqlbuilder.py
===================================================================
--- SQLObject/trunk/sqlobject/sqlbuilder.py	2006-03-21 18:34:42 UTC (rev 1671)
+++ SQLObject/trunk/sqlobject/sqlbuilder.py	2006-03-28 12:17:30 UTC (rev 1672)
@@ -486,20 +486,18 @@
             allowNonDict = False
         if template is not NoDefault:
             insert += " (%s)" % ", ".join(template)
-        first = True
         insert += " VALUES "
+        listToJoin = []
+        listToJoin_app = listToJoin.append
         for value in self.valueList:
-            if first:
-                first = False
-            else:
-                insert += ", "
             if type(value) is type({}):
                 if template is NoDefault:
                     raise TypeError, "You can't mix non-dictionaries with dictionaries in an INSERT if you don't provide a template (%s)" % repr(value)
                 value = dictToList(template, value)
             elif not allowNonDict:
                 raise TypeError, "You can't mix non-dictionaries with dictionaries in an INSERT if you don't provide a template (%s)" % repr(value)
-            insert += "(%s)" % ", ".join([sqlrepr(v, db) for v in value])
+            listToJoin_app("(%s)" % ", ".join([sqlrepr(v, db) for v in value]))
+        insert = "%s%s" % (insert, ", ".join(listToJoin))
         return insert
 
 registerConverter(Insert, SQLExprConverter)

Modified: SQLObject/trunk/sqlobject/tests/test_converters.py
===================================================================
--- SQLObject/trunk/sqlobject/tests/test_converters.py	2006-03-21 18:34:42 UTC (rev 1671)
+++ SQLObject/trunk/sqlobject/tests/test_converters.py	2006-03-28 12:17:30 UTC (rev 1672)
@@ -112,9 +112,39 @@
     assert sqlrepr(instance, 'mysql') == "SELECT 'test'"
 
 def test_insert():
+    # Single column, no keyword arguments.
     instance = Insert('test', [('test',)])
     assert sqlrepr(instance, 'mysql') == "INSERT INTO test VALUES ('test')"
 
+    # Multiple columns, no keyword arguments.
+    instance2 = Insert('test', [('1st', '2nd', '3th', '4th')])
+    assert sqlrepr(instance2, 'postgres') == "INSERT INTO test VALUES ('1st', '2nd', '3th', '4th')"
+
+    # Multiple rows, multiple columns, "valueList" keyword argument.
+    instance3 = Insert('test', valueList=[('a1', 'b1'), ('a2', 'b2'), ('a3', 'b3')])
+    assert sqlrepr(instance3, 'sqlite') == "INSERT INTO test VALUES ('a1', 'b1'), ('a2', 'b2'), ('a3', 'b3')"
+
+    # Multiple columns, "values" keyword argument.
+    instance4 = Insert('test', values=('v1', 'v2', 'v3'))
+    assert sqlrepr(instance4, 'mysql') == "INSERT INTO test VALUES ('v1', 'v2', 'v3')"
+
+    # Single column, "valueList" keyword argument.
+    instance5 = Insert('test', valueList=[('v1',)])
+    assert sqlrepr(instance5, 'mysql') == "INSERT INTO test VALUES ('v1')"
+
+    # Multiple rows, Multiple columns, template.
+    instance6 = Insert('test', valueList=[('a1', 'b1'), ('a2', 'b2')], template=['col1', 'col2'])
+    assert sqlrepr(instance6, 'mysql') == "INSERT INTO test (col1, col2) VALUES ('a1', 'b1'), ('a2', 'b2')"
+
+    # Multiple columns, implicit template (dictionary value).
+    instance7 = Insert('test', valueList=[{'col1': 'a1', 'col2': 'b1'}])
+    assert sqlrepr(instance7, 'mysql') == "INSERT INTO test (col2, col1) VALUES ('b1', 'a1')"
+
+    # Multiple rows, Multiple columns, implicit template.
+    instance8 = Insert('test', valueList=[{'col1': 'a1', 'col2': 'b1'},
+                                        {'col1': 'a2', 'col2': 'b2'}])
+    assert sqlrepr(instance8, 'mysql') == "INSERT INTO test (col2, col1) VALUES ('b1', 'a1'), ('b2', 'a2')"
+
 def test_update():
     instance = Update('test', {'test':'test'})
     assert sqlrepr(instance, 'mysql') == "UPDATE test SET test='test'"

