Skip to content
Projects
Groups
Snippets
Help
Loading...
Help
Support
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in / Register
Toggle navigation
Z
zodb
Project overview
Project overview
Details
Activity
Releases
Repository
Repository
Files
Commits
Branches
Tags
Contributors
Graph
Compare
Issues
0
Issues
0
List
Boards
Labels
Milestones
Merge Requests
0
Merge Requests
0
CI / CD
CI / CD
Pipelines
Jobs
Schedules
Analytics
Analytics
CI / CD
Repository
Value Stream
Wiki
Wiki
Snippets
Snippets
Members
Members
Collapse sidebar
Close sidebar
Activity
Graph
Create a new issue
Jobs
Commits
Issue Boards
Open sidebar
Joshua
zodb
Commits
cfec72a1
Commit
cfec72a1
authored
Apr 13, 2015
by
Jason Madden
Browse files
Options
Browse Files
Download
Email Patches
Plain Diff
Add support for Jython. This required only relatively minor test changes.
parent
c35f2528
Changes
13
Hide whitespace changes
Inline
Side-by-side
Showing
13 changed files
with
100 additions
and
22 deletions
+100
-22
CHANGES.rst
CHANGES.rst
+6
-4
setup.py
setup.py
+1
-0
src/ZODB/DemoStorage.test
src/ZODB/DemoStorage.test
+11
-1
src/ZODB/FileStorage/iterator.test
src/ZODB/FileStorage/iterator.test
+4
-1
src/ZODB/_compat.py
src/ZODB/_compat.py
+11
-2
src/ZODB/tests/dbopen.txt
src/ZODB/tests/dbopen.txt
+6
-5
src/ZODB/tests/testConnectionSavepoint.py
src/ZODB/tests/testConnectionSavepoint.py
+8
-0
src/ZODB/tests/testDB.py
src/ZODB/tests/testDB.py
+4
-1
src/ZODB/tests/testRecover.py
src/ZODB/tests/testRecover.py
+18
-2
src/ZODB/tests/testSerialize.py
src/ZODB/tests/testSerialize.py
+12
-2
src/ZODB/tests/util.py
src/ZODB/tests/util.py
+4
-1
src/ZODB/utils.txt
src/ZODB/utils.txt
+8
-3
tox.ini
tox.ini
+7
-0
No files found.
CHANGES.rst
View file @
cfec72a1
...
...
@@ -8,10 +8,12 @@
- Fix command-line parsing of --verbose and --verify arguments.
(The short versions -v and -V were parsed correctly.)
- Add support for PyPy, and fix the methods in ``ZODB.serialize`` that
find object references under Python 2.7 (used in scripts like
``referrers``, ``netspace``, and ``fsrecover`` among others). This
requires the addition of the ``zodbpickle`` dependency.
- Add support for PyPy and Jython 2.7.
- Fix the methods in ``ZODB.serialize`` that find object references
under Python 2.7 (used in scripts like ``referrers``, ``netspace``,
and ``fsrecover`` among others). This requires the addition of the
``zodbpickle`` dependency.
4.1.0 (2015-01-11)
==================
...
...
setup.py
View file @
cfec72a1
...
...
@@ -60,6 +60,7 @@ Programming Language :: Python :: 3.3
Programming Language :: Python :: 3.4
Programming Language :: Python :: Implementation :: CPython
Programming Language :: Python :: Implementation :: PyPy
Programming Language :: Python :: Implementation :: Jython
Topic :: Database
Topic :: Software Development :: Libraries :: Python Modules
Operating System :: Microsoft :: Windows
...
...
src/ZODB/DemoStorage.test
View file @
cfec72a1
...
...
@@ -14,7 +14,10 @@ existing, base, storage without updating the storage.
...
return
now
>>>
import
time
>>>
real_time_time
=
time
.
time
>>>
time
.
time
=
faux_time_time
>>>
if
isinstance
(
time
,
type
)
:
...
time
.
time
=
staticmethod
(
faux_time_time
)
# Jython
...
else
:
...
time
.
time
=
faux_time_time
To
see
how
this
works
,
we
'll start by creating a base storage and
puting an object (in addition to the root object) in it:
...
...
@@ -45,6 +48,13 @@ and combine the 2 in a demofilestorage:
>>> from ZODB.DemoStorage import DemoStorage
>>> storage = DemoStorage(base=base, changes=changes)
The storage will assign OIDs in a pseudo-random fashion, but for test
purposes we need to control where they start (since the random seeds
can be different on different platforms):
>>> storage._next_oid = 3553260803050964942
If there are no transactions, the storage reports the lastTransaction
of the base database:
...
...
src/ZODB/FileStorage/iterator.test
View file @
cfec72a1
...
