sqlalchemy-tickets — trac tickets and updates forwarded to this list

Re: [Sqlalchemy-tickets] [sqlalchemy] #2863: ParseError on installation, Python 3.3

[sqlalchemy <mi...@zz...>]

#2863: ParseError on installation, Python 3.3
----------------------------------+------------------------------------
      Reporter:  Julian_O         |      Owner:  zzzeek
          Type:  defect           |     Status:  new
      Priority:  medium           |  Milestone:
     Component:  (none)           |   Severity:  no triage selected yet
    Resolution:                   |   Keywords:
Progress State:  awaiting triage  |
----------------------------------+------------------------------------

Comment (by Julian_O):

 The first two lines in question are of the format:

 {{{
     >>> print Session().query(User, User.balance).\\
     ...     join(User.accounts).filter(User.balance > 5000)
 }}}

 i.e. they are doctest strings, using the Python 2.x format.

 I wonder if 2to3 is being confused by the existence of an old format print
 statement within a comment.

 I suggest replacing it with the new style print function (i.e.
 parenthesis) to satisfy 2to3.

 {{{
     >>> print(Session().query(User, User.balance).\\
     ...     join(User.accounts).filter(User.balance > 5000))
 }}}

 If that doesn't work, replace print with str.

 {{{
     >>> str(Session().query(User, User.balance).\\
     ...     join(User.accounts).filter(User.balance > 5000))
 }}}

--
Ticket URL: <http://www.sqlalchemy.org/trac/ticket/2863#comment:1>
sqlalchemy <http://www.sqlalchemy.org/>
The Database Toolkit for Python

[Sqlalchemy-tickets] [sqlalchemy] #2863: ParseError on installation, Python 3.3

[sqlalchemy <mi...@zz...>]

#2863: ParseError on installation, Python 3.3
----------------------+-----------------------------------------
 Reporter:  Julian_O  |           Owner:  zzzeek
     Type:  defect    |          Status:  new
 Priority:  medium    |       Milestone:
Component:  (none)    |        Severity:  no triage selected yet
 Keywords:            |  Progress State:  awaiting triage
----------------------+-----------------------------------------
 When installing in Python 3.3, the setup script gives a !ParseError. It is
 not clear if this is actually interfering with any functionality, but it
 is a concerning flag, and is one of the first experiences users have with
 the software.

 The errors in question are:

 {{{
     Can't parse docstring in build\lib\sqlalchemy\ext\hybrid.py line 247:
 ParseError: bad token: type=55, value='\\', context=('', (247, 42))
     Can't parse docstring in build\lib\sqlalchemy\ext\hybrid.py line 448:
 ParseError: bad token: type=55, value='\\', context=('', (450, 41))
     Can't parse docstring in build\lib\sqlalchemy\sql\expression.py line
 5607: ParseError: bad token: type=55, value='\\', context=('', (5607, 38))
 }}}

 I am not very familiar with the tool, but I think these might be output
 from 2to3.

 It works fine on Python 2.7.5, 32-bit, Windows.
 It gives the errors on Python 3.3, 64-bit, Windows.

 Here is the full output:
 {{{
 >pip install sqlalchemy

 Downloading/unpacking sqlalchemy
   Downloading SQLAlchemy-0.8.3.tar.gz (3.9MB): 3.9MB downloaded
   Running setup.py egg_info for package sqlalchemy

