Differences between revisions 37 and 89 (spanning 52 versions)
⇤ ← Revision 37 as of 2006-09-05 11:35:37 →
Size: 6145
Editor: 211
Comment: More on SQLite
← Revision 89 as of 2008-09-11 18:50:04 → ⇥
Size: 3593
Editor: mail
Comment:
Deletions are marked like this. Additions are marked like this.
Line 1: Line 1:
  ''This page seems to duplicate ChoosingDatabase, wouldn't it be better to merge shelve and the rest there? -- DanielDittmar [[DateTime(2003-02-11T23:15:22)]]''

  ''I don't think so. Now they surely have to be refactored both, but ChoosingDatabase seems to be a good page for one place for categorized comparision to narrow novice's choice while that is merely a catalog. -- MikeRovner [[DateTime(2003-02-11T16:50:45)]]''

  ''I seems missed DbApiModuleComparison page. Probably I better combine them with links from both places. -- MikeRovner [[DateTime()]]''
Line 11: Line 3:
See also HigherLevelDatabaseProgramming The contents of this page are being merged into the ChoosingDatabase page.
Line 13: Line 5:
= DBMS interfaces =
Things you connect to.		 = Relational database =
Line 16: Line 7:
Take a look at
http://www.python.org/topics/database/modules.html and
http://dmoz.org/Computers/Programming/Languages/Python/Modules/Databases_and_Persistence/.

== MySQL ==
 * mysqldb module http://www.mysql.com/downloads/api-python.html
 * SnakeDb (http://www.scriptfoundry.com/modules/snakedb/)		 Databases based on a relational model, with support for SQL.
Line 25: Line 10:
 * psycopg (http://www.zope.org/Members/fog/psycopg) (!) preferable
  *important notes for psycopg
   *for commands that require running outside of a transaction use autocommit
    *ie conn.autocommit()
   *never commit on a cursor, always commit on a connection
 * pyPgSQL (http://pypgsql.sf.net/)
 * psycopg
  * psycopg1: http://initd.org/projects/psycopg1
  * psycopg2: http://initd.org/projects/psycopg2

 * pyPgSQL: http://pypgsql.sourceforge.net/

 * PyGreSQL: http://www.pygresql.org/

 * PoPy: http://sourceforge.net/projects/popy
  * PoPy and PyGreSQL are [http://www.zope.org/Members/tm/Full_Announce merging]

 * pg/python: http://python.projects.postgresql.org/
  * pg_proboscis: DB-API 2.0 and GreenTrunk Interfaces http://python.projects.postgresql.org/project/pg_proboscis.html
  * pg_pqueue: PQ 3.0 Protocol elements http://python.projects.postgresql.org/project/pg_pqueue.html

 * pgasync: http://jamwt.com/pgasync/
  * Asynchronous and pure Python. Speed comparable to C bindings. Special support for Twisted.

 * bpgsql: http://barryp.org/software/bpgsql/
  * Barebones pure-Python PostgreSQL client
Line 32: Line 33:
 * PyGreSQL (http://www.pygresql.org) and PoPy (http://sourceforge.net/projects/popy) are [http://archives.postgresql.org/pgsql-announce/2003-07/msg00002.php merging]
 * PostgresPy (http://python.projects.postgresql.org/)
 * pgasync (http://jamwt.com/pgasync/)
  * Asynchronous and pure Python. Speed comparable to C bindings. Special support for Twisted.
 * mxODBC: http://www.egenix.com/products/python/mxODBC/
  * Supports the [http://www.postgresql.org/ftp/odbc/versions/ PostgreSQL ODBC driver] on both Windows and Unix.
Note that you have to enable the advanced option "Use bytea for lo" in case you want to work with BLOBs.
Line 39: Line 40:
See ["Oracle"] for details, there are two choices:
 * dcOracle
 * cx_Oracle		  * cx_Oracle: http://www.python.net/crew/atuining/cx_Oracle/
Line 43: Line 42:
== Firebird/Interbase ==
 * KInterbasDB (http://kinterbasdb.sourceforge.net/)		  * DCOracle: http://www.zope.org/Products/DCOracle/
  * This is for old Oracle versions (7 and 8).

