[Python-modules-commits] r346 - in /packages/forgetsql: ./
branches/ branches/upstream/
branches/upstream/current/ branches/upstream/current/bin/
branches/upstream/current/lib/ tags/
werner at users.alioth.debian.org
werner at users.alioth.debian.org
Sun Apr 30 23:41:48 UTC 2006
Author: werner
Date: Sun Apr 30 23:40:57 2006
New Revision: 346
URL: http://svn.debian.org/wsvn/python-modules/?sc=1&rev=346
Log:
[svn-inject] Installing original source of forgetsql
Added:
packages/forgetsql/
packages/forgetsql/branches/
packages/forgetsql/branches/upstream/
packages/forgetsql/branches/upstream/current/
packages/forgetsql/branches/upstream/current/BUGS
packages/forgetsql/branches/upstream/current/COPYING
packages/forgetsql/branches/upstream/current/PKG-INFO
packages/forgetsql/branches/upstream/current/README
packages/forgetsql/branches/upstream/current/TODO
packages/forgetsql/branches/upstream/current/bin/
packages/forgetsql/branches/upstream/current/bin/forgetsql-generate (with props)
packages/forgetsql/branches/upstream/current/lib/
packages/forgetsql/branches/upstream/current/lib/forgetSQL.py (with props)
packages/forgetsql/branches/upstream/current/setup.py (with props)
packages/forgetsql/tags/
Added: packages/forgetsql/branches/upstream/current/BUGS
URL: http://svn.debian.org/wsvn/python-modules/packages/forgetsql/branches/upstream/current/BUGS?rev=346&op=file
==============================================================================
--- packages/forgetsql/branches/upstream/current/BUGS (added)
+++ packages/forgetsql/branches/upstream/current/BUGS Sun Apr 30 23:40:57 2006
@@ -1,0 +1,13 @@
+Known bugs for forgetSQL
+========================
+
+See also http://www.sourceforge.net/project/forgetsql for the
+Bug database.
+
+ * getAll* and getChildren does not neccessarily use the
+ cached version.
+
+ * empty/wrong _sqlPrimary for tables with no obvious
+ primary key - should select ALL fields as primary
+ keys to avoid stupid error messages.
+
Added: packages/forgetsql/branches/upstream/current/COPYING
URL: http://svn.debian.org/wsvn/python-modules/packages/forgetsql/branches/upstream/current/COPYING?rev=346&op=file
==============================================================================
--- packages/forgetsql/branches/upstream/current/COPYING (added)
+++ packages/forgetsql/branches/upstream/current/COPYING Sun Apr 30 23:40:57 2006
@@ -1,0 +1,510 @@
+
+ GNU LESSER GENERAL PUBLIC LICENSE
+ Version 2.1, February 1999
+
+ Copyright (C) 1991, 1999 Free Software Foundation, Inc.
+ 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ Everyone is permitted to copy and distribute verbatim copies
+ of this license document, but changing it is not allowed.
+
+[This is the first released version of the Lesser GPL. It also counts
+ as the successor of the GNU Library Public License, version 2, hence
+ the version number 2.1.]
+
+ Preamble
+
+ The licenses for most software are designed to take away your
+freedom to share and change it. By contrast, the GNU General Public
+Licenses are intended to guarantee your freedom to share and change
+free software--to make sure the software is free for all its users.
+
+ This license, the Lesser General Public License, applies to some
+specially designated software packages--typically libraries--of the
+Free Software Foundation and other authors who decide to use it. You
+can use it too, but we suggest you first think carefully about whether
+this license or the ordinary General Public License is the better
+strategy to use in any particular case, based on the explanations
+below.
+
+ When we speak of free software, we are referring to freedom of use,
+not price. Our General Public Licenses are designed to make sure that
+you have the freedom to distribute copies of free software (and charge
+for this service if you wish); that you receive source code or can get
+it if you want it; that you can change the software and use pieces of
+it in new free programs; and that you are informed that you can do
+these things.
+
+ To protect your rights, we need to make restrictions that forbid
+distributors to deny you these rights or to ask you to surrender these
+rights. These restrictions translate to certain responsibilities for
+you if you distribute copies of the library or if you modify it.
+
+ For example, if you distribute copies of the library, whether gratis
+or for a fee, you must give the recipients all the rights that we gave
+you. You must make sure that they, too, receive or can get the source
+code. If you link other code with the library, you must provide
+complete object files to the recipients, so that they can relink them
+with the library after making changes to the library and recompiling
+it. And you must show them these terms so they know their rights.
+
+ We protect your rights with a two-step method: (1) we copyright the
+library, and (2) we offer you this license, which gives you legal
+permission to copy, distribute and/or modify the library.
+
+ To protect each distributor, we want to make it very clear that
+there is no warranty for the free library. Also, if the library is
+modified by someone else and passed on, the recipients should know
+that what they have is not the original version, so that the original
+author's reputation will not be affected by problems that might be
+introduced by others.
+^L
+ Finally, software patents pose a constant threat to the existence of
+any free program. We wish to make sure that a company cannot
+effectively restrict the users of a free program by obtaining a
+restrictive license from a patent holder. Therefore, we insist that
+any patent license obtained for a version of the library must be
+consistent with the full freedom of use specified in this license.
+
+ Most GNU software, including some libraries, is covered by the
+ordinary GNU General Public License. This license, the GNU Lesser
+General Public License, applies to certain designated libraries, and
+is quite different from the ordinary General Public License. We use
+this license for certain libraries in order to permit linking those
+libraries into non-free programs.
+
+ When a program is linked with a library, whether statically or using
+a shared library, the combination of the two is legally speaking a
+combined work, a derivative of the original library. The ordinary
+General Public License therefore permits such linking only if the
+entire combination fits its criteria of freedom. The Lesser General
+Public License permits more lax criteria for linking other code with
+the library.
+
+ We call this license the "Lesser" General Public License because it
+does Less to protect the user's freedom than the ordinary General
+Public License. It also provides other free software developers Less
+of an advantage over competing non-free programs. These disadvantages
+are the reason we use the ordinary General Public License for many
+libraries. However, the Lesser license provides advantages in certain
+special circumstances.
+
+ For example, on rare occasions, there may be a special need to
+encourage the widest possible use of a certain library, so that it
+becomes a de-facto standard. To achieve this, non-free programs must
+be allowed to use the library. A more frequent case is that a free
+library does the same job as widely used non-free libraries. In this
+case, there is little to gain by limiting the free library to free
+software only, so we use the Lesser General Public License.
+
+ In other cases, permission to use a particular library in non-free
+programs enables a greater number of people to use a large body of
+free software. For example, permission to use the GNU C Library in
+non-free programs enables many more people to use the whole GNU
+operating system, as well as its variant, the GNU/Linux operating
+system.
+
+ Although the Lesser General Public License is Less protective of the
+users' freedom, it does ensure that the user of a program that is
+linked with the Library has the freedom and the wherewithal to run
+that program using a modified version of the Library.
+
+ The precise terms and conditions for copying, distribution and
+modification follow. Pay close attention to the difference between a
+"work based on the library" and a "work that uses the library". The
+former contains code derived from the library, whereas the latter must
+be combined with the library in order to run.
+^L
+ GNU LESSER GENERAL PUBLIC LICENSE
+ TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION
+
+ 0. This License Agreement applies to any software library or other
+program which contains a notice placed by the copyright holder or
+other authorized party saying it may be distributed under the terms of
+this Lesser General Public License (also called "this License").
+Each licensee is addressed as "you".
+
+ A "library" means a collection of software functions and/or data
+prepared so as to be conveniently linked with application programs
+(which use some of those functions and data) to form executables.
+
+ The "Library", below, refers to any such software library or work
+which has been distributed under these terms. A "work based on the
+Library" means either the Library or any derivative work under
+copyright law: that is to say, a work containing the Library or a
+portion of it, either verbatim or with modifications and/or translated
+straightforwardly into another language. (Hereinafter, translation is
+included without limitation in the term "modification".)
+
+ "Source code" for a work means the preferred form of the work for
+making modifications to it. For a library, complete source code means
+all the source code for all modules it contains, plus any associated
+interface definition files, plus the scripts used to control
+compilation and installation of the library.
+
+ Activities other than copying, distribution and modification are not
+covered by this License; they are outside its scope. The act of
+running a program using the Library is not restricted, and output from
+such a program is covered only if its contents constitute a work based
+on the Library (independent of the use of the Library in a tool for
+writing it). Whether that is true depends on what the Library does
+and what the program that uses the Library does.
+
+ 1. You may copy and distribute verbatim copies of the Library's
+complete source code as you receive it, in any medium, provided that
+you conspicuously and appropriately publish on each copy an
+appropriate copyright notice and disclaimer of warranty; keep intact
+all the notices that refer to this License and to the absence of any
+warranty; and distribute a copy of this License along with the
+Library.
+
+ You may charge a fee for the physical act of transferring a copy,
+and you may at your option offer warranty protection in exchange for a
+fee.
+
+ 2. You may modify your copy or copies of the Library or any portion
+of it, thus forming a work based on the Library, and copy and
+distribute such modifications or work under the terms of Section 1
+above, provided that you also meet all of these conditions:
+
+ a) The modified work must itself be a software library.
+
+ b) You must cause the files modified to carry prominent notices
+ stating that you changed the files and the date of any change.
+
+ c) You must cause the whole of the work to be licensed at no
+ charge to all third parties under the terms of this License.
+
+ d) If a facility in the modified Library refers to a function or a
+ table of data to be supplied by an application program that uses
+ the facility, other than as an argument passed when the facility
+ is invoked, then you must make a good faith effort to ensure that,
+ in the event an application does not supply such function or
+ table, the facility still operates, and performs whatever part of
+ its purpose remains meaningful.
+
+ (For example, a function in a library to compute square roots has
+ a purpose that is entirely well-defined independent of the
+ application. Therefore, Subsection 2d requires that any
+ application-supplied function or table used by this function must
+ be optional: if the application does not supply it, the square
+ root function must still compute square roots.)
+
+These requirements apply to the modified work as a whole. If
+identifiable sections of that work are not derived from the Library,
+and can be reasonably considered independent and separate works in
+themselves, then this License, and its terms, do not apply to those
+sections when you distribute them as separate works. But when you
+distribute the same sections as part of a whole which is a work based
+on the Library, the distribution of the whole must be on the terms of
+this License, whose permissions for other licensees extend to the
+entire whole, and thus to each and every part regardless of who wrote
+it.
+
+Thus, it is not the intent of this section to claim rights or contest
+your rights to work written entirely by you; rather, the intent is to
+exercise the right to control the distribution of derivative or
+collective works based on the Library.
+
+In addition, mere aggregation of another work not based on the Library
+with the Library (or with a work based on the Library) on a volume of
+a storage or distribution medium does not bring the other work under
+the scope of this License.
