<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="http://www.w3.org/1999/xhtml">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
<title>
What’s New in SQLAlchemy 0.6?
—
SQLAlchemy 0.8 Documentation
</title>
<!-- begin iterate through SQLA + sphinx environment css_files -->
<link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
<link rel="stylesheet" href="../_static/docs.css" type="text/css" />
<link rel="stylesheet" href="../_static/changelog.css" type="text/css" />
<link rel="stylesheet" href="../_static/sphinx_paramlinks.css" type="text/css" />
<!-- end iterate through SQLA + sphinx environment css_files -->
<!-- begin layout.mako headers -->
<script type="text/javascript">
var DOCUMENTATION_OPTIONS = {
URL_ROOT: '../',
VERSION: '0.8.7',
COLLAPSE_MODINDEX: false,
FILE_SUFFIX: '.html'
};
</script>
<!-- begin iterate through sphinx environment script_files -->
<script type="text/javascript" src="../_static/jquery.js"></script>
<script type="text/javascript" src="../_static/underscore.js"></script>
<script type="text/javascript" src="../_static/doctools.js"></script>
<!-- end iterate through sphinx environment script_files -->
<script type="text/javascript" src="../_static/detectmobile.js"></script>
<script type="text/javascript" src="../_static/init.js"></script>
<link rel="index" title="Index" href="../genindex.html" />
<link rel="search" title="Search" href="../search.html" />
<link rel="copyright" title="Copyright" href="../copyright.html" />
<link rel="top" title="SQLAlchemy 0.8 Documentation" href="../index.html" />
<link rel="up" title="Changes and Migration" href="index.html" />
<link rel="next" title="What’s new in SQLAlchemy 0.5?" href="migration_05.html" />
<link rel="prev" title="What’s New in SQLAlchemy 0.7?" href="migration_07.html" />
<!-- end layout.mako headers -->
</head>
<body>
<div id="docs-container">
<div id="docs-top-navigation-container" class="body-background">
<div id="docs-header">
<div id="docs-version-header">
Release: <span class="version-num">0.8.7</span> | Release Date: July 22, 2014
</div>
<h1>SQLAlchemy 0.8 Documentation</h1>
</div>
</div>
<div id="docs-body-container">
<div id="fixed-sidebar" class="withsidebar">
<div id="docs-sidebar-popout">
<h3><a href="../index.html">SQLAlchemy 0.8 Documentation</a></h3>
<p id="sidebar-paginate">
<a href="index.html" title="Changes and Migration">Up</a> |
<a href="migration_07.html" title="What’s New in SQLAlchemy 0.7?">Prev</a> |
<a href="migration_05.html" title="What’s new in SQLAlchemy 0.5?">Next</a>
</p>
<p id="sidebar-topnav">
<a href="../index.html">Contents</a> |
<a href="../genindex.html">Index</a>
</p>
<div id="sidebar-search">
<form class="search" action="../search.html" method="get">
<input type="text" name="q" size="12" /> <input type="submit" value="Search" />
<input type="hidden" name="check_keywords" value="yes" />
<input type="hidden" name="area" value="default" />
</form>
</div>
</div>
<div id="docs-sidebar">
<h3><a href="#">
What’s New in SQLAlchemy 0.6?
</a></h3>
<ul>
<li><a class="reference internal" href="#">What’s New in SQLAlchemy 0.6?</a><ul>
<li><a class="reference internal" href="#platform-support">Platform Support</a></li>
<li><a class="reference internal" href="#new-dialect-system">New Dialect System</a><ul>
<li><a class="reference internal" href="#dialect-imports">Dialect Imports</a></li>
</ul>
</li>
<li><a class="reference internal" href="#expression-language-changes">Expression Language Changes</a><ul>
<li><a class="reference internal" href="#an-important-expression-language-gotcha">An Important Expression Language Gotcha</a></li>
<li><a class="reference internal" href="#stricter-executemany-behavior">Stricter “executemany” Behavior</a></li>
<li><a class="reference internal" href="#union-and-other-compound-constructs-parenthesize-consistently">UNION and other “compound” constructs parenthesize consistently</a></li>
</ul>
</li>
<li><a class="reference internal" href="#c-extensions-for-result-fetching">C Extensions for Result Fetching</a></li>
<li><a class="reference internal" href="#new-schema-capabilities">New Schema Capabilities</a><ul>
<li><a class="reference internal" href="#deprecated-removed-schema-elements">Deprecated/Removed Schema Elements</a></li>
<li><a class="reference internal" href="#other-behavioral-changes">Other Behavioral Changes</a></li>
</ul>
</li>
<li><a class="reference internal" href="#logging-opened-up">Logging opened up</a></li>
<li><a class="reference internal" href="#reflection-inspector-api">Reflection/Inspector API</a></li>
<li><a class="reference internal" href="#returning-support">RETURNING Support</a></li>
<li><a class="reference internal" href="#type-system-changes">Type System Changes</a><ul>
<li><a class="reference internal" href="#new-archicture">New Archicture</a></li>
<li><a class="reference internal" href="#native-unicode-mode">Native Unicode Mode</a></li>
<li><a class="reference internal" href="#generic-enum-type">Generic Enum Type</a></li>
<li><a class="reference internal" href="#reflection-returns-dialect-specific-types">Reflection Returns Dialect-Specific Types</a></li>
<li><a class="reference internal" href="#miscellaneous-api-changes">Miscellaneous API Changes</a></li>
</ul>
</li>
<li><a class="reference internal" href="#orm-changes">ORM Changes</a><ul>
<li><a class="reference internal" href="#new-unit-of-work">New Unit of Work</a></li>
<li><a class="reference internal" href="#changes-to-query-update-and-query-delete">Changes to <tt class="docutils literal"><span class="pre">query.update()</span></tt> and <tt class="docutils literal"><span class="pre">query.delete()</span></tt></a></li>
<li><a class="reference internal" href="#relation-is-officially-named-relationship"><tt class="docutils literal"><span class="pre">relation()</span></tt> is officially named <tt class="docutils literal"><span class="pre">relationship()</span></tt></a></li>
<li><a class="reference internal" href="#subquery-eager-loading">Subquery eager loading</a></li>
<li><a class="reference internal" href="#eagerload-eagerload-all-is-now-joinedload-joinedload-all"><tt class="docutils literal"><span class="pre">`eagerload()``</span></tt>, <tt class="docutils literal"><span class="pre">``eagerload_all()``</span></tt> is now <tt class="docutils literal"><span class="pre">``joinedload()``</span></tt>, <tt class="docutils literal"><span class="pre">``joinedload_all()`</span></tt></a></li>
<li><a class="reference internal" href="#lazy-false-none-true-dynamic-now-accepts-lazy-noload-joined-subquery-select-dynamic"><tt class="docutils literal"><span class="pre">`lazy=False|None|True|'dynamic'``</span></tt> now accepts <tt class="docutils literal"><span class="pre">``lazy='noload'|'joined'|'subquery'|'select'|'dynamic'`</span></tt></a></li>
<li><a class="reference internal" href="#innerjoin-true-on-relation-joinedload">innerjoin=True on relation, joinedload</a></li>
<li><a class="reference internal" href="#many-to-one-enhancements">Many-to-one Enhancements</a></li>
<li><a class="reference internal" href="#mutable-primary-keys-with-joined-table-inheritance">Mutable Primary Keys with Joined Table Inheritance</a></li>
<li><a class="reference internal" href="#beaker-caching">Beaker Caching</a></li>
<li><a class="reference internal" href="#other-changes">Other Changes</a></li>
<li><a class="reference internal" href="#deprecated-removed-orm-elements">Deprecated/Removed ORM Elements</a></li>
</ul>
</li>
<li><a class="reference internal" href="#extensions">Extensions</a><ul>
<li><a class="reference internal" href="#sqlsoup">SQLSoup</a></li>
<li><a class="reference internal" href="#declarative">Declarative</a></li>
</ul>
</li>
</ul>
</li>
</ul>
</div>
</div>
<div id="docs-body" class="withsidebar" >
<div class="section" id="what-s-new-in-sqlalchemy-0-6">
<h1>What’s New in SQLAlchemy 0.6?<a class="headerlink" href="#what-s-new-in-sqlalchemy-0-6" title="Permalink to this headline">¶</a></h1>
<div class="admonition-about-this-document admonition">
<p class="first admonition-title">About this Document</p>
<p>This document describes changes between SQLAlchemy version 0.5,
last released January 16, 2010, and SQLAlchemy version 0.6,
last released May 5, 2012.</p>
<p class="last">Document date: June 6, 2010</p>
</div>
<p>This guide documents API changes which affect users
migrating their applications from the 0.5 series of
SQLAlchemy to 0.6. Note that SQLAlchemy 0.6 removes some
behaviors which were deprecated throughout the span of the
0.5 series, and also deprecates more behaviors specific to
0.5.</p>
<div class="section" id="platform-support">
<h2>Platform Support<a class="headerlink" href="#platform-support" title="Permalink to this headline">¶</a></h2>
<ul class="simple">
<li>cPython versions 2.4 and upwards throughout the 2.xx
series</li>
<li>Jython 2.5.1 - using the zxJDBC DBAPI included with
Jython.</li>
<li>cPython 3.x - see [source:sqlalchemy/trunk/README.py3k]
for information on how to build for python3.</li>
</ul>
</div>
<div class="section" id="new-dialect-system">
<h2>New Dialect System<a class="headerlink" href="#new-dialect-system" title="Permalink to this headline">¶</a></h2>
<p>Dialect modules are now broken up into distinct
subcomponents, within the scope of a single database
backend. Dialect implementations are now in the
<tt class="docutils literal"><span class="pre">sqlalchemy.dialects</span></tt> package. The
<tt class="docutils literal"><span class="pre">sqlalchemy.databases</span></tt> package still exists as a
placeholder to provide some level of backwards compatibility
for simple imports.</p>
<p>For each supported database, a sub-package exists within
<tt class="docutils literal"><span class="pre">sqlalchemy.dialects</span></tt> where several files are contained.