[SQL-CVS] [ sqlobject-Patches-1450568 ] faster sqlbuilder.Insert

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

Patches item #1450568, was opened at 2006-03-15 18:58
Message generated for change (Comment added) made by alberanid
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540674&aid=1450568&group_id=74338

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
Status: Open
Resolution: None
Priority: 5
Submitted By: Davide Alberani (alberanid)
Assigned to: Nobody/Anonymous (nobody)
Summary: faster sqlbuilder.Insert

Initial Comment:
In the attachment, a very simple patch to improve
performances (about 12 times faster, with a list of
~20000 items, tested with python2.3) of the
sqlbuilder.Insert
class, when a very long valueList list is provided.

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

>Comment By: Davide Alberani (alberanid)
Date: 2006-03-27 22:50

Message:
Logged In: YES 
user_id=170840

I've extended the test_insert() function of the
test_converters.py file; you can find the new function in
the attached "insert_tests.py" file.

Now some common conditions are tested.

It seems to works correctly, and the patch was really
simple, but better safe than sorry. :-)

If you need something else, feel free to ask.

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

Comment By: Oleg Broytmann (phd)
Date: 2006-03-27 16:37

Message:
Logged In: YES 
user_id=4799

Please add some tests so I can run them before and after
applying the patch and verify that the patch doesn't break
Insert().

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

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540674&aid=1450568&group_id=74338

[SQL-CVS] [ sqlobject-Bugs-1459482 ] Can't access column called 'class' in SQLObject

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

Bugs item #1459482, was opened at 2006-03-27 09:51
Message generated for change (Comment added) made by nobody
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540672&aid=1459482&group_id=74338

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: General
Group: SQLObject release (specify)
Status: Open
Resolution: None
Priority: 5
Submitted By: Nobody/Anonymous (nobody)
Assigned to: Nobody/Anonymous (nobody)
Summary: Can't access column called 'class' in SQLObject

Initial Comment:
A syntax error is generated when a column named "class"
is accessed using the SQLObject class interface in
SQLObject 0.7.

Example:

The class AltJEDic contains a column named "CLASS" that
should be mapped to the python object AltJEDic.class

class AltJEDic(SQLObject):
    class sqlmeta:
        fromDatabase = True
        #idName = "sqlid"
        table = "ALTJEDIC0111"
        style = HinokiStyle()

However, when I try to use AltJEDic.class in a function
call, it fails with the following error:

# python verify-wordnet.py
  File "verify-wordnet.py", line 190
    if isSynonymRelGT(left_altjdic.class,
right_altjdic.class):
                                       ^
SyntaxError: invalid syntax

It seems that this is interfering with Python keywords.
Do we need to add a restriction on column (or SQLObject
class member names) or can this be worked around?

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

Comment By: Nobody/Anonymous (nobody)
Date: 2006-03-27 09:57

Message:
Logged In: NO 

I forgot to add some context. I get left_altjdic and
right_altjdic by taking the first results from selectBy()
queries. Ex:

left_altjdics = AltJDic.selectBy(entry=left)
left_altjdic = left_altjdics[0]
try:
    right_altjdics = AltJDic.selectBy(entry=right)
    right_altjdic = right_altjdics[0]
    if isSynonymRelGT(left_altjdic.class, right_altjdic.class):

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

Comment By: Nobody/Anonymous (nobody)
Date: 2006-03-27 09:55

Message:
Logged In: NO 

I forgot to add some context. I get left_altjdic and
right_altjdic by taking the first items from the results of
selectBy() queries as follows:

left_altjdics = AltJDic.selectBy(entry=left)
left_altjdic = left_altjdics[0]
try:
    right_altjdics = AltJDic.selectBy(entry=right)
    right_altjdic = right_altjdics[0]
    if isSynonymRelGT(left_altjdic.class, right_altjdic.class):