     warning: no files found matching '*.jpg' under directory 'doc'
     no previously-included directories found matching 'doc\build\output'
 ***************************************************************************
     WARNING: C extensions are not supported on this Python platform,
 speedups are not enabled.
     Plain-Python build succeeded.
 ***************************************************************************
 Installing collected packages: sqlalchemy
   Running setup.py install for sqlalchemy
     Fixing build\lib\sqlalchemy\event.py build\lib\sqlalchemy\events.py
 build\lib\sqlalchemy\exc.py build\lib\sqlalchemy\inspection.py
 build\lib\sqlalchemy\interfaces.py build\lib\sqlal
 \sqlalchemy\dialects\drizzle\__init__.py
 build\lib\sqlalchemy\dialects\firebird\base.py
 build\lib\sqlalchemy\dialects\firebird\fdb.py
 build\lib\sqlalchemy\dialects\firebird\kinterbasdb.
 ts\mysql\base.py build\lib\sqlalchemy\dialects\mysql\cymysql.py
 build\lib\sqlalchemy\dialects\mysql\gaerdbms.py
 build\lib\sqlalchemy\dialects\mysql\mysqlconnector.py build\lib\sqlalchem
 ib\sqlalchemy\dialects\postgresql\pg8000.py
 build\lib\sqlalchemy\dialects\postgresql\psycopg2.py
 build\lib\sqlalchemy\dialects\postgresql\pypostgresql.py
 build\lib\sqlalchemy\dialects\p
 uild\lib\sqlalchemy\engine\interfaces.py
 build\lib\sqlalchemy\engine\reflection.py
 build\lib\sqlalchemy\engine\result.py
 build\lib\sqlalchemy\engine\strategies.py build\lib\sqlalchemy\e
 e\clsregistry.py build\lib\sqlalchemy\ext\declarative\__init__.py
 build\lib\sqlalchemy\orm\attributes.py
 build\lib\sqlalchemy\orm\collections.py
 build\lib\sqlalchemy\orm\dependency.py b
 y build\lib\sqlalchemy\orm\session.py build\lib\sqlalchemy\orm\state.py
 build\lib\sqlalchemy\orm\strategies.py build\lib\sqlalchemy\orm\sync.py
 build\lib\sqlalchemy\orm\unitofwork.py bu
  build\lib\sqlalchemy\testing\mock.py
 build\lib\sqlalchemy\testing\pickleable.py
 build\lib\sqlalchemy\testing\profiling.py
 build\lib\sqlalchemy\testing\requirements.py build\lib\sqlalch
 \test_update_delete.py build\lib\sqlalchemy\testing\suite\__init__.py
 build\lib\sqlalchemy\util\compat.py
 build\lib\sqlalchemy\util\deprecations.py
 build\lib\sqlalchemy\util\langhelpers
     Skipping implicit fixer: buffer
     Skipping implicit fixer: idioms
     Skipping implicit fixer: set_literal
     Skipping implicit fixer: ws_comma
     Fixing build\lib\sqlalchemy\event.py build\lib\sqlalchemy\events.py
 build\lib\sqlalchemy\exc.py build\lib\sqlalchemy\inspection.py
 build\lib\sqlalchemy\interfaces.py build\lib\sqlal
 \sqlalchemy\dialects\drizzle\__init__.py
 build\lib\sqlalchemy\dialects\firebird\base.py
 build\lib\sqlalchemy\dialects\firebird\fdb.py
 build\lib\sqlalchemy\dialects\firebird\kinterbasdb.
 ts\mysql\base.py build\lib\sqlalchemy\dialects\mysql\cymysql.py
 build\lib\sqlalchemy\dialects\mysql\gaerdbms.py
 build\lib\sqlalchemy\dialects\mysql\mysqlconnector.py build\lib\sqlalchem
 ib\sqlalchemy\dialects\postgresql\pg8000.py
 build\lib\sqlalchemy\dialects\postgresql\psycopg2.py
 build\lib\sqlalchemy\dialects\postgresql\pypostgresql.py
 build\lib\sqlalchemy\dialects\p
 uild\lib\sqlalchemy\engine\interfaces.py
 build\lib\sqlalchemy\engine\reflection.py
 build\lib\sqlalchemy\engine\result.py
 build\lib\sqlalchemy\engine\strategies.py build\lib\sqlalchemy\e
 e\clsregistry.py build\lib\sqlalchemy\ext\declarative\__init__.py
 build\lib\sqlalchemy\orm\attributes.py
 build\lib\sqlalchemy\orm\collections.py
 build\lib\sqlalchemy\orm\dependency.py b
 y build\lib\sqlalchemy\orm\session.py build\lib\sqlalchemy\orm\state.py
 build\lib\sqlalchemy\orm\strategies.py build\lib\sqlalchemy\orm\sync.py
 build\lib\sqlalchemy\orm\unitofwork.py bu
  build\lib\sqlalchemy\testing\mock.py
 build\lib\sqlalchemy\testing\pickleable.py
 build\lib\sqlalchemy\testing\profiling.py
 build\lib\sqlalchemy\testing\requirements.py build\lib\sqlalch
 \test_update_delete.py build\lib\sqlalchemy\testing\suite\__init__.py
 build\lib\sqlalchemy\util\compat.py
 build\lib\sqlalchemy\util\deprecations.py
 build\lib\sqlalchemy\util\langhelpers
     Skipping implicit fixer: buffer
     Skipping implicit fixer: idioms
     Skipping implicit fixer: set_literal
     Skipping implicit fixer: ws_comma
     Can't parse docstring in build\lib\sqlalchemy\ext\hybrid.py line 247:
 ParseError: bad token: type=55, value='\\', context=('', (247, 42))
     Can't parse docstring in build\lib\sqlalchemy\ext\hybrid.py line 448:
 ParseError: bad token: type=55, value='\\', context=('', (450, 41))
     Can't parse docstring in build\lib\sqlalchemy\sql\expression.py line
 5607: ParseError: bad token: type=55, value='\\', context=('', (5607, 38))

     warning: no files found matching '*.jpg' under directory 'doc'
     no previously-included directories found matching 'doc\build\output'
 ***************************************************************************
     WARNING: C extensions are not supported on this Python platform,
 speedups are not enabled.
     Plain-Python build succeeded.
 ***************************************************************************
 Successfully installed sqlalchemy
 Cleaning up...


 }}}

--
Ticket URL: <http://www.sqlalchemy.org/trac/ticket/2863>
sqlalchemy <http://www.sqlalchemy.org/>
The Database Toolkit for Python

[Sqlalchemy-tickets] [sqlalchemy] #2862: dogpile helloworld is all wrong

[sqlalchemy <mi...@zz...>]

#2862: dogpile helloworld is all wrong
---------------------------+---------------------------------------
 Reporter:  zzzeek         |           Owner:  zzzeek
     Type:  defect         |          Status:  new
 Priority:  highest        |       Milestone:  0.8.xx
Component:  documentation  |        Severity:  minor - half an hour
 Keywords:                 |  Progress State:  in queue
---------------------------+---------------------------------------
 we removed the "namespace" concept but helloworld.py is all about
 namespacing, with comments that make no sense re: the code.   kind of
 strange.

