sqlobject-discuss — All things SQLObject-related

Re: [SQLObject] Another newbie question: copying objects?

[Mark M. <mar...@me...>]

What about the connection argument? Do I have to just add that to the dict?

On Mar 9, 2011, at 2:43 AM, Oleg Broytman wrote:

> On Tue, Mar 08, 2011 at 03:56:35PM -0800, Mark McWiggins wrote:
>> Is there a way to do this without specifying all the field names? Right now I am just using two fields:
>> 
>> 	zips = zipcode.select()
>> 
>> 	for z in zips:
>>   	 newz = zipcode(id = z.id, city = z.city, connection = peconnection)
>> 
>> This works, but where there are 14 fields in other tables it's going to be a pain. I couldn't find anything about
>> object copy or assignment in the documents, but isn't there a way to do this?
> 
>   You can convert a row to a dict: z.sqlmeta.asDict() - and pass the
> dict back: newz = zipcode(**z.sqlmeta.asDict())
> 
> Oleg.
> -- 
>     Oleg Broytman            http://phdru.name/            ph...@ph...
>           Programmers don't die, they just GOSUB without RETURN.
> 
> ------------------------------------------------------------------------------
> Colocation vs. Managed Hosting
> A question and answer guide to determining the best fit
> for your organization - today and in the future.
> http://p.sf.net/sfu/internap-sfd2d
> _______________________________________________
> sqlobject-discuss mailing list
> sql...@li...
> https://lists.sourceforge.net/lists/listinfo/sqlobject-discuss

Mark McWiggins
Principal Solutions Engineer
mar...@me...
443-656-3311 ext 344

Re: [SQLObject] SQLObject 1.0.0b1

[Oleg B. <ph...@ph...>]

On Wed, Mar 09, 2011 at 05:47:26PM +0300, Oleg Broytman wrote:
> * Major API change: DB URI parser was changed to use urllib.split*() and
>   unquote(). This means any username/password/path are allowed in DB
>   URIs if they are properly %-encoded, and DB URIs are automatically
>   unquoted.

   I have been trying to preserve backward compatibility as much as
possible. Most URIs are not affected at all. Dirty trick with w32 paths
(C|/path instead of C:/path) is still supported, though C:/path is now
the recommended syntax. Old API is preserved:
connectionForURI(uri, oldUri=True), DBConnection.connectionFromOldURI(),
DBConnection.oldURI(). There is a script sqlobject-convertOldURI that
accepts old-style URI at the command line and prints to stdout the URI
in the new format.
   Anyway - a prolonged beta period.

Oleg.
-- 
     Oleg Broytman            http://phdru.name/            ph...@ph...
           Programmers don't die, they just GOSUB without RETURN.

[SQLObject] SQLObject 1.0.0b1

[Oleg B. <ph...@ph...>]

Hello!

I'm pleased to announce version 1.0.0b1, the first beta of the upcoming
release of branch 1.0 of SQLObject.


What is SQLObject
=================

SQLObject is an object-relational mapper.  Your database tables are described
as classes, and rows are instances of those classes.  SQLObject is meant to be
easy to use and quick to get started with.

SQLObject supports a number of backends: MySQL, PostgreSQL, SQLite,
Firebird, Sybase, MSSQL and MaxDB (also known as SAPDB).


Where is SQLObject
==================

Site:
http://sqlobject.org

Development:
http://sqlobject.org/devel/

Mailing list:
https://lists.sourceforge.net/mailman/listinfo/sqlobject-discuss

Archives:
http://news.gmane.org/gmane.comp.python.sqlobject

Download:
http://cheeseshop.python.org/pypi/SQLObject/1.0.0b1dev-r4354

News and changes:
http://sqlobject.org/News.html


What's New
==========

Features & Interface
--------------------

* Major API change: DB URI parser was changed to use urllib.split*() and
  unquote(). This means any username/password/path are allowed in DB
  URIs if they are properly %-encoded, and DB URIs are automatically
  unquoted.

* In SQLite, id columns are made AUTOINCREMENT.