...
@@ -13,7 +13,10 @@ We'll make some assertions about time, so we'll take it over:
...
return
now
>>>
import
time
>>>
time_time
=
time
.
time
>>>
time
.
time
=
faux_time
>>>
if
isinstance
(
time
,
type
)
:
...
time
.
time
=
staticmethod
(
faux_time
)
# Jython
...
else
:
...
time
.
time
=
faux_time
Commit
a
bunch
of
transactions
:
...
...
src/ZODB/_compat.py
View file @
cfec72a1
...
...
@@ -13,6 +13,8 @@
##############################################################################
import
sys
IS_JYTHON
=
sys
.
platform
.
startswith
(
'java'
)
try
:
# Python 2.x
import
cPickle
...
...
@@ -113,8 +115,15 @@ def PersistentUnpickler(find_global, load_persistent, *args, **kwargs):
try
:
# Python 2.x
# XXX: why not just import BytesIO from io?
from
cStringIO
import
StringIO
as
BytesIO
if
IS_JYTHON
:
# Jython 2.7rc2 cStringIO.StringIO class has a bug
# resulting in StringIndexOutOfBoundExceptions
# when repeatedly writing and then seeking back to 0
# http://bugs.jython.org/issue2324
from
io
import
BytesIO
else
:
# XXX: why not just import BytesIO from io?
from
cStringIO
import
StringIO
as
BytesIO
except
ImportError
:
# Python 3.x
from
io
import
BytesIO
...
...
src/ZODB/tests/dbopen.txt
View file @
cfec72a1
...
...
@@ -228,17 +228,18 @@ that are still alive.
3
If a connection object is abandoned (it becomes unreachable), then it
will vanish from pool.all automatically. However, connections are
involved in cycles, so exactly when a connection vanishes from pool.all
isn't predictable. It can be forced by running gc.collect():
will vanish from pool.all automatically. However, connections are
involved in cycles, so exactly when a connection vanishes from
pool.all isn't predictable. It can be forced (on most platforms but
not Jython) by running gc.collect():
>>> import gc
>>> import gc
, sys
>>> dummy = gc.collect()
>>> len(pool.all)
3
>>> c3 = None
>>> dummy = gc.collect() # removes c3 from pool.all
>>> len(pool.all)
>>> len(pool.all)
if not sys.platform.startswith("java") else 2
2
Note that c3 is really gone; in particular it didn't get added back to
...
...
src/ZODB/tests/testConnectionSavepoint.py
View file @
cfec72a1
...
...
@@ -99,6 +99,7 @@ one traditional use for savepoints is simply to free memory space midstream
during a long transaction. Before ZODB 3.4.2, making a savepoint failed
to trigger cache gc, and this test verifies that it now does.
>>> import gc
>>> import ZODB
>>> from ZODB.tests.MinPO import MinPO
>>> from ZODB.MappingStorage import MappingStorage
...
...
@@ -129,6 +130,13 @@ Making a savepoint at this time used to leave the cache holding the same
number of objects. Make sure the cache shrinks now instead.
>>> dummy = transaction.savepoint()
Jython needs a GC, and needs to actually access the map to be sure the size
is updated:
>>> _ = gc.collect()
>>> _ = getattr(cn._cache, 'data', {}).values()
>>> _ = getattr(cn._cache, 'data', {}).keys()
>>> len(cn._cache) <= CACHESIZE + 1
True
...
...
src/ZODB/tests/testDB.py
View file @
cfec72a1
...
...
@@ -125,7 +125,10 @@ def connectionDebugInfo():
... now += .1
... return now
>>> real_time = time.time
>>> time.time = faux_time
>>> if isinstance(time,type):
... time.time = staticmethod(faux_time) # Jython
... else:
... time.time = faux_time
>>> from ZODB.tests.util import DB
>>> import transaction
...
...
src/ZODB/tests/testRecover.py
View file @
cfec72a1
...
...
@@ -69,12 +69,28 @@ class RecoverTest(ZODB.tests.util.TestCase):
def
damage
(
self
,
num
,
size
):
self
.
storage
.
close
()
# Drop size null bytes into num random spots.
for
i
in
range
(
num
):
for
i
in
range
(
num
-
1
):
offset
=
random
.
randint
(
0
,
self
.
storage
.
