Hi, I just replaced the String.intern() mechanism used in DbfFile to save memory heap space by the use of a temporary HashMap.
Please, tell me if you notice any strange behaviour in the shapefile reader. For those of you interested by performance tuning, a short abstract of what we did to try to optimize DbfFile reader : Exemple : loading a 500000 features with attribute A (500000 different values) and attribute B (500000 identical values) version 1 (original jump) : loads 500000 strings for A + 500000 strings for B in the jvm heap space ==> big waste of memory for case B version 2 (optimized by MM and Larry) : loads 500000 strings in permgen memory for attribute A and 1 string in permgen memory for attribute B ==> need to tune max permgen memory when loading big shapefiles with attributes like A ==> permgen memory cannot be cleaned (no way to free the permgen space if the layer is deleted) ==> from my experience, the try/catch block added by Larry to switch from permgen space to heapspace in case or error did not prevent a permgen exception to rise and to stop the process version 3 (today) : loads 500000 strings in the heapspace for A and 1 string in the heapspace for B ==> the loading guaranty unicity of field value for one layer only (string will be duplicated for each loaded layer) ==> all the strings are in the heapspace (and can be free if the layer is deleted) Michaël ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ _______________________________________________ Jump-pilot-devel mailing list Jump-pilot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/jump-pilot-devel