* Parameter ``backend`` in DB URI is no longer supported, use parameter
  ``driver``.

For a more complete list, please see the news:
http://sqlobject.org/News.html

Oleg.
-- 
     Oleg Broytman            http://phdru.name/            ph...@ph...
           Programmers don't die, they just GOSUB without RETURN.

Re: [SQLObject] Another newbie question: copying objects?

[Oleg B. <ph...@ph...>]

On Tue, Mar 08, 2011 at 03:56:35PM -0800, Mark McWiggins wrote:
> Is there a way to do this without specifying all the field names? Right now I am just using two fields:
> 
> 	zips = zipcode.select()
> 
> 	for z in zips:
>    	 newz = zipcode(id = z.id, city = z.city, connection = peconnection)
> 
> This works, but where there are 14 fields in other tables it's going to be a pain. I couldn't find anything about
> object copy or assignment in the documents, but isn't there a way to do this?

   You can convert a row to a dict: z.sqlmeta.asDict() - and pass the
dict back: newz = zipcode(**z.sqlmeta.asDict())

Oleg.
-- 
     Oleg Broytman            http://phdru.name/            ph...@ph...
           Programmers don't die, they just GOSUB without RETURN.

[SQLObject] Another newbie question: copying objects?

[Mark M. <mar...@me...>]

Hi All,

I have the program I'm working on doing basically what I want now, copying a table's data from one DB to another.

But! Is there a way to do this without specifying all the field names? Right now I am just using two fields:

	zips = zipcode.select()

	for z in zips:
   	 newz = zipcode(id = z.id, city = z.city, connection = peconnection)

This works, but where there are 14 fields in other tables it's going to be a pain. I couldn't find anything about
object copy or assignment in the documents, but isn't there a way to do this?

Thanks for any suggestions.
--
Mark McWiggins
Principal Solutions Engineer
mar...@me...
443-656-3311 ext 344

Re: [SQLObject] Object/row instantiation question

[Oleg B. <ph...@ph...>]

On Tue, Mar 08, 2011 at 11:26:19AM -0800, Mark McWiggins wrote:
> I'm a new SQLObject user ...

   Welcome!

> I have this table:
> 
> CREATE TABLE zipcode (
>     zip integer PRIMARY KEY,
>     city text,
> -- and so forth ...
> 
> with Python representing it:
> 
> class zipcode(SQLObject):
>     class sqlmeta:
>         fromDatabase = True
> 	idName = 'zip'
> 
> But when I try to create rows with:
> 
> for myzip in range(1,99999):
>     z = zipcode(zip = myzip, city = 'zipcity' + str(myzip))
> 
> I get an error
> 
>  File "/usr/local/lib/python2.4/site-packages/SQLObject-1.1dev_r4347-py2.4.egg\
> /sqlobject/main.py", line 1099, in set
>     raise TypeError, "%s.set() got an unexpected keyword argument %s" % (self._\
> _class__.__name__, name)
> TypeError: zipcode.set() got an unexpected keyword argument zip

   In SQLObject the primary key is always called 'id'.

for myzip in range(1, 99999):
    z = zipcode(id=myzip, city='zipcity' + str(myzip))

Oleg.
-- 
     Oleg Broytman            http://phdru.name/            ph...@ph...
           Programmers don't die, they just GOSUB without RETURN.

[SQLObject] Object/row instantiation question

[Mark M. <mar...@me...>]

Hi All,

I'm a new SQLObject user ... I have this table:

CREATE TABLE zipcode (
    zip integer PRIMARY KEY,
    city text,
-- and so forth ...

with Python representing it:

class zipcode(SQLObject):
    class sqlmeta:
        fromDatabase = True
	idName = 'zip'

But when I try to create rows with:



for myzip in range(1,99999):
    z = zipcode(zip = myzip, city = 'zipcity' + str(myzip))

