changes: many bugfixes and compatibility fixes
The 2.5.0 version in pkgsrc was broken:
>>> from pysqlite2 import dbapi2
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/usr/pkg/lib/python2.5/site-packages/pysqlite2/dbapi2.py", line 27, in <module>
from pysqlite2._sqlite import *
ImportError: /usr/pkg/lib/python2.5/site-packages/pysqlite2/_sqlite.so: Undefined PLT symbol "sqlite3_enable_load_extension" (symnum = 158)
2.5.0:
- Windows binaries are now cross-built using mingw on Linux
- import various fixes from Python 2.6 version
- Connection has new method iterdump() that allows you to create a script file
that can be used to clone a database
- the docs are now built using Sphinx and were imported from Python 2.6's
sqlite3 module
- Connection.enable_load_extension(enabled) to allow/disallow extension
loading. Allows you to use fulltext search extension, for example ;-)
- Give the remaining C functions used in multiple .c source files the pysqlite_
prefix.
- Release GIL during sqlite3_prepare() calls for better concurrency.
- Automatically download the SQLite amalgamation when building statically.
2.4.1:
- Made unicode strings for the database parameter in connect() work again
- Removed bad defaults from setup.cfg
2.4.0:
- Implemented context managers. pysqlite's connections can now be used as
context managers with Python 2.5 or later:
from __future__ import with_statement
from pysqlite2 import dbapi2 as sqlite
con = sqlite.connect(":memory:")
con.execute("create table person (id integer primary key, firstname varchar unique)")
# Successful, con.commit() is called automatically afterwards
with con:
con.execute("insert into person(firstname) values (?)", ("Joe",))
# con.rollback() is called after the with block finishes with an exception, the
# exception is still raised and must be catched
try:
with con:
con.execute("insert into person(firstname) values (?)", ("Joe",))
except sqlite.IntegrityError:
print "couldn't add Joe twice"
- pysqlite connections can now be created from APSW connections. This enables
users to use APSW functionality in applications using the DB-API from
pysqlite:
from pysqlite2 import dbapi2 as sqlite
import apsw
apsw_con = apsw.Connection(":memory:")
apsw_con.createscalarfunction("times_two", lambda x: 2*x, 1)
# Create pysqlite connection from APSW connection
con = sqlite.connect(apsw_con)
result = con.execute("select times_two(15)").fetchone()[0]
assert result == 30
con.close()
Caveat: This will only work if both pysqlite and APSW are dynamically
linked against the same SQLite shared library. Otherwise you will
experience a segfault.
- Fixed shuffled docstrings for fetchXXX methods.
- Workaround for SQLite 3.5.x versions which apparently return NULL for
"no-operation" statements.
- Disable the test for rollback detection on old SQLite versions. This prevents
test failures on systems that ship outdated SQLite libraries like MacOS X.
- Implemented set_progress_handler for progress callbacks from SQLite. This is
particularly useful to update GUIs during long-running queries. Thanks to
exarkun for the original patch.
changes:
-pysqlite is now easy_install-able
-misc bugfixes
-Allow the size parameter for fetchmany() for better DB-API compliance
-Allow a static build of pysqlite using the SQLite amalgamation
-improve concurrency
-Using mappings and sequences as parameters works now
-Performance optimizations
- self->statement was not checked while fetching data, which could
lead to crashes if you used the pysqlite API in unusual ways.
Closing the cursor and continuing to fetch data was enough.
- Converters are stored in a converters dictionary. The converter name
is uppercased first. The old upper-casing algorithm was wrong and
was replaced by a simple call to the Python string's upper() method
instead.
- Applied patch by Glyph Lefkowitz that fixes the problem with
subsequent SQLITE_SCHEMA errors.
- Improvement to the row type: rows can now be iterated over and have a keys()
method. This improves compatibility with both tuple and dict a lot.
- A bugfix for the subsecond resolution in timestamps.
- Corrected the way the flags PARSE_DECLTYPES and PARSE_COLNAMES are
checked for. Now they work as documented.
- gcc on Linux sucks. It exports all symbols by default in shared
libraries, so if symbols are not unique it can lead to problems with
symbol lookup. pysqlite used to crash under Apache when mod_cache
was enabled because both modules had the symbol cache_init. I fixed
this by applying the prefix pysqlite_ almost everywhere. Sigh.
