Merge commit '7e343b8d39309d2c9d73cc5d1ec2434e666ae48b'
* commit '7e343b8d39309d2c9d73cc5d1ec2434e666ae48b':
fillWindow's start position must be smaller than getCount value
the sql statement is already printed if SqlStatements is set to VERBOSE
anyway. so no need for this extra logging
Change-Id: Id0e6c5acbbbd453ea45f08bcb802213bc22896fc
- Move PackageInfo out of ActivityThread, renaming to LoadedApk.
- Rename some of the other PacakgeInfo inner classes to better
represent what they are.
- Rename HistoryRecord to ActivityRecord.
- Introduce AppGlobals, to eventually let ActivityThread become
package scoped.
Change-Id: Ib714c54ceb3cdbb525dce3db9505f31042e88cf0
an app can execute 'attach database..' statement anytime after WAL is
enabled. since WAL doesn't work with atatched databases,
disable WAL whenever the caller executes 'attach database'
sql statement.
Change-Id: I77dfcb744b59476c357d44296c14d63455985a7b
due to a bug in DefaultDatabaseErrorHandler, if the corruption is
so bad that the list of attached databases can't even be retrieved,
then database is not closed in DefaultDatabaseErrorHandler,
this causes the corrupted file to remain on disk.
this causes corruption detection to occur forever until the stack
overflows.
Change-Id: I9896bee220231cbde0b1620ad0a617420424967c
setting journal_mode to TRUNCATE on memory databases causes an error message
to be displayed by SQLIteDatabase.
Change-Id: Ie8b8ae22cb9fba99cee59dba35b260195c63eadc
two bugs were introduced by the above CL
1. to test if the column value is NULL, getType_native() should check if
the columnindiex is valid. (it seems sometimes callers could ask if a
given non-existent column is null - and the answer should be true.)
2. if a column is null and isBlob_native, isString_native methods should
return true - not false.
Change-Id: I64df75d0a3840a4502c00db8759c66ba4b268a26
cts tests are in Change-Id: Ifcc89b4ff484c7c810fd2d450ded212a43360dda
dependency on: Change-Id: I938c42afc3fb50f5296d01c55ffcf4a102d8b0cb
1. Use sqlite's work-in-progress writeahead logging feature to read old
versions of data and thus increase concurrency of readers
even when there is a writer on the database
2. New API executeQueriesInParallel() sets up a database connecion pool
automatically created and managed by sqlite java layer
3. To increase reader concurrency, add an option to do BEGIN IMMEDIATE xaction
instead of BEGIN EXCLUSIVE
Change-Id: I3ce55a8a7cba538f01f731736e7de8ae1e2a8a1f
sometimes the database can be so corrupt that it cannot even be queried
for attached database list.
Change-Id: Ib8fe3bd94157acab3fbf1011c3f8a532ef5019f4
SQLite is JNI to native code and doesn't go via IFileSystem, so it
needs custom sprinkling, at least for now.
Change-Id: Ic7fded1b64a4f483dfc17b3a7b136c803df1e111
add new method openOrCreateDatabase in Context.java to allow
callers specify a DatabaseErrorHandler object to be used when
database corruption occurs.
add new constructor in SQLiteOpenHelper to accept DatabaseErrorHandler
as an additional param to be used when SQLiteDatabase instance is
created.
Change-Id: I912a0202a74510f9ca0206dd8101c4abab9102ae
A change was made in the last CL to ask sqlite for all unfinalized statements
and then finalizing them before closing the database.
But this crashes sqlite! because sqlite's FTS3
module keeps some prepared statements around and they should not
be finalized before closing the database.
(when sqlite is asked for all unfianlized statements, it also returns
the FTS3's reserved prepared statements which should not be
finalized!!)
Change-Id: I141ab4563985b8cd1305a1228c4cb01bc7281bcb
bug:2683001
implmentation details:
1.close() on any sql statement is should simply queued up for finalization
to be performed later. caller doesn't acquire database lock.
2. the only effect of NOT close immediately is non-release of some memory.
3. all such queued-up-finalizations are performed whenever there is
another execute() of some sql statement on that database from ANY
thread in the process.
4. database close() automatically releases all unfinalized statements
before closing the database.
Change-Id: If4c9c7fde6b8945a41abc6c8b992aa8c69854512
it is trying to get attachedDb list (by executing a pragma) after closing
the database.
also added unittests.
Change-Id: I7dce665ec7354402cdef6fbe055455f5798e123c
- Moved the thread check from AbstractCursor to SQLiteCursor.
- Show warning only once per database
Bug 2633883
Change-Id: I915bfba8715a469c45fb68ba1282231279f2a50c
revert part of the CL: Ia561135e974a44ad3e3774ecb23c6a3d0fc38176
and add it to ContentProvider.query() to force query execution
in worker thread, instead of having the main thread incur the cost
of query execution and potentially get ANR
Change-Id: I0ea8a170bd954a421f4ad825f8090319a83a5b2b
1. when an entry is pushed out of cache, it should be released if it is not
in use by any thread. This didn't have to be done when cache was NOT LRU
because the object either got into the cache while the caller had a reference
to it or it didn't. if it didn't get into cache (because cache is full),
the caller's close() released the object. But in LRU cache, an object
could get pushed out of cache due to LRU policy and if it is NOT in use
by any thread at the time it was pushed out of cache, then it
got GC'ed - which caused warnings to be pronted in the log.
2. also delete some unused methods in SQLiteDatabase.
Change-Id: I7831952647d3a057342bcc8ac186a6a26eb58f33
bug:1648729
print warning if a requery is attemped on main thread
bug:2459293
browser death because of NoSuchElementException. don't use iterator
to march thru the ArrayList. use old style for() loop.
bug:1609474
don't allow downgrades of databases
other stuff
use LRUcache to maintain statement-cache in SQLiteDatabase
Change-Id: I3a42022162f4dfcc75f7d0cfcad8f20f1d193249
let the user specify an interface impl class to specify the actions
to take when db corruption is detected.
this class is specified when the database is opened/created.
Change-Id: I84eb57208c8fedfa7235805b0ec58165efdc1560