I get an error

 File "/usr/local/lib/python2.4/site-packages/SQLObject-1.1dev_r4347-py2.4.egg\
/sqlobject/main.py", line 1099, in set
    raise TypeError, "%s.set() got an unexpected keyword argument %s" % (self._\
_class__.__name__, name)
TypeError: zipcode.set() got an unexpected keyword argument zip


What am I missing?
--
Mark McWiggins
Principal Solutions Engineer
mar...@me...
443-656-3311 ext 344

Re: [SQLObject] Altering/updating a database

[Oleg B. <ph...@ph...>]

On Sat, Mar 05, 2011 at 08:29:54AM +0000, Timothy W. Grove wrote:
> Thanks for the advice, Oleg. I found a solution which may not be  
> elegant, but it works for me! Instead of trying to work with two  
> different class definitions I've just kept one which declares the column  
> I wish to add to the database and run an update function which deletes  
> the column from the class before adding it again to both the class and  
> the database. My example follows.

   Sometimes I do the same.

> Best regards,
> Tim
>
>    class SignEntry(SQLObject):
>         explanatory_map = PickleCol()
>
>    def SignEntryUpdate():
>         cols = [i.name for i in
>    sqlhub.processConnection.columnsFromSchema("sign_entry", SignEntry)]
>         if 'explanatoryMap' not in cols:
>             SignEntry.sqlmeta.delColumn('explanatory_map') # I know
>    this exists in the class definition
>             SignEntry.sqlmeta.addColumn(PickleCol("explanatory_map"),
>    changeSchema=True)

   Minor optimization in case you are going to test many columns - a set
is perhaps faster for repeated 'in' testing:

        cols = set([i.name for i in sqlhub.processConnection.columnsFromSchema("sign_entry", SignEntry)])

Oleg.
-- 
     Oleg Broytman            http://phdru.name/            ph...@ph...
           Programmers don't die, they just GOSUB without RETURN.

Re: [SQLObject] Altering/updating a database

[Timothy W. G. <tim...@si...>]

On 04/03/2011 2:00 PM, Oleg Broytman wrote:
> On Fri, Mar 04, 2011 at 12:49:50PM +0000, Timothy W. Grove wrote:
>> It seems to me that there should be a simple way to add the attribute to
>> the class if the column has been added already, or not if it has been
>> added through updating
>     There is no a simple way. You have to create the table class without
> the column, and then look into the database to determine if the column
> is there. You can use the same mechanism SQLObject uses to implement
> fromDatabase: connection.columnsFromSchema(tableName, soClass); the
> result is a list of SOCol instances.
Thanks for the advice, Oleg. I found a solution which may not be 
elegant, but it works for me! Instead of trying to work with two 
different class definitions I've just kept one which declares the column 
I wish to add to the database and run an update function which deletes 
the column from the class before adding it again to both the class and 
the database. My example follows.

Best regards,
Tim

    class SignEntry(SQLObject):
         explanatory_map = PickleCol()

    def SignEntryUpdate():
         cols = [i.name for i in
    sqlhub.processConnection.columnsFromSchema("sign_entry", SignEntry)]
         if 'explanatoryMap' not in cols:
             SignEntry.sqlmeta.delColumn('explanatory_map') # I know
    this exists in the class definition
             SignEntry.sqlmeta.addColumn(PickleCol("explanatory_map"),
    changeSchema=True)

Re: [SQLObject] Altering/updating a database

[Oleg B. <ph...@ph...>]

On Fri, Mar 04, 2011 at 12:49:50PM +0000, Timothy W. Grove wrote:
> It seems to me that there should be a simple way to add the attribute to  
> the class if the column has been added already, or not if it has been  
> added through updating

   There is no a simple way. You have to create the table class without
the column, and then look into the database to determine if the column
is there. You can use the same mechanism SQLObject uses to implement
fromDatabase: connection.columnsFromSchema(tableName, soClass); the
result is a list of SOCol instances.

> I've also experimented with reading the class description from the  
> database using "class sqlmeta: fromDatabase = True" but I don't think  
> this works with the sqlite backend.