Each package contains a module called <tt class="docutils literal"><span class="pre">base.py</span></tt> which
defines the specific SQL dialect used by that database. It
also contains one or more “driver” modules, each one
corresponding to a specific DBAPI - these files are named
corresponding to the DBAPI itself, such as <tt class="docutils literal"><span class="pre">pysqlite</span></tt>,
<tt class="docutils literal"><span class="pre">cx_oracle</span></tt>, or <tt class="docutils literal"><span class="pre">pyodbc</span></tt>. The classes used by
SQLAlchemy dialects are first declared in the <tt class="docutils literal"><span class="pre">base.py</span></tt>
module, defining all behavioral characteristics defined by
the database. These include capability mappings, such as
“supports sequences”, “supports returning”, etc., type
definitions, and SQL compilation rules. Each “driver”
module in turn provides subclasses of those classes as
needed which override the default behavior to accommodate
the additional features, behaviors, and quirks of that
DBAPI. For DBAPIs that support multiple backends (pyodbc,
zxJDBC, mxODBC), the dialect module will use mixins from the
<tt class="docutils literal"><span class="pre">sqlalchemy.connectors</span></tt> package, which provide
functionality common to that DBAPI across all backends, most
typically dealing with connect arguments. This means that
connecting using pyodbc, zxJDBC or mxODBC (when implemented)
is extremely consistent across supported backends.</p>
<p>The URL format used by <tt class="docutils literal"><span class="pre">create_engine()</span></tt> has been enhanced
to handle any number of DBAPIs for a particular backend,
using a scheme that is inspired by that of JDBC. The
previous format still works, and will select a “default”
DBAPI implementation, such as the Postgresql URL below that
will use psycopg2:</p>
<div class="highlight-python"><div class="highlight"><pre><span class="n">create_engine</span><span class="p">(</span><span class="s">'postgresql://scott:tiger@localhost/test'</span><span class="p">)</span></pre></div>
</div>
<p>However to specify a specific DBAPI backend such as pg8000,
add it to the “protocol” section of the URL using a plus
sign “+”:</p>
<div class="highlight-python"><div class="highlight"><pre><span class="n">create_engine</span><span class="p">(</span><span class="s">'postgresql+pg8000://scott:tiger@localhost/test'</span><span class="p">)</span></pre></div>
</div>
<p>Important Dialect Links:</p>
<ul class="simple">
<li>Documentation on connect arguments:
<a class="reference external" href="http://www.sqlalchemy.org/docs/06/dbengine.html#create">http://www.sqlalchemy.org/docs/06/dbengine.html#create</a>-
engine-url-arguments.</li>
<li>Reference documentation for individual dialects: <a class="reference external" href="http://ww">http://ww</a>
w.sqlalchemy.org/docs/06/reference/dialects/index.html</li>
<li>The tips and tricks at DatabaseNotes.</li>
</ul>
<p>Other notes regarding dialects:</p>
<ul class="simple">
<li>the type system has been changed dramatically in
SQLAlchemy 0.6. This has an impact on all dialects
regarding naming conventions, behaviors, and
implementations. See the section on “Types” below.</li>
<li>the <tt class="docutils literal"><span class="pre">ResultProxy</span></tt> object now offers a 2x speed
improvement in some cases thanks to some refactorings.</li>
<li>the <tt class="docutils literal"><span class="pre">RowProxy</span></tt>, i.e. individual result row object, is
now directly pickleable.</li>
<li>the setuptools entrypoint used to locate external dialects
is now called <tt class="docutils literal"><span class="pre">sqlalchemy.dialects</span></tt>. An external
dialect written against 0.4 or 0.5 will need to be
modified to work with 0.6 in any case so this change does
not add any additional difficulties.</li>
<li>dialects now receive an initialize() event on initial
connection to determine connection properties.</li>
<li>Functions and operators generated by the compiler now use
(almost) regular dispatch functions of the form
“visit_<opname>” and “visit_<funcname>_fn” to provide
customed processing. This replaces the need to copy the
“functions” and “operators” dictionaries in compiler
subclasses with straightforward visitor methods, and also
allows compiler subclasses complete control over
rendering, as the full _Function or _BinaryExpression
object is passed in.</li>
</ul>
<div class="section" id="dialect-imports">
<h3>Dialect Imports<a class="headerlink" href="#dialect-imports" title="Permalink to this headline">¶</a></h3>
<p>The import structure of dialects has changed. Each dialect
now exports its base “dialect” class as well as the full set
of SQL types supported on that dialect via
<tt class="docutils literal"><span class="pre">sqlalchemy.dialects.<name></span></tt>. For example, to import a
set of PG types:</p>
<div class="highlight-python"><div class="highlight"><pre><span class="kn">from</span> <span class="nn">sqlalchemy.dialects.postgresql</span> <span class="kn">import</span> <span class="n">INTEGER</span><span class="p">,</span> <span class="n">BIGINT</span><span class="p">,</span> <span class="n">SMALLINT</span><span class="p">,</span>\
<span class="n">VARCHAR</span><span class="p">,</span> <span class="n">MACADDR</span><span class="p">,</span> <span class="n">DATE</span><span class="p">,</span> <span class="n">BYTEA</span></pre></div>
</div>
<p>Above, <tt class="docutils literal"><span class="pre">INTEGER</span></tt> is actually the plain <tt class="docutils literal"><span class="pre">INTEGER</span></tt> type
from <tt class="docutils literal"><span class="pre">sqlalchemy.types</span></tt>, but the PG dialect makes it
available in the same way as those types which are specific
to PG, such as <tt class="docutils literal"><span class="pre">BYTEA</span></tt> and <tt class="docutils literal"><span class="pre">MACADDR</span></tt>.</p>
</div>
</div>
<div class="section" id="expression-language-changes">
<h2>Expression Language Changes<a class="headerlink" href="#expression-language-changes" title="Permalink to this headline">¶</a></h2>
<div class="section" id="an-important-expression-language-gotcha">
<h3>An Important Expression Language Gotcha<a class="headerlink" href="#an-important-expression-language-gotcha" title="Permalink to this headline">¶</a></h3>
<p>There’s one quite significant behavioral change to the
expression language which may affect some applications.
The boolean value of Python boolean expressions, i.e.
<tt class="docutils literal"><span class="pre">==</span></tt>, <tt class="docutils literal"><span class="pre">!=</span></tt>, and similar, now evaluates accurately with
regards to the two clause objects being compared.</p>
<p>As we know, comparing a <tt class="docutils literal"><span class="pre">ClauseElement</span></tt> to any other
object returns another <tt class="docutils literal"><span class="pre">ClauseElement</span></tt>:</p>
<div class="highlight-python"><div class="highlight"><pre><span class="gp">>>> </span><span class="kn">from</span> <span class="nn">sqlalchemy.sql</span> <span class="kn">import</span> <span class="n">column</span>
<span class="gp">>>> </span><span class="n">column</span><span class="p">(</span><span class="s">'foo'</span><span class="p">)</span> <span class="o">==</span> <span class="mi">5</span>
<span class="go"><sqlalchemy.sql.expression._BinaryExpression object at 0x1252490></span></pre></div>
</div>
<p>This so that Python expressions produce SQL expressions when
converted to strings:</p>
<div class="highlight-python"><div class="highlight"><pre><span class="gp">>>> </span><span class="nb">str</span><span class="p">(</span><span class="n">column</span><span class="p">(</span><span class="s">'foo'</span><span class="p">)</span> <span class="o">==</span> <span class="mi">5</span><span class="p">)</span>
<span class="go">'foo = :foo_1'</span></pre></div>
</div>
<p>But what happens if we say this?</p>
<div class="highlight-python"><div class="highlight"><pre><span class="gp">>>> </span><span class="k">if</span> <span class="n">column</span><span class="p">(</span><span class="s">'foo'</span><span class="p">)</span> <span class="o">==</span> <span class="mi">5</span><span class="p">:</span>
<span class="gp">... </span> <span class="k">print</span> <span class="s">"yes"</span>
<span class="gp">...</span></pre></div>
</div>
<p>In previous versions of SQLAlchemy, the returned
<tt class="docutils literal"><span class="pre">_BinaryExpression</span></tt> was a plain Python object which
evaluated to <tt class="docutils literal"><span class="pre">True</span></tt>. Now it evaluates to whether or not
the actual <tt class="docutils literal"><span class="pre">ClauseElement</span></tt> should have the same hash value
as to that being compared. Meaning:</p>
<div class="highlight-python"><div class="highlight"><pre><span class="gp">>>> </span><span class="nb">bool</span><span class="p">(</span><span class="n">column</span><span class="p">(</span><span class="s">'foo'</span><span class="p">)</span> <span class="o">==</span> <span class="mi">5</span><span class="p">)</span>
<span class="go">False</span>
<span class="gp">>>> </span><span class="nb">bool</span><span class="p">(</span><span class="n">column</span><span class="p">(</span><span class="s">'foo'</span><span class="p">)</span> <span class="o">==</span> <span class="n">column</span><span class="p">(</span><span class="s">'foo'</span><span class="p">))</span>
<span class="go">False</span>
<span class="gp">>>> </span><span class="n">c</span> <span class="o">=</span> <span class="n">column</span><span class="p">(</span><span class="s">'foo'</span><span class="p">)</span>
<span class="gp">>>> </span><span class="nb">bool</span><span class="p">(</span><span class="n">c</span> <span class="o">==</span> <span class="n">c</span><span class="p">)</span>
<span class="go">True</span>
<span class="go">>>></span></pre></div>
</div>
<p>That means code such as the following:</p>
<div class="highlight-python"><div class="highlight"><pre><span class="k">if</span> <span class="n">expression</span><span class="p">:</span>
<span class="k">print</span> <span class="s">"the expression is:"</span><span class="p">,</span> <span class="n">expression</span></pre></div>
</div>
<p>Would not evaluate if <tt class="docutils literal"><span class="pre">expression</span></tt> was a binary clause.