--
Ticket URL: <http://www.sqlalchemy.org/trac/ticket/2862>
sqlalchemy <http://www.sqlalchemy.org/>
The Database Toolkit for Python

Re: [Sqlalchemy-tickets] [sqlalchemy] #2861: document use of version_id_col with versioning example (was: Hazard during versioning records)

[sqlalchemy <mi...@zz...>]

#2861: document use of version_id_col with versioning example
--------------------------------+-----------------------------------
      Reporter:  bikbit         |      Owner:  zzzeek
          Type:  defect         |     Status:  new
      Priority:  medium         |  Milestone:  0.9.0
     Component:  documentation  |   Severity:  trivial - <10 minutes
    Resolution:                 |   Keywords:  versioning
Progress State:  in queue       |
--------------------------------+-----------------------------------
Changes (by zzzeek):

 * milestone:  0.8.xx => 0.9.0
 * component:  utils => documentation
 * severity:  very major - up to 2 days => trivial - <10 minutes
 * status_field:  awaiting triage => in queue


Comment:

 sure, this is a natural effect of the two transactions not being entirely
 isolated.  If you have your isolation set at at least REPEATABLE READ then
 you wouldn't get this issue.

 you can alleviate this condition using a different kind of "versioning"
 which is the mapper-level versioning feature, see
 http://docs.sqlalchemy.org/en/rel_0_9/orm/mapper_config.html#configuring-a
 -version-counter.

 I've committed a proof of concept for this in rac54ba0f2d8df5a76b685284.
 A change could be to make this a flag on the `Versioned` class itself and
 to add documentation.

--
Ticket URL: <http://www.sqlalchemy.org/trac/ticket/2861#comment:1>
sqlalchemy <http://www.sqlalchemy.org/>
The Database Toolkit for Python

[Sqlalchemy-tickets] [sqlalchemy] #2861: Hazard during versioning records

[sqlalchemy <mi...@zz...>]

#2861: Hazard during versioning records
------------------------+--------------------------------------------
 Reporter:  bikbit      |           Owner:  zzzeek
     Type:  defect      |          Status:  new
 Priority:  medium      |       Milestone:  0.8.xx
Component:  utils       |        Severity:  very major - up to 2 days
 Keywords:  versioning  |  Progress State:  awaiting triage
------------------------+--------------------------------------------
 Problem description:
 Two requests from client application works on this same record in short
 period of time and sometimes appears integrity error from database on
 compound key of historical record: id,version.

 My guess/investigations:
 When two requests perform operation on this same database record and for
 example first finished operation, then second have not actual information
 about newest version of historical record and try to save new one with
 this same version as first.

 Versioning used from example of SqlAlchemy and trigered in event
 before_flush.

--
Ticket URL: <http://www.sqlalchemy.org/trac/ticket/2861>
sqlalchemy <http://www.sqlalchemy.org/>
The Database Toolkit for Python

Re: [Sqlalchemy-tickets] [sqlalchemy] #2727: ProgrammingError when autoload from PostgreSQL (Amazon Redshift)

[sqlalchemy <mi...@zz...>]

#2727: ProgrammingError when autoload from PostgreSQL (Amazon Redshift)
---------------------------+----------------------------------
      Reporter:  ashkop    |      Owner:  zzzeek
          Type:  defect    |     Status:  new
      Priority:  medium    |  Milestone:  0.x.xx
     Component:  postgres  |   Severity:  minor - half an hour
    Resolution:            |   Keywords:
Progress State:  on hold   |
---------------------------+----------------------------------

Comment (by zzzeek):

 note potential pullreq at https://bitbucket.org/zzzeek/sqlalchemy/pull-
 request/6/sqlalchemy-to-support-postgresql-80/diff

--
Ticket URL: <http://www.sqlalchemy.org/trac/ticket/2727#comment:13>
sqlalchemy <http://www.sqlalchemy.org/>
The Database Toolkit for Python

Re: [Sqlalchemy-tickets] [sqlalchemy] #2859: Mapper.columns is keyed based on `key` attribute not column name

[sqlalchemy <mi...@zz...>]

#2859: Mapper.columns is keyed based on `key` attribute not column name
-----------------------------------+----------------------------------
      Reporter:  tisdall           |      Owner:  zzzeek
          Type:  defect            |     Status:  closed
      Priority:  medium            |  Milestone:
     Component:  documentation     |   Severity:  minor - half an hour
    Resolution:  worksforme        |   Keywords:
Progress State:  completed/closed  |
-----------------------------------+----------------------------------
Changes (by zzzeek):

 * status:  new => closed
 * resolution:   => worksforme
 * status_field:  awaiting triage => completed/closed


Comment:

 the result you're getting is correct, the mapping links those columns to
 the attribute name specified in declarative first and foremost over that
 of .name or .key of the Column itself.

--
Ticket URL: <http://www.sqlalchemy.org/trac/ticket/2859#comment:2>
sqlalchemy <http://www.sqlalchemy.org/>
The Database Toolkit for Python

Re: [Sqlalchemy-tickets] [sqlalchemy] #2860: spam (was: the incontrovertible fact that it is precisely)

[sqlalchemy <mi...@zz...>]

#2860: spam
----------------------------------+------------------------------------
      Reporter:  spam             |      Owner:  zzzeek
          Type:  defect           |     Status:  closed
      Priority:  medium           |  Milestone:
     Component:  cextensions      |   Severity:  no triage selected yet
    Resolution:  invalid          |   Keywords:
Progress State:  awaiting triage  |
----------------------------------+------------------------------------
Changes (by zzzeek):

 * reporter:  asnmaglot => spam


Old description:

> So everyone is able to uncover precisely the floor lamps, which used is
> normally needed [http://twojalkomat.pl/ alkomaty twojalkomat.pl]?