   Well, database introspection can't be precise; for example, SQLObject
cannot determine on per-column basis if you want StringCol or
UnicodeCol. SQLObject's implementation is also quite simple, it cannot
determine ForeignKeys, e.g. But it can get the list of columns with
their types. Do you have problems with that?

Oleg.
-- 
     Oleg Broytman            http://phdru.name/            ph...@ph...
           Programmers don't die, they just GOSUB without RETURN.

Re: [SQLObject] Altering/updating a database

[Timothy W. G. <tim...@si...>]

On 13/01/2011 10:12 PM, Oleg Broytman wrote:
> On Thu, Jan 13, 2011 at 09:36:03PM +0000, Timothy W. Grove wrote:
>> I want to update an existing database by adding a new column
>> (UnicodeCol) to an existing table. Can anyone suggest a way to
>> automatically accomplish this through sqlobject ?
>     It could be as simple as
>
> Table.sqlmeta.addColumn(UnicodeCol('name'), changeSchema=True)
>
> Oleg.
This has been working for me, but I've come across a problem in my 
approach to using it; perhaps someone can suggest an alternative?

When I first start my application and connect (sqlhub.processConnection) 
to a database (SQLite) I run the following function to update it:

    def SignEntryUpdate():
         try:
             SignEntry.sqlmeta.addColumn(PickleCol("explanatory_map"),
    changeSchema=True)
         except:
             pass #database already has this column

The first time I connect to a database which requires updating all is 
fine; the column is added and I can access it through the class with the 
SignEntry instance attribute 'explanatory_map'. But after the 
application is closed and I open it again connecting to the updated 
database, trying to access the column through the class just leads to 
the AttributeError: 'SignEntry' object has no attribute 
'explanatory_map'. But of course this would happen as it was through the 
updating process that the attribute was added, it would not be added if 
the database didn't need updating.

If I kept the attribute as part of the class definition as in the 
following code, then the update process fails and the new column is not 
added:

    class SignEntry(SQLObject):
         explanatory_map = PickleCol()

It seems to me that there should be a simple way to add the attribute to 
the class if the column has been added already, or not if it has been 
added through updating; the "how to" do that is escaping me at the 
moment. Any and all suggestions welcomed.

I've also experimented with reading the class description from the 
database using "class sqlmeta: fromDatabase = True" but I don't think 
this works with the sqlite backend. The "sqlobject-admin" tool also 
looks promising as a way to update databases, but I'm not too sure how 
far this has been developed after reading the documentation at 
http://sqlobject.org/sqlobject-admin.html ?

Thanks for any help you can give.

Best regards,
Tim

p.s. by the way, I had been using version 0.13 of sqlobject as I 
couldn't find an installer for 0.15 under python 2.7, but I have since 
downloaded and manually installed the version for 2.6 from 
http://pypi.python.org/pypi/SQLObject/0.15.0 and it seems to be working 
fine for me, under Windows 7.

Re: [SQLObject] SQLCipher with SQLite

[Oleg B. <ph...@ph...>]

On Thu, Feb 24, 2011 at 04:04:20PM +0300, Oleg Broytman wrote:
> >    ImportError: DLL load failed: The specified procedure could not be
> >    found.
> 
>    The DLL is broken

   Well, it's not broken, it just isn't a Python module. _sqlite3.dll is
not and SQLite's DLL - it's a Python module statically linking SQLite's
code.

Oleg.
-- 
     Oleg Broytman            http://phdru.name/            ph...@ph...
           Programmers don't die, they just GOSUB without RETURN.

Re: [SQLObject] SQLCipher with SQLite

[Oleg B. <ph...@ph...>]

Please answer to the list.

On Thu, Feb 24, 2011 at 12:52:05PM +0000, Timothy W. Grove wrote:
> On 24/02/2011 12:24 PM, Oleg Broytman wrote:
>>     First thing to try - test the module without SQLObject. Start with
>> "import sqlite3".
> Well, that is the first problem:
>
>    import sqlite3
>    Traceback (most recent call last):
>       File "<input>", line 1, in <module>
>       File "C:\Python27\lib\sqlite3\__init__.py", line 24, in <module>
>         from dbapi2 import *
>       File "C:\Python27\lib\sqlite3\dbapi2.py", line 27, in <module>
>         from _sqlite3 import *
>    ImportError: DLL load failed: The specified procedure could not be
>    found.

