Crash due to database lock

[veetil]

In my android code, SimpleNoSQL ocassionally crashes the app on start. Error message is that database is locked. Since there is not initDB function, how to handle this scenario.

id.database.sqlite.SQLiteDatabaseLockedException: database is locked (code 5)
09-21 18:13:31.008 25314-25521/com.fun.shopping E/AndroidRuntime: at android.database.sqlite.SQLiteConnection.nativeExecuteForChangedRowCount(Native Method)
09-21 18:13:31.008 25314-25521/com.fun.shopping E/AndroidRuntime: at android.database.sqlite.SQLiteConnection.executeForChangedRowCount(SQLiteConnection.java:924)
09-21 18:13:31.008 25314-25521/com.fun.shopping E/AndroidRuntime: at android.database.sqlite.SQLiteSession.executeForChangedRowCount(SQLiteSession.java:754)
09-21 18:13:31.008 25314-25521/com.fun.shopping E/AndroidRuntime: at android.database.sqlite.SQLiteStatement.executeUpdateDelete(SQLiteStatement.java:64)
09-21 18:13:31.008 25314-25521/com.fun.shopping E/AndroidRuntime: at android.database.sqlite.SQLiteDatabase.delete(SQLiteDatabase.java:1618)
09-21 18:13:31.008 25314-25521/com.fun.shopping E/AndroidRuntime: at com.colintmiller.simplenosql.db.SimpleNoSQLDBHelper.deleteBucket(SimpleNoSQLDBHelper.java:84)
09-21 18:13:31.008 25314-25521/com.fun.shopping E/AndroidRuntime: at com.colintmiller.simplenosql.threading.DataDispatcher.delete(DataDispatcher.java:108)
09-21 18:13:31.008 25314-25521/com.fun.shopping E/AndroidRuntime: at com.colintmiller.simplenosql.threading.DataDispatcher.run(DataDispatcher.java:80)

[Jearil]

Interesting. Do you have some sample code to reproduce this? It looks like you're deleting a bucket at some point. When is this?

[veetil]

Sure. Here is a sample code I am using to delete buckets. This I am doing in my MainActivity, before starting any work to clear all cache on new app session.

        NoSQL.with(mcontext).using(Option.class)
                .bucketId("options")
                .delete();
        NoSQL.with(mcontext).using(OptionGroup.class)
                .bucketId("option_group")
                .delete();

I started seeing this issue mostly after using ProGuard. Although I did add the following lines to proguard.
-keep class com.colintmiller.** { ; }
-keepclassmembers class com.colintmiller.* { ; }
-keep interface com.colintmiller.* { ; }
-keep enum com.colintmiller.* { *; }

[Jearil]

Can you reproduce it without proguard? If it's a progaurd problem, that's a bit different than an error in the library. I'll look into it though, it may there is something wrong with the locking code which should prevent multiple writes from happening at a time.

[Jearil]

Also, as a temporary work around, you can use:
NoSQL.with(mContext, 1).using(....
to prevent the error (since it seems to be a threading issue).