sqlobject-cvs — CVS notification

[SQL-CVS] r1895 - SQLObject/docs

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

Author: phd
Date: 2006-09-01 07:14:02 -0600 (Fri, 01 Sep 2006)
New Revision: 1895

Modified:
   SQLObject/docs/News.txt
   SQLObject/docs/SQLObject.txt
Log:
Updated docs.


Modified: SQLObject/docs/News.txt
===================================================================
--- SQLObject/docs/News.txt	2006-09-01 13:12:58 UTC (rev 1894)
+++ SQLObject/docs/News.txt	2006-09-01 13:14:02 UTC (rev 1895)
@@ -109,6 +109,25 @@
 Bug Fixes
 ---------
 
+* Fixed a longstanding problem with UnicodeCol - at last you can use unicode
+  strings in .select() and .selectBy() queries. There are some limitations,
+  though:
+
+  - only simple q-magic fields are supported; no expressions;
+  - only == and <> operators are supported;
+
+  The following code works::
+
+    MyTable.select(u'value' == MyTable.q.name)
+    MyTable.select(MyTable.q.name <> u'value')
+    MyTable.select(OR(MyTable.q.col1 == u'value1', MyTable.q.col2 <> u'value2'))
+    MyTable.selectBy(name = u'value')
+    MyTable.selectBy(col1=u'value1', col2=u'value2')
+
+  The following still does not work::
+
+    MyTable.select((MyTable.q.name + MyTable.q.surname) == u'value')
+
 * Cull patch (clear cache).
 
 * .destroySelf() inside a transaction.

Modified: SQLObject/docs/SQLObject.txt
===================================================================
--- SQLObject/docs/SQLObject.txt	2006-09-01 13:12:58 UTC (rev 1894)
+++ SQLObject/docs/SQLObject.txt	2006-09-01 13:14:02 UTC (rev 1895)
@@ -1128,11 +1128,28 @@
     A subclass of `StringCol`.  Also accepts a dbEncoding keyword
     argument, which defaults to ``"UTF-8"``.  Values coming in and
     out from the database will be encoded and decoded.  **Note**:
-    parameters in queries will not be automatically encoded, so if
-    you do a query matching a UnicodeCol column you must apply the
-    encoding yourself.
+    there are some limitations on using UnicodeCol in queries:
 
+    - only simple q-magic fields are supported; no expressions;
+    - only == and <> operators are supported;
 
+    The following code works::
+
+        MyTable.select(u'value' == MyTable.q.name)
+        MyTable.select(MyTable.q.name <> u'value')
+        MyTable.select(OR(MyTable.q.col1 == u'value1', MyTable.q.col2 <> u'value2'))
+        MyTable.selectBy(name = u'value')
+        MyTable.selectBy(col1=u'value1', col2=u'value2')
+
+    The following does not work::
+
+        MyTable.select((MyTable.q.name + MyTable.q.surname) == u'value')
+
+    In that case you must apply the encoding yourself::
+
+        MyTable.select((MyTable.q.name + MyTable.q.surname) == u'value'.encode(dbEncoding))
+
+
 .. Relationships_:
 
 Relationships Between Classes/Tables

[SQL-CVS] r1894 - in SQLObject/branches/0.7-bugfix: docs sqlobject sqlobject/tests

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

Author: phd
Date: 2006-09-01 07:12:58 -0600 (Fri, 01 Sep 2006)
New Revision: 1894

Modified:
   SQLObject/branches/0.7-bugfix/docs/News.txt
   SQLObject/branches/0.7-bugfix/docs/SQLObject.txt
   SQLObject/branches/0.7-bugfix/sqlobject/main.py
   SQLObject/branches/0.7-bugfix/sqlobject/sqlbuilder.py
   SQLObject/branches/0.7-bugfix/sqlobject/tests/test_unicode.py
Log:
Fixed a longstanding problem with UnicodeCol - at last you can use unicode
strings in .select() and .selectBy() queries (with some limitations).


Modified: SQLObject/branches/0.7-bugfix/docs/News.txt
===================================================================
--- SQLObject/branches/0.7-bugfix/docs/News.txt	2006-09-01 13:12:23 UTC (rev 1893)
+++ SQLObject/branches/0.7-bugfix/docs/News.txt	2006-09-01 13:12:58 UTC (rev 1894)
@@ -49,6 +49,25 @@
 Bug Fixes
 ---------
 
+* Fixed a longstanding problem with UnicodeCol - at last you can use unicode
+  strings in .select() and .selectBy() queries. There are some limitations,
+  though:
+
+  - only simple q-magic fields are supported; no expressions;
+  - only == and <> operators are supported;
+
+  The following code works::
+
+    MyTable.select(u'value' == MyTable.q.name)
+    MyTable.select(MyTable.q.name <> u'value')
+    MyTable.select(OR(MyTable.q.col1 == u'value1', MyTable.q.col2 <> u'value2'))
+    MyTable.selectBy(name = u'value')
+    MyTable.selectBy(col1=u'value1', col2=u'value2')
+
+  The following still does not work::
+
+    MyTable.select((MyTable.q.name + MyTable.q.surname) == u'value')
+
 * Cull patch (clear cache).
 
 * .destroySelf() inside a transaction.

Modified: SQLObject/branches/0.7-bugfix/docs/SQLObject.txt
===================================================================
--- SQLObject/branches/0.7-bugfix/docs/SQLObject.txt	2006-09-01 13:12:23 UTC (rev 1893)
+++ SQLObject/branches/0.7-bugfix/docs/SQLObject.txt	2006-09-01 13:12:58 UTC (rev 1894)
@@ -1127,11 +1127,28 @@
     A subclass of `StringCol`.  Also accepts a dbEncoding keyword
     argument, which defaults to ``"UTF-8"``.  Values coming in and
     out from the database will be encoded and decoded.  **Note**:
-    parameters in queries will not be automatically encoded, so if
-    you do a query matching a UnicodeCol column you must apply the
-    encoding yourself.
+    there are some limitations on using UnicodeCol in queries:
 
+    - only simple q-magic fields are supported; no expressions;
+    - only == and <> operators are supported;
 
+    The following code works::
+
+        MyTable.select(u'value' == MyTable.q.name)
+        MyTable.select(MyTable.q.name <> u'value')
+        MyTable.select(OR(MyTable.q.col1 == u'value1', MyTable.q.col2 <> u'value2'))
+        MyTable.selectBy(name = u'value')
+        MyTable.selectBy(col1=u'value1', col2=u'value2')
+
+    The following does not work::
+
+        MyTable.select((MyTable.q.name + MyTable.q.surname) == u'value')
+
+    In that case you must apply the encoding yourself::
+
+        MyTable.select((MyTable.q.name + MyTable.q.surname) == u'value'.encode(dbEncoding))
+
+
 .. Relationships_:
 
 Relationships Between Classes/Tables

Modified: SQLObject/branches/0.7-bugfix/sqlobject/main.py
===================================================================
--- SQLObject/branches/0.7-bugfix/sqlobject/main.py	2006-09-01 13:12:23 UTC (rev 1893)
+++ SQLObject/branches/0.7-bugfix/sqlobject/main.py	2006-09-01 13:12:58 UTC (rev 1894)
@@ -1297,6 +1297,11 @@
     select = classmethod(select)
 
     def selectBy(cls, connection=None, **kw):
+        for key, column in cls.sqlmeta.columns.items():
+            if (key in kw) and isinstance(column, col.SOUnicodeCol):
+                value = kw[key]
+                if isinstance(value, unicode):
+                    kw[key] = value.encode(column.dbEncoding)
         conn = connection or cls._connection
         return cls.SelectResultsClass(cls,
                              conn._SO_columnClause(cls, kw),

Modified: SQLObject/branches/0.7-bugfix/sqlobject/sqlbuilder.py
===================================================================
--- SQLObject/branches/0.7-bugfix/sqlobject/sqlbuilder.py	2006-09-01 13:12:23 UTC (rev 1893)
+++ SQLObject/branches/0.7-bugfix/sqlobject/sqlbuilder.py	2006-09-01 13:12:58 UTC (rev 1894)
@@ -323,6 +323,23 @@
 
 registerConverter(SQLObjectField, SQLExprConverter)
 
+
+class UnicodeField(SQLObjectField):
+    def __init__(self, tableName, fieldName, original, column):
+        SQLObjectField.__init__(self, tableName, fieldName, original)
+        self.column = column
+    def __eq__(self, other):
+        if isinstance(other, unicode):
+            other = other.encode(self.column.dbEncoding)
+        return SQLOp('=', self, other)
+    def __ne__(self, other):
+        if isinstance(other, unicode):
+            other = other.encode(self.column.dbEncoding)
+        return SQLOp('<>', self, other)
+
+registerConverter(UnicodeField, SQLExprConverter)
+
+
 class Table(SQLExpression):
     FieldClass = Field
 
@@ -339,6 +356,7 @@
 
 class SQLObjectTable(Table):
     FieldClass = SQLObjectField
+    UnicodeFieldClass = UnicodeField
 
     def __init__(self, soClass):
         self.soClass = soClass
@@ -355,9 +373,12 @@
         elif attr not in self.soClass.sqlmeta.columns:
             raise AttributeError("%s instance has no attribute '%s'" % (self.soClass.__name__, attr))
         else:
-            return self.FieldClass(self.tableName,
-                                   self.soClass.sqlmeta.columns[attr].dbName,
-                                   attr)
+            column = self.soClass.sqlmeta.columns[attr]
+            if hasattr(column, "dbEncoding"):
+                return self.UnicodeFieldClass(self.tableName, column.dbName,
+                    attr, column)
+            else:
+                return self.FieldClass(self.tableName, column.dbName, attr)
 
 class TableSpace:
     TableClass = Table

Modified: SQLObject/branches/0.7-bugfix/sqlobject/tests/test_unicode.py
===================================================================
--- SQLObject/branches/0.7-bugfix/sqlobject/tests/test_unicode.py	2006-09-01 13:12:23 UTC (rev 1893)
+++ SQLObject/branches/0.7-bugfix/sqlobject/tests/test_unicode.py	2006-09-01 13:12:58 UTC (rev 1894)
@@ -5,7 +5,7 @@
 ## Unicode columns
 ########################################
 
-class Unicode1(SQLObject):
+class TestUnicode(SQLObject):
     count = IntCol(alternateID=True)
     col1 = UnicodeCol()
     col2 = UnicodeCol(dbEncoding='latin-1')
@@ -16,24 +16,54 @@
     def enumerate(lst):
         return [(i, lst[i]) for i in range(len(lst))]
 
-def test_create():
-    setupClass(Unicode1)
-    data = [u'\u00f0', u'test', 'ascii test']
+data = [u'\u00f0', u'test', 'ascii test']
+items = []
 
+def setup():
+    global items
     items = []
+    setupClass(TestUnicode)
     for i, n in enumerate(data):
-        items.append(Unicode1(count=i, col1=n, col2=n))
+        items.append(TestUnicode(count=i, col1=n, col2=n))
+
+def test_create():
+    setup()
     for n, item in zip(data, items):
-        item.col1 = item.col2 = n
-    for n, item in zip(data, items):
         assert item.col1 == item.col2
         assert item.col1 == n
-    conn = Unicode1._connection
+
+    conn = TestUnicode._connection
     rows = conn.queryAll("""
     SELECT count, col1, col2
-    FROM unicode1
+    FROM test_unicode
     ORDER BY count
     """)
     for count, col1, col2 in rows:
         assert data[count].encode('utf-8') == col1
         assert data[count].encode('latin1') == col2
+
+def test_select():
+    setup()
+    for value in data:
+        rows = list(TestUnicode.select(TestUnicode.q.col1 == value))
+        assert len(rows) == 1
+        rows = list(TestUnicode.select(TestUnicode.q.col2 == value))
+        assert len(rows) == 1
+        rows = list(TestUnicode.select(AND(
+            TestUnicode.q.col1 == value,
+            TestUnicode.q.col2 == value
+        )))
+        assert len(rows) == 1
+        rows = list(TestUnicode.selectBy(col1=value))
+        assert len(rows) == 1
+        rows = list(TestUnicode.selectBy(col2=value))
+        assert len(rows) == 1
+        rows = list(TestUnicode.selectBy(col1=value, col2=value))
+        assert len(rows) == 1
+    rows = list(TestUnicode.select(OR(
+        TestUnicode.q.col1 == u'\u00f0',
+        TestUnicode.q.col2 == u'test'
+    )))
+    assert len(rows) == 2
+    rows = list(TestUnicode.selectBy(col1=u'\u00f0', col2=u'test'))
+    assert len(rows) == 0

[SQL-CVS] r1893 - in SQLObject/trunk/sqlobject: . tests

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

Author: phd
Date: 2006-09-01 07:12:23 -0600 (Fri, 01 Sep 2006)
New Revision: 1893

Modified:
   SQLObject/trunk/sqlobject/main.py
   SQLObject/trunk/sqlobject/sqlbuilder.py
   SQLObject/trunk/sqlobject/tests/test_unicode.py
Log:
Fixed a longstanding problem with UnicodeCol - at last you can use unicode
strings in .select() and .selectBy() queries (with some limitations).


Modified: SQLObject/trunk/sqlobject/main.py
===================================================================
--- SQLObject/trunk/sqlobject/main.py	2006-09-01 12:59:15 UTC (rev 1892)
+++ SQLObject/trunk/sqlobject/main.py	2006-09-01 13:12:23 UTC (rev 1893)
@@ -1328,6 +1328,11 @@
     select = classmethod(select)
 
     def selectBy(cls, connection=None, **kw):
+        for key, column in cls.sqlmeta.columns.items():
+            if (key in kw) and isinstance(column, col.SOUnicodeCol):
+                value = kw[key]
+                if isinstance(value, unicode):
+                    kw[key] = value.encode(column.dbEncoding)
         conn = connection or cls._connection
         return cls.SelectResultsClass(cls,
                              conn._SO_columnClause(cls, kw),

Modified: SQLObject/trunk/sqlobject/sqlbuilder.py
===================================================================
--- SQLObject/trunk/sqlobject/sqlbuilder.py	2006-09-01 12:59:15 UTC (rev 1892)
+++ SQLObject/trunk/sqlobject/sqlbuilder.py	2006-09-01 13:12:23 UTC (rev 1893)
@@ -326,6 +326,23 @@
 
 registerConverter(SQLObjectField, SQLExprConverter)
 
+
+class UnicodeField(SQLObjectField):
+    def __init__(self, tableName, fieldName, original, column):
+        SQLObjectField.__init__(self, tableName, fieldName, original)
+        self.column = column
+    def __eq__(self, other):
+        if isinstance(other, unicode):
+            other = other.encode(self.column.dbEncoding)
+        return SQLOp('=', self, other)
+    def __ne__(self, other):
+        if isinstance(other, unicode):
+            other = other.encode(self.column.dbEncoding)
+        return SQLOp('<>', self, other)
+
+registerConverter(UnicodeField, SQLExprConverter)
+
+
 class Table(SQLExpression):
     FieldClass = Field
 
@@ -342,6 +359,7 @@
 
 class SQLObjectTable(Table):
     FieldClass = SQLObjectField
+    UnicodeFieldClass = UnicodeField
 
     def __init__(self, soClass):
         self.soClass = soClass
@@ -358,9 +376,12 @@
         elif attr not in self.soClass.sqlmeta.columns:
             raise AttributeError("%s instance has no attribute '%s'" % (self.soClass.__name__, attr))
         else:
-            return self.FieldClass(self.tableName,
-                                   self.soClass.sqlmeta.columns[attr].dbName,
-                                   attr)
+            column = self.soClass.sqlmeta.columns[attr]
+            if hasattr(column, "dbEncoding"):
+                return self.UnicodeFieldClass(self.tableName, column.dbName,
+                    attr, column)
+            else:
+                return self.FieldClass(self.tableName, column.dbName, attr)
 
 class TableSpace:
     TableClass = Table

Modified: SQLObject/trunk/sqlobject/tests/test_unicode.py
===================================================================
--- SQLObject/trunk/sqlobject/tests/test_unicode.py	2006-09-01 12:59:15 UTC (rev 1892)
+++ SQLObject/trunk/sqlobject/tests/test_unicode.py	2006-09-01 13:12:23 UTC (rev 1893)
@@ -5,7 +5,7 @@
 ## Unicode columns
 ########################################
 
-class Unicode1(SQLObject):
+class TestUnicode(SQLObject):
     count = IntCol(alternateID=True)
     col1 = UnicodeCol()
     col2 = UnicodeCol(dbEncoding='latin-1')
@@ -16,24 +16,54 @@
     def enumerate(lst):
         return [(i, lst[i]) for i in range(len(lst))]
 
-def test_create():
-    setupClass(Unicode1)
-    data = [u'\u00f0', u'test', 'ascii test']
+data = [u'\u00f0', u'test', 'ascii test']
+items = []
 
+def setup():
+    global items
     items = []
+    setupClass(TestUnicode)
     for i, n in enumerate(data):
-        items.append(Unicode1(count=i, col1=n, col2=n))
+        items.append(TestUnicode(count=i, col1=n, col2=n))
+
+def test_create():
+    setup()
     for n, item in zip(data, items):
-        item.col1 = item.col2 = n
-    for n, item in zip(data, items):
         assert item.col1 == item.col2
         assert item.col1 == n
-    conn = Unicode1._connection
+
+    conn = TestUnicode._connection
     rows = conn.queryAll("""
     SELECT count, col1, col2
-    FROM unicode1
+    FROM test_unicode
     ORDER BY count
     """)
     for count, col1, col2 in rows:
         assert data[count].encode('utf-8') == col1
         assert data[count].encode('latin1') == col2
+
+def test_select():
+    setup()
+    for value in data:
+        rows = list(TestUnicode.select(TestUnicode.q.col1 == value))
+        assert len(rows) == 1
+        rows = list(TestUnicode.select(TestUnicode.q.col2 == value))
+        assert len(rows) == 1
+        rows = list(TestUnicode.select(AND(
+            TestUnicode.q.col1 == value,
+            TestUnicode.q.col2 == value
+        )))
+        assert len(rows) == 1
+        rows = list(TestUnicode.selectBy(col1=value))
+        assert len(rows) == 1
+        rows = list(TestUnicode.selectBy(col2=value))
+        assert len(rows) == 1
+        rows = list(TestUnicode.selectBy(col1=value, col2=value))
+        assert len(rows) == 1
+    rows = list(TestUnicode.select(OR(
+        TestUnicode.q.col1 == u'\u00f0',
+        TestUnicode.q.col2 == u'test'
+    )))
+    assert len(rows) == 2
+    rows = list(TestUnicode.selectBy(col1=u'\u00f0', col2=u'test'))
+    assert len(rows) == 0

[SQL-CVS] r1892 - SQLObject/branches/0.7-bugfix/sqlobject

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

Author: phd
Date: 2006-09-01 06:59:15 -0600 (Fri, 01 Sep 2006)
New Revision: 1892

Modified:
   SQLObject/branches/0.7-bugfix/sqlobject/dbconnection.py
Log:
Fixed compatibility: Python 2.3 doesn't have .rsplit().


Modified: SQLObject/branches/0.7-bugfix/sqlobject/dbconnection.py
===================================================================
--- SQLObject/branches/0.7-bugfix/sqlobject/dbconnection.py	2006-09-01 12:58:42 UTC (rev 1891)
+++ SQLObject/branches/0.7-bugfix/sqlobject/dbconnection.py	2006-09-01 12:59:15 UTC (rev 1892)
@@ -95,7 +95,8 @@
             else:
                 host, rest = rest.split('/', 1)
         if host and host.find('@') != -1:
-            user, host = host.rsplit('@', 1)
+            user = host[:host.rfind('@')] # Python 2.3 doesn't have .rsplit()
+            host = host[host.rfind('@')+1:] # !!!
             if user.find(':') != -1:
                 user, password = user.split(':', 1)
             else:

[SQL-CVS] r1891 - SQLObject/trunk/sqlobject

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

Author: phd
Date: 2006-09-01 06:58:42 -0600 (Fri, 01 Sep 2006)
New Revision: 1891

Modified:
   SQLObject/trunk/sqlobject/dbconnection.py
Log:
Fixed compatibility: Python 2.3 doesn't have .rsplit().


Modified: SQLObject/trunk/sqlobject/dbconnection.py
===================================================================
--- SQLObject/trunk/sqlobject/dbconnection.py	2006-08-29 17:07:41 UTC (rev 1890)
+++ SQLObject/trunk/sqlobject/dbconnection.py	2006-09-01 12:58:42 UTC (rev 1891)
@@ -95,7 +95,8 @@
             else:
                 host, rest = rest.split('/', 1)
         if host and host.find('@') != -1:
-            user, host = host.rsplit('@', 1)
+            user = host[:host.rfind('@')] # Python 2.3 doesn't have .rsplit()
+            host = host[host.rfind('@')+1:] # !!!
             if user.find(':') != -1:
                 user, password = user.split(':', 1)
             else:

[SQL-CVS] r1889 - in home/phd/SQLObject/paramstyles/sqlobject: . tests

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

Author: phd
Date: 2006-08-29 08:26:02 -0600 (Tue, 29 Aug 2006)
New Revision: 1889

Modified:
   home/phd/SQLObject/paramstyles/sqlobject/dbconnection.py
   home/phd/SQLObject/paramstyles/sqlobject/tests/test_parse.py
Log:
Merged patches from the revisions 1886:1888 from the trunk

Modified: home/phd/SQLObject/paramstyles/sqlobject/dbconnection.py
===================================================================
--- home/phd/SQLObject/paramstyles/sqlobject/dbconnection.py	2006-08-29 14:20:29 UTC (rev 1888)
+++ home/phd/SQLObject/paramstyles/sqlobject/dbconnection.py	2006-08-29 14:26:02 UTC (rev 1889)
@@ -96,7 +96,7 @@
             else:
                 host, rest = rest.split('/', 1)
         if host and host.find('@') != -1:
-            user, host = host.split('@', 1)
+            user, host = host.rsplit('@', 1)
             if user.find(':') != -1:
                 user, password = user.split(':', 1)
             else:

Modified: home/phd/SQLObject/paramstyles/sqlobject/tests/test_parse.py
===================================================================
--- home/phd/SQLObject/paramstyles/sqlobject/tests/test_parse.py	2006-08-29 14:20:29 UTC (rev 1888)
+++ home/phd/SQLObject/paramstyles/sqlobject/tests/test_parse.py	2006-08-29 14:26:02 UTC (rev 1889)
@@ -48,6 +48,14 @@
     assert path == "/full/path/to/socket/database"
     assert args == {}
 
+    user, password, host, port, path, args = _parseURI("postgres://user:at@inpwd@host/database")
+    assert user == "user"
+    assert password == "at@inpwd"
+    assert host == "host"
+    assert port is None
+    assert path == "/database"
+    assert args == {}
+
     user, password, host, port, path, args = _parseURI("sqlite:///full/path/to/database")
     assert user is None
     assert password is None

[SQL-CVS] [ sqlobject-Bugs-1426574 ] Database passwords with '@' don't parse correctly.

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

Bugs item #1426574, was opened at 2006-02-07 23:43
Message generated for change (Comment added) made by phd
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540672&aid=1426574&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: Closed
>Resolution: Fixed
Priority: 5
Submitted By: Nobody/Anonymous (nobody)
>Assigned to: Oleg Broytmann (phd)
Summary: Database passwords with '@' don't parse correctly.

Initial Comment:
As the title says, if your database password contains
an '@' character, then the connection string is parsed
incorrectly. To remidy this, I've used host.rfind('@')
instead of host.find('@'). A patch for this would look
something like (I hope the indentation works):

 dbconnection.py.ORIG dbconnection.py
94,99c94,102
<         if host and host.find('@') != -1:
<             user, host = host.split('@', 1)
<             if user.find(':') != -1:
<                 user, password = user.split(':', 1)
<             else:
<                 password = None
---
>         if host:
>           at = host.rfind('@')
>           if at != -1:
>               user = host[:at]
>               host = host[at+1:]
>               if user.find(':') != -1:
>                   user, password = user.split(':', 1)
>               else:
>                   password = None



gk...@gm...

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

>Comment By: Oleg Broytmann (phd)
Date: 2006-08-29 18:23

Message:
Logged In: YES 
user_id=4799

Fixed in the revision 1887 in the trunk, rev. 1888 in the
0.7-branch.

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

Comment By: Nobody/Anonymous (nobody)
Date: 2006-02-07 23:45

Message:
Logged In: NO 

Doesn't look like the indentation worked, but it should be
easy to see that rather than using the split function, use
the reverse find() to split the string.

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

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540672&aid=1426574&group_id=74338

[SQL-CVS] r1888 - in SQLObject/branches/0.7-bugfix/sqlobject: . tests

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

Author: phd
Date: 2006-08-29 08:20:29 -0600 (Tue, 29 Aug 2006)
New Revision: 1888

Modified:
   SQLObject/branches/0.7-bugfix/sqlobject/dbconnection.py
   SQLObject/branches/0.7-bugfix/sqlobject/tests/test_parse.py
Log:
Applied a patch by Ivan Horvath <i.h...@al...> to fix the bug 1426574:
Database passwords with '@' don't parse correctly.


Modified: SQLObject/branches/0.7-bugfix/sqlobject/dbconnection.py
===================================================================
--- SQLObject/branches/0.7-bugfix/sqlobject/dbconnection.py	2006-08-29 14:20:14 UTC (rev 1887)
+++ SQLObject/branches/0.7-bugfix/sqlobject/dbconnection.py	2006-08-29 14:20:29 UTC (rev 1888)
@@ -95,7 +95,7 @@
             else:
                 host, rest = rest.split('/', 1)
         if host and host.find('@') != -1:
-            user, host = host.split('@', 1)
+            user, host = host.rsplit('@', 1)
             if user.find(':') != -1:
                 user, password = user.split(':', 1)
             else:

Modified: SQLObject/branches/0.7-bugfix/sqlobject/tests/test_parse.py
===================================================================
--- SQLObject/branches/0.7-bugfix/sqlobject/tests/test_parse.py	2006-08-29 14:20:14 UTC (rev 1887)
+++ SQLObject/branches/0.7-bugfix/sqlobject/tests/test_parse.py	2006-08-29 14:20:29 UTC (rev 1888)
@@ -48,6 +48,14 @@
     assert path == "/full/path/to/socket/database"
     assert args == {}
 
+    user, password, host, port, path, args = _parseURI("postgres://user:at@inpwd@host/database")
+    assert user == "user"
+    assert password == "at@inpwd"
+    assert host == "host"
+    assert port is None
+    assert path == "/database"
+    assert args == {}
+
     user, password, host, port, path, args = _parseURI("sqlite:///full/path/to/database")
     assert user is None
     assert password is None

[SQL-CVS] r1887 - in SQLObject/trunk/sqlobject: . tests

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

Author: phd
Date: 2006-08-29 08:20:14 -0600 (Tue, 29 Aug 2006)
New Revision: 1887

Modified:
   SQLObject/trunk/sqlobject/dbconnection.py
   SQLObject/trunk/sqlobject/tests/test_parse.py
Log:
Applied a patch by Ivan Horvath <i.h...@al...> to fix the bug 1426574:
Database passwords with '@' don't parse correctly.


Modified: SQLObject/trunk/sqlobject/dbconnection.py
===================================================================
--- SQLObject/trunk/sqlobject/dbconnection.py	2006-08-28 16:28:28 UTC (rev 1886)
+++ SQLObject/trunk/sqlobject/dbconnection.py	2006-08-29 14:20:14 UTC (rev 1887)
@@ -95,7 +95,7 @@
             else:
                 host, rest = rest.split('/', 1)
         if host and host.find('@') != -1:
-            user, host = host.split('@', 1)
+            user, host = host.rsplit('@', 1)
             if user.find(':') != -1:
                 user, password = user.split(':', 1)
             else:

Modified: SQLObject/trunk/sqlobject/tests/test_parse.py
===================================================================
--- SQLObject/trunk/sqlobject/tests/test_parse.py	2006-08-28 16:28:28 UTC (rev 1886)
+++ SQLObject/trunk/sqlobject/tests/test_parse.py	2006-08-29 14:20:14 UTC (rev 1887)
@@ -48,6 +48,14 @@
     assert path == "/full/path/to/socket/database"
     assert args == {}
 
+    user, password, host, port, path, args = _parseURI("postgres://user:at@inpwd@host/database")
+    assert user == "user"
+    assert password == "at@inpwd"
+    assert host == "host"
+    assert port is None
+    assert path == "/database"
+    assert args == {}
+
     user, password, host, port, path, args = _parseURI("sqlite:///full/path/to/database")
     assert user is None
     assert password is None

[SQL-CVS] [ sqlobject-Bugs-1547743 ] sqlobject-0.7.1dev_r1884, Mysql and unicode

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

Bugs item #1547743, was opened at 2006-08-28 10:57
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=1547743&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: MySQL
Group: SQLObject release (specify)
Status: Open
Resolution: None
Priority: 5
Submitted By: dado1945 (dado1945)
Assigned to: Nobody/Anonymous (nobody)
Summary: sqlobject-0.7.1dev_r1884, Mysql and unicode

Initial Comment:
Please, check if suplied string is not unicode already
in mysqlconnection.py. There is extract how it could
look like:

import types
.....
                if self.need_unicode:
                    # For MysqlDB 1.2.1 and later, we go
                    # encoding->unicode->charset (in
the mysql db)
                    if type(query) != types.UnicodeType:
                        query = unicode(query,
self.encoding)
                    return cursor.execute(query)
                else:
                    return cursor.execute(query)
.....

Or use "isinstance(query, unicode)". I don't know what
different we will have and which one is better to use
however.

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

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540672&aid=1547743&group_id=74338

[SQL-CVS] [ sqlobject-Patches-1539325 ] fromDatabase and delColumn for sqlite

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

Patches item #1539325, was opened at 2006-08-12 14:42
Message generated for change (Comment added) made by abelcodemonk
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540674&aid=1539325&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: Arlo Belshee (abelcodemonk)
Assigned to: Oleg Broytmann (phd)
Summary: fromDatabase and delColumn for sqlite

Initial Comment:
This patch implements both fromDatabase and delColumn 
for sqlite. I added sqlite to the appropriate 
existing tests; it passes them.

fromDatabase is done by parsing the table creation 
sql from the sqlite_master table. This will work with 
any sqlite 3, and I think dates back into sqlite 2, 
but I'm not sure.

delColumn is implemented by renaming the table, 
creating a new table without the one column, then 
copying all the data over. This is obviously not as 
efficient as on DBs that support drop column 
directly, but it works.

This implementation of dropColumn does deal correctly 
with triggers, foreign keys, and indices that 
reference the table whose column is being dropped. It 
does not deal correctly with triggers & indices 
attached directly to the table being altered. 
Attached items will be lost. Extending the 
implementation to support re-creating those indices & 
triggers should be straightforward, but is beyond my 
needs. You are welcome to do so.

Finally, this patch slightly improves the 
implementation of addColumn for sqlite. It performs a 
vacuum operation after each column add. See the 
sqlite documentation for why this is a good idea.


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

>Comment By: Arlo Belshee (abelcodemonk)
Date: 2006-08-27 09:20

Message:
Logged In: YES 
user_id=1027257

I've slightly update my previous patch. Feel free to use 
whichever version you'd like - I just want to get these 
new functions into the main code line as quickly as 
possible. I need them in order to support migrations in 
TurboGears.

This update does two things:
 1. svn:ignore *.pyc in a bunch of directories.
 2. Add sqlite to the -transactions list in the 
supportsMatrix.

Why:

1: I find it easier to see what needs to get added to svn 
if I don't have clutter on the various "modified files" 
lists. Thus, I make sure to always tell svn how to handle 
each file: ignore it or commit it. I find it reduces the 
chances of bad checkins.

2: sqlite fails several tests that require transactions. 
It really doesn't handle them well. This may be dues to my 
adding vacuuming to addColumn, or it may be new failures 
because I added sqlite to the list of DBs that support 
fromDatabase() and delColumn. In any case, I found this to 
be the fastest reasonable way to express what SQLite can 
do. Feel free to modify if you want to go another way.

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

Comment By: Arlo Belshee (abelcodemonk)
Date: 2006-08-27 09:09

Message:
Logged In: YES 
user_id=1027257

Why vacuum after addColumn:

According to SQLite's docs, alter table add column doesn't 
add the column to the table's main location in the file, 
but rather creates it at the end of the file and the DB 
then takes care of pretending that the column adjoins all 
the other columns.

Unfortunately, older versions of SQLite don't do this 
pretending, so can't read a table that has had a column 
added. Vacuum re-creates the table as a single table, 
meaning that all versions can then read it.

Delete column, being not supported directly by SQLite, 
doesn't do anything funny. Since my implementation is just 
to create a new table without the column being dropped, 
all the tables remain together without any weird extra 
space, and a vacuum is not needed.

Why I added semicolons:

No, they are not needed, if you are using pysqlite. 
However, they are required in the interactive SQLite 
interpreter. I added the semicolons so that I could 
quickly paste generated SQL statements into the 
interactive console while I was trying to get things 
working.