+
+ 3. You may opt to apply the terms of the ordinary GNU General Public
+License instead of this License to a given copy of the Library. To do
+this, you must alter all the notices that refer to this License, so
+that they refer to the ordinary GNU General Public License, version 2,
+instead of to this License. (If a newer version than version 2 of the
+ordinary GNU General Public License has appeared, then you can specify
+that version instead if you wish.) Do not make any other change in
+these notices.
+^L
+ Once this change is made in a given copy, it is irreversible for
+that copy, so the ordinary GNU General Public License applies to all
+subsequent copies and derivative works made from that copy.
+
+ This option is useful when you wish to copy part of the code of
+the Library into a program that is not a library.
+
+ 4. You may copy and distribute the Library (or a portion or
+derivative of it, under Section 2) in object code or executable form
+under the terms of Sections 1 and 2 above provided that you accompany
+it with the complete corresponding machine-readable source code, which
+must be distributed under the terms of Sections 1 and 2 above on a
+medium customarily used for software interchange.
+
+ If distribution of object code is made by offering access to copy
+from a designated place, then offering equivalent access to copy the
+source code from the same place satisfies the requirement to
+distribute the source code, even though third parties are not
+compelled to copy the source along with the object code.
+
+ 5. A program that contains no derivative of any portion of the
+Library, but is designed to work with the Library by being compiled or
+linked with it, is called a "work that uses the Library". Such a
+work, in isolation, is not a derivative work of the Library, and
+therefore falls outside the scope of this License.
+
+ However, linking a "work that uses the Library" with the Library
+creates an executable that is a derivative of the Library (because it
+contains portions of the Library), rather than a "work that uses the
+library". The executable is therefore covered by this License.
+Section 6 states terms for distribution of such executables.
+
+ When a "work that uses the Library" uses material from a header file
+that is part of the Library, the object code for the work may be a
+derivative work of the Library even though the source code is not.
+Whether this is true is especially significant if the work can be
+linked without the Library, or if the work is itself a library. The
+threshold for this to be true is not precisely defined by law.
+
+ If such an object file uses only numerical parameters, data
+structure layouts and accessors, and small macros and small inline
+functions (ten lines or less in length), then the use of the object
+file is unrestricted, regardless of whether it is legally a derivative
+work. (Executables containing this object code plus portions of the
+Library will still fall under Section 6.)
+
+ Otherwise, if the work is a derivative of the Library, you may
+distribute the object code for the work under the terms of Section 6.
+Any executables containing that work also fall under Section 6,
+whether or not they are linked directly with the Library itself.
+^L
+ 6. As an exception to the Sections above, you may also combine or
+link a "work that uses the Library" with the Library to produce a
+work containing portions of the Library, and distribute that work
+under terms of your choice, provided that the terms permit
+modification of the work for the customer's own use and reverse
+engineering for debugging such modifications.
+
+ You must give prominent notice with each copy of the work that the
+Library is used in it and that the Library and its use are covered by
+this License. You must supply a copy of this License. If the work
+during execution displays copyright notices, you must include the
+copyright notice for the Library among them, as well as a reference
+directing the user to the copy of this License. Also, you must do one
+of these things:
+
+ a) Accompany the work with the complete corresponding
+ machine-readable source code for the Library including whatever
+ changes were used in the work (which must be distributed under
+ Sections 1 and 2 above); and, if the work is an executable linked
+ with the Library, with the complete machine-readable "work that
+ uses the Library", as object code and/or source code, so that the
+ user can modify the Library and then relink to produce a modified
+ executable containing the modified Library. (It is understood
+ that the user who changes the contents of definitions files in the
+ Library will not necessarily be able to recompile the application
+ to use the modified definitions.)
+
+ b) Use a suitable shared library mechanism for linking with the
+ Library. A suitable mechanism is one that (1) uses at run time a
+ copy of the library already present on the user's computer system,
+ rather than copying library functions into the executable, and (2)
+ will operate properly with a modified version of the library, if
+ the user installs one, as long as the modified version is
+ interface-compatible with the version that the work was made with.
+
+ c) Accompany the work with a written offer, valid for at least
+ three years, to give the same user the materials specified in
+ Subsection 6a, above, for a charge no more than the cost of
+ performing this distribution.
+
+ d) If distribution of the work is made by offering access to copy
+ from a designated place, offer equivalent access to copy the above
+ specified materials from the same place.
+
+ e) Verify that the user has already received a copy of these
+ materials or that you have already sent this user a copy.
+
+ For an executable, the required form of the "work that uses the
+Library" must include any data and utility programs needed for
+reproducing the executable from it. However, as a special exception,
+the materials to be distributed need not include anything that is
+normally distributed (in either source or binary form) with the major
+components (compiler, kernel, and so on) of the operating system on
+which the executable runs, unless that component itself accompanies
+the executable.
+
+ It may happen that this requirement contradicts the license
+restrictions of other proprietary libraries that do not normally
+accompany the operating system. Such a contradiction means you cannot
+use both them and the Library together in an executable that you
+distribute.
+^L
+ 7. You may place library facilities that are a work based on the
+Library side-by-side in a single library together with other library
+facilities not covered by this License, and distribute such a combined
+library, provided that the separate distribution of the work based on
+the Library and of the other library facilities is otherwise
+permitted, and provided that you do these two things:
+
+ a) Accompany the combined library with a copy of the same work
+ based on the Library, uncombined with any other library
+ facilities. This must be distributed under the terms of the
+ Sections above.
+
+ b) Give prominent notice with the combined library of the fact
+ that part of it is a work based on the Library, and explaining
+ where to find the accompanying uncombined form of the same work.
+
+ 8. You may not copy, modify, sublicense, link with, or distribute
+the Library except as expressly provided under this License. Any
+attempt otherwise to copy, modify, sublicense, link with, or
+distribute the Library is void, and will automatically terminate your
+rights under this License. However, parties who have received copies,
+or rights, from you under this License will not have their licenses
+terminated so long as such parties remain in full compliance.
+
+ 9. You are not required to accept this License, since you have not
+signed it. However, nothing else grants you permission to modify or
+distribute the Library or its derivative works. These actions are
+prohibited by law if you do not accept this License. Therefore, by
+modifying or distributing the Library (or any work based on the
+Library), you indicate your acceptance of this License to do so, and
+all its terms and conditions for copying, distributing or modifying
+the Library or works based on it.
+
+ 10. Each time you redistribute the Library (or any work based on the
+Library), the recipient automatically receives a license from the
+original licensor to copy, distribute, link with or modify the Library
+subject to these terms and conditions. You may not impose any further
+restrictions on the recipients' exercise of the rights granted herein.
+You are not responsible for enforcing compliance by third parties with
+this License.
+^L
+ 11. If, as a consequence of a court judgment or allegation of patent
+infringement or for any other reason (not limited to patent issues),
+conditions are imposed on you (whether by court order, agreement or
+otherwise) that contradict the conditions of this License, they do not
+excuse you from the conditions of this License. If you cannot
+distribute so as to satisfy simultaneously your obligations under this
+License and any other pertinent obligations, then as a consequence you
+may not distribute the Library at all. For example, if a patent
+license would not permit royalty-free redistribution of the Library by
+all those who receive copies directly or indirectly through you, then
+the only way you could satisfy both it and this License would be to
+refrain entirely from distribution of the Library.
+
+If any portion of this section is held invalid or unenforceable under
+any particular circumstance, the balance of the section is intended to
+apply, and the section as a whole is intended to apply in other
+circumstances.
+
+It is not the purpose of this section to induce you to infringe any
+patents or other property right claims or to contest validity of any
+such claims; this section has the sole purpose of protecting the
+integrity of the free software distribution system which is
+implemented by public license practices. Many people have made
+generous contributions to the wide range of software distributed
+through that system in reliance on consistent application of that
+system; it is up to the author/donor to decide if he or she is willing
+to distribute software through any other system and a licensee cannot
+impose that choice.
+
+This section is intended to make thoroughly clear what is believed to
+be a consequence of the rest of this License.
+
+ 12. If the distribution and/or use of the Library is restricted in
+certain countries either by patents or by copyrighted interfaces, the
+original copyright holder who places the Library under this License
+may add an explicit geographical distribution limitation excluding those
+countries, so that distribution is permitted only in or among
+countries not thus excluded. In such case, this License incorporates
+the limitation as if written in the body of this License.
+
+ 13. The Free Software Foundation may publish revised and/or new
+versions of the Lesser General Public License from time to time.
+Such new versions will be similar in spirit to the present version,
+but may differ in detail to address new problems or concerns.
+
+Each version is given a distinguishing version number. If the Library
+specifies a version number of this License which applies to it and
+"any later version", you have the option of following the terms and
+conditions either of that version or of any later version published by
+the Free Software Foundation. If the Library does not specify a
+license version number, you may choose any version ever published by
+the Free Software Foundation.
+^L
+ 14. If you wish to incorporate parts of the Library into other free
+programs whose distribution conditions are incompatible with these,
+write to the author to ask for permission. For software which is
+copyrighted by the Free Software Foundation, write to the Free
+Software Foundation; we sometimes make exceptions for this. Our
+decision will be guided by the two goals of preserving the free status
+of all derivatives of our free software and of promoting the sharing
+and reuse of software generally.
+
+ NO WARRANTY
+
+ 15. BECAUSE THE LIBRARY IS LICENSED FREE OF CHARGE, THERE IS NO
+WARRANTY FOR THE LIBRARY, TO THE EXTENT PERMITTED BY APPLICABLE LAW.
+EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR
+OTHER PARTIES PROVIDE THE LIBRARY "AS IS" WITHOUT WARRANTY OF ANY
+KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
+PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE
+LIBRARY IS WITH YOU. SHOULD THE LIBRARY PROVE DEFECTIVE, YOU ASSUME
+THE COST OF ALL NECESSARY SERVICING, REPAIR OR CORRECTION.
+
+ 16. IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN
+WRITING WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY
+AND/OR REDISTRIBUTE THE LIBRARY AS PERMITTED ABOVE, BE LIABLE TO YOU
+FOR DAMAGES, INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR
+CONSEQUENTIAL DAMAGES ARISING OUT OF THE USE OR INABILITY TO USE THE
+LIBRARY (INCLUDING BUT NOT LIMITED TO LOSS OF DATA OR DATA BEING
+RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD PARTIES OR A
+FAILURE OF THE LIBRARY TO OPERATE WITH ANY OTHER SOFTWARE), EVEN IF
+SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH
+DAMAGES.
+
+ END OF TERMS AND CONDITIONS
+^L
+ How to Apply These Terms to Your New Libraries
+
+ If you develop a new library, and you want it to be of the greatest
+possible use to the public, we recommend making it free software that
+everyone can redistribute and change. You can do so by permitting
+redistribution under these terms (or, alternatively, under the terms
+of the ordinary General Public License).