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

Comment By: Nobody/Anonymous (nobody)
Date: 2006-03-27 09:54

Message:
Logged In: NO 

I forgot to add some context. I get left_altjdic and
right_altjdic by taking the first items from the results of
selectBy() queries as follows:

left_altjdics = AltJDic.selectBy(entry=left)
left_altjdic = left_altjdics[0]
try:
    right_altjdics = AltJDic.selectBy(entry=right)
    right_altjdic = right_altjdics[0]
    if isSynonymRelGT(left_altjdic.class, right_altjdic.class):

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

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540672&aid=1459482&group_id=74338

[SQL-CVS] [ sqlobject-Bugs-1459482 ] Can't access column called 'class' in SQLObject

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

Bugs item #1459482, was opened at 2006-03-27 09:51
Message generated for change (Comment added) made by nobody
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540672&aid=1459482&group_id=74338

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: General
Group: SQLObject release (specify)
Status: Open
Resolution: None
Priority: 5
Submitted By: Nobody/Anonymous (nobody)
Assigned to: Nobody/Anonymous (nobody)
Summary: Can't access column called 'class' in SQLObject

Initial Comment:
A syntax error is generated when a column named "class"
is accessed using the SQLObject class interface in
SQLObject 0.7.

Example:

The class AltJEDic contains a column named "CLASS" that
should be mapped to the python object AltJEDic.class

class AltJEDic(SQLObject):
    class sqlmeta:
        fromDatabase = True
        #idName = "sqlid"
        table = "ALTJEDIC0111"
        style = HinokiStyle()

However, when I try to use AltJEDic.class in a function
call, it fails with the following error:

# python verify-wordnet.py
  File "verify-wordnet.py", line 190
    if isSynonymRelGT(left_altjdic.class,
right_altjdic.class):
                                       ^
SyntaxError: invalid syntax

It seems that this is interfering with Python keywords.
Do we need to add a restriction on column (or SQLObject
class member names) or can this be worked around?

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

Comment By: Nobody/Anonymous (nobody)
Date: 2006-03-27 09:55

Message:
Logged In: NO 

I forgot to add some context. I get left_altjdic and
right_altjdic by taking the first items from the results of
selectBy() queries as follows:

left_altjdics = AltJDic.selectBy(entry=left)
left_altjdic = left_altjdics[0]
try:
    right_altjdics = AltJDic.selectBy(entry=right)
    right_altjdic = right_altjdics[0]
    if isSynonymRelGT(left_altjdic.class, right_altjdic.class):

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

Comment By: Nobody/Anonymous (nobody)
Date: 2006-03-27 09:54

Message:
Logged In: NO 

I forgot to add some context. I get left_altjdic and
right_altjdic by taking the first items from the results of
selectBy() queries as follows:

left_altjdics = AltJDic.selectBy(entry=left)
left_altjdic = left_altjdics[0]
try:
    right_altjdics = AltJDic.selectBy(entry=right)
    right_altjdic = right_altjdics[0]
    if isSynonymRelGT(left_altjdic.class, right_altjdic.class):

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

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540672&aid=1459482&group_id=74338

[SQL-CVS] [ sqlobject-Bugs-1459482 ] Can't access column called 'class' in SQLObject

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

Bugs item #1459482, was opened at 2006-03-27 09:51
Message generated for change (Comment added) made by nobody
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540672&aid=1459482&group_id=74338

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: General
Group: SQLObject release (specify)
Status: Open
Resolution: None
Priority: 5
Submitted By: Nobody/Anonymous (nobody)
Assigned to: Nobody/Anonymous (nobody)
Summary: Can't access column called 'class' in SQLObject

Initial Comment:
A syntax error is generated when a column named "class"
is accessed using the SQLObject class interface in
SQLObject 0.7.

Example:

The class AltJEDic contains a column named "CLASS" that
should be mapped to the python object AltJEDic.class

class AltJEDic(SQLObject):
    class sqlmeta:
        fromDatabase = True
        #idName = "sqlid"
        table = "ALTJEDIC0111"
        style = HinokiStyle()

However, when I try to use AltJEDic.class in a function
call, it fails with the following error:

# python verify-wordnet.py
  File "verify-wordnet.py", line 190
    if isSynonymRelGT(left_altjdic.class,
right_altjdic.class):
                                       ^
SyntaxError: invalid syntax

It seems that this is interfering with Python keywords.
Do we need to add a restriction on column (or SQLObject
class member names) or can this be worked around?