Since the semicolons have no effect with pysqlite, I just 
left them in for the next time I wan't to do this. Feel 
free to take them out if you'd like.

On the name of delColumn():

Feel free to name it delColumn. I gave the longer and 
clear name to the low-level implementation primarily to 
highlight the performance implications involved. The 
public-facing objects still expose delColumn; that is 
translated to a call to recreateTableWithoutColumn() if 
the sqliteconnection is being used.

So, external programs use it as deleteColumn(), but my 
patch is easier to understand and the performance 
implications are clearer.

Mostly, I was hoping that that would speed patch 
approval. ;) Feel free to name the function as you will.

On transactions:

Yeah, you can't vacuum in transactions. However, sqlite's 
support for transactions is already fairly problematic. I 
find I can't use transactions with SQLite on any real 
problems anyway, so I don't worry about it. When I need 
transactions, then I generally need a more full-featured 
DBMS than SQLite.

If you have a different opinion on this, you could remove 
the call to vacuum in addColumn(). You are trading off 
read speed & the ability for older versions of sqlite to 
read such a modified table. Yer call.



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

Comment By: Oleg Broytmann (phd)
Date: 2006-08-15 06:08

Message:
Logged In: YES 
user_id=4799

Thank you very much! recreateTableWithoutColumn() is an
excellent solution!

I have some questions about the patch. Why have you added
semicolons? I think they are not needed.

Why VACUUM after addColumn()? It is more logical to do
VACUUM after delColumn(), I suppose. And, BTW, "This command
will fail if there is an active transaction." (from the
SQLite documentation).

Your patch deletes .delColumn(). But I have a number of
programs that relies on a presence of the method (even if
it's empty). I think you could just call
recreateTableWithoutColumn() from delColumn() or raise a
warning or an exception.

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

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540674&aid=1539325&group_id=74338

[SQL-CVS] [ sqlobject-Patches-1539325 ] fromDatabase and delColumn for sqlite

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

Patches item #1539325, was opened at 2006-08-12 14:42
Message generated for change (Comment added) made by abelcodemonk
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540674&aid=1539325&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: Arlo Belshee (abelcodemonk)
Assigned to: Oleg Broytmann (phd)
Summary: fromDatabase and delColumn for sqlite

Initial Comment:
This patch implements both fromDatabase and delColumn 
for sqlite. I added sqlite to the appropriate 
existing tests; it passes them.

fromDatabase is done by parsing the table creation 
sql from the sqlite_master table. This will work with 
any sqlite 3, and I think dates back into sqlite 2, 
but I'm not sure.

delColumn is implemented by renaming the table, 
creating a new table without the one column, then 
copying all the data over. This is obviously not as 
efficient as on DBs that support drop column 
directly, but it works.

This implementation of dropColumn does deal correctly 
with triggers, foreign keys, and indices that 
reference the table whose column is being dropped. It 
does not deal correctly with triggers & indices 
attached directly to the table being altered. 
Attached items will be lost. Extending the 
implementation to support re-creating those indices & 
triggers should be straightforward, but is beyond my 
needs. You are welcome to do so.

Finally, this patch slightly improves the 
implementation of addColumn for sqlite. It performs a 
vacuum operation after each column add. See the 
sqlite documentation for why this is a good idea.


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

>Comment By: Arlo Belshee (abelcodemonk)
Date: 2006-08-27 09:09

Message:
Logged In: YES 
user_id=1027257

Why vacuum after addColumn:

According to SQLite's docs, alter table add column doesn't 
add the column to the table's main location in the file, 
but rather creates it at the end of the file and the DB 
then takes care of pretending that the column adjoins all 
the other columns.

Unfortunately, older versions of SQLite don't do this 
pretending, so can't read a table that has had a column 
added. Vacuum re-creates the table as a single table, 
meaning that all versions can then read it.

Delete column, being not supported directly by SQLite, 
doesn't do anything funny. Since my implementation is just 
to create a new table without the column being dropped, 
all the tables remain together without any weird extra 
space, and a vacuum is not needed.

Why I added semicolons:

No, they are not needed, if you are using pysqlite. 
However, they are required in the interactive SQLite 
interpreter. I added the semicolons so that I could 
quickly paste generated SQL statements into the 
interactive console while I was trying to get things 
working.

Since the semicolons have no effect with pysqlite, I just 
left them in for the next time I wan't to do this. Feel 
free to take them out if you'd like.

On the name of delColumn():

Feel free to name it delColumn. I gave the longer and 
clear name to the low-level implementation primarily to 
highlight the performance implications involved. The 
public-facing objects still expose delColumn; that is 
translated to a call to recreateTableWithoutColumn() if 
the sqliteconnection is being used.

So, external programs use it as deleteColumn(), but my 
patch is easier to understand and the performance 
implications are clearer.

Mostly, I was hoping that that would speed patch 
approval. ;) Feel free to name the function as you will.

On transactions:

Yeah, you can't vacuum in transactions. However, sqlite's 
support for transactions is already fairly problematic. I 
find I can't use transactions with SQLite on any real 
problems anyway, so I don't worry about it. When I need 
transactions, then I generally need a more full-featured 
DBMS than SQLite.

If you have a different opinion on this, you could remove 
the call to vacuum in addColumn(). You are trading off 
read speed & the ability for older versions of sqlite to 
read such a modified table. Yer call.



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

Comment By: Oleg Broytmann (phd)
Date: 2006-08-15 06:08

Message:
Logged In: YES 
user_id=4799

Thank you very much! recreateTableWithoutColumn() is an
excellent solution!

I have some questions about the patch. Why have you added
semicolons? I think they are not needed.

Why VACUUM after addColumn()? It is more logical to do
VACUUM after delColumn(), I suppose. And, BTW, "This command
will fail if there is an active transaction." (from the
SQLite documentation).

Your patch deletes .delColumn(). But I have a number of
programs that relies on a presence of the method (even if
it's empty). I think you could just call
recreateTableWithoutColumn() from delColumn() or raise a
warning or an exception.

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

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540674&aid=1539325&group_id=74338

[SQL-CVS] [ sqlobject-Bugs-1547004 ] SQLObject does not reconnect

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

Bugs item #1547004, was opened at 2006-08-26 00:05
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=1547004&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: DBM
Group: SQLObject from repository
Status: Open
Resolution: None
Priority: 5
Submitted By: elpargo (elpargo)
Assigned to: Nobody/Anonymous (nobody)
Summary: SQLObject does not reconnect

Initial Comment:
this is from the 0.7bug fix branch

a lot of background has been created on this in
Turbogears bug tracker.

mainly in
http://trac.turbogears.org/turbogears/ticket/872
http://trac.turbogears.org/turbogears/ticket/781

Basically the connection after being idle is drop and
SO engine is not aware of this. 

Many patches have been proposed but none accepted.



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

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540672&aid=1547004&group_id=74338

[SQL-CVS] [ sqlobject-Bugs-1305062 ] 0.7a: RelatedJoin() glue table has no constraints

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

Bugs item #1305062, was opened at 2005-09-26 14:07
Message generated for change (Comment added) made by l33tminion
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540672&aid=1305062&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: Andy Dustman (adustman)
Assigned to: Nobody/Anonymous (nobody)
Summary: 0.7a: RelatedJoin() glue table has no constraints

Initial Comment:
Consider the example from the online 0.6.1
documentation, with the User and Role objects, where
each object does a RelatedJoin() to the other. This
results in a glue table that looks like this:

CREATE TABLE user_role (
user_id INT NOT NULL,
role_id INT NOT NULL
)

The schema lacks any FOREIGN KEY constraints, and a
PRIMARY KEY. It should probalby look more like this:

CREATE TABLE user_role (
user_id INT NOT NULL, CONSTRAINT user_id_exists FOREIGN
KEY (user_id) REFERENCES user (id),
role_id INT NOT NULL, CONSTRAINT role_id_exists FOREIGN
KEY (role_id) REFERENCES role (id),
PRIMARY KEY (user_id, role_id)
)

ON DELETE CASCADE may be needed as well.

Note that I am using patch #1299109 to add FOREIGN KEY
support for MySQL. However, looking at join.py, it does
not seem to do anything with foreign keys, so I do not
think this is a dependency on MySQL; I think this
behavior should be reproducable on other databases. The
intermediate table schema is created in
dbconnection.DBAPI._SO_createJoinTableSQL(), and it
calls self.joinSQLType(join) to get the SQL for the
column. In all the implementations (except Sybase which
is slightly different), it returns 'INT NOT NULL'

The table creation order looks like it could be a
problem, as the glue/intermediate table seems to be
created before the second of the other two tables,
which would produce referential integrity problems
(i.e.reference to a non-existent table).

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

Comment By: L33tminion (l33tminion)
Date: 2006-08-25 14:25

Message:
Logged In: YES 
user_id=1583993

At the very least, adding ON DELETE CASCADE, etc., to the
join table should be an option.  Working around that is
_really_ annoying.

I second adustman's comments.

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

Comment By: nyenyec (nyenyec)
Date: 2006-02-15 01:31

Message:
Logged In: YES 
user_id=6142

This is still a problem in sqlobject 0.7.1dev-r1457

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

Comment By: Andy Dustman (adustman)
Date: 2005-10-31 17:34

Message:
Logged In: YES 
user_id=71372

BTW, the table creation order comment is really more of a
problem for Turbogears.

I've done a bit of testing by altering an existing glue
table to use ON DELETE CASCADE and it seems to work as
expected/desired. Without ON DELETE CASCADE, you have to
delete the glue entry tables manually, (typically with
obj.removeXXX(other_id)); otherwise your glue table has a
bunch of dead entries, though if you have the FOREIGN KEY
entry without ON CASCADE DELETE, then you can't remove the
parent objects without first deleting the glue records.

In other words, having FOREIGN KEY ... REFERENCES ... ON
DELETE CASCADE keeps your glue table clean when you remove
either object of the RelatedJoin().

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

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540672&aid=1305062&group_id=74338

[SQL-CVS] [ sqlobject-Bugs-1546294 ] inheritance and cascade='null' do not work together

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

Bugs item #1546294, was opened at 2006-08-24 18:11
Message generated for change (Comment added) made by rick446
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540672&aid=1546294&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: Rick Copeland (rick446)
Assigned to: Nobody/Anonymous (nobody)
Summary: inheritance and cascade='null' do not work together

Initial Comment:
(Based on development version 0.8dev_r1883)

Start with the following schema:

class A(InheritableSQLObject):
    dummy=IntCol()

class B(A):
    c=ForeignKey('C', cascade='null')

class C(SQLObject):
    dummy=IntCol()

C.createTable()
A.createTable()
B.createTable()


c=C(dummy=1)
b=B(cID=c.id, dummy=1)
c.destroySelf()
***exception***

This can be fixed by updating SQLObject.destroySelf to
use SQLBuilder by the following changes:

in sqlobject/main.py:
(insert at top)
from sqlbuilder import OR

(replacing line starting with "query.append("%s = (%...
query.append(getattr(k.q, col.name)==self.id)

(replacing line "' OR '.join(query)")
query=OR(*query)


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

>Comment By: Rick Copeland (rick446)
Date: 2006-08-24 18:46

Message:
Logged In: YES 
user_id=1583356

I have created patch #1546314 to address this bug.

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

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540672&aid=1546294&group_id=74338

[SQL-CVS] [ sqlobject-Patches-1546314 ] Fix bug 1546294

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

Patches item #1546314, was opened at 2006-08-24 18:44
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=1546314&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: Rick Copeland (rick446)
Assigned to: Nobody/Anonymous (nobody)
Summary: Fix bug 1546294 

Initial Comment:
Here is the patch that fixes bug 1546294.

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

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540674&aid=1546314&group_id=74338

[SQL-CVS] [ sqlobject-Bugs-1546294 ] inheritance and cascade='null' do not work together

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

Bugs item #1546294, was opened at 2006-08-24 18:11
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=1546294&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: Rick Copeland (rick446)
Assigned to: Nobody/Anonymous (nobody)
Summary: inheritance and cascade='null' do not work together

Initial Comment:
(Based on development version 0.8dev_r1883)

Start with the following schema:

class A(InheritableSQLObject):
    dummy=IntCol()

class B(A):
    c=ForeignKey('C', cascade='null')

class C(SQLObject):
    dummy=IntCol()

C.createTable()
A.createTable()
B.createTable()


c=C(dummy=1)
b=B(cID=c.id, dummy=1)
c.destroySelf()
***exception***

This can be fixed by updating SQLObject.destroySelf to
use SQLBuilder by the following changes:

in sqlobject/main.py:
(insert at top)
from sqlbuilder import OR

(replacing line starting with "query.append("%s = (%...
query.append(getattr(k.q, col.name)==self.id)

(replacing line "' OR '.join(query)")
query=OR(*query)


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

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540672&aid=1546294&group_id=74338

[SQL-CVS] r1885 - in home/phd/SQLObject/paramstyles/sqlobject: . tests

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

Author: phd
Date: 2006-08-24 09:41:31 -0600 (Thu, 24 Aug 2006)
New Revision: 1885

Modified:
   home/phd/SQLObject/paramstyles/sqlobject/sqlbuilder.py
   home/phd/SQLObject/paramstyles/sqlobject/tests/test_aliases.py
Log:
Merged patches from the revisions 1881:1884 from the trunk

Modified: home/phd/SQLObject/paramstyles/sqlobject/sqlbuilder.py
===================================================================
--- home/phd/SQLObject/paramstyles/sqlobject/sqlbuilder.py	2006-08-24 15:38:41 UTC (rev 1884)
+++ home/phd/SQLObject/paramstyles/sqlobject/sqlbuilder.py	2006-08-24 15:41:31 UTC (rev 1885)
@@ -419,6 +419,8 @@
 ########################################
 
 class AliasField(Field, SQLListMixin):
+    as_string = '' # set it to "AS" if your database requires it
+
     def __init__(self, tableName, fieldName, alias):
         Field.__init__(self, tableName, fieldName)
         self.alias = alias
@@ -427,7 +429,7 @@
         return self.alias + "." + self.fieldName
 
     def tablesUsedImmediate(self):
-        return ["%s AS %s" % (self.tableName, self.alias)]
+        return ["%s %s %s" % (self.tableName, self.as_string, self.alias)]
 
 class AliasTable(Table):
     FieldClass = AliasField

Modified: home/phd/SQLObject/paramstyles/sqlobject/tests/test_aliases.py
===================================================================
--- home/phd/SQLObject/paramstyles/sqlobject/tests/test_aliases.py	2006-08-24 15:38:41 UTC (rev 1884)
+++ home/phd/SQLObject/paramstyles/sqlobject/tests/test_aliases.py	2006-08-24 15:41:31 UTC (rev 1885)
@@ -15,7 +15,7 @@
     alias = Alias(JoinAlias)
     select = JoinAlias.select(JoinAlias.q.parent == alias.q.name)
     assert str(select) == \
-        "SELECT join_alias.id, join_alias.name, join_alias.parent FROM join_alias AS join_alias_alias1, join_alias WHERE ((join_alias.parent) = (join_alias_alias1.name))"
+        "SELECT join_alias.id, join_alias.name, join_alias.parent FROM join_alias  join_alias_alias1, join_alias WHERE ((join_alias.parent) = (join_alias_alias1.name))"
 
 def test_2perform_join():
     setupClass(JoinAlias)

[SQL-CVS] [ sqlobject-Bugs-1516394 ] SQLbuilder aliases fail for Firebird

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

Bugs item #1516394, was opened at 2006-07-03 18:56
Message generated for change (Comment added) made by phd
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540672&aid=1516394&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: Firebird/Interbase
Group: SQLObject release (specify)
>Status: Closed
>Resolution: Fixed
Priority: 5
Submitted By: Henrik Weber (hweber)
>Assigned to: Oleg Broytmann (phd)
Summary: SQLbuilder aliases fail for Firebird

Initial Comment:
Creating a table alias with SQLBuilder will alway 
create alias clauses of the form:

orig_table AS alias_name

SQLbuilder does not discriminate between different SQL 
dialects here.

For Firebird this syntax variant fails, because 
aliases are defined without the AS keyword there:

orig_table alias_name

The AS keyword will generate a syntax error with 
Firebird.

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

>Comment By: Oleg Broytmann (phd)
Date: 2006-08-24 19:38

Message:
Logged In: YES 
user_id=4799

I found most database (I tested MySQL, PostgreSQL and
SQLite) ignore that noisy "AS" so I just removed it from the
queries for all backends. Committed in the revision 1883 to
the trunk, rev. 1884 to the 0.7-branch.

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

Comment By: Henrik Weber (hweber)
Date: 2006-07-04 22:37

Message:
Logged In: YES 
user_id=121229

Forgot to mention: this is with SQLobject 0.7.0

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

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540672&aid=1516394&group_id=74338

[SQL-CVS] r1884 - in SQLObject/branches/0.7-bugfix/sqlobject: . tests

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

Author: phd
Date: 2006-08-24 09:38:41 -0600 (Thu, 24 Aug 2006)
New Revision: 1884

Modified:
   SQLObject/branches/0.7-bugfix/sqlobject/sqlbuilder.py
   SQLObject/branches/0.7-bugfix/sqlobject/tests/test_aliases.py
Log:
Fixed the bug 1516394: SQLbuilder aliases fail for Firebird.


Modified: SQLObject/branches/0.7-bugfix/sqlobject/sqlbuilder.py
===================================================================
--- SQLObject/branches/0.7-bugfix/sqlobject/sqlbuilder.py	2006-08-24 15:37:53 UTC (rev 1883)
+++ SQLObject/branches/0.7-bugfix/sqlobject/sqlbuilder.py	2006-08-24 15:38:41 UTC (rev 1884)
@@ -379,6 +379,8 @@
 ########################################
 
 class AliasField(Field):
+    as_string = '' # set it to "AS" if your database requires it
+
     def __init__(self, tableName, fieldName, alias):
         Field.__init__(self, tableName, fieldName)
         self.alias = alias
@@ -387,7 +389,7 @@
         return self.alias + "." + self.fieldName
 
     def tablesUsedImmediate(self):
-        return ["%s AS %s" % (self.tableName, self.alias)]
+        return ["%s %s %s" % (self.tableName, self.as_string, self.alias)]
 
 class AliasTable(Table):
     FieldClass = AliasField

Modified: SQLObject/branches/0.7-bugfix/sqlobject/tests/test_aliases.py
===================================================================
--- SQLObject/branches/0.7-bugfix/sqlobject/tests/test_aliases.py	2006-08-24 15:37:53 UTC (rev 1883)
+++ SQLObject/branches/0.7-bugfix/sqlobject/tests/test_aliases.py	2006-08-24 15:38:41 UTC (rev 1884)
@@ -15,7 +15,7 @@
     alias = Alias(JoinAlias)
     select = JoinAlias.select(JoinAlias.q.parent == alias.q.name)
     assert str(select) == \
-        "SELECT join_alias.id, join_alias.name, join_alias.parent FROM join_alias AS join_alias_alias1, join_alias WHERE ((join_alias.parent) = (join_alias_alias1.name))"
+        "SELECT join_alias.id, join_alias.name, join_alias.parent FROM join_alias  join_alias_alias1, join_alias WHERE ((join_alias.parent) = (join_alias_alias1.name))"
 
 def test_2perform_join():
     setupClass(JoinAlias)

[SQL-CVS] r1883 - in SQLObject/trunk/sqlobject: . tests

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

Author: phd
Date: 2006-08-24 09:37:53 -0600 (Thu, 24 Aug 2006)
New Revision: 1883

Modified:
   SQLObject/trunk/sqlobject/sqlbuilder.py
   SQLObject/trunk/sqlobject/tests/test_aliases.py
Log:
In the course of fixing the bug 1516394: SQLbuilder aliases fail for Firebird
I found most database (I tested MySQL, PostgreSQL and SQLite) ignore
that noisy "AS" so I just removed it from the queries.


Modified: SQLObject/trunk/sqlobject/sqlbuilder.py
===================================================================
--- SQLObject/trunk/sqlobject/sqlbuilder.py	2006-08-24 14:38:27 UTC (rev 1882)
+++ SQLObject/trunk/sqlobject/sqlbuilder.py	2006-08-24 15:37:53 UTC (rev 1883)
@@ -382,6 +382,8 @@
 ########################################
 
 class AliasField(Field):
+    as_string = '' # set it to "AS" if your database requires it
+
     def __init__(self, tableName, fieldName, alias):
         Field.__init__(self, tableName, fieldName)
         self.alias = alias
@@ -390,7 +392,7 @@
         return self.alias + "." + self.fieldName
 
     def tablesUsedImmediate(self):
-        return ["%s AS %s" % (self.tableName, self.alias)]
+        return ["%s %s %s" % (self.tableName, self.as_string, self.alias)]
 
 class AliasTable(Table):
     FieldClass = AliasField

Modified: SQLObject/trunk/sqlobject/tests/test_aliases.py
===================================================================
--- SQLObject/trunk/sqlobject/tests/test_aliases.py	2006-08-24 14:38:27 UTC (rev 1882)
+++ SQLObject/trunk/sqlobject/tests/test_aliases.py	2006-08-24 15:37:53 UTC (rev 1883)
@@ -15,7 +15,7 @@
     alias = Alias(JoinAlias)
     select = JoinAlias.select(JoinAlias.q.parent == alias.q.name)
     assert str(select) == \
-        "SELECT join_alias.id, join_alias.name, join_alias.parent FROM join_alias AS join_alias_alias1, join_alias WHERE ((join_alias.parent) = (join_alias_alias1.name))"
+        "SELECT join_alias.id, join_alias.name, join_alias.parent FROM join_alias  join_alias_alias1, join_alias WHERE ((join_alias.parent) = (join_alias_alias1.name))"
 
 def test_2perform_join():
     setupClass(JoinAlias)

[SQL-CVS] r1881 - in home/phd/SQLObject/paramstyles/sqlobject: . firebird mysql postgres

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

Author: phd
Date: 2006-08-22 09:04:33 -0600 (Tue, 22 Aug 2006)
New Revision: 1881

Modified:
   home/phd/SQLObject/paramstyles/sqlobject/firebird/firebirdconnection.py
   home/phd/SQLObject/paramstyles/sqlobject/index.py
   home/phd/SQLObject/paramstyles/sqlobject/main.py
   home/phd/SQLObject/paramstyles/sqlobject/mysql/mysqlconnection.py
   home/phd/SQLObject/paramstyles/sqlobject/postgres/pgconnection.py
Log:
Merged patches from the revisions 1868:1880 from the trunk

Modified: home/phd/SQLObject/paramstyles/sqlobject/firebird/firebirdconnection.py
===================================================================
--- home/phd/SQLObject/paramstyles/sqlobject/firebird/firebirdconnection.py	2006-08-22 14:59:53 UTC (rev 1880)
+++ home/phd/SQLObject/paramstyles/sqlobject/firebird/firebirdconnection.py	2006-08-22 15:04:33 UTC (rev 1881)
@@ -90,8 +90,8 @@
         for each table this method to work."""
         table = soInstance.sqlmeta.table
         idName = soInstance.sqlmeta.idName
-        sequenceName = getattr(soInstance, '_idSequence',
-                               'GEN_%s' % table)
+        sequenceName = soInstance.sqlmeta.idSequence or \
+                               'GEN_%s' % table
         c = conn.cursor()
         if id is None:
             c.execute('SELECT gen_id(%s,1) FROM rdb$database'

Modified: home/phd/SQLObject/paramstyles/sqlobject/index.py
===================================================================
--- home/phd/SQLObject/paramstyles/sqlobject/index.py	2006-08-22 14:59:53 UTC (rev 1880)
+++ home/phd/SQLObject/paramstyles/sqlobject/index.py	2006-08-22 15:04:33 UTC (rev 1881)
@@ -156,7 +156,7 @@
         self.kw['name'] = value
 
     def _get_name(self):
-        return kw['name']
+        return self.kw['name']
 
     def _set_name(self, value):
         self.setName(value)

Modified: home/phd/SQLObject/paramstyles/sqlobject/main.py
===================================================================
--- home/phd/SQLObject/paramstyles/sqlobject/main.py	2006-08-22 14:59:53 UTC (rev 1880)
+++ home/phd/SQLObject/paramstyles/sqlobject/main.py	2006-08-22 15:04:33 UTC (rev 1881)
@@ -180,6 +180,7 @@
 
     table = None
     idName = None
+    idSequence = None
     # This function is used to coerce IDs into the proper format,
     # so you should replace it with str, or another function, if you
     # aren't using integer IDs
@@ -879,7 +880,7 @@
         """
         for attr in ['_table', '_lazyUpdate', '_style', '_idName',
                      '_defaultOrder', '_cacheValues', '_registry',
-                     '_idType', '_fromDatabase']:
+                     '_idType', '_fromDatabase', '_idSequence']:
             if new_attrs.has_key(attr):
                 new_name = attr[1:]
                 deprecated("%r is deprecated; please set the %r "
@@ -1414,7 +1415,7 @@
                 # This join has requested not to be created
                 continue
             sql.append(conn._SO_createJoinTableSQL(join))
-        return '\n'.join(sql)
+        return ';\n'.join(sql)
     createJoinTablesSQL = classmethod(createJoinTablesSQL)
 
     def createIndexes(cls, ifNotExists=False, connection=None):
@@ -1432,7 +1433,7 @@
             if not index:
                 continue
             sql.append(conn.createIndexSQL(cls, index))
-        return '\n'.join(sql)
+        return ';\n'.join(sql)
     createIndexesSQL = classmethod(createIndexesSQL)
 
     def _getJoinsToCreate(cls):