   The DLL is broken, forget about it (until you are going to fix and
recompile it).

Oleg.
-- 
     Oleg Broytman            http://phdru.name/            ph...@ph...
           Programmers don't die, they just GOSUB without RETURN.

Re: [SQLObject] SQLCipher with SQLite

[Oleg B. <ph...@ph...>]

On Wed, Feb 23, 2011 at 05:43:52PM +0000, Timothy W. Grove wrote:
>    File
>    "C:\Python27\lib\site-packages\sqlobject-0.13.0-py2.7.egg\sqlobject\sqlite\sqliteconnection.py",
>    line 48, in __init__
>    raise ImportError('Cannot find an SQLite driver, tried %s' % drivers)
>    ImportError: Cannot find an SQLite driver, tried
>    pysqlite2,sqlite3,sqlite

   First thing to try - test the module without SQLObject. Start with
"import sqlite3".

   Also look at sqlobject/include/hashcol.py. You can use it as a model
to try to implement an encrypting column or an encrypting validator that
could be applied to all columns.

Oleg.
-- 
     Oleg Broytman            http://phdru.name/            ph...@ph...
           Programmers don't die, they just GOSUB without RETURN.

[SQLObject] SQLCipher with SQLite

[Timothy W. G. <tim...@si...>]

Has anyone had success using SQLCipher (http://sqlcipher.net/) with 
SQLite through SQLObject? If you have, then I'd love to hear your 
solution. If you haven't but think it could work, I'd like to here your 
suggestions. If you think that I haven't got a chance, then please tell 
me that too! The real question is how to work with encryption through 
sqlobject; SQLCipher "might" be one way...

I've found the following thread 
http://groups.google.com/group/sqlcipher/browse_thread/thread/b85f42dc9921df46 
and downloaded the attached file at the bottom of the page. After 
unpacking "sqlcipher.7z", on my system I replaced 
"C:\Python27\DLLs\sqlite3.dll" with 
"C:\Downloads\sqlcipher\1.1.1\sqlite-3.6.22\MSVC\StaticLib\Release\sqlite3.dll" 
but haven't gotten any further. I'm not sure if there is any way to pass 
an encryption key through the current sqlobject sqlite connector. 
Without making any other changes to my program or databases I now get 
the message when attempting to connect:

    File
    "C:\Python27\lib\site-packages\sqlobject-0.13.0-py2.7.egg\sqlobject\sqlite\sqliteconnection.py",
    line 48, in __init__
    raise ImportError('Cannot find an SQLite driver, tried %s' % drivers)
    ImportError: Cannot find an SQLite driver, tried
    pysqlite2,sqlite3,sqlite

Thanks for any help,
Tim

Re: [SQLObject] Setting encoding on sqlite (on disk) DB

[Jason H. <jas...@gm...>]

On 18 February 2011 16:12, Oleg Broytman <ph...@ph...> wrote:
>   Does SQLite support anything other than UTF-8? I think it's always
> UTF-8 so you don't need (and actually, cannot) change anything.

I thought it supported UTF-16 too, but uses UTF-8 by default.
Presumably then, if I just leave it alone, it'll use UTF-8 (I just
like to be explicit).

Cheers,
Jason

Re: [SQLObject] Setting encoding on sqlite (on disk) DB

[Oleg B. <ph...@ph...>]

On Fri, Feb 18, 2011 at 11:12:29AM +0300, Oleg Broytman wrote:
> On Fri, Feb 18, 2011 at 03:50:41PM +0800, Jason Heeris wrote:
> > When using MySQL with SQLObject, I can use a URI like:
> > "mysql://.../db?charset=utf8" to indicate the encoding. What's the
> > equivalent for SQLite? What if I create the tables using the
> > .createTable() class method ??? how do I specify it there?
> 
>    Does SQLite support anything other than UTF-8? I think it's always
> UTF-8 so you don't need (and actually, cannot) change anything.