_pos
-
size
)
with
open
(
self
.
path
,
"a+b"
)
as
f
:
# Note that we open the file as r+, not a+. Seeking a file
# open in append mode is effectively a no-op *depending on
# platform*, as the write may simply append to the file. An
# earlier version of this code opened the file is a+ mode,
# meaning on some platforms it was only writing to the end of the
# file, and so the test cases were always finding that bad data.
# For compatibility with that, we do one write outside the loop
# at the end.
with
open
(
self
.
path
,
"r+b"
)
as
f
:
f
.
seek
(
offset
)
f
.
write
(
b"
\
0
"
*
size
)
with
open
(
self
.
path
,
'rb'
)
as
f
:
f
.
seek
(
offset
)
v
=
f
.
read
(
size
)
self
.
assertEqual
(
b"
\
0
"
*
size
,
v
)
with
open
(
self
.
path
,
'a+b'
)
as
f
:
f
.
write
(
b"
\
0
"
*
size
)
ITERATIONS
=
5
# Run recovery, from self.path to self.dest. Return whatever
...
...
src/ZODB/tests/testSerialize.py
View file @
cfec72a1
...
...
@@ -17,7 +17,7 @@ import unittest
import
ZODB.tests.util
from
ZODB
import
serialize
from
ZODB._compat
import
Pickler
,
BytesIO
,
_protocol
from
ZODB._compat
import
Pickler
,
BytesIO
,
_protocol
,
IS_JYTHON
class
ClassWithNewargs
(
int
):
...
...
@@ -139,7 +139,17 @@ class SerializerFunctestCase(unittest.TestCase):
# buildout doesn't arrange for the sys.path to be exported,
# so force it ourselves
environ
=
os
.
environ
.
copy
()
environ
[
'PYTHONPATH'
]
=
os
.
pathsep
.
join
(
sys
.
path
)
if
IS_JYTHON
:
# Jython 2.7rc2 has a bug; if it's Lib directory is
# specifically put on the PYTHONPATH, then it doesn't add
# it itself, which means it fails to 'import site' because
# it can't import '_jythonlib' and the whole process fails
# We would use multiprocessing here, but it doesn't exist on jython
sys_path
=
[
x
for
x
in
sys
.
path
if
not
x
.
endswith
(
'Lib'
)
and
x
!=
'__classpath__'
and
x
!=
'__pyclasspath__/'
]
else
:
sys_path
=
sys
.
path
environ
[
'PYTHONPATH'
]
=
os
.
pathsep
.
join
(
sys_path
)
subprocess
.
check_call
(
prep_args
,
env
=
environ
)
load_args
=
[
sys
.
executable
,
'-c'
,
'from ZODB.tests.testSerialize import _functest_load; '
...
...
src/ZODB/tests/util.py
View file @
cfec72a1
...
...
@@ -182,4 +182,7 @@ def mess_with_time(test=None, globs=None, now=1278864701.5):
import
time
zope
.
testing
.
setupstack
.
register
(
test
,
setattr
,
time
,
'time'
,
time
.
time
)
time
.
time
=
faux_time
if
isinstance
(
time
,
type
):
time
.
time
=
staticmethod
(
faux_time
)
# jython
else
:
time
.
time
=
faux_time
src/ZODB/utils.txt
View file @
cfec72a1
...
...
@@ -41,7 +41,12 @@ To see this work (in a predictable way), we'll first hack time.time:
>>> import time
>>> old_time = time.time
>>> time.time = lambda : 1224825068.12
>>> time_value = 1224825068.12
>>> faux_time = lambda: time_value
>>> if isinstance(time,type):
... time.time = staticmethod(faux_time) # Jython
... else:
... time.time = faux_time
Now, if we ask for a new time stamp, we'll get one based on our faux
time:
...
...
@@ -71,7 +76,7 @@ Here, since we called it at the same time, we got a time stamp that
was only slightly larger than the previos one. Of course, at a later
time, the time stamp we get will be based on the time:
>>> time
.time = lambda :
1224825069.12
>>> time
_value =
1224825069.12
>>> tid = ZODB.utils.newTid(tid2)
>>> print(ZODB.TimeStamp.TimeStamp(tid))
2008-10-24 05:11:09.120000
...
...
@@ -194,4 +199,4 @@ supports optional method preconditions [1]_.
locked. Combining preconditions with locking provides both
efficiency and concise expressions. A more general-purpose
facility would almost certainly provide separate descriptors for
preconditions.
preconditions.
tox.ini
View file @
cfec72a1
[tox]
# Jython 2.7rc2 does work, but unfortunately has an issue running
# with Tox 1.9.2 (http://bugs.jython.org/issue2325)
#envlist = py26,py27,py32,py33,py34,pypy,simple,jython,pypy3
envlist
=
py26,py27,py32,py33,py34,pypy,simple,pypy3
[testenv]
...
...
@@ -29,6 +32,10 @@ commands =
python
setup.py
test
-q
deps
=
{[testenv]deps}
[testenv:jython]
commands
=
jython
setup.py
test
-q
[testenv:coverage]
basepython
=
python2.7
usedevelop
=
true
...
...
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment