I'm sorry Willoughby but I don't understand your question.

With this csv export/import the problem is that id's change.

For example if you take a look to the auth_user table in my previous post, 
table auth_membership will have wrong auth_user id's:
auth_membership.id 
<https://dev.tasko.it/mompala/appadmin/select/db?orderby=auth_membership.id>  
           auth_membership.user_id 
<https://dev.tasko.it/mompala/appadmin/select/db?orderby=auth_membership.group_id>1
 
                                                  3 (John)
2                                                   7 (Sally)

And in each application that I have most of them are in relation, so that's 
a problem.
I've take a look to pgloader but it seems complicated to make a migration.
If someone have a detailed example it could be useful.

Il giorno martedì 31 marzo 2015 20:52:18 UTC+2, Willoughby ha scritto:
>
> Well it has a bunch of log statements, were all of those OK?
>
> On Tuesday, March 31, 2015 at 2:19:41 PM UTC-4, Gael Princivalle wrote:
>>
>> Thanks Ron but:with CSV export:
>> auth_user_id            user
>> 3                             John
>> 7                             Sally
>>
>> And then you import your CSV file you will have:
>> auth_user_id            user
>> 1                             John
>> 2                             Sally
>>
>> And so all relations between tables will be broken.
>> Someone have a solution?
>>
>> Il giorno martedì 31 marzo 2015 15:10:53 UTC+2, Ron Chatterjee ha scritto:
>>>
>>> As an alrenative. you can upload a csv file into web2py. Not a solution 
>>> if you want to do things dynamically and all the time. But you can use the 
>>> csv option.
>>>
>>> On Tuesday, March 31, 2015 at 4:11:18 AM UTC-4, Gael Princivalle wrote:
>>>>
>>>> Hello.
>>>>
>>>> Like a lot of users I need to clone an existing SQLlite db to new 
>>>> PostgreSQL db.
>>>> I've read this post in web2py-developers by Alan Etkin:
>>>>
>>>> https://groups.google.com/forum/#!searchin/web2py-developers/sqlite$20postgres/web2py-developers/QxeJNByj6qc/cpBHsa1ymUkJ
>>>>
>>>> However I've tried to use this script and my PosgreSQL db still empty.
>>>>
>>>> My db:
>>>> dbsq = DAL('sqlite://storage.sqlite',pool_size=1,check_reserved=['all'
>>>> ])
>>>> dbpg = DAL('postgres://user:pass@localhost:5432/postg_myapp',pool_size=
>>>> 1,check_reserved=['all'])
>>>>
>>>> My controller:
>>>> def migrate_to_pg():
>>>>     migrate(dbsq, dbpg)
>>>>     return dict()
>>>>
>>>> def disable_triggers(pgdb):
>>>>     for tablename in pgdb.tables():
>>>>         pgdb.executesql("ALTER TABLE %s DISABLE TRIGGER ALL;" % \
>>>>                         tablename)
>>>>
>>>> def enable_triggers(pgdb):
>>>>     for tablename in pgdb.tables():
>>>>         pgdb.executesql("ALTER TABLE %s ENABLE TRIGGER ALL;" % \
>>>>                         tablename)
>>>>
>>>> def migrate(sqlitedb, pgdb):
>>>>     """Transfer data to PostgreSQL
>>>>     
>>>>     Arguments:
>>>>         sqlitedb: the Sqlite db  source connection
>>>>         pgdb: the PostgreSQL db target connection
>>>>     """
>>>>     import logging
>>>>     setval = "select setval('%s'::regclass"
>>>>     setval += ", (SELECT MAX(%s) FROM %s));"
>>>>     logging.debug("Disabling pg triggers temporarily")
>>>>     disable_triggers(pgdb)
>>>>     for table in pgdb.tables:
>>>>         logging.debug("copying %s" % table)
>>>>         query = pgdb[table]._id > 0
>>>>         rows = pgdb(query).select()
>>>>         colnames = [k[k.index(".")+1:] for k in rows.colnames]
>>>>         logging.debug(pgdb._lastsql)
>>>>         logging.debug(colnames)
>>>>         rows = sqlitedb.executesql(pgdb._lastsql)
>>>>         # clean up table
>>>>         pgdb.executesql("TRUNCATE %s" % table)
>>>>         for i, row in enumerate(rows):
>>>>           logging.debug("inserting", i, len(rows))
>>>>           sql = "INSERT INTO %s (%s) VALUES (%s)" % (
>>>>                   table,
>>>>                   ','.join(colnames),
>>>>                   ','.join(["%s" for k in colnames]),
>>>>                   )
>>>>           logging.debug(sql)
>>>>           pgdb.executesql(sql, row)
>>>>         # update serials
>>>>         sequence = pgdb[table]._sequence_name
>>>>         if sequence in ["", None]:
>>>>             sequence = "%s_%s_seq" % (table, colnames[0])
>>>>         pgdb.executesql(setval % (sequence,
>>>>                                   colnames[0],
>>>>                                   table))
>>>>     logging.debug("Re-enabling pg triggers")
>>>>     enable_triggers(pgdb)
>>>>     pgdb.commit()
>>>>     logging.debug("Migration done")
>>>>
>>>> When I run migrate_to_pg I don't have any kind of error but the 
>>>> PosgreSQL db still empty.
>>>>
>>>> Someone can give me a help, or another solution for this migration?
>>>> I think it's necessary for web2py having a tool for this migration, 
>>>> giving an easy way for fresh web2py users like me that understand that 
>>>> they 
>>>> must migrate some existing applications to PostgreSQL to do it without 
>>>> headaches.
>>>>
>>>> Thanks, regards.
>>>>
>>>>

-- 
Resources:
- http://web2py.com
- http://web2py.com/book (Documentation)
- http://github.com/web2py/web2py (Source code)
- https://code.google.com/p/web2py/issues/list (Report Issues)
--- 
You received this message because you are subscribed to the Google Groups 
"web2py-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to web2py+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to