   Aha, it does, to some extent:
http://www.sqlite.org/pragma.html#pragma_encoding

   "It is not possible to change the text encoding of a database after
it has been created and any attempt to do so will be silently ignored."

   So you only need this PRAGMA once - before you create you database.

Oleg.
-- 
     Oleg Broytman            http://phdru.name/            ph...@ph...
           Programmers don't die, they just GOSUB without RETURN.

Re: [SQLObject] Setting encoding on sqlite (on disk) DB

[Oleg B. <ph...@ph...>]

On Fri, Feb 18, 2011 at 03:50:41PM +0800, Jason Heeris wrote:
> When using MySQL with SQLObject, I can use a URI like:
> "mysql://.../db?charset=utf8" to indicate the encoding. What's the
> equivalent for SQLite? What if I create the tables using the
> .createTable() class method ??? how do I specify it there?

   Does SQLite support anything other than UTF-8? I think it's always
UTF-8 so you don't need (and actually, cannot) change anything.

Oleg.
-- 
     Oleg Broytman            http://phdru.name/            ph...@ph...
           Programmers don't die, they just GOSUB without RETURN.

[SQLObject] Setting encoding on sqlite (on disk) DB

[Jason H. <jas...@gm...>]

When using MySQL with SQLObject, I can use a URI like:
"mysql://.../db?charset=utf8" to indicate the encoding. What's the
equivalent for SQLite? What if I create the tables using the
.createTable() class method — how do I specify it there?

Thanks,
Jason

Re: [SQLObject] Use a column other than the primary key for a foreignkey...

[Sam's L. <sam...@gm...>]

Cool. Well if it gets done I'll definitely use it. :)

I'll try and take a look myself at the parts you mentioned when I have some
free time in the next few weeks.

Thanks
Sam

On Wed, Feb 16, 2011 at 2:38 AM, Oleg Broytman <ph...@ph...> wrote:

> On Wed, Feb 16, 2011 at 02:10:41AM -0800, Sam's Lists wrote:
> > Are there any plans to change that?
>
>    There is no just because this is the first time the question is
> raised. I've added it to the TODO list:
>
> http://sourceforge.net/mailarchive/forum.php?thread_name=E1PnpY6-00009p-IX%40webwareforpython.org&forum_name=sqlobject-cvs
>
> > I assume it's a pretty complicated change, right?
>
>    At the first glance it seems one needs to change SOForeignKey (and
> perhaps SOCol), sqlmeta.add/delColumn(); SQLObject._SO_foreignKey() must
> take into account the id could be any column, not only an id; the
> current code could be preserved as a shortcut in case otherColumn is id.
>
> Oleg.
> --
>     Oleg Broytman            http://phdru.name/            ph...@ph...
>           Programmers don't die, they just GOSUB without RETURN.
>
>
> ------------------------------------------------------------------------------
> The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE:
> Pinpoint memory and threading errors before they happen.
> Find and fix more than 250 security defects in the development cycle.
> Locate bottlenecks in serial and parallel code that limit performance.
> http://p.sf.net/sfu/intel-dev2devfeb
> _______________________________________________
> sqlobject-discuss mailing list
> sql...@li...
> https://lists.sourceforge.net/lists/listinfo/sqlobject-discuss
>

Re: [SQLObject] Filtering selections via foreign keys

[Oleg B. <ph...@ph...>]

On Thu, Feb 17, 2011 at 09:49:46AM +0800, Jason Heeris wrote:
> On 17 February 2011 09:23, Jason Heeris <jas...@gm...> wrote:
> > Any other suggestions?
> 
> Oooooh I'm a bit of an idiot. I failed to realise that ".filter"
> RETURNS a filtered result set, it doesn't mutate the existing one!
> 
> Your initial suggestion works fine :)

   Nice!