Since the above pattern should never be used, the base
<tt class="docutils literal"><span class="pre">ClauseElement</span></tt> now raises an exception if called in a
boolean context:</p>
<div class="highlight-python"><div class="highlight"><pre><span class="gp">>>> </span><span class="nb">bool</span><span class="p">(</span><span class="n">c</span><span class="p">)</span>
<span class="gt">Traceback (most recent call last):</span>
File <span class="nb">"<stdin>"</span>, line <span class="m">1</span>, in <span class="n"><module></span>
<span class="c">...</span>
<span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="s">"Boolean value of this clause is not defined"</span><span class="p">)</span>
<span class="gr">TypeError</span>: <span class="n">Boolean value of this clause is not defined</span></pre></div>
</div>
<p>Code that wants to check for the presence of a
<tt class="docutils literal"><span class="pre">ClauseElement</span></tt> expression should instead say:</p>
<div class="highlight-python"><div class="highlight"><pre><span class="k">if</span> <span class="n">expression</span> <span class="ow">is</span> <span class="ow">not</span> <span class="bp">None</span><span class="p">:</span>
<span class="k">print</span> <span class="s">"the expression is:"</span><span class="p">,</span> <span class="n">expression</span></pre></div>
</div>
<p>Keep in mind, <strong>this applies to Table and Column objects
too</strong>.</p>
<p>The rationale for the change is twofold:</p>
<ul class="simple">
<li>Comparisons of the form <tt class="docutils literal"><span class="pre">if</span> <span class="pre">c1</span> <span class="pre">==</span> <span class="pre">c2:</span> <span class="pre"><do</span> <span class="pre">something></span></tt>
can actually be written now</li>
<li>Support for correct hashing of <tt class="docutils literal"><span class="pre">ClauseElement</span></tt> objects
now works on alternate platforms, namely Jython. Up until
this point SQLAlchemy relied heavily on the specific
behavior of cPython in this regard (and still had
occasional problems with it).</li>
</ul>
</div>
<div class="section" id="stricter-executemany-behavior">
<h3>Stricter “executemany” Behavior<a class="headerlink" href="#stricter-executemany-behavior" title="Permalink to this headline">¶</a></h3>
<p>An “executemany” in SQLAlchemy corresponds to a call to
<tt class="docutils literal"><span class="pre">execute()</span></tt>, passing along a collection of bind parameter
sets:</p>
<div class="highlight-python"><div class="highlight"><pre><span class="n">connection</span><span class="o">.</span><span class="n">execute</span><span class="p">(</span><span class="n">table</span><span class="o">.</span><span class="n">insert</span><span class="p">(),</span> <span class="p">{</span><span class="s">'data'</span><span class="p">:</span><span class="s">'row1'</span><span class="p">},</span> <span class="p">{</span><span class="s">'data'</span><span class="p">:</span><span class="s">'row2'</span><span class="p">},</span> <span class="p">{</span><span class="s">'data'</span><span class="p">:</span><span class="s">'row3'</span><span class="p">})</span></pre></div>
</div>
<p>When the <tt class="docutils literal"><span class="pre">Connection</span></tt> object sends off the given
<tt class="docutils literal"><span class="pre">insert()</span></tt> construct for compilation, it passes to the
compiler the keynames present in the first set of binds
passed along to determine the construction of the
statement’s VALUES clause. Users familiar with this
construct will know that additional keys present in the
remaining dictionaries don’t have any impact. What’s
different now is that all subsequent dictionaries need to
include at least <em>every</em> key that is present in the first
dictionary. This means that a call like this no longer
works:</p>
<div class="highlight-python"><div class="highlight"><pre><span class="n">connection</span><span class="o">.</span><span class="n">execute</span><span class="p">(</span><span class="n">table</span><span class="o">.</span><span class="n">insert</span><span class="p">(),</span>
<span class="p">{</span><span class="s">'timestamp'</span><span class="p">:</span><span class="n">today</span><span class="p">,</span> <span class="s">'data'</span><span class="p">:</span><span class="s">'row1'</span><span class="p">},</span>
<span class="p">{</span><span class="s">'timestamp'</span><span class="p">:</span><span class="n">today</span><span class="p">,</span> <span class="s">'data'</span><span class="p">:</span><span class="s">'row2'</span><span class="p">},</span>
<span class="p">{</span><span class="s">'data'</span><span class="p">:</span><span class="s">'row3'</span><span class="p">})</span></pre></div>
</div>
<p>Because the third row does not specify the ‘timestamp’
column. Previous versions of SQLAlchemy would simply insert
NULL for these missing columns. However, if the
<tt class="docutils literal"><span class="pre">timestamp</span></tt> column in the above example contained a
Python-side default value or function, it would <em>not</em> be
used. This because the “executemany” operation is optimized
for maximum performance across huge numbers of parameter
sets, and does not attempt to evaluate Python-side defaults
for those missing keys. Because defaults are often
implemented either as SQL expressions which are embedded
inline with the INSERT statement, or are server side
expressions which again are triggered based on the structure
of the INSERT string, which by definition cannot fire off
conditionally based on each parameter set, it would be
inconsistent for Python side defaults to behave differently
vs. SQL/server side defaults. (SQL expression based
defaults are embedded inline as of the 0.5 series, again to
minimize the impact of huge numbers of parameter sets).</p>
<p>SQLAlchemy 0.6 therefore establishes predictable consistency
by forbidding any subsequent parameter sets from leaving any
fields blank. That way, there’s no more silent failure of
Python side default values and functions, which additionally
are allowed to remain consistent in their behavior versus
SQL and server side defaults.</p>
</div>
<div class="section" id="union-and-other-compound-constructs-parenthesize-consistently">
<h3>UNION and other “compound” constructs parenthesize consistently<a class="headerlink" href="#union-and-other-compound-constructs-parenthesize-consistently" title="Permalink to this headline">¶</a></h3>
<p>A rule that was designed to help SQLite has been removed,
that of the first compound element within another compound
(such as, a <tt class="docutils literal"><span class="pre">union()</span></tt> inside of an <tt class="docutils literal"><span class="pre">except_()</span></tt>) wouldn’t
be parenthesized. This is inconsistent and produces the
wrong results on Postgresql, which has precedence rules
regarding INTERSECTION, and its generally a surprise. When
using complex composites with SQLite, you now need to turn
the first element into a subquery (which is also compatible
on PG). A new example is in the SQL expression tutorial at
the end of
[<a class="reference external" href="http://www.sqlalchemy.org/docs/06/sqlexpression.html">http://www.sqlalchemy.org/docs/06/sqlexpression.html</a>
#unions-and-other-set-operations]. See <a class="reference external" href="http://www.sqlalchemy.org/trac/ticket/1665">#1665</a> and
r6690 for more background.</p>
</div>
</div>
<div class="section" id="c-extensions-for-result-fetching">
<h2>C Extensions for Result Fetching<a class="headerlink" href="#c-extensions-for-result-fetching" title="Permalink to this headline">¶</a></h2>
<p>The <tt class="docutils literal"><span class="pre">ResultProxy</span></tt> and related elements, including most
common “row processing” functions such as unicode
conversion, numerical/boolean conversions and date parsing,
have been re-implemented as optional C extensions for the
purposes of performance. This represents the beginning of
SQLAlchemy’s path to the “dark side” where we hope to
continue improving performance by reimplementing critical
sections in C. The extensions can be built by specifying
<tt class="docutils literal"><span class="pre">--with-cextensions</span></tt>, i.e. <tt class="docutils literal"><span class="pre">python</span> <span class="pre">setup.py</span> <span class="pre">--with-</span>
<span class="pre">cextensions</span> <span class="pre">install</span></tt>.</p>
<p>The extensions have the most dramatic impact on result
fetching using direct <tt class="docutils literal"><span class="pre">ResultProxy</span></tt> access, i.e. that
which is returned by <tt class="docutils literal"><span class="pre">engine.execute()</span></tt>,
<tt class="docutils literal"><span class="pre">connection.execute()</span></tt>, or <tt class="docutils literal"><span class="pre">session.execute()</span></tt>. Within
results returned by an ORM <tt class="docutils literal"><span class="pre">Query</span></tt> object, result fetching
is not as high a percentage of overhead, so ORM performance
improves more modestly, and mostly in the realm of fetching
large result sets. The performance improvements highly
depend on the dbapi in use and on the syntax used to access
the columns of each row (eg <tt class="docutils literal"><span class="pre">row['name']</span></tt> is much faster
than <tt class="docutils literal"><span class="pre">row.name</span></tt>). The current extensions have no impact
on the speed of inserts/updates/deletes, nor do they improve
the latency of SQL execution, that is, an application that
spends most of its time executing many statements with very
small result sets will not see much improvement.</p>
<p>Performance has been improved in 0.6 versus 0.5 regardless
of the extensions. A quick overview of what connecting and
fetching 50,000 rows looks like with SQLite, using mostly
direct SQLite access, a <tt class="docutils literal"><span class="pre">ResultProxy</span></tt>, and a simple mapped
ORM object:</p>
<div class="highlight-python"><pre>sqlite select/native: 0.260s
0.6 / C extension
sqlalchemy.sql select: 0.360s
sqlalchemy.orm fetch: 2.500s
0.6 / Pure Python
sqlalchemy.sql select: 0.600s
sqlalchemy.orm fetch: 3.000s
0.5 / Pure Python
sqlalchemy.sql select: 0.790s
sqlalchemy.orm fetch: 4.030s</pre>
</div>
<p>Above, the ORM fetches the rows 33% faster than 0.5 due to
in-python performance enhancements. With the C extensions
we get another 20%. However, <tt class="docutils literal"><span class="pre">ResultProxy</span></tt> fetches
