On Sun, 30 Jun 2013, Andi Vajda wrote:
On Sun, 30 Jun 2013, Michael McCandless wrote:
On Sun, Jun 30, 2013 at 3:20 PM, Andi Vajda <va...@apache.org> wrote:
On Sun, 30 Jun 2013, Michael McCandless wrote:
Hmm I see two test failures, on Linux, Python 2.7.3, Java 1.7.0_07
:
ERROR: testCachingWorks (__main__.CachingWrapperFilterTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
File "test/test_CachingWrapperFilter.py", line 53, in testCachingWorks
strongRef = cacher.getDocIdSet(context,
context.reader().getLiveDocs())
AttributeError: 'IndexReader' object has no attribute 'getLiveDocs'
and:
ERROR: testPayloadsPos0 (__main__.PositionIncrementTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
File "test/test_PositionIncrement.py", line 257, in testPayloadsPos0
pspans = MultiSpansWrapper.wrap(searcher.getTopReaderContext(), snq)
File "/home/mike/src/pylucene-4.3.1-1/test/MultiSpansWrapper.py",
line 49, in wrap
return query.getSpans(ctx, ctx.reader().getLiveDocs(), termContexts)
AttributeError: 'IndexReader' object has no attribute 'getLiveDocs'
I can't reproduce this, the tests pass for me.
I suspect there is some differennce in version somewhere.
The clue is the error message:
AttributeError: 'IndexReader' object has no attribute 'getLiveDocs'
This is correct. What is not is that ctx.reader() is returning an
IndexReader instance. I put a breakpoint at one of the failures and I'm
getting an AtomicReader back from ctx.reader() and that object does have
getLiveDocs() defined. Below is an excerpt of my debugging session:
yuzu:vajda> _install/bin/python test/test_PositionIncrement.py
> /Users/vajda/apache/pylucene-4.3.1/test/MultiSpansWrapper.py(50)wrap()
-> return query.getSpans(ctx, ctx.reader().getLiveDocs(), termContexts)
(Pdb) p ctx.reader()
<AtomicReader: _0(4.3.1):C1>
(Pdb) !from org.apache.lucene.index import IndexReader
(Pdb) p IndexReader.getLiveDocs
(Pdb) !from org.apache.lucene.index import AtomicReader
(Pdb) p AtomicReader.getLiveDocs
<method 'getLiveDocs' of 'AtomicReader' objects>
I built and tested this release on Mac OS X with Apple's Java 6.
I suspect the different to be either Java 7 or you have an old version of
jcc, maybe ?
(Pdb) !import lucene
(Pdb) p lucene.JCC_VERSION
'2.16'
(Pdb) p lucene.VERSION
'4.3.1'
I need to either renew my VMware Fusion license or install VirtualBox to
test on Linux with Java 7.
I tried Oracle's Java 7 on Mac and it was a big pain as it doesn't seem to
blend with the existing Java Frameworks and I kept getting a JRE mismatch
(building jcc with Java 7 but then having Java 6 run it and complain about
incompatible bytecode formats). I'm sure this can be eventually resolved
but I was short on time.
Can you please confirm that you see the same jcc version in your build ?
If that's the case, then the next thing is for me to try Java 7 again.
I see 2.16 as the JCC_VERSION and 4.3.1 as the VERSION, so that matches.
Also, I went and built with Java 1.6.0_32, and tests now pass.
I'm not sure why java 7 seems to lead to the failures ...
Me neither, but at least I know now how to proceed...
Thank you for checking that this was indeed related to Java 7 !
The plot thickens.
I was able to install Orache Java 7 on my Mac and configure it for use with
JCC. I now have PyLucene 4.3.1 built with Java 7 that passes its tests.
I wasn't able to reproduce the problem you found :-(
Python 2.7 (r27:82500, Jul 7 2010, 03:32:44)
[GCC 4.2.1 (Apple Inc. build 5659)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> import lucene
>>> lucene.initVM()
<jcc.JCCEnv object at 0x10029c0f0>
>>> from java.lang import System
>>> System.getProperty('java.runtime.version')
u'1.7.0_25-b15'
yuzu:vajda> _install/bin/python test/test_PositionIncrement.py
..
----------------------------------------------------------------------
Ran 2 tests in 0.161s
OK
Could it be that you're picking up an old version of JCC when building
PyLucene ?
In any case, this exercise was worth it since I can add support for Java 7
on Mac now in JCC and roll another release candidate.
Andi..
