| 27604018f783bf6354a13870b3e7785edca69b5f |
|
28-Sep-2016 |
Paul Duffin <paulduffin@google.com> |
Detect and fix resource leakages in tests
Tests that test those classes that use CloseGuard to detect
reosurce leakages are switched to use TestCaseWithRules and
ResourceLeakageDetector.getRule() to detect any resource
leakages and the leaks fixed.
A number of tests could not be fixed due to bugs in the core
classes so a new DisableResourceLeakageDetection annotation was
added as part of this change. A method annotated with that does
not perform resource leakage detection. The affected tests were
annotated with a description and a reference to an associated
bug.
This uses try-with-resources to aid in closing resources. While
that does provide some benefits it is all too easy to change the
behavior of the test unless care is taken. To avoid that it is
important not to merge it with an existing try block such as
those which follow the following pattern:
try {... fail()} catch(..) {}
Merging them could result in the test catching and ignoring
exceptions thrown during the creation of the resource which is
almost certainly not what the test was intending.
Test: Tested with vogar and CTS
Bug: 31542223
Change-Id: Id32bb40fe04eb9719d7e6cd367abb53118b21832
|
| 5ccb2b3621f633708eb3ae1fd0f27f88bdb32a9d |
|
29-Jan-2015 |
Narayan Kamath <narayan@google.com> |
Fix deflater in / out counts for inputs > 4G.
The underlying inflate & deflate can process any amount of data
correctly, since the protocol is chunk based (the size of each
chunk is limited to 32 bits, usually) and there is no limit to
the number of chunks.
On LP32, the stream counters are limited to 32 bit counts and
the manual recommends that the caller maintain their own counters
to support larger counts.
<quote>
Note however that the strm.total_in and strm_total_out counters may be
limited to 4 GB. These counters are provided as a convenience and are
not used internally by inflate() or deflate(). The application can
easily set up its own counters updated after each call of inflate()
or deflate() to count beyond 4 GB.
</quote>
Change-Id: I066e1141935154773360c1b9f94c56bd604f8474
|
| b814b1850f5ed997ea1e7f574c8fc0df6a693c36 |
|
09-Sep-2010 |
Jesse Wilson <jessewilson@google.com> |
Address test failures in java.util.zip.
Some deflater tests fail because they expect a specific byte sequence in
the compressed form. Our compressed form is valid but has a different
sequence. I've written new tests to cover the related APIs and suppressed
the broken tests.
The other deleted tests are duplicated exactly from Harmony. I'm removing
our copy.
Change-Id: If480d54948e634e4f6b1018da30951a1d4d07b49
|
| 4557728efb66c455a52b7669a8eefef7a9e54854 |
|
11-Aug-2010 |
Jesse Wilson <jessewilson@google.com> |
Moving tests to be under the libcore.* package.
This is indended to make it easier to run on VMs that restrict the packages
from which application classes can be loaded. For example, on the RI you need
to use the bootclasspath to load these tests.
Change-Id: I52193f35c5fcca18b5a3e1d280505b1e29b388af
|