improve by 67% with the C extension versus not. Other
tests report as much as a 200% speed improvement for some
scenarios, such as those where lots of string conversions
are occurring.</p>
</div>
<div class="section" id="new-schema-capabilities">
<h2>New Schema Capabilities<a class="headerlink" href="#new-schema-capabilities" title="Permalink to this headline">¶</a></h2>
<p>The <tt class="docutils literal"><span class="pre">sqlalchemy.schema</span></tt> package has received some long-
needed attention. The most visible change is the newly
expanded DDL system. In SQLAlchemy, it was possible since
version 0.5 to create custom DDL strings and associate them
with tables or metadata objects:</p>
<div class="highlight-python"><div class="highlight"><pre><span class="kn">from</span> <span class="nn">sqlalchemy.schema</span> <span class="kn">import</span> <span class="n">DDL</span>
<span class="n">DDL</span><span class="p">(</span><span class="s">'CREATE TRIGGER users_trigger ...'</span><span class="p">)</span><span class="o">.</span><span class="n">execute_at</span><span class="p">(</span><span class="s">'after-create'</span><span class="p">,</span> <span class="n">metadata</span><span class="p">)</span></pre></div>
</div>
<p>Now the full suite of DDL constructs are available under the
same system, including those for CREATE TABLE, ADD
CONSTRAINT, etc.:</p>
<div class="highlight-python"><div class="highlight"><pre><span class="kn">from</span> <span class="nn">sqlalchemy.schema</span> <span class="kn">import</span> <span class="n">Constraint</span><span class="p">,</span> <span class="n">AddConstraint</span>
<span class="n">AddContraint</span><span class="p">(</span><span class="n">CheckConstraint</span><span class="p">(</span><span class="s">"value > 5"</span><span class="p">))</span><span class="o">.</span><span class="n">execute_at</span><span class="p">(</span><span class="s">'after-create'</span><span class="p">,</span> <span class="n">mytable</span><span class="p">)</span></pre></div>
</div>
<p>Additionally, all the DDL objects are now regular
<tt class="docutils literal"><span class="pre">ClauseElement</span></tt> objects just like any other SQLAlchemy
expression object:</p>
<div class="highlight-python"><div class="highlight"><pre><span class="kn">from</span> <span class="nn">sqlalchemy.schema</span> <span class="kn">import</span> <span class="n">CreateTable</span>
<span class="n">create</span> <span class="o">=</span> <span class="n">CreateTable</span><span class="p">(</span><span class="n">mytable</span><span class="p">)</span>
<span class="c"># dumps the CREATE TABLE as a string</span>
<span class="k">print</span> <span class="n">create</span>
<span class="c"># executes the CREATE TABLE statement</span>
<span class="n">engine</span><span class="o">.</span><span class="n">execute</span><span class="p">(</span><span class="n">create</span><span class="p">)</span></pre></div>
</div>
<p>and using the <tt class="docutils literal"><span class="pre">sqlalchemy.ext.compiler</span></tt> extension you can
make your own:</p>
<div class="highlight-python"><div class="highlight"><pre><span class="kn">from</span> <span class="nn">sqlalchemy.schema</span> <span class="kn">import</span> <span class="n">DDLElement</span>
<span class="kn">from</span> <span class="nn">sqlalchemy.ext.compiler</span> <span class="kn">import</span> <span class="n">compiles</span>
<span class="k">class</span> <span class="nc">AlterColumn</span><span class="p">(</span><span class="n">DDLElement</span><span class="p">):</span>
<span class="k">def</span> <span class="nf">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">column</span><span class="p">,</span> <span class="n">cmd</span><span class="p">):</span>
<span class="bp">self</span><span class="o">.</span><span class="n">column</span> <span class="o">=</span> <span class="n">column</span>
<span class="bp">self</span><span class="o">.</span><span class="n">cmd</span> <span class="o">=</span> <span class="n">cmd</span>
<span class="nd">@compiles</span><span class="p">(</span><span class="n">AlterColumn</span><span class="p">)</span>
<span class="k">def</span> <span class="nf">visit_alter_column</span><span class="p">(</span><span class="n">element</span><span class="p">,</span> <span class="n">compiler</span><span class="p">,</span> <span class="o">**</span><span class="n">kw</span><span class="p">):</span>
<span class="k">return</span> <span class="s">"ALTER TABLE </span><span class="si">%s</span><span class="s"> ALTER COLUMN </span><span class="si">%s</span><span class="s"> </span><span class="si">%s</span><span class="s"> ..."</span> <span class="o">%</span> <span class="p">(</span>
<span class="n">element</span><span class="o">.</span><span class="n">column</span><span class="o">.</span><span class="n">table</span><span class="o">.</span><span class="n">name</span><span class="p">,</span>
<span class="n">element</span><span class="o">.</span><span class="n">column</span><span class="o">.</span><span class="n">name</span><span class="p">,</span>
<span class="n">element</span><span class="o">.</span><span class="n">cmd</span>
<span class="p">)</span>
<span class="n">engine</span><span class="o">.</span><span class="n">execute</span><span class="p">(</span><span class="n">AlterColumn</span><span class="p">(</span><span class="n">table</span><span class="o">.</span><span class="n">c</span><span class="o">.</span><span class="n">mycolumn</span><span class="p">,</span> <span class="s">"SET DEFAULT 'test'"</span><span class="p">))</span></pre></div>
</div>
<div class="section" id="deprecated-removed-schema-elements">
<h3>Deprecated/Removed Schema Elements<a class="headerlink" href="#deprecated-removed-schema-elements" title="Permalink to this headline">¶</a></h3>
<p>The schema package has also been greatly streamlined. Many
options and methods which were deprecated throughout 0.5
have been removed. Other little known accessors and methods
have also been removed.</p>
<ul class="simple">
<li>the “owner” keyword argument is removed from <tt class="docutils literal"><span class="pre">Table</span></tt>.
Use “schema” to represent any namespaces to be prepended
to the table name.</li>
<li>deprecated <tt class="docutils literal"><span class="pre">MetaData.connect()</span></tt> and
<tt class="docutils literal"><span class="pre">ThreadLocalMetaData.connect()</span></tt> have been removed - send
the “bind” attribute to bind a metadata.</li>
<li>deprecated metadata.table_iterator() method removed (use
sorted_tables)</li>
<li>the “metadata” argument is removed from
<tt class="docutils literal"><span class="pre">DefaultGenerator</span></tt> and subclasses, but remains locally
present on <tt class="docutils literal"><span class="pre">Sequence</span></tt>, which is a standalone construct
in DDL.</li>
<li>deprecated <tt class="docutils literal"><span class="pre">PassiveDefault</span></tt> - use <tt class="docutils literal"><span class="pre">DefaultClause</span></tt>.</li>
<li>Removed public mutability from <tt class="docutils literal"><span class="pre">Index</span></tt> and
<tt class="docutils literal"><span class="pre">Constraint</span></tt> objects:<ul>
<li><tt class="docutils literal"><span class="pre">ForeignKeyConstraint.append_element()</span></tt></li>
<li><tt class="docutils literal"><span class="pre">Index.append_column()</span></tt></li>
<li><tt class="docutils literal"><span class="pre">UniqueConstraint.append_column()</span></tt></li>
<li><tt class="docutils literal"><span class="pre">PrimaryKeyConstraint.add()</span></tt></li>
<li><tt class="docutils literal"><span class="pre">PrimaryKeyConstraint.remove()</span></tt></li>
</ul>
</li>
</ul>
<p>These should be constructed declaratively (i.e. in one
construction).</p>
<ul class="simple">
<li>Other removed things:<ul>
<li><tt class="docutils literal"><span class="pre">Table.key</span></tt> (no idea what this was for)</li>
<li><tt class="docutils literal"><span class="pre">Column.bind</span></tt> (get via column.table.bind)</li>
<li><tt class="docutils literal"><span class="pre">Column.metadata</span></tt> (get via column.table.metadata)</li>
<li><tt class="docutils literal"><span class="pre">Column.sequence</span></tt> (use column.default)</li>
</ul>
</li>
</ul>
</div>
<div class="section" id="other-behavioral-changes">
<h3>Other Behavioral Changes<a class="headerlink" href="#other-behavioral-changes" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><tt class="docutils literal"><span class="pre">UniqueConstraint</span></tt>, <tt class="docutils literal"><span class="pre">Index</span></tt>, <tt class="docutils literal"><span class="pre">PrimaryKeyConstraint</span></tt>
all accept lists of column names or column objects as
arguments.</li>
<li>The <tt class="docutils literal"><span class="pre">use_alter</span></tt> flag on <tt class="docutils literal"><span class="pre">ForeignKey</span></tt> is now a shortcut
option for operations that can be hand-constructed using
the <tt class="docutils literal"><span class="pre">DDL()</span></tt> event system. A side effect of this refactor
is that <tt class="docutils literal"><span class="pre">ForeignKeyConstraint</span></tt> objects with
<tt class="docutils literal"><span class="pre">use_alter=True</span></tt> will <em>not</em> be emitted on SQLite, which
does not support ALTER for foreign keys. This has no
effect on SQLite’s behavior since SQLite does not actually
honor FOREIGN KEY constraints.</li>
<li><tt class="docutils literal"><span class="pre">Table.primary_key</span></tt> is not assignable - use
<tt class="docutils literal"><span class="pre">table.append_constraint(PrimaryKeyConstraint(...))</span></tt></li>
<li>A <tt class="docutils literal"><span class="pre">Column</span></tt> definition with a <tt class="docutils literal"><span class="pre">ForeignKey</span></tt> and no type,
e.g. <tt class="docutils literal"><span class="pre">Column(name,</span> <span class="pre">ForeignKey(sometable.c.somecol))</span></tt>
used to get the type of the referenced column. Now support
for that automatic type inference is partial and may not
work in all cases.</li>
</ul>
</div>
</div>
<div class="section" id="logging-opened-up">
<h2>Logging opened up<a class="headerlink" href="#logging-opened-up" title="Permalink to this headline">¶</a></h2>
<p>At the expense of a few extra method calls here and there,
you can set log levels for INFO and DEBUG after an engine,
pool, or mapper has been created, and logging will commence.