+
+ To apply these terms, attach the following notices to the library.
+It is safest to attach them to the start of each source file to most
+effectively convey the exclusion of warranty; and each file should
+have at least the "copyright" line and a pointer to where the full
+notice is found.
+
+
+ <one line to give the library's name and a brief idea of what it does.>
+ Copyright (C) <year> <name of author>
+
+ This library is free software; you can redistribute it and/or
+ modify it under the terms of the GNU Lesser General Public
+ License as published by the Free Software Foundation; either
+ version 2.1 of the License, or (at your option) any later version.
+
+ This library is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ Lesser General Public License for more details.
+
+ You should have received a copy of the GNU Lesser General Public
+ License along with this library; if not, write to the Free Software
+ Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+
+Also add information on how to contact you by electronic and paper mail.
+
+You should also get your employer (if you work as a programmer) or
+your school, if any, to sign a "copyright disclaimer" for the library,
+if necessary. Here is a sample; alter the names:
+
+ Yoyodyne, Inc., hereby disclaims all copyright interest in the
+ library `Frob' (a library for tweaking knobs) written by James
+ Random Hacker.
+
+ <signature of Ty Coon>, 1 April 1990
+ Ty Coon, President of Vice
+
+That's all there is to it!
+
+
Added: packages/forgetsql/branches/upstream/current/PKG-INFO
URL: http://svn.debian.org/wsvn/python-modules/packages/forgetsql/branches/upstream/current/PKG-INFO?rev=346&op=file
==============================================================================
--- packages/forgetsql/branches/upstream/current/PKG-INFO (added)
+++ packages/forgetsql/branches/upstream/current/PKG-INFO Sun Apr 30 23:40:57 2006
@@ -1,0 +1,13 @@
+Metadata-Version: 1.0
+Name: forgetSQL
+Version: 0.5.1
+Summary: forgetSQL is a Python module for accessing SQL databases by creating
+classes that maps SQL tables to objects, normally one class pr. SQL
+table. The idea is to forget everything about SQL and just worrying
+about normal classes and objects.
+Home-page: http://forgetsql.sourceforge.net/
+Author: Stian Soiland
+Author-email: stian at soiland.no
+License: LGPL
+Description: UNKNOWN
+Platform: UNKNOWN
Added: packages/forgetsql/branches/upstream/current/README
URL: http://svn.debian.org/wsvn/python-modules/packages/forgetsql/branches/upstream/current/README?rev=346&op=file
==============================================================================
--- packages/forgetsql/branches/upstream/current/README (added)
+++ packages/forgetsql/branches/upstream/current/README Sun Apr 30 23:40:57 2006
@@ -1,0 +1,620 @@
+================
+forgetSQL readme
+================
+
+:Author: Stian Soiland <stian at soiland.no>
+:WWW: http://forgetsql.sourceforge.net/
+:License: GNU Lesser General Public License (LGPL)
+ See the file COPYING for details.
+:Status: unfinished
+:Abstract:
+ forgetSQL is a Python module for accessing SQL databases by creating
+ classes that maps SQL tables to objects, normally one class pr. SQL
+ table. The idea is to forget everything about SQL and just worrying
+ about normal classes and objects.
+
+.. contents:: Contents
+
+
+Installation
+============
+
+Installation of the forgetSQL module is pretty straight
+forward:
+
+ python setup.py install
+
+This will install forgetSQL.py into site-packages/ of your
+Python distribution.
+
+
+Dependencies
+------------
+
+* Python 2.2.1 (True/False, new style classes, classmethods, iterators)
+* Some database module (tested: MySQLdb, psycopg)
+
+If using psycopg, mx.DateTime is needed to avoid a psycopg
+bug related to re-inserting dates. psycopg depends on mx.DateTime, so
+that shouldn't be a problem.
+
+
+What is forgetSQL?
+==================
+
+Why forgetSQL?
+--------------
+
+Let's start by showing an example using an imaginary database
+``mydatabase``:
+
+This example is based on these SQL tables:
+
+account
+~~~~~~~
+
+=========== ================= =======
+accountid fullname groupid
+=========== ================= =======
+stain Stian Soiland 15
+magnun Magnus Nordseth 15
+stornes Sverre Stornes 17
+mjaavatt Erlend Mjaavatten 15
+=========== ================= =======
+
+
+group
+~~~~~
+
+======= ====
+groupid name
+======= ====
+15 unix
+17 tie
+======= ====
+
+And should output something like::
+
+ Account details for Stian Søiland
+ Group unix (15)
+ Other members:
+ Magnus Nordseth
+ Erlend Mjaavatten
+
+
+In regular SQL programming, this could be done something like this::
+
+ cursor = dbconn.cursor()
+ cursor.execute("SELECT fullname,groupid FROM account WHERE accountid=%s",
+ ('stain',))
+ fullname,groupid = cursor.fetchone()
+ print "Account details for", fullname
+
+ cursor.execute("SELECT name FROM group WHERE groupid=%s" % groupid)
+ (groupname,) = cursor.fetchone()
+ print "Group %s (%s)" % (groupid, name)
+
+ cursor.execute("""SELECT fullname
+ FROM account JOIN group USING (groupid)
+ WHERE group.groupid=%s AND
+ NOT account.accountid=%s""",
+ (groupid, accountid))
+ print "Other members:"
+ for (membername,) in cursor.fetchall():
+ print membername
+
+
+Now, using forgetSQL::
+
+ from mydatabase import *
+ account = Account("stain") # primary key
+ print "Account details for", account.fullname
+ group = account.group
+ print "Group %s (%s)" % (group.name, group.groupid)
+ print "Other members: "
+
+ for member in group.getChildren(Account):
+ # find Account with group as foreign key
+ if member <> account:
+ print member.fullname
+
+
+Notice the difference in size and complexity of these two examples.
+
+The first example is tightly bound against SQL. The programmer is forced
+to think about SQL instead of the real code. This programming style
+tends to move high-level details to SQL, even if it is not neccessary.
+In this example, when getting "other members", the detail of skipping
+the active user is done in SQL.
+
+This would hardly save any CPU time on modern computers, but has made
+the code more complex. Thinking in SQL makes your program very large,
+as everything can be solved by some specialized SQL. Trying to change
+your program or database structure at a later time would be a nightmare.
+
+Now, forgetSQL removes all those details for the every-day-SQL tasks. It
+will not be hyper-effective or give you points in the
+largest-join-ever-possible-contest, but it will help you focus on what
+you should be thinking of, making your program work.
+
+If you at a later point (when everything runs without failures)
+discovers that you need to optimize something with a mega-query in SQL,
+you could just replace that code with regular SQL operations.
+Of course, if you've been using test-driven development (like in
+http://c2.com/cgi/wiki?ExtremeProgramming ) your tests will show if
+the replaced code works.
+
+Another alternative could be to use views and stored procedure, and
+layer forgetSQL on top of those views and procedures. This has never
+been tested, though. =)
+
+What does forgetSQL do?
+-----------------------
+
+For each table in your database, a class is created. Each instance
+created of these classes refer to a row in the given table. Each
+instance have attributes that refer to the fields in the database.
+Note that the instance is not created until you access that particular
+row.
+
+So accessing a column of a row is simply accessing the attribute
+``row.column``. Now, if this column is a reference to another table, a
+foreign key, instead of an identifier you will in ``row.column`` find an
+instance from the other table, ie. from the other class.
+
+This is what happens in the example above, ``group = account.group``
+retrieves this instance. Further attribute access within this instance
+is resolved from the matching row in the group table.
+
+If you want to change some value, you could just change the attribute
+value. In the example, if you want to change my name, simply run
+``account.fullname = "Knut Carlsen"`` (my boss).
+
+You can retrieve every row in some table that refers to the current
+object. This is what happens in ``group.getChildren(Account)``, which will
+return a list of those Accounts that have a foreign key refering to
+``group``.
+
+If you retrieve the objects several times, the constructor will return
+the same object the second time (unless some timeout has expired). This
+means that changes done to the object is immediately visible to all
+instances. This is to reflect normal behaviour in object oriented
+programming.
+
+ >>> stain = Account("stain")
+ >>> stain2 = Account("stain")
+ >>> stain.fullname = "Knut Carlsen"
+ >>> print stain2.fullname
+ Knut Carlsen
+
+
+What does forgetSQL not do?
+---------------------------
+
+forgetSQL is not a way to store objects in a database. It is a way to
+use databases as objects. You cannot store arbitrary objects in the
+database unless you use pickling.
+
+forgetSQL does not help you with database design, although you might
+choose a development style that uses regular classes and objects at
+first, and then design the database afterwards. You could then change
+your classes to use forgetSQL for data retrieval and storage, and later
+possibly replace forgetSQL classes with even more advanced objects.
+
+forgetSQL does not remove the need of heavy duty SQL. In some
+situations, SQL is simply the best solution. forgetSQL might involve
+many SQL operations for something that could be done in a single
+operations with a large magic query. If something does not scale up
+with forgetSQL, even if you refactored your code, you might try using
+SQL instead. This example would use excessive time in a table with a
+million rows::
+
+ for row in table.getAll():
+ row.backedUp = True
+ row.save()
+
+This would involve creating one million object instances (each row), one
+million SELECTs (to get the other values that needs to be saved), and
+one million UPDATEs. By using ``getAllIterator`` you could reduce this to
+just one million UPDATEs (one SELECT, reusing the same object), but
+still it would be far much slower than ``UPDATE table SET
+backedUp=true``.
+
+forgetSQL does not support commits/rollback. This might be implemented
+later, but I'm still unsure of how to actually use this in programming.
+Any suggestions are welcome.
+
+Keeping in sync
+~~~~~~~~~~~~~~~
+
+forgetSQL does not ensure that objects in memory are in sync with what
+is stored in the database. The values in the object will be a snapshot
+of how the row were at the time you first tried to retrieve an
+attribute. If you change some value, and then save the object, the row
+is updated to your version, no matter what has happened in the database
+meanwhile. An object does not timeout while in memory, it does not
+refresh it's values unless you call ``_loadDB()`` manually, as
+automatically updating could confuse programmers. However, a timeout
+value is set, and if exceeded, *new* objects retrieved from database
+(ie. ``Account("stain")`` will be fresh.
+
+It is not easy to make a general way to ensure objects are updated. For
+instance, always checking it could be heavy. It could also confuse some
+programs if an object suddenly changes some of it's attributes without
+telling, this could fuck up any updates the program is attempting to do.
+On the other hand, saving a changed object as forgetSQL is now, will
+overwrite *all* attributes, not just the changed ones.
+
+
+Usage
+=====
+
+forgetsql-generate
+------------------
+
+Before you can use forgetSQL, you will need to generate a module
+containg the classes representing database tables. Luckily, forgetSQL
+ships with a program that can do this for you by guessing.
+
+The program is called ``forgetsql-generate`` and should be installed by
+``setup.py`` or the packaging system. You might need the devel-version
+of the forgetSQL package.
+
+Create a file ``tables.txt``, with a list of database tables, one per
+line. (This is needed since there is no consistent way to query a
+database about it's tables)
+
+Then generate the module representing your tables::
+
+ forgetsql-generate --dbmodule psycopg --username=johndoe
+ --password=Jens1PuLe --database=genious
+ --tables tables.txt --output Genious.py
+
+Alternative, you could pipe the table list to ``forgetsql-generate``
+and avoid ``--tables`` -- and likewise drop ``--output`` and capture stdout
+from forgetsql-generate.
+
+The generated module is ready for use, except that you need should
+set database connecting details. One possible way is included in the
+generated code, commented out and without a password.
+
+It is recommended to set connection details from the outside instead,
+since the tables might be used by different parts of a system using
+different database passwords, connection details could be in a
+configuration file, you need persistent database connections, etc.
+
+The way to do this is to set Genious._Wrapper.cursor to a cursor
+method, and Genious._Wrapper._dbModule to the database module used::
+
+ import Genious
+ import psycopg
+ conn = psycopg.connect(user="blal", pass="sdlksdlk", database="blabla")
+ Genious._Wrapper.cursor = conn.cursor()
+ Genious._Wrapper._dbModule = psycopg
+
+This should be refactored to a more userfriendly interface.
+
+
+Normal use
+----------
+
+We'll call a class that is a representation of a database table a
+forgetter, because it inherits forgetSQL.Forgetter.
+This chapter will present normal usage of such forgetters by examples.
+
+Getting a row by giving primary key
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Example::
+
+ account = Account("stain")
+ print account.fullname
+
+If the primary key is wrong (ie. the row does not exist) accessing
+``account.fullname`` will raise ``forgetSQL.NotFound``. The object is
+actually not loaded from the database until a attribute is read.
+(delayed loading) One problem with that is that ``forgetSQL.NotFound``
+will not be raised until the attribute is read.
+
+To test if the primary key is valid, force a load::
+
+ account = Account("stain")
+ try:
+ account.load()
+ except forgetSQL.NotFound():
+ print "Cannot find stain"
+ return
+
+
+Getting all rows in a table
+~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Example::
+
+ allAccounts = Account.getAll()
+ for account in allAccounts:
+ print account.accountid, account.fullname
+
+Note that ``getAll`` is a class method, so it is available even before
+creating some ``Account``. The returned list will be empty if nothing is
+found.
+
+Also note that if what you want to do is to iterate, using
+``getAllIterator()`` would work well. This avoids creating all objects
+at once.
+
+To create a new row in a table
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Example::
+
+ account = Account()
+ account.accountid = "jennyme" # primary key
+ account.fullname = "Jenny Marie Ellingsaeter"
+ account.save()
+
+If you have forgotten to set some required fields, save() will fail. If
+you don't set the primary key, forgetSQL will try to guess the sequence
+name (tablename_primarykey_seq) to retrieve a new one. This might or
+might not work. For MySQL some other magic is involved, but it should
+work.
+
+
+Change some attribute
+~~~~~~~~~~~~~~~~~~~~~
+
+Example::
+
+ account = Account("stain")
+ account.fullname = "Stian Stornes" # got married to a beautiful man
+
+You can choose wether you want to call ``save()`` or not. If you don't call
+``save()``, the object will be saved when the object reference disappaers
+(ie. del account, end of function, etc.) and collected by the garbage
+collector. Note that this might be delayed, and that any errors
+will be disgarded.
+
+If you are unsure if you have used the correct datatype or want to
+catch save-errors, use ``save()``::
+
+ group = Group(17)
+ group.accountid = 'itil' # a string won't work in a integer field
+ try:
+ group.save()
+ except Exception, e:
+ print "Could not save group %s: %s" % (group, e)
+
+The exception raised will be database module specific, like
+``psycopg.ProgrammingError``, possible containing some useful information.
+
+``save()`` will return ``True`` if successful.
+
+Undoing an attribute change
+~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+If you changed an attribute, and you don't want to save the change to
+the database (as this will happen when the garbage collector kicks in),
+you have two choices:
+
+* reset the instance to a blank state::
+
+ group.reset()
+
+ This sets everything to None, including the primary key.
+ If you have referenced the instance anywhere else, they
+ will now experience a blank instance.
+
+* reload from database::
+
+ group.load()
+
+ Note, ``load()`` will perform a new SELECT.
+
+Note that you don't have to ``reset()`` if you haven't changed any
+attributes, the instance will only save if anything has changed.
+
+
+Access foreign keys
+~~~~~~~~~~~~~~~~~~~
+
+Example::
+
+ account = Account("stain")
+ print account.group.accountid
+ print account.group.name
+
+An attribute which is a foreign key to some other table will be
+identified by forgetsql-generate if it's name is something like
+``other_table_id``. If the generator could not identify foreign keys
+correctly, modify ``_userClasses`` in the generated Forgetter
+definition. (See `Specializing the forgetters`_).
+
+To access the real primary key, use account.group.accountid or
+account.group._getID(). Note that the latter will return a tupple
+(in case the primary key contained several columns).o
+
+You can set a foreign key attribute to a new object from the
+foreign class::
+
+ import random
+ allGroups = Group.getAll()
+ for account in Account.getAll():
+ # Set the group to one of the Group instances
+ # in allGroups
+ account.group = random.choice(allGroups)
+ del account
+ # Note that by reusing the account variable all of these
+ # will be saved by the garbage collector
+
+or to just the foreign primary key::
+
+ account.group = 18
+
+Note that this referencing magic makes JOIN unneccessary in many cases,
+but be aware that due to lazy loading (attributes are not loaded from
+database before they are accessed for the first time), in some cases
+this might result in many SELECT-calls. There are ways to avoid this,
+see `Wrapping SQL queries`_.
+
+
+Finding foreign keys
+~~~~~~~~~~~~~~~~~~~~
+
+You might want to walk in reverse, finding all accounts that have a
+given group as a foreign key::
+
+ group = Group(15)
+ members = group.getChildren(Account)
+
+This is equivalent to SQL::
+
+ SELECT * FROM account WHERE groupid=15
+
+
+Deleting an instance
+~~~~~~~~~~~~~~~~~~~~
+
+Note that although rows are represented as instances, they will not be
+deleted from the database by dereferencing. Simply removing a name
+binding only removes the representation. (and actually forces a
+``save()`` if anything has changed).
+
+To remove a row from the database::
+
+ account = Account("stornes")
+ account.delete()
+
+``delete()`` might fail if your database claims reference integrity but
+does not cascade delete::
+
+ group = Group(17)
+ group.delete()
+
+
+Advanced use
+------------
+
+WHERE-clasules
+~~~~~~~~~~~~~~
+
+You may specify a where-sentence to be inserted into the SELECT-call of
+``getAll``-methods::
+
+ members = Account.getAll(where="groupid=17")
+
+Note that you must take care of proper escaping on your own by using
+this approach. Most database modules have some form of escape functions.
+
+In many cases, what you want to do with WHERE is probably the
+same as with ``getChildren()``::
+
+ group = Group(17)
+ members = group.getChildren(Account)
+
+This will be as effective as generating a WHERE-clasule, since
+``group.load()`` won't be run (no attributes accessed, only the primary
+key).
+
+The sentence is directly inserted, so you need to use the actual SQL
+column names, not the attribute names. You can use AND and OR as you
+like.
+
+If you have several clauses to be AND-ed together, forgetSQL can do this
+for you, as the where-parameter can be a list::
+
+ where = []
+ where.append("groupid=17")
+ if something:
+ where.append("fullname like 'Stian%'")
+ Account.getAll(where=where)
+
+
+Sorting
+~~~~~~~
+
+If you have specified ``_orderBy`` (see `Specializing the forgetters`_),
+the results of ``getAll*`` and ``getChildren`` will be ordered by those
+attributes.
+
+If you want to specify ordering manually, you can supply a keyword
+argument to getAll::
+
+ all = Account.getAll(orderBy="fullname")
+
+The value of ``orderBy`` could be either a string (representing the
+object attribute to be sorted) or a tupple of strings (order by A, then
+B, etc.). Note that you can only order by attributes defined in the
+given table.
+
+If you want some other fancy sorting, sort the list after retrieval
+using regular ``list.sort()``::
+
+ all = Account.getAll()
+ all.sort(lambda a,b:
+ cmp(a.split()[-1],
+ b.split()[-1]))
+ # order by last name! :=)
+
+
+More getAll
+~~~~~~~~~~~
+
+There are specialized ``getAll`` methods for different situations.
+
+If you just want the IDs in a table::
+
+ >>> all = Account.getAllIDs()
+ ['stornes', 'stain', 'magnun', 'mjaavatt']
+
+The regular ``getAll()`` actually runs ``getAllIDs()``, and returns a
+list of instances based on those IDs. The real data is not loaded
+until attribute access. In some cases, this might be OK, for instance if
+you want to call getChildren and really don't care about the attribute
+values.
+
+If you are going to iterate through the list, a common case, use
+instead::
+
+ for account in Account.getAllIterator():
+ print account.fullname
+
+This will return an iterator, not a list, returning ``Account`` objects.
+For each iteration, a new instance is returned, with all fields
+loaded. Internally in the iterator, a buffer of results from SELECT * is
+contained.
+
+In Python, object creation is a bit expensive, so you might reuse the
+same object for each iteration by creating it first and specifying it
+as the keyword argument ``useObject``::
+
+ for account in Account.getAllIterator(useObject=Account()):
+ print account.fullname
+
+Note that changes made to account in this case will be flushed unless
+you manually call ``save()``. Do not pass this instance on, as it's content
+will change for each iteration.
+
+Finally, ``getAllText()`` will use ``_shortView`` (See `Specializing
+the forgetters`_) and return tupples of (id, text). This is useful for
+a dropdown-list of selectors.
+
+Specializing the forgetters
+---------------------------
+.. About specifying and correcting _sqlFields, etc.
+
+Sorry, this section is currently unfinished.
+
+Wrapping SQL queries
+--------------------
+.. About joins, views, functions.
+
+Sorry, this section is currently unfinished.
+
+
+Framework suggestion
+--------------------
+.. My suggestions for how you should wrap up things nicely,
+.. How to deal with database connections and extensions.
+
+Sorry, this section is currently unfinished.
+
Added: packages/forgetsql/branches/upstream/current/TODO
URL: http://svn.debian.org/wsvn/python-modules/packages/forgetsql/branches/upstream/current/TODO?rev=346&op=file
==============================================================================
--- packages/forgetsql/branches/upstream/current/TODO (added)
+++ packages/forgetsql/branches/upstream/current/TODO Sun Apr 30 23:40:57 2006
@@ -1,0 +1,46 @@
+TODO for forgetSQL
+==================
+
+ * rollback/commits
+ how to do it in programs using forgetSQL?