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

Comment By: Nobody/Anonymous (nobody)
Date: 2006-03-27 09:54

Message:
Logged In: NO 

I forgot to add some context. I get left_altjdic and
right_altjdic by taking the first items from the results of
selectBy() queries as follows:

left_altjdics = AltJDic.selectBy(entry=left)
left_altjdic = left_altjdics[0]
try:
    right_altjdics = AltJDic.selectBy(entry=right)
    right_altjdic = right_altjdics[0]
    if isSynonymRelGT(left_altjdic.class, right_altjdic.class):

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

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540672&aid=1459482&group_id=74338

[SQL-CVS] [ sqlobject-Bugs-1459482 ] Can't access column called 'class' in SQLObject

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

Bugs item #1459482, was opened at 2006-03-27 09:51
Message generated for change (Tracker Item Submitted) made by Item Submitter
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540672&aid=1459482&group_id=74338

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: General
Group: SQLObject release (specify)
Status: Open
Resolution: None
Priority: 5
Submitted By: Nobody/Anonymous (nobody)
Assigned to: Nobody/Anonymous (nobody)
Summary: Can't access column called 'class' in SQLObject

Initial Comment:
A syntax error is generated when a column named "class"
is accessed using the SQLObject class interface in
SQLObject 0.7.

Example:

The class AltJEDic contains a column named "CLASS" that
should be mapped to the python object AltJEDic.class

class AltJEDic(SQLObject):
    class sqlmeta:
        fromDatabase = True
        #idName = "sqlid"
        table = "ALTJEDIC0111"
        style = HinokiStyle()

However, when I try to use AltJEDic.class in a function
call, it fails with the following error:

# python verify-wordnet.py
  File "verify-wordnet.py", line 190
    if isSynonymRelGT(left_altjdic.class,
right_altjdic.class):
                                       ^
SyntaxError: invalid syntax

It seems that this is interfering with Python keywords.
Do we need to add a restriction on column (or SQLObject
class member names) or can this be worked around?

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

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540672&aid=1459482&group_id=74338

[SQL-CVS] [ sqlobject-Patches-1458380 ] Let admin create tables with no foreign keys first

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

Patches item #1458380, was opened at 2006-03-25 19:08
Message generated for change (Comment added) made by dennis
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540674&aid=1458380&group_id=74338

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
Status: Open
Resolution: Invalid
Priority: 5
Submitted By: Dennis Brakhane (dennis)
Assigned to: Nobody/Anonymous (nobody)
Summary: Let admin create tables with no foreign keys first

Initial Comment:
Suppose I have the following model:

class A(SQLObject):
    foo = StringCol()
    b = ForeignKey("B")

class B(SQLObject):
    bar = StringCol()

If I now let sqlobject-admin create the tables for me,
the order in with these are created is not defined.
If table A is created before table B, PostgreSQL will fail:

CREATE TABLE a (
    id SERIAL PRIMARY KEY,
    foo TEXT,
    b_id INT, CONSTRAINT b_id_exists FOREIGN KEY (b_id)
REFERENCES b (id)
)

Fails with "Relation >b< does not exist"

So, I've patched manager/command.py to create tables in
the order of classes with fewest foreign keys created
first.

This might not catch all cases, but it "works for me(TM)"

Patch is attached


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

>Comment By: Dennis Brakhane (dennis)
Date: 2006-03-27 17:47

Message:
Logged In: YES 
user_id=26932

> I think I'd be better to use this feature instead of
> unreliable "fewest foreign keys". Would you mind
> recreating your patch?

Erm... What exactly should I recreate? It seems like
manager.py already has this functionality built-in.

I'm using TurboGears svn and didn't realize the external
reference to SQLObject's SVN was to the "0.7-bugfix" branch
and not the trunk. So I think the bug is already fixed.