The <tt class="docutils literal"><span class="pre">isEnabledFor(INFO)</span></tt> method is now called
per-<tt class="docutils literal"><span class="pre">Connection</span></tt> and <tt class="docutils literal"><span class="pre">isEnabledFor(DEBUG)</span></tt>
per-<tt class="docutils literal"><span class="pre">ResultProxy</span></tt> if already enabled on the parent
connection. Pool logging sends to <tt class="docutils literal"><span class="pre">log.info()</span></tt> and
<tt class="docutils literal"><span class="pre">log.debug()</span></tt> with no check - note that pool
checkout/checkin is typically once per transaction.</p>
</div>
<div class="section" id="reflection-inspector-api">
<h2>Reflection/Inspector API<a class="headerlink" href="#reflection-inspector-api" title="Permalink to this headline">¶</a></h2>
<p>The reflection system, which allows reflection of table
columns via <tt class="docutils literal"><span class="pre">Table('sometable',</span> <span class="pre">metadata,</span> <span class="pre">autoload=True)</span></tt>
has been opened up into its own fine-grained API, which
allows direct inspection of database elements such as
tables, columns, constraints, indexes, and more. This API
expresses return values as simple lists of strings,
dictionaries, and <tt class="docutils literal"><span class="pre">TypeEngine</span></tt> objects. The internals of
<tt class="docutils literal"><span class="pre">autoload=True</span></tt> now build upon this system such that the
translation of raw database information into
<tt class="docutils literal"><span class="pre">sqlalchemy.schema</span></tt> constructs is centralized and the
contract of individual dialects greatly simplified, vastly
reducing bugs and inconsistencies across different backends.</p>
<p>To use an inspector:</p>
<div class="highlight-python"><div class="highlight"><pre><span class="kn">from</span> <span class="nn">sqlalchemy.engine.reflection</span> <span class="kn">import</span> <span class="n">Inspector</span>
<span class="n">insp</span> <span class="o">=</span> <span class="n">Inspector</span><span class="o">.</span><span class="n">from_engine</span><span class="p">(</span><span class="n">my_engine</span><span class="p">)</span>
<span class="k">print</span> <span class="n">insp</span><span class="o">.</span><span class="n">get_schema_names</span><span class="p">()</span></pre></div>
</div>
<p>the <tt class="docutils literal"><span class="pre">from_engine()</span></tt> method will in some cases provide a
backend-specific inspector with additional capabilities,
such as that of Postgresql which provides a
<tt class="docutils literal"><span class="pre">get_table_oid()</span></tt> method:</p>
<div class="highlight-python"><div class="highlight"><pre><span class="n">my_engine</span> <span class="o">=</span> <span class="n">create_engine</span><span class="p">(</span><span class="s">'postgresql://...'</span><span class="p">)</span>
<span class="n">pg_insp</span> <span class="o">=</span> <span class="n">Inspector</span><span class="o">.</span><span class="n">from_engine</span><span class="p">(</span><span class="n">my_engine</span><span class="p">)</span>
<span class="k">print</span> <span class="n">pg_insp</span><span class="o">.</span><span class="n">get_table_oid</span><span class="p">(</span><span class="s">'my_table'</span><span class="p">)</span></pre></div>
</div>
</div>
<div class="section" id="returning-support">
<h2>RETURNING Support<a class="headerlink" href="#returning-support" title="Permalink to this headline">¶</a></h2>
<p>The <tt class="docutils literal"><span class="pre">insert()</span></tt>, <tt class="docutils literal"><span class="pre">update()</span></tt> and <tt class="docutils literal"><span class="pre">delete()</span></tt> constructs
now support a <tt class="docutils literal"><span class="pre">returning()</span></tt> method, which corresponds to
the SQL RETURNING clause as supported by Postgresql, Oracle,
MS-SQL, and Firebird. It is not supported for any other
backend at this time.</p>
<p>Given a list of column expressions in the same manner as
that of a <tt class="docutils literal"><span class="pre">select()</span></tt> construct, the values of these
columns will be returned as a regular result set:</p>
<div class="highlight-python"><div class="highlight"><pre><span class="n">result</span> <span class="o">=</span> <span class="n">connection</span><span class="o">.</span><span class="n">execute</span><span class="p">(</span>
<span class="n">table</span><span class="o">.</span><span class="n">insert</span><span class="p">()</span><span class="o">.</span><span class="n">values</span><span class="p">(</span><span class="n">data</span><span class="o">=</span><span class="s">'some data'</span><span class="p">)</span><span class="o">.</span><span class="n">returning</span><span class="p">(</span><span class="n">table</span><span class="o">.</span><span class="n">c</span><span class="o">.</span><span class="n">id</span><span class="p">,</span> <span class="n">table</span><span class="o">.</span><span class="n">c</span><span class="o">.</span><span class="n">timestamp</span><span class="p">)</span>
<span class="p">)</span>
<span class="n">row</span> <span class="o">=</span> <span class="n">result</span><span class="o">.</span><span class="n">first</span><span class="p">()</span>
<span class="k">print</span> <span class="s">"ID:"</span><span class="p">,</span> <span class="n">row</span><span class="p">[</span><span class="s">'id'</span><span class="p">],</span> <span class="s">"Timestamp:"</span><span class="p">,</span> <span class="n">row</span><span class="p">[</span><span class="s">'timestamp'</span><span class="p">]</span></pre></div>
</div>
<p>The implementation of RETURNING across the four supported
backends varies wildly, in the case of Oracle requiring an
intricate usage of OUT parameters which are re-routed into a
“mock” result set, and in the case of MS-SQL using an
awkward SQL syntax. The usage of RETURNING is subject to
limitations:</p>
<ul class="simple">
<li>it does not work for any “executemany()” style of
execution. This is a limitation of all supported DBAPIs.</li>
<li>Some backends, such as Oracle, only support RETURNING that
returns a single row - this includes UPDATE and DELETE
statements, meaning the update() or delete() construct
must match only a single row, or an error is raised (by
Oracle, not SQLAlchemy).</li>
</ul>
<p>RETURNING is also used automatically by SQLAlchemy, when
available and when not otherwise specified by an explicit
<tt class="docutils literal"><span class="pre">returning()</span></tt> call, to fetch the value of newly generated
primary key values for single-row INSERT statements. This
means there’s no more “SELECT nextval(sequence)” pre-
execution for insert statements where the primary key value
is required. Truth be told, implicit RETURNING feature
does incur more method overhead than the old “select
nextval()” system, which used a quick and dirty
cursor.execute() to get at the sequence value, and in the
case of Oracle requires additional binding of out
parameters. So if method/protocol overhead is proving to be
more expensive than additional database round trips, the
feature can be disabled by specifying
<tt class="docutils literal"><span class="pre">implicit_returning=False</span></tt> to <tt class="docutils literal"><span class="pre">create_engine()</span></tt>.</p>
</div>
<div class="section" id="type-system-changes">
<h2>Type System Changes<a class="headerlink" href="#type-system-changes" title="Permalink to this headline">¶</a></h2>
<div class="section" id="new-archicture">
<h3>New Archicture<a class="headerlink" href="#new-archicture" title="Permalink to this headline">¶</a></h3>
<p>The type system has been completely reworked behind the
scenes to provide two goals:</p>
<ul class="simple">
<li>Separate the handling of bind parameters and result row
values, typically a DBAPI requirement, from the SQL
specification of the type itself, which is a database
requirement. This is consistent with the overall dialect
refactor that separates database SQL behavior from DBAPI.</li>
<li>Establish a clear and consistent contract for generating
DDL from a <tt class="docutils literal"><span class="pre">TypeEngine</span></tt> object and for constructing
<tt class="docutils literal"><span class="pre">TypeEngine</span></tt> objects based on column reflection.</li>
</ul>
<p>Highlights of these changes include:</p>
<ul class="simple">
<li>The construction of types within dialects has been totally
overhauled. Dialects now define publically available types
as UPPERCASE names exclusively, and internal
implementation types using underscore identifiers (i.e.
are private). The system by which types are expressed in
SQL and DDL has been moved to the compiler system. This
has the effect that there are much fewer type objects
within most dialects. A detailed document on this
architecture for dialect authors is in [source:/lib/sqlalc
hemy/dialects/type_migration_guidelines.txt].</li>
<li>Reflection of types now returns the exact UPPERCASE type
within types.py, or the UPPERCASE type within the dialect
itself if the type is not a standard SQL type. This means
reflection now returns more accurate information about
reflected types.</li>
<li>User defined types that subclass <tt class="docutils literal"><span class="pre">TypeEngine</span></tt> and wish
to provide <tt class="docutils literal"><span class="pre">get_col_spec()</span></tt> should now subclass
<tt class="docutils literal"><span class="pre">UserDefinedType</span></tt>.</li>
<li>The <tt class="docutils literal"><span class="pre">result_processor()</span></tt> method on all type classes now
accepts an additional argument <tt class="docutils literal"><span class="pre">coltype</span></tt>. This is the
DBAPI type object attached to cursor.description, and
should be used when applicable to make better decisions on
what kind of result-processing callable should be
returned. Ideally result processor functions would never
need to use <tt class="docutils literal"><span class="pre">isinstance()</span></tt>, which is an expensive call
at this level.</li>
</ul>
</div>
<div class="section" id="native-unicode-mode">
<h3>Native Unicode Mode<a class="headerlink" href="#native-unicode-mode" title="Permalink to this headline">¶</a></h3>
<p>As more DBAPIs support returning Python unicode objects
directly, the base dialect now performs a check upon the
first connection which establishes whether or not the DBAPI
returns a Python unicode object for a basic select of a
VARCHAR value. If so, the <tt class="docutils literal"><span class="pre">String</span></tt> type and all
subclasses (i.e. <tt class="docutils literal"><span class="pre">Text</span></tt>, <tt class="docutils literal"><span class="pre">Unicode</span></tt>, etc.) will skip the
“unicode” check/conversion step when result rows are
received. This offers a dramatic performance increase for
large result sets. The “unicode mode” currently is known to
work with:</p>
<ul class="simple">
<li>sqlite3 / pysqlite</li>
<li>psycopg2 - SQLA 0.6 now uses the “UNICODE” type extension
by default on each psycopg2 connection object</li>
<li>pg8000</li>
<li>cx_oracle (we use an output processor - nice feature !)</li>
</ul>
<p>Other types may choose to disable unicode processing as
needed, such as the <tt class="docutils literal"><span class="pre">NVARCHAR</span></tt> type when used with MS-SQL.</p>
<p>In particular, if porting an application based on a DBAPI
that formerly returned non-unicode strings, the “native
unicode” mode has a plainly different default behavior -
columns that are declared as <tt class="docutils literal"><span class="pre">String</span></tt> or <tt class="docutils literal"><span class="pre">VARCHAR</span></tt> now
return unicode by default whereas they would return strings
before. This can break code which expects non-unicode
strings. The psycopg2 “native unicode” mode can be
disabled by passing <tt class="docutils literal"><span class="pre">use_native_unicode=False</span></tt> to
<tt class="docutils literal"><span class="pre">create_engine()</span></tt>.</p>
<p>A more general solution for string columns that explicitly
do not want a unicode object is to use a <tt class="docutils literal"><span class="pre">TypeDecorator</span></tt>
that converts unicode back to utf-8, or whatever is desired:</p>
<div class="highlight-python"><div class="highlight"><pre><span class="k">class</span> <span class="nc">UTF8Encoded</span><span class="p">(</span><span class="n">TypeDecorator</span><span class="p">):</span>
<span class="sd">"""Unicode type which coerces to utf-8."""</span>
<span class="n">impl</span> <span class="o">=</span> <span class="n">sa</span><span class="o">.</span><span class="n">VARCHAR</span>
<span class="k">def</span> <span class="nf">process_result_value</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">value</span><span class="p">,</span> <span class="n">dialect</span><span class="p">):</span>
<span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">value</span><span class="p">,</span> <span class="nb">unicode</span><span class="p">):</span>
<span class="n">value</span> <span class="o">=</span> <span class="n">value</span><span class="o">.</span><span class="n">encode</span><span class="p">(</span><span class="s">'utf-8'</span><span class="p">)</span>
<span class="k">return</span> <span class="n">value</span></pre></div>
</div>
<p>Note that the <tt class="docutils literal"><span class="pre">assert_unicode</span></tt> flag is now deprecated.