New description:

 spam

--

--
Ticket URL: <http://www.sqlalchemy.org/trac/ticket/2860#comment:2>
sqlalchemy <http://www.sqlalchemy.org/>
The Database Toolkit for Python

Re: [Sqlalchemy-tickets] [sqlalchemy] #2860: the incontrovertible fact that it is precisely

[sqlalchemy <mi...@zz...>]

#2860: the incontrovertible fact that it is precisely
----------------------------------+------------------------------------
      Reporter:  asnmaglot        |      Owner:  zzzeek
          Type:  defect           |     Status:  closed
      Priority:  medium           |  Milestone:
     Component:  cextensions      |   Severity:  no triage selected yet
    Resolution:  invalid          |   Keywords:
Progress State:  awaiting triage  |
----------------------------------+------------------------------------
Changes (by zzzeek):

 * status:  new => closed
 * resolution:   => invalid


--
Ticket URL: <http://www.sqlalchemy.org/trac/ticket/2860#comment:1>
sqlalchemy <http://www.sqlalchemy.org/>
The Database Toolkit for Python

[Sqlalchemy-tickets] [sqlalchemy] #2860: the incontrovertible fact that it is precisely

[sqlalchemy <mi...@zz...>]

#2860: the incontrovertible fact that it is precisely
-------------------------+-----------------------------------------
 Reporter:  asnmaglot    |           Owner:  zzzeek
     Type:  defect       |          Status:  new
 Priority:  medium       |       Milestone:
Component:  cextensions  |        Severity:  no triage selected yet
 Keywords:               |  Progress State:  awaiting triage
-------------------------+-----------------------------------------
 So everyone is able to uncover precisely the floor lamps, which used is
 normally needed [http://twojalkomat.pl/ alkomaty twojalkomat.pl]?

--
Ticket URL: <http://www.sqlalchemy.org/trac/ticket/2860>
sqlalchemy <http://www.sqlalchemy.org/>
The Database Toolkit for Python

Re: [Sqlalchemy-tickets] [sqlalchemy] #2859: Mapper.columns is keyed based on `key` attribute not column name

[sqlalchemy <mi...@zz...>]

#2859: Mapper.columns is keyed based on `key` attribute not column name
----------------------------------+----------------------------------
      Reporter:  tisdall          |      Owner:  zzzeek
          Type:  defect           |     Status:  new
      Priority:  medium           |  Milestone:
     Component:  documentation    |   Severity:  minor - half an hour
    Resolution:                   |   Keywords:
Progress State:  awaiting triage  |
----------------------------------+----------------------------------

Comment (by tisdall):

 sorry, I think this is a non-issue but I have no idea how to close the
 ticket...

--
Ticket URL: <http://www.sqlalchemy.org/trac/ticket/2859#comment:1>
sqlalchemy <http://www.sqlalchemy.org/>
The Database Toolkit for Python

[Sqlalchemy-tickets] [sqlalchemy] #2859: Mapper.columns is keyed based on `key` attribute not column name

[sqlalchemy <mi...@zz...>]

#2859: Mapper.columns is keyed based on `key` attribute not column name
---------------------------+---------------------------------------
 Reporter:  tisdall        |           Owner:  zzzeek
     Type:  defect         |          Status:  new
 Priority:  medium         |       Milestone:
Component:  documentation  |        Severity:  minor - half an hour
 Keywords:                 |  Progress State:  awaiting triage
---------------------------+---------------------------------------
 I'm not sure if this is an issue with the documentation or the
 implementation.  Basically they don't match.

 [http://docs.sqlalchemy.org/en/latest/orm/mapper_config.html#sqlalchemy.orm.mapper.Mapper.columns
 The Mapper docs] say "are keyed based on the attribute name defined in the
 mapping, not necessarily the key attribute of the Column itself" but I
 have a case where this doesn't appear to be true.

 Using [http://docs.sqlalchemy.org/en/latest/orm/mapper_config.html#using-
 descriptors-and-hybrids the 2nd EmailAddress class in the hybrid docs]
 (the first would likely give the same result) I end up getting the "email"
 column keyed as "_email".


 {{{
 inspector = inspect(EmailAddress)
 print(dict(inspector.columns))
 }}}

 results in:

 {{{
 {'_email': Column('email', String(), table=<email_address>), 'id':
 Column('id', Integer(), table=<email_address>, primary_key=True,
 nullable=False)}
 }}}

--
Ticket URL: <http://www.sqlalchemy.org/trac/ticket/2859>
sqlalchemy <http://www.sqlalchemy.org/>
The Database Toolkit for Python

Re: [Sqlalchemy-tickets] [sqlalchemy] #2858: regression w/ sqlite join rewriting

[sqlalchemy <mi...@zz...>]

#2858: regression w/ sqlite join rewriting
-----------------------------------+-------------------------------
      Reporter:  zzzeek            |      Owner:  zzzeek
          Type:  defect            |     Status:  closed
      Priority:  highest           |  Milestone:  0.9.0
     Component:  sqlite            |   Severity:  major - 1-3 hours
    Resolution:  fixed             |   Keywords:
Progress State:  completed/closed  |
-----------------------------------+-------------------------------
Changes (by zzzeek):