+ how to do it internally?
+
+ * objects should timeout as well, but what should happen?
+ raise TimeOutException ?
+ Try a passive reload, and only raise exception if something has
+ changed in the database?
+
+ * generator should take parameters for database connection details
+
+ * generator should try to fetch table lists from mysql/postgresql
+ if possible
+
+ * generated framework (_Wrapper and it's like) needs to be
+ refactored - to make it easier to set database cursor and
+ module. Could the set-module be avoided?
+
+ Framework should include the suggestion of specializing in a
+ subclassing module - to allow regeneration.
+
+ * getAll* should try to use the _cache - but what about
+ getAllIterator and the useObject?
+
+ Maybe the getAll-things should be changed, when using getAll but not
+ loading datas be appropriate?
+
+ What about loading data immediately to _values, but skip
+ instanciating _userClasses until __getattr__ ?
+
+ * Include attributes in dir()
+ -- how to do this? Skip the _values dictionary? There is another
+ method almost like __getattr__ that I don't remember, it is
+ called for EVERY attribute access, not just the non-existing ones.
+ Use this?
+
+ * Only fields changed should be stored in an UPDATE
+
+ * connection details should be made easier to set
+
+ * documentation, documentation, documentation!
+ More examples, please!
+ -- README added, but needs more work
Added: packages/forgetsql/branches/upstream/current/bin/forgetsql-generate
URL: http://svn.debian.org/wsvn/python-modules/packages/forgetsql/branches/upstream/current/bin/forgetsql-generate?rev=346&op=file
==============================================================================
--- packages/forgetsql/branches/upstream/current/bin/forgetsql-generate (added)
+++ packages/forgetsql/branches/upstream/current/bin/forgetsql-generate Sun Apr 30 23:40:57 2006
@@ -1,0 +1,257 @@
+#!/usr/bin/env python
+
+# $Id: forgetsql-generate,v 1.5 2004/03/08 11:04:28 stain Exp $
+
+## Distributed under LGPL
+## (c) Stian Søiland 2002-2004
+## stian at soiland.no
+## http://forgetsql.sourceforge.net/
+
+# __version__ should really come from setup.py.. hmm
+__version__ = "0.5.1"
+
+import exceptions, time, re, types, pprint, sys
+
+import forgetSQL
+
+# backwards compatibility
+try:
+ True,False
+except NameError:
+ (True, False) = (1==1, 1==0)
+
+# Taken from http://www.python.org/doc/current/lib/built-in-funcs.html
+def my_import(name):
+ mod = __import__(name)
+ components = name.split('.')
+ # Takes care of things like pyPgSQL.PgSQL
+ for comp in components[1:]:
+ mod = getattr(mod, comp)
+ return mod
+
+def generateFromTables(tables, cursor, getLinks=1, code=0):
+ """Generates python code (or class objects if code is false)
+ based on SQL queries on the table names given in the list
+ tables.
+ code - if given - should be an dictionary containing these
+ keys to be inserted into generated code:
+ 'database': database name
+ 'module': database module name
+ 'connect': string to be inserted into module.connect()
+ """
+ curs = cursor()
+ forgetters = {}
+ class _Wrapper(forgetSQL.Forgetter):
+ pass
+ _Wrapper.cursor = cursor
+ for table in tables:
+ # capitalize the table name to make it look like a class
+ name = table.capitalize()
+ # Define the class by instanciating the meta class to
+ # the given name (requires Forgetter to be new style)
+ forgetter = _Wrapper.__class__(name, (_Wrapper,), {})
+ # Register it
+ forgetters[name] = forgetter
+ forgetter._sqlTable = table
+ forgetter._sqlLinks = {}
+ forgetter._sqlFields = {}
+ forgetter._shortView = ()
+ forgetter._descriptions = {}
+ forgetter._userClasses = {}
+
+ # Get columns
+ curs.execute("SELECT * FROM %s LIMIT 1" % table)
+ columns = [column[0] for column in curs.description]
+ # convert to dictionary and register in forgetter
+ for column in columns:
+ forgetter._sqlFields[column] = column
+
+ if getLinks:
+ # Try to find links between tables (!)
+ # Note the big O factor with this ...
+
+ for (tableName, forgetter) in forgetters.items():
+ for (key, column) in forgetter._sqlFields.items():
+ # A column refering to another table would most likely
+ # be called otherColumnID or just otherColumn. We'll
+ # lowercase below when performing the test.
+ possTable = re.sub(r'_?id$', '', column)
+
+ # all tables (ie. one of the forgetters) are candidates
+ foundLink = False
+ for candidate in forgetters.keys():
+ if candidate.lower() == possTable.lower():
+ if possTable.lower() == tableName.lower():
+ # It's our own primary key!
+ forgetter._sqlPrimary = (column,)
+ break
+
+ # Woooh! First - let's replace 'blapp_id' with 'blapp'
+ # as the attribute name to indicate that it would
+ # contain the Blapp instance, not just
+ # some ID.
+ del forgetter._sqlFields[key]
+ forgetter._sqlFields[possTable] = column
+
+ # And.. we'll need to know which class we refer to
+ forgetter._userClasses[possTable] = candidate
+ break # we've found our candidate
+
+ if code:
+ if code['module'] == "MySQLdb":
+ code['class'] = 'forgetSQL.MysqlForgetter'
+ else:
+ code['class'] = 'forgetSQL.Forgetter'
+ code['date'] = time.strftime('%Y-%m-%d')
+ print '''
+"""Database wrappers %(database)s
+Autogenerated by forgetsql-generate %(date)s.
+"""
+
+import forgetSQL
+
+#import %(module)s
+
+class _Wrapper(%(class)s):
+ """Just a simple wrapper class so that you may
+ easily change stuff for all forgetters. Typically
+ this involves subclassing MysqlForgetter instead."""
+
+ # Example database connection (might miss password)
+ #_dbModule = %(module)s
+ #_dbConnection = %(module)s.connect(%(connect)s)
+ #def cursor(self):
+ # return self._dbConnection.cursor()
+
+''' % code
+ items = forgetters.items()
+ items.sort()
+ for (name, forgetter) in items:
+ print "class %s(_Wrapper):" % name
+ for (key, value) in forgetter.__dict__.items():
+ if key.find('__') == 0:
+ continue
+ nice = pprint.pformat(value)
+ # Get some indention
+ nice = nice.replace('\n', '\n ' + ' '*len(key))
+ print ' %s = ' % key, nice
+ print ""
+ print '''
+
+# Prepare them all. We need to send in our local
+# namespace.
+forgetSQL.prepareClasses(locals())
+'''
+ else:
+ forgetSQL.prepareClasses(forgetters)
+ return forgetters
+
+def main():
+ try:
+ # Should
+ from optparse import OptionParser
+ except ImportError:
+ print >>sys.stderr, "optik 1.4.1 or Python 2.3 or later needed for command line usage"
+ print >>sys.stderr, "Download optik from http://optik.sourceforge.net/"
+ print >>sys.stderr, "or upgrade Python."
+ sys.exit(1)
+
+ usage = """usage: %prog [options]
+Generates Python code for using forgetSQL to access database tables.
+You need to include a line-seperated list of table names to either
+stdin or as a file using option --tables."""
+
+ parser = OptionParser(version="%prog " + __version__, usage=usage)
+ parser.add_option("-t", "--tables", dest="tables",
+ help="read list of tables from FILE instead of stdin",
+ metavar="FILE")
+ parser.add_option("-o", "--output", dest="output",
+ help="write generated code to OUTPUT instead of stdout")
+ parser.add_option("-m", "--dbmodule", dest="dbmodule",
+ help="database module to use")
+ parser.add_option("-H", "--host", dest="host",
+ help="hostname of database server")
+ parser.add_option("-d", "--database", dest="database",
+ help="database to connect to")
+ parser.add_option("-u", "--username", dest="username",
+ help="database username")
+ parser.add_option("-p", "--password", dest="password",
+ help="database password")
+ parser.add_option("-c", "--connect", dest="connect",
+ help="database connect string (instead of host/database/user/password")
+
+ (options, args) = parser.parse_args()
+ if options.tables:
+ try:
+ file = open(options.tables)
+ except IOError, e:
+ print >>sys.stderr, "%s: %s" % (e.strerror, e.filename)
+ sys.exit(2)
+ else:
+ file = sys.stdin
+
+ if options.output:
+ try:
+ # Override print.. dirty.
+ sys.stdout = open(output, "w")
+ except IOError, e:
+ print >>sys.stderr, "%s: %s" % (e.strerror, e.filename)
+ sys.exit(3)
+
+ if not options.dbmodule:
+ print >>sys.stderr, "Missing required option --dbmodule"
+ parser.print_help(file=sys.stderr)
+ sys.exit(4)
+
+ try:
+ dbmodule = my_import(options.dbmodule)
+ except ImportError:
+ print >>sys.stderr, "Unknown database module", options.dbmodule
+ sys.exit(5)
+
+ if options.connect:
+ connectstring = options.connect
+ try:
+ connection = dbmodule.connect(options.connect)
+ except Exception, e:
+ print >>sys.stderr, "Could not connect to database using", \
+ options.connect
+ sys.exit(6)
+ else:
+ params = {}
+ if options.database:
+ params['database'] = options.database
+ else:
+ print >>sys.stderr, "Missing required option --database or --connect"
+ sys.exit(7)
+ if options.host:
+ params['host'] = options.host
+ if options.username:
+ params['user'] = options.username
+ if options.password:
+ params['password'] = options.password
+ connectstring = ", ".join(["%s=%r" % (key, value)
+ for (key,value) in params.items()
+ # filter out password for 'security reasons'
+ if key != "password"])
+ try:
+ connection = dbmodule.connect(**params)
+ except Exception, e:
+ print >>sys.stderr, "Could not connect to database using", \
+ connectstring
+ print >>sys.stderr, e
+ sys.exit(8)
+ cursor = connection.cursor
+ tables = file.read().split()
+ if not tables:
+ print >>sys.stderr, "No table names supplied"
+ sys.exit(9)
+ # collect useful strings for generated code
+ code = {}
+ code['connect'] = connectstring
+ code['module'] = options.dbmodule
+ code['database'] = options.database or '(unknown)'
+ generateFromTables(tables, cursor, code=code)
+
+if __name__=='__main__':
+ main()
Propchange: packages/forgetsql/branches/upstream/current/bin/forgetsql-generate
------------------------------------------------------------------------------
svn:executable =
Added: packages/forgetsql/branches/upstream/current/lib/forgetSQL.py
URL: http://svn.debian.org/wsvn/python-modules/packages/forgetsql/branches/upstream/current/lib/forgetSQL.py?rev=346&op=file
==============================================================================
--- packages/forgetsql/branches/upstream/current/lib/forgetSQL.py (added)
+++ packages/forgetsql/branches/upstream/current/lib/forgetSQL.py Sun Apr 30 23:40:57 2006
@@ -1,0 +1,903 @@
+#!/usr/bin/env python
+
+__version__ = "0.5.1"
+
+## Distributed under LGPL
+## (c) Stian Søiland 2002-2003
+## stian at soiland.no
+## http://forgetsql.sourceforge.net/
+
+
+import exceptions, time, re, types, sys
+
+# from nav import database
+
+try:
+ from mx import DateTime
+except:
+ DateTime = None
+
+try:
+ True,False
+except NameError:
+ (True,False) = (1==1, 0==1)
+
+import weakref
+
+class NotFound(exceptions.Exception):
+ pass
+
+
+class Forgetter(object):
+ """SQL to object database wrapper.