SQLAlchemy allows the DBAPI and backend database in use to
handle Unicode parameters when available, and does not add
operational overhead by checking the incoming type; modern
systems like sqlite and Postgresql will raise an encoding
error on their end if invalid data is passed. In those
cases where SQLAlchemy does need to coerce a bind parameter
from Python Unicode to an encoded string, or when the
Unicode type is used explicitly, a warning is raised if the
object is a bytestring. This warning can be suppressed or
converted to an exception using the Python warnings filter
documented at: <a class="reference external" href="http://docs.python.org/library/warnings.html">http://docs.python.org/library/warnings.html</a></p>
</div>
<div class="section" id="generic-enum-type">
<h3>Generic Enum Type<a class="headerlink" href="#generic-enum-type" title="Permalink to this headline">¶</a></h3>
<p>We now have an <tt class="docutils literal"><span class="pre">Enum</span></tt> in the <tt class="docutils literal"><span class="pre">types</span></tt> module. This is a
string type that is given a collection of “labels” which
constrain the possible values given to those labels. By
default, this type generates a <tt class="docutils literal"><span class="pre">VARCHAR</span></tt> using the size of
the largest label, and applies a CHECK constraint to the
table within the CREATE TABLE statement. When using MySQL,
the type by default uses MySQL’s ENUM type, and when using
Postgresql the type will generate a user defined type using
<tt class="docutils literal"><span class="pre">CREATE</span> <span class="pre">TYPE</span> <span class="pre"><mytype></span> <span class="pre">AS</span> <span class="pre">ENUM</span></tt>. In order to create the
type using Postgresql, the <tt class="docutils literal"><span class="pre">name</span></tt> parameter must be
specified to the constructor. The type also accepts a
<tt class="docutils literal"><span class="pre">native_enum=False</span></tt> option which will issue the
VARCHAR/CHECK strategy for all databases. Note that
Postgresql ENUM types currently don’t work with pg8000 or
zxjdbc.</p>
</div>
<div class="section" id="reflection-returns-dialect-specific-types">
<h3>Reflection Returns Dialect-Specific Types<a class="headerlink" href="#reflection-returns-dialect-specific-types" title="Permalink to this headline">¶</a></h3>
<p>Reflection now returns the most specific type possible from
the database. That is, if you create a table using
<tt class="docutils literal"><span class="pre">String</span></tt>, then reflect it back, the reflected column will
likely be <tt class="docutils literal"><span class="pre">VARCHAR</span></tt>. For dialects that support a more
specific form of the type, that’s what you’ll get. So a
<tt class="docutils literal"><span class="pre">Text</span></tt> type would come back as <tt class="docutils literal"><span class="pre">oracle.CLOB</span></tt> on Oracle,
a <tt class="docutils literal"><span class="pre">LargeBinary</span></tt> might be an <tt class="docutils literal"><span class="pre">mysql.MEDIUMBLOB</span></tt> etc. The
obvious advantage here is that reflection preserves as much
information possible from what the database had to say.</p>
<p>Some applications that deal heavily in table metadata may
wish to compare types across reflected tables and/or non-
reflected tables. There’s a semi-private accessor available
on <tt class="docutils literal"><span class="pre">TypeEngine</span></tt> called <tt class="docutils literal"><span class="pre">_type_affinity</span></tt> and an
associated comparison helper <tt class="docutils literal"><span class="pre">_compare_type_affinity</span></tt>.
This accessor returns the “generic” <tt class="docutils literal"><span class="pre">types</span></tt> class which
the type corresponds to:</p>
<div class="highlight-python"><div class="highlight"><pre><span class="gp">>>> </span><span class="n">String</span><span class="p">(</span><span class="mi">50</span><span class="p">)</span><span class="o">.</span><span class="n">_compare_type_affinity</span><span class="p">(</span><span class="n">postgresql</span><span class="o">.</span><span class="n">VARCHAR</span><span class="p">(</span><span class="mi">50</span><span class="p">))</span>
<span class="go">True</span>
<span class="gp">>>> </span><span class="n">Integer</span><span class="p">()</span><span class="o">.</span><span class="n">_compare_type_affinity</span><span class="p">(</span><span class="n">mysql</span><span class="o">.</span><span class="n">REAL</span><span class="p">)</span>
<span class="go">False</span></pre></div>
</div>
</div>
<div class="section" id="miscellaneous-api-changes">
<h3>Miscellaneous API Changes<a class="headerlink" href="#miscellaneous-api-changes" title="Permalink to this headline">¶</a></h3>
<p>The usual “generic” types are still the general system in
use, i.e. <tt class="docutils literal"><span class="pre">String</span></tt>, <tt class="docutils literal"><span class="pre">Float</span></tt>, <tt class="docutils literal"><span class="pre">DateTime</span></tt>. There’s a
few changes there:</p>
<ul class="simple">
<li>Types no longer make any guesses as to default parameters.
In particular, <tt class="docutils literal"><span class="pre">Numeric</span></tt>, <tt class="docutils literal"><span class="pre">Float</span></tt>, as well as
subclasses NUMERIC, FLOAT, DECIMAL don’t generate any
length or scale unless specified. This also continues to
include the controversial <tt class="docutils literal"><span class="pre">String</span></tt> and <tt class="docutils literal"><span class="pre">VARCHAR</span></tt> types
(although MySQL dialect will pre-emptively raise when
asked to render VARCHAR with no length). No defaults are
assumed, and if they are used in a CREATE TABLE statement,
an error will be raised if the underlying database does
not allow non-lengthed versions of these types.</li>
<li>the <tt class="docutils literal"><span class="pre">Binary</span></tt> type has been renamed to <tt class="docutils literal"><span class="pre">LargeBinary</span></tt>,
for BLOB/BYTEA/similar types. For <tt class="docutils literal"><span class="pre">BINARY</span></tt> and
<tt class="docutils literal"><span class="pre">VARBINARY</span></tt>, those are present directly as
<tt class="docutils literal"><span class="pre">types.BINARY</span></tt>, <tt class="docutils literal"><span class="pre">types.VARBINARY</span></tt>, as well as in the
MySQL and MS-SQL dialects.</li>
<li><tt class="docutils literal"><span class="pre">PickleType</span></tt> now uses == for comparison of values when
mutable=True, unless the “comparator” argument with a
comparison function is specified to the type. If you are
pickling a custom object you should implement an
<tt class="docutils literal"><span class="pre">__eq__()</span></tt> method so that value-based comparisons are
accurate.</li>
<li>The default “precision” and “scale” arguments of Numeric
and Float have been removed and now default to None.
NUMERIC and FLOAT will be rendered with no numeric
arguments by default unless these values are provided.</li>
<li>DATE, TIME and DATETIME types on SQLite can now take
optional “storage_format” and “regexp” argument.
“storage_format” can be used to store those types using a
custom string format. “regexp” allows to use a custom
regular expression to match string values from the
database.</li>
<li><tt class="docutils literal"><span class="pre">__legacy_microseconds__</span></tt> on SQLite <tt class="docutils literal"><span class="pre">Time</span></tt> and
<tt class="docutils literal"><span class="pre">DateTime</span></tt> types is not supported anymore. You should
use the new “storage_format” argument instead.</li>
<li><tt class="docutils literal"><span class="pre">DateTime</span></tt> types on SQLite now use by a default a
stricter regular expression to match strings from the
database. Use the new “regexp” argument if you are using
data stored in a legacy format.</li>
</ul>
</div>
</div>
<div class="section" id="orm-changes">
<h2>ORM Changes<a class="headerlink" href="#orm-changes" title="Permalink to this headline">¶</a></h2>
<p>Upgrading an ORM application from 0.5 to 0.6 should require
little to no changes, as the ORM’s behavior remains almost
identical. There are some default argument and name
changes, and some loading behaviors have been improved.</p>
<div class="section" id="new-unit-of-work">
<h3>New Unit of Work<a class="headerlink" href="#new-unit-of-work" title="Permalink to this headline">¶</a></h3>
<p>The internals for the unit of work, primarily
<tt class="docutils literal"><span class="pre">topological.py</span></tt> and <tt class="docutils literal"><span class="pre">unitofwork.py</span></tt>, have been
completely rewritten and are vastly simplified. This
should have no impact on usage, as all existing behavior
during flush has been maintained exactly (or at least, as
far as it is exercised by our testsuite and the handful of
production environments which have tested it heavily). The
performance of flush() now uses 20-30% fewer method calls
and should also use less memory. The intent and flow of the
source code should now be reasonably easy to follow, and the
architecture of the flush is fairly open-ended at this
point, creating room for potential new areas of
sophistication. The flush process no longer has any
reliance on recursion so flush plans of arbitrary size and
complexity can be flushed. Additionally, the mapper’s
“save” process, which issues INSERT and UPDATE statements,
now caches the “compiled” form of the two statements so that
callcounts are further dramatically reduced with very large
flushes.</p>
<p>Any changes in behavior observed with flush versus earlier
versions of 0.6 or 0.5 should be reported to us ASAP - we’ll
make sure no functionality is lost.</p>
</div>
<div class="section" id="changes-to-query-update-and-query-delete">
<h3>Changes to <tt class="docutils literal"><span class="pre">query.update()</span></tt> and <tt class="docutils literal"><span class="pre">query.delete()</span></tt><a class="headerlink" href="#changes-to-query-update-and-query-delete" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li>the ‘expire’ option on query.update() has been renamed to
‘fetch’, thus matching that of query.delete()</li>
<li><tt class="docutils literal"><span class="pre">query.update()</span></tt> and <tt class="docutils literal"><span class="pre">query.delete()</span></tt> both default to
‘evaluate’ for the synchronize strategy.</li>
<li>the ‘synchronize’ strategy for update() and delete()
raises an error on failure. There is no implicit fallback
onto “fetch”. Failure of evaluation is based on the
structure of criteria, so success/failure is deterministic