 * status:  new => closed
 * resolution:   => fixed
 * status_field:  in progress => completed/closed


Comment:

 the fix was easier than it first seemed it would be:

 r5d0e84434f116460d64d3078f650d9

--
Ticket URL: <http://www.sqlalchemy.org/trac/ticket/2858#comment:1>
sqlalchemy <http://www.sqlalchemy.org/>
The Database Toolkit for Python

Re: [Sqlalchemy-tickets] [sqlalchemy] #2857: PostgreSQL LATERAL JOIN support

[sqlalchemy <mi...@zz...>]

#2857: PostgreSQL LATERAL JOIN support
------------------------------------+----------------------------------
      Reporter:  jcigar             |      Owner:  zzzeek
          Type:  enhancement        |     Status:  new
      Priority:  low                |  Milestone:  0.x.xx
     Component:  postgres           |   Severity:  minor - half an hour
    Resolution:                     |   Keywords:  lateral
Progress State:  needs a volunteer  |
------------------------------------+----------------------------------
Changes (by zzzeek):

 * priority:  medium => low
 * status_field:  awaiting triage => needs a volunteer
 * severity:  no triage selected yet => minor - half an hour
 * milestone:  0.9.xx => 0.x.xx


Comment:

 this is pretty easy to do yourself, feel free to turn into a pullreq (with
 unit tests of course):

 {{{
 #!python

 class Lateral(Alias):
    # ...

 @compiles(Lateral, "postgresql")
 def compile(element, compiler, **kw):
     return "LATERAL %s" % compiler.visit_alias(element, **kw)

 }}}

--
Ticket URL: <http://www.sqlalchemy.org/trac/ticket/2857#comment:1>
sqlalchemy <http://www.sqlalchemy.org/>
The Database Toolkit for Python

[Sqlalchemy-tickets] [sqlalchemy] #2858: regression w/ sqlite join rewriting

[sqlalchemy <mi...@zz...>]

#2858: regression w/ sqlite join rewriting
---------------------+------------------------------------
 Reporter:  zzzeek   |           Owner:  zzzeek
     Type:  defect   |          Status:  new
 Priority:  highest  |       Milestone:  0.9.0
Component:  sqlite   |        Severity:  major - 1-3 hours
 Keywords:           |  Progress State:  in progress
---------------------+------------------------------------
 {{{
 #!python
 from sqlalchemy import *
 from sqlalchemy.orm import *
 from sqlalchemy.ext.declarative import *


 engine = create_engine('sqlite://', echo=True)
 #engine = create_engine('postgresql://scott:tiger@localhost/test',
 echo=True)

 Base = declarative_base(bind=engine)
 metadata = Base.metadata


 account = Table('account', metadata,
     Column('account_id', Integer, key='id', primary_key=True),
     Column('name', String(255), nullable=False),
 )

 account_address_map = Table('account_address_map', metadata,
     Column('account_id', Integer, ForeignKey('account.id'),
 nullable=False),
     Column('address_id', Integer, ForeignKey('address.id'),
 nullable=False),
     Column('role_id', Integer, nullable=False)
 )

 address = Table('address', metadata,
     Column('address_id', Integer, key='id', primary_key=True),
     Column('address1', String(255), nullable=False),
 )


 mapping = account_address_map.select().where(account_address_map.c.role_id
 == 1)


 class Account(Base):
     __table__ = account

     service_address = relationship('Address',
         secondary=mapping,
         uselist=False,
         viewonly=True)


 class Address(Base):
     __table__ = address


 metadata.drop_all()
 metadata.create_all()
 session = Session()
 session.query(Account).options(joinedload_all('service_address')).all()
 }}}

--
Ticket URL: <http://www.sqlalchemy.org/trac/ticket/2858>
sqlalchemy <http://www.sqlalchemy.org/>
The Database Toolkit for Python

[Sqlalchemy-tickets] [sqlalchemy] #2857: PostgreSQL LATERAL JOIN support

[sqlalchemy <mi...@zz...>]

#2857: PostgreSQL LATERAL JOIN support
-------------------------+-----------------------------------------
 Reporter:  jcigar       |           Owner:  zzzeek
     Type:  enhancement  |          Status:  new
 Priority:  medium       |       Milestone:  0.9.xx
Component:  postgres     |        Severity:  no triage selected yet
 Keywords:  lateral      |  Progress State:  awaiting triage
-------------------------+-----------------------------------------
 all is in the title ... :)
 [http://wiki.postgresql.org/wiki/What%27s_new_in_PostgreSQL_9.3#LATERAL_JOIN]

--
Ticket URL: <http://www.sqlalchemy.org/trac/ticket/2857>
sqlalchemy <http://www.sqlalchemy.org/>
The Database Toolkit for Python

Re: [Sqlalchemy-tickets] [sqlalchemy] #2856: ORM mapping of reflected col names that are not ASCII

[sqlalchemy <mi...@zz...>]

#2856: ORM mapping of reflected col names that are not ASCII
-------------------------+-------------------------------------------------
      Reporter:  teddy   |      Owner:  zzzeek
          Type:  defect  |     Status:  closed
      Priority:  medium  |  Milestone:
     Component:          |   Severity:  no triage selected yet
  declarative            |   Keywords:  oracle, cyrillic, declarative,
    Resolution:          |  reflexion
  worksforme             |
Progress State:          |
  completed/closed       |
-------------------------+-------------------------------------------------

Comment (by teddy):

 Thank you!