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

Comment By: Oleg Broytmann (phd)
Date: 2006-03-27 16:32

Message:
Logged In: YES 
user_id=4799

.createTable() in the trunk has got a feature - it can delay
creating foreign keys and instead return SQL query to create
them later. So one can create all tables, collects these
delayed SQL queries and run them later, when all tables are
in place. Look in the FAQ:
http://svn.colorstudy.com/SQLObject/docs/FAQ.txt , question
"Mutually referencing tables".

I think I'd be better to use this feature instead of
unreliable "fewest foreign keys". Would you mind recreating
your patch?

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

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540674&aid=1458380&group_id=74338

[SQL-CVS] [ sqlobject-Patches-1458925 ] Fix for bug 1458595 (destroySelf in Transaction = deadlock)

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

Patches item #1458925, was opened at 2006-03-26 14:35
Message generated for change (Comment added) made by cpisto
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540674&aid=1458925&group_id=74338

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
Status: Open
Resolution: None
Priority: 5
Submitted By: Cody Pisto (cpisto)
Assigned to: Nobody/Anonymous (nobody)
Summary: Fix for bug 1458595 (destroySelf in Transaction = deadlock)

Initial Comment:
This patch fixes bug 1458595, in which actions that
resulted in Transaction._SO_delete being called
resulted in an additional database connection being
opened, causing a deadlock while multiple connections
await on eachothers commit.

This patch is against svn trunk,
r1668. (dbconnection.py)

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

>Comment By: Cody Pisto (cpisto)
Date: 2006-03-27 08:45

Message:
Logged In: YES 
user_id=118227

Sure, ill refactor-
Revised patch will follow this evening.

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

Comment By: Oleg Broytmann (phd)
Date: 2006-03-27 07:46

Message:
Logged In: YES 
user_id=4799

Your patch duplicates the query from DBAPI._SO_delete() to
Transaction._SO_delete(). Would you mind to refactor your
patch - split DBAPI._SO_delete() into two methods - one to
generate a query string, and another to execute the query;
then use the first method to generate a query in the
Transaction._SO_delete()?

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

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540674&aid=1458925&group_id=74338

[SQL-CVS] [ sqlobject-Bugs-1455251 ] dbConnection.query() opens another connection

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

Bugs item #1455251, was opened at 2006-03-21 12:47
Message generated for change (Settings changed) made by phd
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540672&aid=1455251&group_id=74338

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
>Status: Deleted
>Resolution: Wont Fix
Priority: 5
Submitted By: Oleg Broytmann (phd)
>Assigned to: Oleg Broytmann (phd)
Summary: dbConnection.query() opens another connection

Initial Comment:
DBConnection often uses self.query() to run additional
queries. That's acceptable for DDL queries
(CREATE/DROP) but is very bad for DML queries
(SELECT/INSERT/UPDATE/DELETE) because the new query is
executed in a different connection, outside of the
current transaction. Even worse, for row.destroySelf()
SQLite often fails with the error "Database is locked".

DBConnection mustn't use self.query(); the query must
be run via the current low-level (raw) connection.

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

>Comment By: Oleg Broytmann (phd)
Date: 2006-03-27 18:58

Message:
Logged In: YES 
user_id=4799

It seems it's me overgeneralizing a problem with
.destroySelf(). See
http://sourceforge.net/mailarchive/forum.php?thread_id=10043224&forum_id=30269
for a discussion.

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

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540672&aid=1455251&group_id=74338

[SQL-CVS] [ sqlobject-Bugs-1429248 ] tableExists query issues with uper case table names

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

Bugs item #1429248, was opened at 2006-02-10 20:00
Message generated for change (Comment added) made by phd
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540672&aid=1429248&group_id=74338

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: Postgres
Group: None
Status: Open
>Resolution: Wont Fix
Priority: 5
Submitted By: Hootbah (hootbah)
>Assigned to: Oleg Broytmann (phd)
Summary: tableExists query issues with uper case table names

Initial Comment:
Its is possible for the tableExists query not to find a
table when it does actually exist. 