based on code structure.</li>
</ul>
</div>
<div class="section" id="relation-is-officially-named-relationship">
<h3><tt class="docutils literal"><span class="pre">relation()</span></tt> is officially named <tt class="docutils literal"><span class="pre">relationship()</span></tt><a class="headerlink" href="#relation-is-officially-named-relationship" title="Permalink to this headline">¶</a></h3>
<p>This to solve the long running issue that “relation” means a
“table or derived table” in relational algebra terms. The
<tt class="docutils literal"><span class="pre">relation()</span></tt> name, which is less typing, will hang around
for the foreseeable future so this change should be entirely
painless.</p>
</div>
<div class="section" id="subquery-eager-loading">
<h3>Subquery eager loading<a class="headerlink" href="#subquery-eager-loading" title="Permalink to this headline">¶</a></h3>
<p>A new kind of eager loading is added called “subquery”
loading. This is a load that emits a second SQL query
immediately after the first which loads full collections for
all the parents in the first query, joining upwards to the
parent using INNER JOIN. Subquery loading is used simlarly
to the current joined-eager loading, using the
<tt class="docutils literal"><span class="pre">`subqueryload()``</span></tt> and <tt class="docutils literal"><span class="pre">``subqueryload_all()``</span></tt> options
as well as the <tt class="docutils literal"><span class="pre">``lazy='subquery'``</span></tt> setting on
<tt class="docutils literal"><span class="pre">``relationship()`</span></tt>. The subquery load is usually much
more efficient for loading many larger collections as it
uses INNER JOIN unconditionally and also doesn’t re-load
parent rows.</p>
</div>
<div class="section" id="eagerload-eagerload-all-is-now-joinedload-joinedload-all">
<h3><tt class="docutils literal"><span class="pre">`eagerload()``</span></tt>, <tt class="docutils literal"><span class="pre">``eagerload_all()``</span></tt> is now <tt class="docutils literal"><span class="pre">``joinedload()``</span></tt>, <tt class="docutils literal"><span class="pre">``joinedload_all()`</span></tt><a class="headerlink" href="#eagerload-eagerload-all-is-now-joinedload-joinedload-all" title="Permalink to this headline">¶</a></h3>
<p>To make room for the new subquery load feature, the existing
<tt class="docutils literal"><span class="pre">`eagerload()``</span></tt>/<tt class="docutils literal"><span class="pre">``eagerload_all()``</span></tt> options are now
superseded by <tt class="docutils literal"><span class="pre">``joinedload()``</span></tt> and
<tt class="docutils literal"><span class="pre">``joinedload_all()``</span></tt>. The old names will hang around
for the foreseeable future just like <tt class="docutils literal"><span class="pre">``relation()`</span></tt>.</p>
</div>
<div class="section" id="lazy-false-none-true-dynamic-now-accepts-lazy-noload-joined-subquery-select-dynamic">
<h3><tt class="docutils literal"><span class="pre">`lazy=False|None|True|'dynamic'``</span></tt> now accepts <tt class="docutils literal"><span class="pre">``lazy='noload'|'joined'|'subquery'|'select'|'dynamic'`</span></tt><a class="headerlink" href="#lazy-false-none-true-dynamic-now-accepts-lazy-noload-joined-subquery-select-dynamic" title="Permalink to this headline">¶</a></h3>
<p>Continuing on the theme of loader strategies opened up, the
standard keywords for the <tt class="docutils literal"><span class="pre">`lazy``</span></tt> option on
<tt class="docutils literal"><span class="pre">``relationship()``</span></tt> are now <tt class="docutils literal"><span class="pre">``select``</span></tt> for lazy
loading (via a SELECT issued on attribute access),
<tt class="docutils literal"><span class="pre">``joined``</span></tt> for joined-eager loading, <tt class="docutils literal"><span class="pre">``subquery``</span></tt>
for subquery-eager loading, <tt class="docutils literal"><span class="pre">``noload``</span></tt> for no loading
should occur, and <tt class="docutils literal"><span class="pre">``dynamic``</span></tt> for a “dynamic”
relationship. The old <tt class="docutils literal"><span class="pre">``True``</span></tt>, <tt class="docutils literal"><span class="pre">``False``</span></tt>,
<tt class="docutils literal"><span class="pre">``None`</span></tt> arguments are still accepted with the identical
behavior as before.</p>
</div>
<div class="section" id="innerjoin-true-on-relation-joinedload">
<h3>innerjoin=True on relation, joinedload<a class="headerlink" href="#innerjoin-true-on-relation-joinedload" title="Permalink to this headline">¶</a></h3>
<p>Joined-eagerly loaded scalars and collections can now be
instructed to use INNER JOIN instead of OUTER JOIN. On
Postgresql this is observed to provide a 300-600% speedup on
some queries. Set this flag for any many-to-one which is
on a NOT NULLable foreign key, and similarly for any
collection where related items are guaranteed to exist.</p>
<p>At mapper level:</p>
<div class="highlight-python"><div class="highlight"><pre><span class="n">mapper</span><span class="p">(</span><span class="n">Child</span><span class="p">,</span> <span class="n">child</span><span class="p">)</span>
<span class="n">mapper</span><span class="p">(</span><span class="n">Parent</span><span class="p">,</span> <span class="n">parent</span><span class="p">,</span> <span class="n">properties</span><span class="o">=</span><span class="p">{</span>
<span class="s">'child'</span><span class="p">:</span><span class="n">relationship</span><span class="p">(</span><span class="n">Child</span><span class="p">,</span> <span class="n">lazy</span><span class="o">=</span><span class="s">'joined'</span><span class="p">,</span> <span class="n">innerjoin</span><span class="o">=</span><span class="bp">True</span><span class="p">)</span>
<span class="p">})</span></pre></div>
</div>
<p>At query time level:</p>
<div class="highlight-python"><div class="highlight"><pre><span class="n">session</span><span class="o">.</span><span class="n">query</span><span class="p">(</span><span class="n">Parent</span><span class="p">)</span><span class="o">.</span><span class="n">options</span><span class="p">(</span><span class="n">joinedload</span><span class="p">(</span><span class="n">Parent</span><span class="o">.</span><span class="n">child</span><span class="p">,</span> <span class="n">innerjoin</span><span class="o">=</span><span class="bp">True</span><span class="p">))</span><span class="o">.</span><span class="n">all</span><span class="p">()</span></pre></div>
</div>
<p>The <tt class="docutils literal"><span class="pre">innerjoin=True</span></tt> flag at the <tt class="docutils literal"><span class="pre">relationship()</span></tt> level
will also take effect for any <tt class="docutils literal"><span class="pre">joinedload()</span></tt> option which
does not override the value.</p>
</div>
<div class="section" id="many-to-one-enhancements">
<h3>Many-to-one Enhancements<a class="headerlink" href="#many-to-one-enhancements" title="Permalink to this headline">¶</a></h3>
<ul>
<li><p class="first">many-to-one relations now fire off a lazyload in fewer
cases, including in most cases will not fetch the “old”
value when a new one is replaced.</p>
</li>
<li><p class="first">many-to-one relation to a joined-table subclass now uses
get() for a simple load (known as the “use_get”
condition), i.e. <tt class="docutils literal"><span class="pre">Related</span></tt>->``Sub(Base)``, without the
need to redefine the primaryjoin condition in terms of the
base table. [ticket:1186]</p>
</li>
<li><p class="first">specifying a foreign key with a declarative column, i.e.
<tt class="docutils literal"><span class="pre">ForeignKey(MyRelatedClass.id)</span></tt> doesn’t break the
“use_get” condition from taking place [ticket:1492]</p>
</li>
<li><p class="first">relationship(), joinedload(), and joinedload_all() now
feature an option called “innerjoin”. Specify <tt class="docutils literal"><span class="pre">True</span></tt> or
<tt class="docutils literal"><span class="pre">False</span></tt> to control whether an eager join is constructed
as an INNER or OUTER join. Default is <tt class="docutils literal"><span class="pre">False</span></tt> as always.
The mapper options will override whichever setting is
specified on relationship(). Should generally be set for
many-to-one, not nullable foreign key relations to allow
improved join performance. [ticket:1544]</p>
</li>
<li><p class="first">the behavior of joined eager loading such that the main
query is wrapped in a subquery when LIMIT/OFFSET are
present now makes an exception for the case when all eager
loads are many-to-one joins. In those cases, the eager
joins are against the parent table directly along with the
limit/offset without the extra overhead of a subquery,
since a many-to-one join does not add rows to the result.</p>
<p>For example, in 0.5 this query:</p>
<div class="highlight-python"><div class="highlight"><pre><span class="n">session</span><span class="o">.</span><span class="n">query</span><span class="p">(</span><span class="n">Address</span><span class="p">)</span><span class="o">.</span><span class="n">options</span><span class="p">(</span><span class="n">eagerload</span><span class="p">(</span><span class="n">Address</span><span class="o">.</span><span class="n">user</span><span class="p">))</span><span class="o">.</span><span class="n">limit</span><span class="p">(</span><span class="mi">10</span><span class="p">)</span></pre></div>
</div>
<p>would produce SQL like:</p>
<div class="highlight-python"><pre>SELECT * FROM
(SELECT * FROM addresses LIMIT 10) AS anon_1
LEFT OUTER JOIN users AS users_1 ON users_1.id = anon_1.addresses_user_id</pre>
</div>
<p>This because the presence of any eager loaders suggests
that some or all of them may relate to multi-row
collections, which would necessitate wrapping any kind of
rowcount-sensitive modifiers like LIMIT inside of a
subquery.</p>
<p>In 0.6, that logic is more sensitive and can detect if all
eager loaders represent many-to-ones, in which case the
eager joins don’t affect the rowcount:</p>
<div class="highlight-python"><pre>SELECT * FROM addresses LEFT OUTER JOIN users AS users_1 ON users_1.id = addresses.user_id LIMIT 10</pre>
</div>
</li>
</ul>
</div>
<div class="section" id="mutable-primary-keys-with-joined-table-inheritance">
<h3>Mutable Primary Keys with Joined Table Inheritance<a class="headerlink" href="#mutable-primary-keys-with-joined-table-inheritance" title="Permalink to this headline">¶</a></h3>
<p>A joined table inheritance config where the child table has
a PK that foreign keys to the parent PK can now be updated
on a CASCADE-capable database like Postgresql.