 * DCOracle2: http://www.zope.org/Members/matt/dco2
  * For Oracle 8i and up.

 * mxODBC: http://www.egenix.com/products/python/mxODBC/
  * Supports the [http://www.oracle.com/technology/tech/oci/instantclient/index.html Oracle Instant Client] which is available for Windows and many popular Unix platforms.

== IBM DB2 ==

 * More info on ["DB2"]
Line 47: Line 56:
module developed by Dave Cole http://www.object-craft.com.au/projects/sybase/
Line 49: Line 57:
== MSSQL ==
 * by Dave Cole: http://www.object-craft.com.au/projects/mssql/
 * pymssql: http://pymssql.sourceforge.net/		  * sybase: http://www.object-craft.com.au/projects/sybase/
Line 53: Line 59:
== DB2 ==
 * PyDB2: http://sourceforge.net/projects/pydb2		  * mxODBC: http://www.egenix.com/products/python/mxODBC/
  * Supports Sybase ASE and Sybase Anywhere.
Line 56: Line 62:
== BerkeleyDb == == MaxDB/SAPDB ==
Line 58: Line 64:
== ThinkSQL ==
 * Pure Python DB-API 2.0 module http://www.thinksql.co.uk		  * sapdb: http://dev.mysql.com/doc/maxdb/interfaces.html

 * mxODBC: http://www.egenix.com/products/python/mxODBC/
  * MaxDB/SAPDB's native CLI is ODBC compatible and mxODBC can link directly against the CLI libs on Unix. It also supports the ODBC driver on Windows.
Line 62: Line 70:
 * DB-API 2.0 module available here: http://informixdb.sourceforge.net/
 * InformixDB: http://informixdb.sourceforge.net/

 * mxODBC: http://www.egenix.com/products/python/mxODBC/
  * Note: The Informix ODBC drivers are included in the Informix CSDK.

== Ingres ==

 * ingresdbi: http://www.ingres.com
Line 66: Line 82:
== Gadfly ==
Gadfly is a simple relational database system implemented in Python based on the SQL Structured Query Language.		 == buzhug ==
Line 69: Line 84:
http://gadfly.sourceforge.net/ [http://buzhug.sourceforge.net/ buzhug] is a pure-Python database engine, using a Pythonic, no-SQL syntax.

The data is stored and accessed on disk (it is not an in-memory database). The implementation has been designed to make all operations, and especially selection, as fast as possible with an interpreted language.

A limited benchmark using the same use cases as SQLite's author shows that buzhug is much faster than other pure-Python modules (KirbyBase, gadfly). SQLite, which is implemented in C, is faster, but only less than 3 times on the average.
Line 72: Line 91:
SnakeSQL is a pure Python SQL database written to remove the dependence of the Python Web Modules on 3rd party drivers for non-Python databases like MySQL but designed to be a useful database in its own right.
Line 74: Line 92:
http://www.pythonweb.org/projects/snakesql/ [http://www.pythonweb.org/projects/snakesql/ SnakeSQL] is a pure Python SQL database written to remove the dependence of the Python Web Modules on 3rd party drivers for non-Python databases like MySQL but designed to be a useful database in its own right.
Line 76: Line 94:
== ZODB ==
http://www.zope.org/Wikis/ZODB Zope Object DB

== shelve ==
A [http://www.python.org/doc/current/lib/module-shelve.html ''shelf''] is a persistent, dictionary-like object. The difference with ''dbm'' databases is that the values (not the keys!) in a shelf can be essentially arbitrary Python objects -- anything that the [http://www.python.org/doc/current/lib/module-pickle.html pickle] module can handle. This includes most class instances, recursive data types, and objects containing lots of shared sub-objects. The keys are ordinary strings.

== KirbyBase ==
http://www.netpromi.com/kirbybase.html


= Datafiles interfaces =
Things you open.