When using uppercase names e.g.

class sqlmeta:
    table = 'MYTABLE'


The tableExists query becomes:

SELECT COUNT(relname) FROM pg_class WHERE relname =
'MYTABLE'

This fails to find the table as the table names stored
within pg_class are all lower case.

I propose to change the query to be:

SELECT COUNT(relname) FROM pg_class WHERE relname =
lower('MYTABLE')


This would mean a change to pgconnection.py:

 result = self.queryOne("SELECT COUNT(relname) FROM
pg_class WHERE relname = lower(%s)" 
        % self.sqlrepr(tableName))

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

>Comment By: Oleg Broytmann (phd)
Date: 2006-03-27 18:51

Message:
Logged In: YES 
user_id=4799

The fix would break case-sensitive names. You don't want to
make this lowercase:

class sqlmeta:
   table = '"MYTABLE"'

Instead just write the name lower-cased:

class sqlmeta:
   table = 'mytable'

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

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540672&aid=1429248&group_id=74338

[SQL-CVS] [ sqlobject-Patches-1458925 ] Fix for bug 1458595 (destroySelf in Transaction = deadlock)

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

Patches item #1458925, was opened at 2006-03-27 01:35
Message generated for change (Comment added) made by phd
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540674&aid=1458925&group_id=74338

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
Status: Open
Resolution: None
Priority: 5
Submitted By: Cody Pisto (cpisto)
Assigned to: Nobody/Anonymous (nobody)
Summary: Fix for bug 1458595 (destroySelf in Transaction = deadlock)

Initial Comment:
This patch fixes bug 1458595, in which actions that
resulted in Transaction._SO_delete being called
resulted in an additional database connection being
opened, causing a deadlock while multiple connections
await on eachothers commit.

This patch is against svn trunk,
r1668. (dbconnection.py)

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

>Comment By: Oleg Broytmann (phd)
Date: 2006-03-27 18:46

Message:
Logged In: YES 
user_id=4799

Your patch duplicates the query from DBAPI._SO_delete() to
Transaction._SO_delete(). Would you mind to refactor your
patch - split DBAPI._SO_delete() into two methods - one to
generate a query string, and another to execute the query;
then use the first method to generate a query in the
Transaction._SO_delete()?

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

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540674&aid=1458925&group_id=74338

[SQL-CVS] [ sqlobject-Patches-1450568 ] faster sqlbuilder.Insert

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

Patches item #1450568, was opened at 2006-03-15 20:58
Message generated for change (Comment added) made by phd
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540674&aid=1450568&group_id=74338

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
Status: Open
Resolution: None
Priority: 5
Submitted By: Davide Alberani (alberanid)
Assigned to: Nobody/Anonymous (nobody)
Summary: faster sqlbuilder.Insert

Initial Comment:
In the attachment, a very simple patch to improve
performances (about 12 times faster, with a list of
~20000 items, tested with python2.3) of the
sqlbuilder.Insert
class, when a very long valueList list is provided.

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

>Comment By: Oleg Broytmann (phd)
Date: 2006-03-27 18:37

Message:
Logged In: YES 
user_id=4799

Please add some tests so I can run them before and after
applying the patch and verify that the patch doesn't break
Insert().

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

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540674&aid=1450568&group_id=74338

[SQL-CVS] [ sqlobject-Patches-1458380 ] Let admin create tables with no foreign keys first

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

Patches item #1458380, was opened at 2006-03-25 21:08
Message generated for change (Comment added) made by phd
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540674&aid=1458380&group_id=74338

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
Status: Open
>Resolution: Invalid
Priority: 5
Submitted By: Dennis Brakhane (dennis)
Assigned to: Nobody/Anonymous (nobody)
Summary: Let admin create tables with no foreign keys first

Initial Comment:
Suppose I have the following model:

class A(SQLObject):
    foo = StringCol()
    b = ForeignKey("B")

class B(SQLObject):
    bar = StringCol()

If I now let sqlobject-admin create the tables for me,
the order in with these are created is not defined.
If table A is created before table B, PostgreSQL will fail:

CREATE TABLE a (
    id SERIAL PRIMARY KEY,
    foo TEXT,
    b_id INT, CONSTRAINT b_id_exists FOREIGN KEY (b_id)
REFERENCES b (id)
)

Fails with "Relation >b< does not exist"

So, I've patched manager/command.py to create tables in
the order of classes with fewest foreign keys created
first.