<tt class="docutils literal"><span class="pre">mapper()</span></tt> now has an option <tt class="docutils literal"><span class="pre">passive_updates=True</span></tt>
which indicates this foreign key is updated automatically.
If on a non-cascading database like SQLite or MySQL/MyISAM,
set this flag to <tt class="docutils literal"><span class="pre">False</span></tt>. A future feature enhancement
will try to get this flag to be auto-configuring based on
dialect/table style in use.</p>
</div>
<div class="section" id="beaker-caching">
<h3>Beaker Caching<a class="headerlink" href="#beaker-caching" title="Permalink to this headline">¶</a></h3>
<p>A promising new example of Beaker integration is in
<tt class="docutils literal"><span class="pre">examples/beaker_caching</span></tt>. This is a straightforward
recipe which applies a Beaker cache within the result-
generation engine of <tt class="docutils literal"><span class="pre">Query</span></tt>. Cache parameters are
provided via <tt class="docutils literal"><span class="pre">query.options()</span></tt>, and allows full control
over the contents of the cache. SQLAlchemy 0.6 includes
improvements to the <tt class="docutils literal"><span class="pre">Session.merge()</span></tt> method to support
this and similar recipes, as well as to provide
significantly improved performance in most scenarios.</p>
</div>
<div class="section" id="other-changes">
<h3>Other Changes<a class="headerlink" href="#other-changes" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li>the “row tuple” object returned by <tt class="docutils literal"><span class="pre">Query</span></tt> when multiple
column/entities are selected is now picklable as well as
higher performing.</li>
<li><tt class="docutils literal"><span class="pre">query.join()</span></tt> has been reworked to provide more
consistent behavior and more flexibility (includes
[ticket:1537])</li>
<li><tt class="docutils literal"><span class="pre">query.select_from()</span></tt> accepts multiple clauses to
produce multiple comma separated entries within the FROM
clause. Useful when selecting from multiple-homed join()
clauses.</li>
<li>the “dont_load=True” flag on <tt class="docutils literal"><span class="pre">Session.merge()</span></tt> is
deprecated and is now “load=False”.</li>
<li>added “make_transient()” helper function which transforms
a persistent/ detached instance into a transient one (i.e.
deletes the instance_key and removes from any session.)
[ticket:1052]</li>
<li>the allow_null_pks flag on mapper() is deprecated and has
been renamed to allow_partial_pks. It is turned “on” by
default. This means that a row which has a non-null value
for any of its primary key columns will be considered an
identity. The need for this scenario typically only occurs
when mapping to an outer join. When set to False, a PK
that has NULLs in it will not be considered a primary key
- in particular this means a result row will come back as
None (or not be filled into a collection), and new in 0.6
also indicates that session.merge() won’t issue a round
trip to the database for such a PK value. [ticket:1680]</li>
<li>the mechanics of “backref” have been fully merged into the
finer grained “back_populates” system, and take place
entirely within the <tt class="docutils literal"><span class="pre">_generate_backref()</span></tt> method of
<tt class="docutils literal"><span class="pre">RelationProperty</span></tt>. This makes the initialization
procedure of <tt class="docutils literal"><span class="pre">RelationProperty</span></tt> simpler and allows
easier propagation of settings (such as from subclasses of
<tt class="docutils literal"><span class="pre">RelationProperty</span></tt>) into the reverse reference. The
internal <tt class="docutils literal"><span class="pre">BackRef()</span></tt> is gone and <tt class="docutils literal"><span class="pre">backref()</span></tt> returns a
plain tuple that is understood by <tt class="docutils literal"><span class="pre">RelationProperty</span></tt>.</li>
<li>the keys attribute of <tt class="docutils literal"><span class="pre">ResultProxy</span></tt> is now a method, so
references to it (<tt class="docutils literal"><span class="pre">result.keys</span></tt>) must be changed to
method invocations (<tt class="docutils literal"><span class="pre">result.keys()</span></tt>)</li>
<li><tt class="docutils literal"><span class="pre">ResultProxy.last_inserted_ids</span></tt> is now deprecated, use
<tt class="docutils literal"><span class="pre">ResultProxy.inserted_primary_key</span></tt> instead.</li>
</ul>
</div>
<div class="section" id="deprecated-removed-orm-elements">
<h3>Deprecated/Removed ORM Elements<a class="headerlink" href="#deprecated-removed-orm-elements" title="Permalink to this headline">¶</a></h3>
<p>Most elements that were deprecated throughout 0.5 and raised
deprecation warnings have been removed (with a few
exceptions). All elements that were marked “pending
deprecation” are now deprecated and will raise a warning
upon use.</p>
<ul class="simple">
<li>‘transactional’ flag on sessionmaker() and others is
removed. Use ‘autocommit=True’ to indicate
‘transactional=False’.</li>
<li>‘polymorphic_fetch’ argument on mapper() is removed.
Loading can be controlled using the ‘with_polymorphic’
option.</li>
<li>‘select_table’ argument on mapper() is removed. Use
‘with_polymorphic=(“*”, <some selectable>)’ for this
functionality.</li>
<li>‘proxy’ argument on synonym() is removed. This flag did
nothing throughout 0.5, as the “proxy generation”
behavior is now automatic.</li>
<li>Passing a single list of elements to joinedload(),
joinedload_all(), contains_eager(), lazyload(), defer(),
and undefer() instead of multiple positional *args is
deprecated.</li>
<li>Passing a single list of elements to query.order_by(),
query.group_by(), query.join(), or query.outerjoin()
instead of multiple positional *args is deprecated.</li>
<li><tt class="docutils literal"><span class="pre">query.iterate_instances()</span></tt> is removed. Use
<tt class="docutils literal"><span class="pre">query.instances()</span></tt>.</li>
<li><tt class="docutils literal"><span class="pre">Query.query_from_parent()</span></tt> is removed. Use the
sqlalchemy.orm.with_parent() function to produce a
“parent” clause, or alternatively <tt class="docutils literal"><span class="pre">query.with_parent()</span></tt>.</li>
<li><tt class="docutils literal"><span class="pre">query._from_self()</span></tt> is removed, use
<tt class="docutils literal"><span class="pre">query.from_self()</span></tt> instead.</li>
<li>the “comparator” argument to composite() is removed. Use
“comparator_factory”.</li>
<li><tt class="docutils literal"><span class="pre">RelationProperty._get_join()</span></tt> is removed.</li>
<li>the ‘echo_uow’ flag on Session is removed. Use logging
on the “sqlalchemy.orm.unitofwork” name.</li>
<li><tt class="docutils literal"><span class="pre">session.clear()</span></tt> is removed. use
<tt class="docutils literal"><span class="pre">session.expunge_all()</span></tt>.</li>
<li><tt class="docutils literal"><span class="pre">session.save()</span></tt>, <tt class="docutils literal"><span class="pre">session.update()</span></tt>,
<tt class="docutils literal"><span class="pre">session.save_or_update()</span></tt> are removed. Use
<tt class="docutils literal"><span class="pre">session.add()</span></tt> and <tt class="docutils literal"><span class="pre">session.add_all()</span></tt>.</li>
<li>the “objects” flag on session.flush() remains deprecated.</li>
<li>the “dont_load=True” flag on session.merge() is deprecated
in favor of “load=False”.</li>
<li><tt class="docutils literal"><span class="pre">ScopedSession.mapper</span></tt> remains deprecated. See the
usage recipe at <a class="reference external" href="http://www.sqlalchemy.org/trac/wiki/Usag">http://www.sqlalchemy.org/trac/wiki/Usag</a>
eRecipes/SessionAwareMapper</li>
<li>passing an <tt class="docutils literal"><span class="pre">InstanceState</span></tt> (internal SQLAlchemy state
object) to <tt class="docutils literal"><span class="pre">attributes.init_collection()</span></tt> or
<tt class="docutils literal"><span class="pre">attributes.get_history()</span></tt> is deprecated. These
functions are public API and normally expect a regular
mapped object instance.</li>
<li>the ‘engine’ parameter to <tt class="docutils literal"><span class="pre">declarative_base()</span></tt> is
removed. Use the ‘bind’ keyword argument.</li>
</ul>
</div>
</div>
<div class="section" id="extensions">
<h2>Extensions<a class="headerlink" href="#extensions" title="Permalink to this headline">¶</a></h2>
<div class="section" id="sqlsoup">
<h3>SQLSoup<a class="headerlink" href="#sqlsoup" title="Permalink to this headline">¶</a></h3>
<p>SQLSoup has been modernized and updated to reflect common
0.5/0.6 capabilities, including well defined session
integration. Please read the new docs at [<a class="reference external" href="http://www.sqlalc">http://www.sqlalc</a>
hemy.org/docs/06/reference/ext/sqlsoup.html].</p>
</div>
<div class="section" id="declarative">
<h3>Declarative<a class="headerlink" href="#declarative" title="Permalink to this headline">¶</a></h3>
<p>The <tt class="docutils literal"><span class="pre">DeclarativeMeta</span></tt> (default metaclass for
<tt class="docutils literal"><span class="pre">declarative_base</span></tt>) previously allowed subclasses to
modify <tt class="docutils literal"><span class="pre">dict_</span></tt> to add class attributes (e.g. columns).
This no longer works, the <tt class="docutils literal"><span class="pre">DeclarativeMeta</span></tt> constructor
now ignores <tt class="docutils literal"><span class="pre">dict_</span></tt>. Instead, the class attributes should
be assigned directly, e.g. <tt class="docutils literal"><span class="pre">cls.id=Column(...)</span></tt>, or the
<a class="reference external" href="http://www.sqlalchemy.org/docs/reference/ext/declarative.html#mix-in-classes">MixIn class</a> approach should be used
instead of the metaclass approach.</p>
</div>
</div>
</div>
</div>
</div>
<div id="docs-bottom-navigation" class="docs-navigation-links">
Previous:
<a href="migration_07.html" title="previous chapter">What’s New in SQLAlchemy 0.7?</a>
Next:
<a href="migration_05.html" title="next chapter">What’s new in SQLAlchemy 0.5?</a>
<div id="docs-copyright">
© <a href="../copyright.html">Copyright</a> 2007-2014, the SQLAlchemy authors and contributors.
Created using <a href="http://sphinx.pocoo.org/">Sphinx</a> 1.2b1.
</div>
</div>
</div>
</body>
</html>