== xBase ==
Which stands for .dbf files interface.[[BR]]
.dbf files were produced by several old systems like dBase(II,III,IV), Fox(Base,Pro)
 * xBase (http://linux.techass.com/projects/xdb/) - Python interface in plans
 * http://www.fiby.at/dbfpy/index.html - working now. or (take it [http://starship.python.net/crew/jjkunce/python/dbfpy.tgz here])
 * http://www.sequiter.com/products/Python/
 * http://cbbrowne.com/info/xbase.html
 * http://www.e-bachmann.dk/docs/xbase.htm - dead link

== dbm ==
A family of old unix plain hash tables. Has varieties like dbm, ndbm, gdbm, dbmdb185.[[BR]]
See [http://www.python.org/doc/current/lib/module-anydbm.html anydbm],
[http://www.python.org/doc/current/lib/module-dumbdbm.html dumbdbm],
[http://www.python.org/doc/current/lib/module-dbhash.html dbhash],
[http://www.python.org/doc/current/lib/module-bsddb.html bsddb],
[http://www.python.org/doc/current/lib/module-dbm.html dbm],
[http://www.python.org/doc/current/lib/module-gdbm.html gdbm]
in Python Standard Library.

== MetaKit ==
http://www.equi4.com/metakit/python.html

== SQLite ==
 * pysqlite: http://pysqlite.org/
  * Have versions for both SQLite 2 and 3.
 * APSW: http://www.rogerbinns.com/apsw.html
  * SQLite 3 only.
  * Not DBAPI compliant. Focus on wrapping all SQLite APIs.

= XML Databases or Interfaces =

== Forest ==

[http://cvs.infrae.com/forest/ Forest] is a (native) XML database written in Python. It is intended to
support fast queries of XML data.


== 4ODS ==

http://www.4suite.org/


= Object-Relational Mappers =

[http://sqlobject.org SQLObject] is an object-relational mapper. It allows you to translate
RDBMS table rows into Python objects, and manipulate those objects to
transparently manipulate the database.

[http://www.tux4web.de/orm ORM] The Object Relational Membrane is a Python package that provides the functionality of an object relational layer like EJB or other persistence storage systems. It is a thin compatibility layer between SQL table layouts and Object Oriented Python. While providing a good deal of functionality, it tries to be as small and simple as possible. It works with PostgreSQL and MySQL.

[http://www.qlime.org/ QLime] Easy to use, transparent data access to relational databases or other data sources. See examples here: http://www.qlime.org/example.rst

See also HigherLevelDatabaseProgramming.

= Special file interface =

 * http://python-dsv.sourceforge.net/ CSV or any separated file (see also PEP:0305)
 * ConfigParser.py - Windows .ini format
 * gzip.py
 * zipfile.py
 * tar
 * pdf http://www.pythonware.com/
 * PyTables		 == MySQL ==
mysql-python: http://sourceforge.net/projects/mysql-python

TableOfContents

The contents of this page are being merged into the ChoosingDatabase page.

Relational database

Databases based on a relational model, with support for SQL.

PostgreSQL

Note that you have to enable the advanced option "Use bytea for lo" in case you want to work with BLOBs.

Oracle

IBM DB2

  • More info on ["DB2"]

Sybase

MaxDB/SAPDB

Informix

Ingres

Native Python Databases

buzhug

[http://buzhug.sourceforge.net/ buzhug] is a pure-Python database engine, using a Pythonic, no-SQL syntax.

The data is stored and accessed on disk (it is not an in-memory database). The implementation has been designed to make all operations, and especially selection, as fast as possible with an interpreted language.

A limited benchmark using the same use cases as SQLite's author shows that buzhug is much faster than other pure-Python modules (KirbyBase, gadfly). SQLite, which is implemented in C, is faster, but only less than 3 times on the average.

SnakeSQL

[http://www.pythonweb.org/projects/snakesql/ SnakeSQL] is a pure Python SQL database written to remove the dependence of the Python Web Modules on 3rd party drivers for non-Python databases like MySQL but designed to be a useful database in its own right.

MySQL

mysql-python: http://sourceforge.net/projects/mysql-python

DatabaseInterfaces (last edited 2020-12-09 09:29:13 by MarcAndreLemburg)

Unable to edit the page? See the FrontPage for instructions.