--
Ticket URL: <http://www.sqlalchemy.org/trac/ticket/2856#comment:8>
sqlalchemy <http://www.sqlalchemy.org/>
The Database Toolkit for Python

Re: [Sqlalchemy-tickets] [sqlalchemy] #2856: ORM mapping of reflected col names that are not ASCII

[sqlalchemy <mi...@zz...>]

#2856: ORM mapping of reflected col names that are not ASCII
-------------------------+-------------------------------------------------
      Reporter:  teddy   |      Owner:  zzzeek
          Type:  defect  |     Status:  closed
      Priority:  medium  |  Milestone:
     Component:          |   Severity:  no triage selected yet
  declarative            |   Keywords:  oracle, cyrillic, declarative,
    Resolution:          |  reflexion
  worksforme             |
Progress State:          |
  completed/closed       |
-------------------------+-------------------------------------------------

Comment (by zzzeek):

 here's a doc section on that:
 http://docs.sqlalchemy.org/en/rel_0_9/orm/mapper_config.html#automating-
 column-naming-schemes-from-reflected-tables

--
Ticket URL: <http://www.sqlalchemy.org/trac/ticket/2856#comment:7>
sqlalchemy <http://www.sqlalchemy.org/>
The Database Toolkit for Python

Re: [Sqlalchemy-tickets] [sqlalchemy] #2856: ORM mapping of reflected col names that are not ASCII

[sqlalchemy <mi...@zz...>]

#2856: ORM mapping of reflected col names that are not ASCII
-------------------------+-------------------------------------------------
      Reporter:  teddy   |      Owner:  zzzeek
          Type:  defect  |     Status:  closed
      Priority:  medium  |  Milestone:
     Component:          |   Severity:  no triage selected yet
  declarative            |   Keywords:  oracle, cyrillic, declarative,
    Resolution:          |  reflexion
  worksforme             |
Progress State:          |
  completed/closed       |
-------------------------+-------------------------------------------------

Comment (by teddy):

 Thank's for your explanation, really great. When i'm started with my
 experiments, i decided that there is a tricky problem with naming =)
 I'm not a very experienced or skilled developer, especially in
 architecture and smart separation of components (decomposition), so this a
 very good demonstration for me =)

 Thanks!

--
Ticket URL: <http://www.sqlalchemy.org/trac/ticket/2856#comment:6>
sqlalchemy <http://www.sqlalchemy.org/>
The Database Toolkit for Python

Re: [Sqlalchemy-tickets] [sqlalchemy] #2856: ORM mapping of reflected col names that are not ASCII

[sqlalchemy <mi...@zz...>]

#2856: ORM mapping of reflected col names that are not ASCII
-------------------------+-------------------------------------------------
      Reporter:  teddy   |      Owner:  zzzeek
          Type:  defect  |     Status:  closed
      Priority:  medium  |  Milestone:
     Component:          |   Severity:  no triage selected yet
  declarative            |   Keywords:  oracle, cyrillic, declarative,
    Resolution:          |  reflexion
  worksforme             |
Progress State:          |
  completed/closed       |
-------------------------+-------------------------------------------------

Comment (by zzzeek):

 I'm changing the mapper documentation now to clarify this case.

--
Ticket URL: <http://www.sqlalchemy.org/trac/ticket/2856#comment:5>
sqlalchemy <http://www.sqlalchemy.org/>
The Database Toolkit for Python

Re: [Sqlalchemy-tickets] [sqlalchemy] #2856: ORM mapping of reflected col names that are not ASCII (was: Oracle error with reflect on table with cyrillic column names in)

[sqlalchemy <mi...@zz...>]

#2856: ORM mapping of reflected col names that are not ASCII
-------------------------+-------------------------------------------------
      Reporter:  teddy   |      Owner:  zzzeek
          Type:  defect  |     Status:  closed
      Priority:  medium  |  Milestone:
     Component:          |   Severity:  no triage selected yet
  declarative            |   Keywords:  oracle, cyrillic, declarative,
    Resolution:          |  reflexion
  worksforme             |
Progress State:          |
  completed/closed       |
-------------------------+-------------------------------------------------
Changes (by zzzeek):

 * status:  new => closed
 * resolution:   => worksforme
 * status_field:  awaiting triage => completed/closed


Comment:

 OK when i first saw this I assumed the oracle dialect couldn't reflect
 those names, but it seems like it is, so wow on that.

 Second issue, note that the ORM is an optional component here.  The
 reflection is working fine as you're getting a `Table` with the names you
 want.    The issue is that you're mapping that `Table` to a Python class,
 a process which by default links the columns to same-named attributes on a
 class - and Python 2.x does not accept non-ASCII identifiers - this is not
 a SQLAlchemy limitation, it's a Python 2.x one.

 So if you wanted to do #1, e.g. `CyrTest.Поле1.`, you could use Python 3
 (and you'd definitely want to use 0.9, which is released as beta1 so far,
 for the best Py3K compat), otherwise it's not possible.  In Python 3 the
 reflection + mapping works as is.

 Now using a system like a transliteration library, great, that's not a
 built-in for SQLAlchemy, you do in fact need to go with the "key"
 approach.  We provide the "column_reflect" event to most effectively
 redefine keys as needed.