+ Given a welldefined database, by subclassing Forgetter
+ and supplying some attributes, you may wrap your SQL tables
+ into objects that are easier to program with.
+
+ You must define all fields in the database table that you want
+ to expose, and you may refine the names to suit your
+ object oriented programming style. (ie. customerID -> customer)
+
+ Objects will be created without loading from database,
+ loading will occur when you try to read or write some of the
+ attributes defined as a SQL field. If you change some attributes the
+ object will be saved to the database by save() or garbage
+ collection. (be aware that GC in Py >= 2.2 is not immediate)
+
+ If you want to create new objects, just supply them with blank
+ ID-fields, and _nextSequence() will be called to fetch a new
+ ID used for insertion.
+
+ The rule is one class pr. table, although it is possible
+ to join several table into one class, as long as the
+ identificator is unique.
+
+ By defining _userClasses you can resolve links to other
+ tables, a field in this table would be an id in another
+ table, ie. another class. In practical use this means that
+ behind attributes pointing to other classes (tables)
+ you will find instances of that class.
+
+ Short example usage of forgetterobjects:
+
+ # Process all
+ for user in User.getAllIterator():
+ # Access attributes
+ print user.name
+ print "Employed at:"
+ # Access the Employed-class/table
+ print user.employed.name, user.employed.address
+ # fire him, setting employed reference to SQL NULL
+ user.employed = None
+
+ # Retrieve some ID
+ shop = Shop(552)
+ shop.name = 'Corrected name'
+ shop.save() # Save now instead of waiting for garbage collactor
+
+ # Include SQL where-statements in selections
+ myIDs = User.getAllIDs(("name='soiland'", 'salary > 5'))
+
+
+ Requirements: The attributes 'cursor' and '_dbModule' should
+ be set from the outside.
+ The cursor should be DB 2.0 complient, preferably
+ with autocommit turned on. (Transactions are not
+ within the scope of this module yet)
+
+ Python 2.2 (iterators, methodclasses)
+ """
+ # How long to keep objects in cache?
+ _timeout = 60
+ # Will be 1 once prepare() is called
+ _prepared = 0
+
+ # The default table containing our fields
+ # _sqlTable = 'shop'
+ _sqlTable = ''
+
+ # A mapping between our fields and the database fields.
+ #
+ # You must include all fields needed here. You may specify
+ # other names if you want to make the sql name more approriate
+ # for object oriented programming. (Like calling a field 'location'
+ # instead of 'location_id', because we wrap the location in a seperate
+ # object and don't really care about the id)
+ #
+ # You may reference to other tables with a dot, all
+ # other db fields will be related to _sqlTable.
+ # If you reference other tables, don't forget to
+ # modify _sqlLinks.
+ #
+ # _sqlFields = {
+ # 'id': 'shop_id',
+ # 'name': 'name',
+ # 'location': 'location_id',
+ # 'chain': 'shop_chain_id',
+ # 'address': 'address.address_id',
+ # }
+ _sqlFields = {}
+
+ # A list of attribute names (in the object, not database)
+ # that are the primary key in the database. Normally
+ # 'id' is sufficient. It is legal to have
+ # multiple fields as primary key, but it won't work
+ # properly with _userClasses and getChildren().
+ #
+ # If your table is a link table or something, ALL fields
+ # should be in _sqlPrimary. (all fields are needed to define
+ # a unique row to be deleted/updated)
+ _sqlPrimary = ('id',)
+
+ # When using several tables, you should include a
+ # 'link' statement, displaying which fields link the
+ # two tables together. Note that these are sql names.
+ # _sqlLinks = (
+ # ('shop_id', 'address.shop_id'),
+ # )
+ _sqlLinks = ()
+
+ # The name of the sequence used by _nextSequence
+ # - if None, a guess will be made based on _sqlTable
+ # and _sqlPrimary.
+ _sqlSequence = None
+
+ # Order by this attribute by default, if specified
+ # _orderBy = 'name' - this could also be a tupple
+ _orderBy = None
+
+ # _userClasses can be used to trigger creation of a field
+ # with an instance of the class. The given database field
+ # will be sent to the constructor as an objectID
+ # (ie. as self.id in this object) (ie. the class does not
+ # neccessary need to be a subclass of Forgetter)
+ #
+ # This means that the attribute will be an instance of that
+ # class, not the ID. The object will not be loaded from the
+ # database until you try to read any of it's attributes,
+ # though. (to prevent unneccessary database overload and
+ # recursions)
+ #
+ # Notice that _userClasses must be a name resolvable, ie.
+ # from the same module as your other classes.
+ # _userClasses = {
+ # 'location': 'Location',
+ # 'chain': 'Chain',
+ # 'address': 'Address',
+ # }
+ _userClasses = {}
+
+ # If you want userClasses to work properly with strings instead of
+ # instances, you must also 'prepare' your classes to resolve the
+ # names. This must be done from the same module you are defining the
+ # classes: forgetSQL.prepareClasses(locals())
+
+ # A list of fields that are suitable for a textual
+ # representation (typical a one liner).
+ #
+ # Fields will be joint together with spaces or
+ # simular.
+ # _shortView = ('name')
+ _shortView = ()
+
+ # Description for the fields (ie. labels)
+ # Note that these fields will be translated with the _ function.
+ # If a field is undescribe, a capitalized version of the field name
+ # will be presented.
+ #_descriptions = {
+ # 'name': 'Full name',
+ # 'description': 'Description of thingie',
+ #}
+ _descriptions = {}
+
+ def cursor(cls):
+ try:
+ import database
+ return database.cursor()
+ except:
+ raise "cursor method undefined, no database connection could be made"
+ cursor = classmethod(cursor)
+
+ # a reference to the database module object used, ie.
+ # MySQLdb, psycopg etc.
+ # Use MyClass._dbModule = MySQLdb - not "MySQLdb"
+ #
+ _dbModule = None
+
+ def __new__(cls, *args):
+ if not hasattr(cls, '_cache'):
+ cls._cache = {}
+ try: # to implement 'goto' in Python.. UGH
+ if not cls._cache.has_key(args):
+ # unknown
+ raise "NotFound"
+ (ref, updated) = cls._cache[args]
+ realObject = ref()
+ if realObject is None:
+ # No more real references to it, dead object
+ raise "NotFound"
+ age = time.time() - updated
+ if age > cls._timeout:
+ # Too old!
+ raise "NotFound"
+ updated = time.time()
+ except "NotFound":
+ # We'll need to create it
+ realObject = object.__new__(cls, *args)
+ ref = weakref.ref(realObject)
+ updated = time.time()
+ # store a weak reference
+ cls._cache[args] = (ref, updated)
+ return realObject
+
+ def __init__(self, *id):
+ """Initialize, possibly with a database id. A forgetter with
+ multivalue primary key (ie. _sqlPrimary more than 1 in length),
+ may be initalized by using several parameters to this constructor.
+ Note that the object will not be loaded before you call load()."""
+ self._values = {}
+ self.reset()
+ if not id:
+ self._resetID()
+ else:
+ self._setID(id)
+ def _setID(self, id):
+ """Sets the ID, id can be either a list, following the
+ _sqlPrimary, or some other type, that will be set
+ as the singleton ID (requires 1-length sqlPrimary).
+
+ """
+ if type(id) in (types.ListType, types.TupleType):
+ try:
+ for key in self._sqlPrimary:
+ value = id[0]
+ self.__dict__[key] = value
+ id = id[1:] # rest, go revursive
+ except IndexError:
+ raise 'Not enough id fields, required: %s' % len(self._sqlPrimary)
+ elif len(self._sqlPrimary) <= 1:
+ # It's a simple value
+ key = self._sqlPrimary[0]
+ self.__dict__[key] = id
+ else:
+ raise 'Not enough id fields, required: %s' % len(self._sqlPrimary)
+ self._new = False
+
+ def _getID(self):
+ """Gets the ID values as a tupple annotated by sqlPrimary"""
+ id = []
+ for key in self._sqlPrimary:
+ value = self.__dict__[key]
+ if isinstance(value, Forgetter):
+ # It's another object, we store only the ID
+ if value._new:
+ # It's a new object too, it must be saved!
+ value.save()
+ try:
+ (value,) = value._getID()
+ except:
+ raise "Unsupported: Part %s of %s primary key is a reference to %s, with multiple-primary-key %s " % (key, self.__class__, value.__class__, value)
+ id.append(value)
+ return id
+
+ def _resetID(self):
+ """Resets all ID fields."""
+ # Dirty.. .=))
+ self._setID((None,) * len(self._sqlPrimary))
+ self._new = True
+
+ def _validID(self):
+ """Is all ID fields with values, ie. not None?"""
+ return not None in self._getID()
+
+ def __getattr__(self, key):
+ """Will be called when an unknown key is to be
+ retrieved, ie. most likely one of our database
+ fields."""
+ if self._sqlFields.has_key(key):
+ if not self._updated:
+ self.load()
+ return self._values[key]
+ else:
+ raise AttributeError, key
+
+ def __setattr__(self, key, value):
+ """Will be called whenever something needs to be set, so
+ we store the value as a SQL-thingie unless the key
+ is not listed in sqlFields."""
+ if key not in self._sqlPrimary and self._sqlFields.has_key(key):
+ if not self._updated:
+ self.load()
+ self._values[key] = value
+ self._changed = time.time()
+ else:
+ # It's a normal thingie
+ self.__dict__[key] = value
+
+ def __del__(self):
+ """Saves the object on deletion. Be aware of this. If
+ you want to undo some change, use reset() first.
+
+ Be aware of Python 2.2's garbage collector, that
+ might run in the background. This means that
+ unless you call save() changes might not
+ be done immediately in the database.
+
+ Not calling save() also means that you cannot catch
+ errors caused by wrong insertion/update (ie. wrong
+ datatype for a field)
+ """
+ try:
+ self.save()
+ except Exception, e:
+ pass
+
+ def _checkTable(cls, field):
+ """Splits a field from _sqlFields into table, column.