This might not catch all cases, but it "works for me(TM)"

Patch is attached


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

>Comment By: Oleg Broytmann (phd)
Date: 2006-03-27 18:32

Message:
Logged In: YES 
user_id=4799

.createTable() in the trunk has got a feature - it can delay
creating foreign keys and instead return SQL query to create
them later. So one can create all tables, collects these
delayed SQL queries and run them later, when all tables are
in place. Look in the FAQ:
http://svn.colorstudy.com/SQLObject/docs/FAQ.txt , question
"Mutually referencing tables".

I think I'd be better to use this feature instead of
unreliable "fewest foreign keys". Would you mind recreating
your patch?

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

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540674&aid=1458380&group_id=74338

[SQL-CVS] [ sqlobject-Patches-1458925 ] Fix for bug 1458595 (destroySelf in Transaction = deadlock)

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

Patches item #1458925, was opened at 2006-03-26 14:35
Message generated for change (Tracker Item Submitted) made by Item Submitter
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540674&aid=1458925&group_id=74338

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
Status: Open
Resolution: None
Priority: 5
Submitted By: Cody Pisto (cpisto)
Assigned to: Nobody/Anonymous (nobody)
Summary: Fix for bug 1458595 (destroySelf in Transaction = deadlock)

Initial Comment:
This patch fixes bug 1458595, in which actions that
resulted in Transaction._SO_delete being called
resulted in an additional database connection being
opened, causing a deadlock while multiple connections
await on eachothers commit.

This patch is against svn trunk,
r1668. (dbconnection.py)

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

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540674&aid=1458925&group_id=74338

[SQL-CVS] [ sqlobject-Bugs-1458595 ] destroySelf in transaction causes lockup

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

Bugs item #1458595, was opened at 2006-03-25 18:52
Message generated for change (Comment added) made by cpisto
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540672&aid=1458595&group_id=74338

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: SQLObject from repository
Status: Open
Resolution: None
Priority: 5
Submitted By: Cody Pisto (cpisto)
Assigned to: Nobody/Anonymous (nobody)
Summary: destroySelf in transaction causes lockup

Initial Comment:
Calling destroySelf() on an sqlobject acquired in the
same explicit transaction (using conn.transaction()...)
causes the entire python interpreter to lockup if the
sqlobject schema in question has any SQLRelatedJoin's
and the accompaying remove[JoinCol]() function is
called in the transaction before destroySelf().

PDB stepping seems to be of no use,
the interpreter locks up on line 306 of dbconnection.py
"return cursor.execute(query)"

The same sequence of calls works without problems
outside a transaction.

Version details:

Python 2.4.2
PostgreSQL 8.1.2 (via psycopg2 2.0b8 or psycopg 1.1.21)
SQLObject 0.8dev r1668 and r1615


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

>Comment By: Cody Pisto (cpisto)
Date: 2006-03-26 14:32

Message:
Logged In: YES 
user_id=118227

OK, found the problem.

The _SO_delete method on the Transaction class had no way of
specifying which connection to use for the query, thus a new
one was being created.

the attached patch fixes that.

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

Comment By: Cody Pisto (cpisto)
Date: 2006-03-26 14:14

Message:
Logged In: YES 
user_id=118227

OK, after further investigation,

It seems that for some reason a second database connection
is being created during the lifetime of the transaction (by
one of the SQLObjects instantiated inside doInTransaction),
so a deadlock is occuring while this second connection waits
for a commit on the first connection, I am still investating
why this second connection is being created...


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

Comment By: Cody Pisto (cpisto)
Date: 2006-03-25 21:27

Message:
Logged In: YES 
user_id=118227

The following link seems to reference the exact same problem,
only on a slightly older version of sqlobject, and using mysql.

http://www.mail-archive.com/sql...@li.../msg00226.html

After contacting the poster, reverting to a much older
version of sqlobject (r1547) seems to work around the issue

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

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540672&aid=1458595&group_id=74338