 Below is an example program against SQLite illustrating both a Python 3
 and a Python 2 approach.   I don't have a transliteration library
 available so I am using `"attr_%s" % hexlify(name.encode('utf-8')` here,
 replace that with whatever you'd like.

 If the reflection is working then this is as far as we can go, so I'm
 closing this as "worksforme".

 {{{
 #!python
 #!coding: utf-8

 from sqlalchemy import create_engine, Table
 from sqlalchemy.ext.declarative import declarative_base
 from sqlalchemy.orm import Session
 import sys

 e = create_engine("sqlite://", echo=True)

 e.execute(u"""
     CREATE TABLE test (

         "ИД" INTEGER PRIMARY KEY,
         "ПОЛЕ1" VARCHAR(20),
         "ЕЩЕПОЛЕ" VARCHAR(20)
     )
 """)

 e.execute(u"""
     INSERT INTO test ("ИД", "ПОЛЕ1", "ЕЩЕПОЛЕ") VALUES (1, 'v1', 'v2')
 """)


 py3 = sys.version_info >= (3, )

 if not py3:
     from sqlalchemy import event
     import binascii

     @event.listens_for(Table, "column_reflect")
     def column_reflect(inspector, table, column_info):
         # do whatever encoding/transliteration you'd like here
         column_info['key'] = "attr_%s" %
 binascii.hexlify(column_info['name'].encode('utf-8'))


 Base = declarative_base()

 class Foo(Base):
     __table__ = Table('test', Base.metadata, autoload=True,
 autoload_with=e)

 s = Session(e)
 f1 = s.query(Foo).first()

 if py3:
     # using exec so this script runs in py2k as well,
     # you can use non-ascii identifiers in py3k.
     exec("print(f1.ПОЛЕ1)")
 else:
     # our hexlified (or transliteterated, or whatever)
     print(f1.attr_d09fd09ed09bd09531)
 }}}

--
Ticket URL: <http://www.sqlalchemy.org/trac/ticket/2856#comment:4>
sqlalchemy <http://www.sqlalchemy.org/>
The Database Toolkit for Python

Re: [Sqlalchemy-tickets] [sqlalchemy] #2856: Oracle error with reflect on table with cyrillic column names in

[sqlalchemy <mi...@zz...>]

#2856: Oracle error with reflect on table with cyrillic column names in
-------------------------+-------------------------------------------------
      Reporter:  teddy   |      Owner:  zzzeek
          Type:  defect  |     Status:  new
      Priority:  medium  |  Milestone:
     Component:          |   Severity:  no triage selected yet
  declarative            |   Keywords:  oracle, cyrillic, declarative,
    Resolution:          |  reflexion
Progress State:          |
  awaiting triage        |
-------------------------+-------------------------------------------------

Comment (by teddy):

 Hmmm, did some experiments.
 I've put some changes to instrumentation.py:
 {{{#!python
 line 244 setattr(self.class_, key, inst)
 }}}
 to
 {{{#!python
 line 244 setattr(self.class_, key.encode('utf-8'), inst)
 }}}

 and to
 {{{#!python
 line 1 import trans
 ...
 line 244 setattr(self.class_, key.encode('trans'), inst)
 }}}

 If we use transliteration, there is a problems such a
 {{{SomeMappedClass.__table__.c.keys()}}} returns native field names.
 If we use native names, there is a big question with manipulation in
 interactive mode...

--
Ticket URL: <http://www.sqlalchemy.org/trac/ticket/2856#comment:3>
sqlalchemy <http://www.sqlalchemy.org/>
The Database Toolkit for Python

Re: [Sqlalchemy-tickets] [sqlalchemy] #2856: Oracle error with reflect on table with cyrillic column names in

[sqlalchemy <mi...@zz...>]

#2856: Oracle error with reflect on table with cyrillic column names in
-------------------------+-------------------------------------------------
      Reporter:  teddy   |      Owner:  zzzeek
          Type:  defect  |     Status:  new
      Priority:  medium  |  Milestone:
     Component:          |   Severity:  no triage selected yet
  declarative            |   Keywords:  oracle, cyrillic, declarative,
    Resolution:          |  reflexion
Progress State:          |
  awaiting triage        |
-------------------------+-------------------------------------------------

Comment (by teddy):

 Another little addition, sorry for my memory =)

 If we declare mapped class a bit more explicitly:
 {{{#!python
 class CyrTest(Base):
     __table__ = Table("cyr_test", Base.metadata,
                     Column(u'ИД', Integer, primary_key=True, key='id'),
                     Column(u'ПОЛЕ1', String, key='field1'),
                     Column(u'ЕЩЕПОЛЕ', String, key='field2'),
                     oracle_resolve_synonyms=True,
                     autoload=True
                 )
 }}}
 It will reflect on table successfully. The problem in keys, of course.
 I dont't know how to resolve this situation, but i see 2 ways:
 At first, we can just use colimn names as is, so we can write something
 like {{{CyrTest.Поле1}}}.
 At second, we can trasliterate names to latin, and use something like
 {{{CyrTest.Pole1}}}.
 Transliteration libraries is present for Python, for example
 https://pypi.python.org/pypi/trans, which i use in my projects.