Modified: home/phd/SQLObject/paramstyles/sqlobject/mysql/mysqlconnection.py
===================================================================
--- home/phd/SQLObject/paramstyles/sqlobject/mysql/mysqlconnection.py	2006-08-22 14:59:53 UTC (rev 1880)
+++ home/phd/SQLObject/paramstyles/sqlobject/mysql/mysqlconnection.py	2006-08-22 15:04:33 UTC (rev 1881)
@@ -33,7 +33,7 @@
                 "read_default_file", "read_default_group", "charset"):
             if key in kw:
                 self.kw[key] = col.popKey(kw, key)
-        for key in ("connect_time", "compress", "named_pipe", "use_unicode",
+        for key in ("connect_timeout", "compress", "named_pipe", "use_unicode",
                 "client_flag", "local_infile"):
             if key in kw:
                 self.kw[key] = int(col.popKey(kw, key))

Modified: home/phd/SQLObject/paramstyles/sqlobject/postgres/pgconnection.py
===================================================================
--- home/phd/SQLObject/paramstyles/sqlobject/postgres/pgconnection.py	2006-08-22 14:59:53 UTC (rev 1880)
+++ home/phd/SQLObject/paramstyles/sqlobject/postgres/pgconnection.py	2006-08-22 15:04:33 UTC (rev 1881)
@@ -119,8 +119,8 @@
     def _queryInsertID(self, conn, soInstance, id, names, values):
         table = soInstance.sqlmeta.table
         idName = soInstance.sqlmeta.idName
-        sequenceName = getattr(soInstance, '_idSequence',
-                               '%s_%s_seq' % (table, idName))
+        sequenceName = soInstance.sqlmeta.idSequence or \
+                               '%s_%s_seq' % (table, idName)
         c = conn.cursor()
         if id is None:
             c.execute("SELECT NEXTVAL('%s')" % sequenceName)

[SQL-CVS] [ sqlobject-Bugs-1541095 ] Capital letters in table/column names

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

Bugs item #1541095, was opened at 2006-08-16 11:19
Message generated for change (Settings changed) made by phd
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540672&aid=1541095&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: SQLObject release (specify)
>Status: Closed
>Resolution: Wont Fix
Priority: 5
Submitted By: Michel Albert (exhuma)
>Assigned to: Oleg Broytmann (phd)
Summary: Capital letters in table/column names

Initial Comment:
Postgre requires table names, column names and so on 
to be enclosed in double quotes if they contain 
capital letters. Otherwise it assumes that they are 
all lower-case. In order to be more DBMS-agnostic, 
SQLObject should deal with that.

It is possible to do something like:
   lastName = StringCol(dbName='"lastName"')

It would be nicer to be able to do this:
   lastName = StringCol(dbName='lastName')

The current solution is error-prone. Also, when doing 
this, it get's a bit confusing as you have to leave 
out the quotes when doing joins:

   wrong:
      address = SingleJoin('"Address"')
   correct:
      address = SingleJoin('Address')

as the joins relate to Python objects, not directly to 
tables.

I have to deal with this, as I am migrating a web-
application from PHP to Python, and the database 
already exists and modifying it is not an option.


Best regards,

Michel Albert

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

>Comment By: Oleg Broytmann (phd)
Date: 2006-08-22 19:03

Message:
Logged In: YES 
user_id=4799

Use sqlmeta.idSequence to name your sequence.

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

Comment By: Nobody/Anonymous (nobody)
Date: 2006-08-16 20:45

Message:
Logged In: NO 

This issue actually *breaks* postgres!:

--- In postgres: -----------------------------------
database=> SELECT NEXTVAL('"Hotel_idHotel_seq"');
 nextval
---------
       1
(1 row)
----------------------------------------------------


--- Using SQLObject --------------------------------
>>> x = Hotel(name="foo", address=None, owner=None, 
establishment=None)
Traceback (most recent call last):
  File "<console>", line 1, in ?
  File "/usr/lib/python2.4/site-packages/SQLObject-
0.7.1dev_r1675-py2.4.egg/sqlobject/declarative.py", line 
92, in _wrapper
    return_value = fn(self, *args, **kwargs)
  File "/usr/lib/python2.4/site-packages/SQLObject-
0.7.1dev_r1675-py2.4.egg/sqlobject/main.py", line 1197, in 
__init__
    self._create(id, **kw)
  File "/usr/lib/python2.4/site-packages/SQLObject-
0.7.1dev_r1675-py2.4.egg/sqlobject/main.py", line 1224, in 
_create
    self._SO_finishCreate(id)
  File "/usr/lib/python2.4/site-packages/SQLObject-
0.7.1dev_r1675-py2.4.egg/sqlobject/main.py", line 1248, in 
_SO_finishCreate
    id, names, values)
  File "/usr/lib/python2.4/site-packages/SQLObject-
0.7.1dev_r1675-py2.4.egg/sqlobject/dbconnection.py", line 
759, in queryInsertID
    return self._dbConnection._queryInsertID(
  File "/usr/lib/python2.4/site-packages/SQLObject-
0.7.1dev_r1675-py2.4.egg/sqlobject/postgres/
pgconnection.py", line 126, in _queryInsertID
    c.execute("SELECT NEXTVAL('%s')" % sequenceName)
ProgrammingError: ERROR:  invalid name syntax

SELECT NEXTVAL('"Hotel"_"idHotel"_seq')
----------------------------------------------------

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

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540672&aid=1541095&group_id=74338

[SQL-CVS] [ sqlobject-Bugs-1539087 ] Link to Japanese translation has moved

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

Bugs item #1539087, was opened at 2006-08-12 10:31
Message generated for change (Comment added) made by phd
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540672&aid=1539087&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: None
>Status: Closed
>Resolution: Fixed
Priority: 5
Submitted By: [N/A] (ymasuda)
Assigned to: Nobody/Anonymous (nobody)
Summary: Link to Japanese translation has moved

Initial Comment:
Hello,

I have moved Japanese translation of SQLObject main doc from
http://www.python.jp/~ymasuda/sqlobject/SQLObject.html which is 
listed on http://sqlobject.org/links.html, to http://ymasuda.jp/python/
sqlobject/doc_0.7/SQLObject.html .
Also I'm going underway to translate all doc entries.
These will be available at http://ymasuda.jp/python/sqlobject/doc_0.7/ .

If you have any questions or advices, please tell me.

Thanks 

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

>Comment By: Oleg Broytmann (phd)
Date: 2006-08-22 19:01

Message:
Logged In: YES 
user_id=4799

I have fixed this in the trunk. Don't know when Ian will
update links.html.

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

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=540672&aid=1539087&group_id=74338

[SQL-CVS] r1880 - SQLObject/docs

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

Author: phd
Date: 2006-08-22 08:59:53 -0600 (Tue, 22 Aug 2006)
New Revision: 1880

Modified:
   SQLObject/docs/links.txt
Log:
Fixed the bug 1539087: Link to Japanese translation has moved.


Modified: SQLObject/docs/links.txt
===================================================================
--- SQLObject/docs/links.txt	2006-08-22 14:53:23 UTC (rev 1879)
+++ SQLObject/docs/links.txt	2006-08-22 14:59:53 UTC (rev 1880)
@@ -28,7 +28,7 @@
   <http://formencode.org>`_.
 
 * `Translation of SQLObject documentation to Japanese
-  <http://www.python.jp/~ymasuda/sqlobject/SQLObject.html>`_.
+  <http://ymasuda.jp/python/sqlobject/doc_0.7/>`_.
 
 * `How to Use Database Agnostic SQL in SQLObject
   <http://groovie.org/articles/2005/11/01/how-to-use-database-agnostic-sql-in-sqlobject>`_.