+ Registers the table in cls._tables, and returns
+ a fully qualified table.column
+ (default table: cls._sqlTable)"""
+ # Get table part
+ try:
+ (table, field) = field.split('.')
+ except ValueError:
+ table = cls._sqlTable
+ # clean away white space
+ table = table.strip()
+ field = field.strip()
+ # register table
+ cls._tables[table] = None
+ # and return in proper shape
+ return table + '.' + field
+
+ _checkTable = classmethod(_checkTable)
+
+ def reset(self):
+ """Reset all fields, almost like creating a new object.
+ Note: Forgets changes you have made not saved to database!
+ (Remember: Others might reference the object already,
+ expecting something else!)
+ Override this method if you add properties
+ not defined in _sqlFields"""
+ self._resetID()
+ self._new = None
+ self._updated = None
+ self._changed = None
+ self._values = {}
+ # initially create fields
+ for field in self._sqlFields.keys():
+ self._values[field] = None
+
+ def load(self, id=None):
+ """Loads from database. Old values will be discarded."""
+ if id is not None:
+ # We are asked to change our ID to something else
+ self.reset()
+ self._setID(id)
+ if not self._new and self._validID():
+ self._loadDB()
+ self._updated = time.time()
+
+ def save(self):
+ """Saves to database if anything has changed since last load"""
+ if ( self._new or
+ (self._validID() and self._changed) or
+ (self._updated and self._changed > self._updated) ):
+ # Don't save if we have not loaded existing data!
+ self._saveDB()
+ return True
+ return False
+
+ def delete(self):
+ """Marks this object for deletion in the database.
+ The object will then be reset and ready for use
+ again with a new id."""
+ (sql, ) = self._prepareSQL("DELETE")
+ curs = self.cursor()
+ curs.execute(sql, self._getID())
+ curs.close()
+ self.reset()
+
+ def _prepareSQL(cls, operation="SELECT", where=None, selectfields=None, orderBy=None):
+ """Returns a sql for the given operation.
+ Possible operations:
+ SELECT read data for this id
+ SELECTALL read data for all ids
+ INSERT insert data, create new id
+ UPDATE update data for this id
+ DELETE remove data for this id
+
+ SQL will be built by data from _sqlFields, and will
+ contain 0 or several %s for you to sprintf-format in later:
+
+ SELECT --> len(cls._sqlPrimary)
+ SELECTALL --> 0 %s
+ INSERT --> len(cls._sqlFields) %s (including id)
+ UPDATE --> len(cls._sqlFields) %s (including id)
+ DELETE --> len(cls._sqlPrimary)
+
+ (Note: INSERT and UPDATE will only change values in _sqlTable, so
+ the actual number of fields for substitutions might be lower
+ than len(cls._sqlFields) )
+
+ For INSERT you should use cls._nextSequence() to retrieve
+ a new 'id' number. Note that if your sequences are not named
+ tablename_primarykey_seq (ie. for table 'blapp' with primary key
+ 'john_id', sequence name blapp_john_id_seq) you must give the sequence
+ name as an optional argument to _nextSequence)
+
+ Additional note: cls._nextSequence() MUST be overloaded
+ for multi _sqlPrimary classes. Return a tupple.
+
+ Return values will always be tuples:
+ SELECT --> (sql, fields)
+ SELECTALL -> sql, fields)
+ INSERT -> (sql, fields)
+ UPDATE -> (sql, fields)
+ DELETE -> (sql,) -- for consistency
+
+ fields will be object properties as a list, ie. the keys from
+ cls._sqlFields. The purpose of this list is to give the programmer
+ an idea of which order the keys are inserted in the SQL, giving
+ help for retreiving (SELECT, SELECTALL) or inserting for %s
+ (INSERT, DELETE).
+
+ Why? Well, the keys are stored in a hash, and we cannot be sure
+ about the order of hash.keys() from time to time, not even with
+ the same instance.
+
+ Optional where-parameter applies to SELECT, SELECTALL and DELETE.
+ where should be a list or string of where clauses.
+
+ """
+ # Normalize parameter for later comparissions
+ operation = operation.upper()
+ # Convert where to a list if it is a string
+ if type(where) in (types.StringType, types.UnicodeType):
+ where = (where,)
+ if orderBy is None:
+ orderBy = cls._orderBy
+
+ if operation in ('SELECT', 'SELECTALL'):
+ # Get the object fields and sql fields in the same
+ # order to be able to reconstruct later.
+ fields = []
+ sqlfields = []
+ for (field, sqlfield) in cls._sqlFields.items():
+ if selectfields is None or field in selectfields:
+ fields.append(field)
+ sqlfields.append(sqlfield)
+ if not fields:
+ # dirrrrrty!
+ raise """ERROR: No fields defined, cannot create SQL.
+Maybe sqlPrimary is invalid?
+Fields asked: %s
+My fields: %s""" % (selectfields, cls._sqlFields)
+
+ sql = "SELECT\n "
+ sql += ', '.join(sqlfields)
+ sql += "\nFROM\n "
+ tables = cls._tables.keys()
+ if not tables:
+ raise "REALITY ERROR: No tables defined"
+ sql += ', '.join(tables)
+ tempWhere = ["%s=%s" % linkPair for linkPair in cls._sqlLinks]
+ # this MUST be here.
+ if operation <> 'SELECTALL':
+ for key in cls._sqlPrimary:
+ tempWhere.append(cls._sqlFields[key] + "=%s")
+ if where:
+ tempWhere += where
+ if(tempWhere):
+ # Make sure to use paranteses in case someone has used
+ # ORs in the WHERE-list..
+ sql += "\nWHERE\n ("
+ sql += ') AND\n ('.join(tempWhere)
+ sql += ')'
+ if operation == 'SELECTALL' and orderBy:
+ sql += '\nORDER BY\n '
+ if type(orderBy) in (types.TupleType, types.ListType):
+ orderBy = [cls._sqlFields[x] for x in orderBy]
+ orderBy = ',\n '.join(orderBy)
+ else:
+ orderBy = cls._sqlFields[orderBy]
+ sql += orderBy
+ return (sql, fields)
+
+ elif operation in ('INSERT', 'UPDATE'):
+ if operation == 'UPDATE':
+ sql = 'UPDATE %s SET\n ' % cls._sqlTable
+ else:
+ sql = 'INSERT INTO %s (\n ' % cls._sqlTable
+
+ set = []
+ fields = []
+ sqlfields = []
+ for (field, sqlfield) in cls._sqlFields.items():
+ if operation == 'UPDATE' and field in cls._sqlPrimary:
+ continue
+ if sqlfield.find(cls._sqlTable + '.') == 0:
+ # It's a local field, chop of the table part
+ sqlfield = sqlfield[len(cls._sqlTable)+1:]
+ fields.append(field)
+ sqlfields.append(sqlfield)
+ set.append(sqlfield + '=%s')
+ if operation == 'UPDATE':
+ sql += ',\n '.join(set)
+ sql += '\nWHERE\n '
+ tempWhere = []
+ for key in cls._sqlPrimary:
+ tempWhere.append(cls._sqlFields[key] + "=%s")
+ fields.append(key)
+ sql += ' AND\n '.join(tempWhere)
+ else:
+ sql += ',\n '.join(sqlfields)
+ sql += ')\nVALUES (\n '
+ sql += ',\n '.join(('%s',) * len(sqlfields))
+ sql += ')'
+
+ return (sql, fields)
+
+ elif operation == 'DELETE':
+ sql = 'DELETE FROM ' + cls._sqlTable + ' WHERE '
+ if where:
+ sql += " AND\n ".join(where)
+ else:
+ for key in cls._sqlPrimary:
+ tempWhere = []
+ for key in cls._sqlPrimary:
+ tempWhere.append(cls._sqlFields[key] + "=%s")
+ sql += ' AND\n '.join(tempWhere)
+ return (sql, )
+ else:
+ raise "Unknown operation", operation
+
+ _prepareSQL = classmethod(_prepareSQL)
+
+ def _nextSequence(cls, name=None):
+ """Returns a new sequence number for insertion in self._sqlTable.
+ Note that if your sequences are not named
+ tablename_primarykey_seq (ie. for table 'blapp' with primary key
+ 'john_id', sequence name blapp_john_id_seq) you must give the full
+ sequence name as an optional argument to _nextSequence)
+ """
+ if not name:
+ name = cls._sqlSequence
+ if not name:
+ # Assume it's tablename_primarykey_seq
+ if len(cls._sqlPrimary) <> 1:
+ raise "Could not guess sequence name for multi-primary-key"
+ primary = cls._sqlPrimary[0]
+ name = '%s_%s_seq' % (cls._sqlTable, primary.replace('.','_'))
+ # Don't have . as a tablename or column name! =)
+ curs = cls.cursor()
+ curs.execute("SELECT nextval('%s')" % name)
+ value = curs.fetchone()[0]
+ curs.close()
+ return value
+
+ _nextSequence = classmethod(_nextSequence)
+
+ def _loadFromRow(self, result, fields, cursor):
+ """Load from a database row, described by fields.
+ fields should be the attribute names that
+ will be set. Note that userclasses will be
+ created (but not loaded). """
+ position = 0
+ for elem in fields:
+ value = result[position]
+ valueType = cursor.description[position][1]
+ if hasattr(self._dbModule, 'BOOLEAN') and \
+ valueType == self._dbModule.BOOLEAN and \
+ (value is not True or value is not False):
+ # convert to a python boolean
+ value = value and True or False
+ if value and self._userClasses.has_key(elem):
+ userClass = self._userClasses[elem]
+ # create an instance
+ value = userClass(value)
+
+ self._values[elem] = value
+ position += 1
+
+ def _loadDB(self):
+ """Connects to the database to load myself"""
+ if not self._validID():
+ raise NotFound, self._getID()
+ (sql, fields) = self._prepareSQL("SELECT")
+ curs = self.cursor()
+ curs.execute(sql, self._getID())
+ result = curs.fetchone()
+ curs.close()
+ if not result:
+ raise NotFound, self._getID()
+ self._loadFromRow(result, fields, curs)
+ self._updated = time.time()
+
+ def _saveDB(self):
+ """Inserts or updates into the database. Note that every field
+ will be updated, not just the changed one."""
+ # We're a "fresh" copy now
+ self._updated = time.time()
+ if self._new:
+ operation = 'INSERT'
+ if not self._validID():
+ self._setID(self._nextSequence())
+ # Note that we assign this ID to our self
+ # BEFORE possibly saving any of our attribute
+ # objects that might be new as well. This means
+ # that they might have references to us, as long
+ # as the database does not require our existence
+ # yet.
+ #
+ # Since mysql does not have Sequences, this will
+ # not work as smoothly there. See class
+ # MysqlForgetter below.