--
Ticket URL: <http://www.sqlalchemy.org/trac/ticket/2856#comment:2>
sqlalchemy <http://www.sqlalchemy.org/>
The Database Toolkit for Python

Re: [Sqlalchemy-tickets] [sqlalchemy] #2856: Oracle error with reflect on table with cyrillic column names in

[sqlalchemy <mi...@zz...>]

#2856: Oracle error with reflect on table with cyrillic column names in
-------------------------+-------------------------------------------------
      Reporter:  teddy   |      Owner:  zzzeek
          Type:  defect  |     Status:  new
      Priority:  medium  |  Milestone:
     Component:          |   Severity:  no triage selected yet
  declarative            |   Keywords:  oracle, cyrillic, declarative,
    Resolution:          |  reflexion
Progress State:          |
  awaiting triage        |
-------------------------+-------------------------------------------------

Comment (by teddy):

 Little addition. This bug is present in a 0.7.x version too, just forgot
 to post it then.

--
Ticket URL: <http://www.sqlalchemy.org/trac/ticket/2856#comment:1>
sqlalchemy <http://www.sqlalchemy.org/>
The Database Toolkit for Python

[Sqlalchemy-tickets] [sqlalchemy] #2856: Oracle error with reflect on table with cyrillic column names in

[sqlalchemy <mi...@zz...>]

#2856: Oracle error with reflect on table with cyrillic column names in
-------------------------------------+-------------------------------------
 Reporter:  teddy                    |           Owner:  zzzeek
     Type:  defect                   |          Status:  new
 Priority:  medium                   |       Milestone:
Component:  declarative              |        Severity:  no triage selected
 Keywords:  oracle, cyrillic,        |  yet
  declarative, reflexion             |  Progress State:  awaiting triage
-------------------------------------+-------------------------------------
 There is a bug with reflexion on cyrillic column names  in Oracle.
 Lets assume we have a table with such DDL:
 {{{
 #!sql
 CREATE TABLE cyr_test (
     "ИД" NUMBER CONSTRAINT id_demo NOT NULL,
     "ПОЛЕ1"    VARCHAR2,
     "ЕЩЕПОЛЕ"   VARCHAR2,
     CONSTRAINT    с_id_pk_demo PRIMARY KEY ("ИД")
 );
 }}}

 So we try to reflect on this table in sqlalchemy:
 {{{
 #!python
 #!/usr/bin/env python
 # -*- coding: utf-8 -*-

 from sqlalchemy import *
 from sqlalchemy.orm import sessionmaker

 engine = create_engine('oracle+cx_oracle://<some_conn_string>,
 encoding="windows-1251", optimize_limits=True, use_binds_for_limits=False,
 echo='debug', coerce_to_decimal=False)
 from sqlalchemy.ext.declarative import declarative_base
 Base = declarative_base(bind=engine)

 S = sessionmaker(bind=engine)
 s = S()
 class CyrTest(Base):
     __table__ = Table("cyr_test", Base.metadata,
                     Column(u'ИД', Integer, primary_key=True),
                     oracle_resolve_synonyms=True,
                     autoload=True
                 )
 }}}

 And, getting this error:
 {{{#!python
 Traceback (most recent call last):
   File "<stdin>", line 1, in <module>
   File "test.py", line 14, in <module>
     class CyrTest(Base):
   File "C:\Python27\lib\site-packages\sqlalchemy\ext\declarative\api.py",
 line 5
 0, in __init__
     _as_declarative(cls, classname, cls.__dict__)
   File "C:\Python27\lib\site-packages\sqlalchemy\ext\declarative\base.py",
 line
 292, in _as_declarative
     mt.map()
   File "C:\Python27\lib\site-packages\sqlalchemy\ext\declarative\base.py",
 line
 376, in map
     **mapper_args
   File "C:\Python27\lib\site-packages\sqlalchemy\orm\__init__.py", line
 1229, in
  mapper
     return Mapper(class_, local_table, *args, **params)
   File "C:\Python27\lib\site-packages\sqlalchemy\orm\mapper.py", line 219,
 in __
 init__
     self._configure_properties()
   File "C:\Python27\lib\site-packages\sqlalchemy\orm\mapper.py", line 904,
 in _c
 onfigure_properties
     setparent=True)
   File "C:\Python27\lib\site-packages\sqlalchemy\orm\mapper.py", line
 1181, in _
 configure_property
     prop.instrument_class(self)
   File "C:\Python27\lib\site-packages\sqlalchemy\orm\properties.py", line
 128, i
 n instrument_class
     doc=self.doc
   File "C:\Python27\lib\site-packages\sqlalchemy\orm\attributes.py", line
 1449,
 in register_descriptor
     manager.instrument_attribute(key, descriptor)
   File "C:\Python27\lib\site-packages\sqlalchemy\orm\instrumentation.py",
 line 1
 92, in instrument_attribute
     self.install_descriptor(key, inst)
   File "C:\Python27\lib\site-packages\sqlalchemy\orm\instrumentation.py",
 line 2
 44, in install_descriptor
     setattr(self.class_, key, inst)
 UnicodeEncodeError: 'ascii' codec can't encode characters in position 0-1:
 ordin
 al not in range(128)
 }}}
 With another table, without cyrillic column names, everything works fine.

 Tested on sqlalchemy 0.8.1/0.8.3/0.9.0b1 in Python 2.7.2 (ActiveState
 ActivePython 2.7.2.5), Windows 7 x64, Oracle 11g.

--
Ticket URL: <http://www.sqlalchemy.org/trac/ticket/2856>
sqlalchemy <http://www.sqlalchemy.org/>
The Database Toolkit for Python