Oleg.
-- 
     Oleg Broytman            http://phdru.name/            ph...@ph...
           Programmers don't die, they just GOSUB without RETURN.

Re: [SQLObject] Filtering selections via foreign keys

[Jason H. <jas...@gm...>]

On 17 February 2011 09:23, Jason Heeris <jas...@gm...> wrote:
> Any other suggestions?

Oooooh I'm a bit of an idiot. I failed to realise that ".filter"
RETURNS a filtered result set, it doesn't mutate the existing one!

Your initial suggestion works fine :)

Thanks,
Jason

Re: [SQLObject] Filtering selections via foreign keys

[Jason H. <jas...@gm...>]

On 16 February 2011 17:05, Oleg Broytman <ph...@ph...> wrote:
>   If SVNDocument is actually Document in this example, and type to
> filter must be compared equal to DocumentType.dbType

Ah yes, sorry about those errors.

>    docs.filter(
>        (Document.q.dbDocumentTypeID==DocumentType.q.id) &
>        (DocumentType.q.dbType==type)
>    )

I still can't get this to work — even if I set "type" to something
that doesn't match anything in the table, the filter has no effect —
the count of results is the same before and after, when I'd expect it
to be zero for a type not in the table. (I've checked that it goes
through that "if", too.)

I also changed the dbDocumentTypeID to dbDocumentType with no luck
(what's the difference between the two? I notice the same naming
change in the j-magic section of the docs).

One of the other things I tried is to use a j-magic expression:
Document.j.dbDocumentType & (DocumentType.q.dbType==type) ... but
still nothing happens to the select result. I've been trying to
construct an "IN" query, but haven't really gotten anywhere with that
yet either.

Any other suggestions?

Cheers,
Jason

Re: [SQLObject] Use a column other than the primary key for a foreignkey...

[Oleg B. <ph...@ph...>]

On Wed, Feb 16, 2011 at 02:10:41AM -0800, Sam's Lists wrote:
> Are there any plans to change that?

   There is no just because this is the first time the question is
raised. I've added it to the TODO list:
http://sourceforge.net/mailarchive/forum.php?thread_name=E1PnpY6-00009p-IX%40webwareforpython.org&forum_name=sqlobject-cvs

> I assume it's a pretty complicated change, right?

   At the first glance it seems one needs to change SOForeignKey (and
perhaps SOCol), sqlmeta.add/delColumn(); SQLObject._SO_foreignKey() must
take into account the id could be any column, not only an id; the
current code could be preserved as a shortcut in case otherColumn is id.

Oleg.
-- 
     Oleg Broytman            http://phdru.name/            ph...@ph...
           Programmers don't die, they just GOSUB without RETURN.

Re: [SQLObject] Use a column other than the primary key for a foreignkey...

[Sam's L. <sam...@gm...>]

I see.

Are there any plans to change that?

I assume it's a pretty complicated change, right?

-Sam

On Fri, Feb 11, 2011 at 1:05 AM, Oleg Broytman <ph...@ph...> wrote:

> On Thu, Feb 10, 2011 at 10:24:43PM -0800, Sam's Lists wrote:
> > I'd like to have a foreign key use a unique column in another table which
> is
> > not the primary key.  This works fine in Postgres, but I can't figure out
> > how to do it in SQLObject.  Any idea?
>
>    SQLObject can only references an 'id' column. That's, probably, a
> deficiency.
>
> Oleg.
> --
>     Oleg Broytman            http://phdru.name/            ph...@ph...
>           Programmers don't die, they just GOSUB without RETURN.
>
>
> ------------------------------------------------------------------------------
> The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE:
> Pinpoint memory and threading errors before they happen.
> Find and fix more than 250 security defects in the development cycle.
> Locate bottlenecks in serial and parallel code that limit performance.
> http://p.sf.net/sfu/intel-dev2devfeb
> _______________________________________________
> sqlobject-discuss mailing list
> sql...@li...
> https://lists.sourceforge.net/lists/listinfo/sqlobject-discuss
>