+ else:
+ operation = 'UPDATE'
+ (sql, fields) = self._prepareSQL(operation)
+ values = []
+ for field in fields:
+ value = getattr(self, field)
+ # First some dirty datatype hacks
+ if DateTime and type(value) == DateTime.DateTimeType:
+ # stupid psycopg does not support it's own return type..
+ # lovely..
+ value = str(value)
+ if DateTime and type(value) == DateTime.DateTimeDeltaType:
+ # Format delta as days, hours, minutes seconds
+ # NOTE: includes value.second directly to get the
+ # whole floating number
+ value = value.strftime("%d %H:%M:") + str(value.second)
+ if value is True or value is False:
+ # We must store booleans as 't' and 'f' ...
+ value = value and 't' or 'f'
+ if isinstance(value, Forgetter):
+ # It's another object, we store only the ID
+ if value._new:
+ # It's a new object too, it must be saved!
+ value.save()
+ try:
+ (value,) = value._getID()
+ except:
+ raise "Unsupported: Can't reference multiple-primary-key: %s" % value
+ values.append(value)
+ cursor = self.cursor()
+ cursor.execute(sql, values)
+ # cursor.commit()
+ cursor.close()
+ self._new = False
+
+ def getAll(cls, where=None, orderBy=None):
+ """Retrieves all the objects, possibly matching
+ the where list of clauses, that will be AND-ed.
+ This will not load everything out
+ from the database, but will create a large amount
+ of objects with only the ID inserted.
+ The data will be loaded from the objects
+ when needed by the regular load()-autocall."""
+ ids = cls.getAllIDs(where, orderBy=orderBy)
+ # Instansiate a lot of them
+ if len(cls._sqlPrimary) > 1:
+ return [cls(*id) for id in ids]
+ else:
+ return [cls(id) for id in ids]
+
+
+ getAll = classmethod(getAll)
+
+ def getAllIterator(cls, where=None, buffer=100,
+ useObject=None, orderBy=None):
+ """Retrieves every object, possibly limitted by the where
+ list of clauses that will be AND-ed). Since this an
+ iterator is returned, only buffer rows are loaded
+ from the database at once. This is useful if you need
+ to process all objects. If useObject is given, this object
+ is returned each time, but with new data.
+ """
+ (sql, fields) = cls._prepareSQL("SELECTALL", where, orderBy=orderBy)
+ curs = cls.cursor()
+ fetchedAt = time.time()
+ curs.execute(sql)
+
+ # We might start eating memory at this point
+
+ def getNext(rows=[]):
+ forgetter = cls
+ if not rows:
+ rows += curs.fetchmany(buffer)
+ if not rows:
+ curs.close()
+ return None
+ row = rows[0]
+ del rows[0]
+ try:
+ idPositions = [fields.index(key) for key in cls._sqlPrimary]
+ except ValueError:
+ raise "Bad sqlPrimary, should be a list or tupple: %s" % cls._sqlPrimary
+ ids = [row[pos] for pos in idPositions]
+ if useObject:
+ result = useObject
+ result.reset()
+ result._setID(ids)
+ else:
+ result = forgetter(*ids)
+ result._loadFromRow(row, fields, curs)
+ result._updated = fetchedAt
+ return result
+
+ return iter(getNext, None)
+
+ getAllIterator = classmethod(getAllIterator)
+
+ def getAllIDs(cls, where=None, orderBy=None):
+ """Retrives all the IDs, possibly matching the
+ where clauses. Where should be some list of
+ where clauses that will be joined with AND). Note
+ that the result might be tuples if this table
+ has a multivalue _sqlPrimary."""
+
+ (sql, fields) = cls._prepareSQL("SELECTALL", where,
+ cls._sqlPrimary, orderBy=orderBy)
+ curs = cls.cursor()
+ curs.execute(sql)
+ # We might start eating memory at this point
+ rows = curs.fetchall()
+ curs.close()
+ result = []
+ idPositions = [fields.index(key) for key in cls._sqlPrimary]
+ for row in rows:
+ ids = [row[pos] for pos in idPositions]
+ if len(idPositions) > 1:
+ ids = tuple(ids)
+ else:
+ ids = ids[0]
+ result.append((ids))
+ return result
+
+ getAllIDs = classmethod(getAllIDs)
+
+ def getAllText(cls, where=None, SEPERATOR=' ', orderBy=None):
+ """Retrieves a list of of all possible instances of this class.
+ The list is composed of tupples in the format (id, description) -
+ where description is a string composed by the fields from
+ cls._shortView, joint with SEPERATOR.
+
+ """
+ (sql, fields) = cls._prepareSQL("SELECTALL", where, orderBy=orderBy)
+ curs = cls.cursor()
+ curs.execute(sql)
+ # We might start eating memory at this point
+ rows = curs.fetchall()
+ curs.close()
+ result = []
+ idPositions = [fields.index(key) for key in cls._sqlPrimary]
+ shortPos = [fields.index(short) for short in cls._shortView]
+ for row in rows:
+ ids = [row[pos] for pos in idPositions]
+ if len(idPositions) > 1:
+ ids = tuple(ids)
+ else:
+ ids = ids[0]
+ text = SEPERATOR.join([str(row[pos]) for pos in shortPos])
+ result.append((ids, text))
+ return result
+
+ getAllText = classmethod(getAllText)
+
+ def getChildren(self, forgetter, field=None, where=None, orderBy=None):
+ """Returns the children that links to me. That means that I have
+ to be listed in their _userClasses somehow. If field is
+ specified, that field in my children is used as the pointer
+ to me. Use this if you have multiple fields referring to
+ my class."""
+ if type(where) in (types.StringType, types.UnicodeType):
+ where = (where,)
+
+ if not field:
+ for (i_field, i_class) in forgetter._userClasses.items():
+ if isinstance(self, i_class):
+ field = i_field
+ break # first one found is ok :=)
+ if not field:
+ raise "No field found, check forgetter's _userClasses"
+ sqlname = forgetter._sqlFields[field]
+ myID = self._getID()[0] # assuming single-primary !
+
+ whereList = ["%s='%s'" % (sqlname, myID)]
+ if where:
+ whereList.extend(where)
+
+ return forgetter.getAll(whereList, orderBy=orderBy)
+
+ def __repr__(self):
+ return self.__class__.__name__ + ' %s' % self._getID()
+
+ def __str__(self):
+ shortView = self._shortView or self._sqlPrimary
+ short = [str(getattr(self, short)) for short in shortView]
+ text = ', '.join(short)
+ # return repr(self) + ': ' + text
+ return text
+
+ def __eq__(self, obj):
+ """Simple comparsion of objects."""
+ return self.__class__.__name__ == obj.__class__.__name__ \
+ and self._getID() == obj._getID()
+
+
+class MysqlForgetter(Forgetter):
+ """MYSQL-compatible Forgetter"""
+ def _saveDB(self):
+ """Overloaded - we dont have nextval() in mysql"""
+ # We're a "fresh" copy now
+ self._updated = time.time()
+ if self._new:
+ operation = 'INSERT'
+ else:
+ operation = 'UPDATE'
+ (sql, fields) = self._prepareSQL(operation)
+ values = []
+ for field in fields:
+ value = getattr(self, field)
+ if isinstance(value, Forgetter):
+ # It's another object, we store only the ID
+ if value._new:
+ # It's a new object too, it must be saved!
+ value.save()
+ try:
+ (value,) = value._getID()
+ except:
+ raise "Can't reference multiple-primary-key: %s" % value
+ values.append(value)
+ cursor = self.cursor()
+ cursor.execute(sql, values)
+ # cursor.commit()
+
+ if not self._validID():
+ if not len(self._getID()) == 1:
+ raise "Can't retrieve auto-inserted ID for multiple-primary-key"
+ # Here's the mysql magic to get the new ID
+ self._setID(cursor.insert_id())
+ cursor.close()
+ self._new = False
+
+def prepareClasses(locals):
+ """Fix _userClasses and some stuff in classes.
+ Traverses locals, which is a locals() dictionary from
+ the namespace where Forgetter subclasses have been
+ defined, and resolves names in _userClasses to real
+ class-references.
+
+ Normally you would call forgettSQL.prepareClasses(locals())
+ after defining all classes in your local module.
+ prepareClasses will only touch objects in the name space
+ that is a subclassed of Forgetter."""
+ for (name, forgetter) in locals.items():
+ if not (type(forgetter) is types.TypeType and
+ issubclass(forgetter, Forgetter)):
+ # Only care about Forgetter objects
+ continue
+
+ # Resolve classes
+ for (key, userclass) in forgetter._userClasses.items():
+ if type(userclass) is types.StringType:
+ # resolve from locals
+ resolved = locals[userclass]
+ forgetter._userClasses[key] = resolved
+
+ forgetter._tables = {}
+ # Update all fields with proper names
+ for (field, sqlfield) in forgetter._sqlFields.items():
+ forgetter._sqlFields[field] = forgetter._checkTable(sqlfield)
+
+ newLinks = []
+ for linkpair in forgetter._sqlLinks:
+ (link1, link2) = linkpair
+ link1=forgetter._checkTable(link1)
+ link2=forgetter._checkTable(link2)
+ newLinks.append((link1, link2))
+
+ forgetter._sqlLinks = newLinks
+ forgetter._prepared = 1
+
Propchange: packages/forgetsql/branches/upstream/current/lib/forgetSQL.py
------------------------------------------------------------------------------
svn:executable =
Added: packages/forgetsql/branches/upstream/current/setup.py
URL: http://svn.debian.org/wsvn/python-modules/packages/forgetsql/branches/upstream/current/setup.py?rev=346&op=file
==============================================================================
--- packages/forgetsql/branches/upstream/current/setup.py (added)
+++ packages/forgetsql/branches/upstream/current/setup.py Sun Apr 30 23:40:57 2006
@@ -1,0 +1,26 @@
+#!/usr/bin/env python
+
+from distutils.core import setup
+
+# Whyyyyyyy oh whyyy doesn't distutils do this !??
+# (if you forget this - root with umask 0077
+# will install non-readable libraries)
+import os
+if os.geteuid() == 0:
+ os.umask(0022)
+
+setup(name="forgetSQL",
+ version="0.5.1",
+ author="Stian Soiland",
+ author_email="stian at soiland.no",
+ url="http://forgetsql.sourceforge.net/",
+ license="LGPL",
+ description=
+"""forgetSQL is a Python module for accessing SQL databases by creating
+classes that maps SQL tables to objects, normally one class pr. SQL
+table. The idea is to forget everything about SQL and just worrying
+about normal classes and objects.""",
+ py_modules=['forgetSQL'],
+ scripts = ['bin/forgetsql-generate'],
+ package_dir = {'': 'lib'},
+ )
Propchange: packages/forgetsql/branches/upstream/current/setup.py
------------------------------------------------------------------------------
svn:executable =
More information about the Python-modules-commits
mailing list