From: Andrea A. (JIRA) <ji...@co...> - 2009-11-09 16:05:04
|
[ http://jira.codehaus.org/browse/GEOT-2683?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Andrea Aime reopened GEOT-2683: ------------------------------- Reopen after Ivan reporting issues with some one-byte charsets > Speed up dbf reading when the target charset is not multibyte > ------------------------------------------------------------- > > Key: GEOT-2683 > URL: http://jira.codehaus.org/browse/GEOT-2683 > Project: GeoTools > Issue Type: Improvement > Components: data shapefile > Affects Versions: 2.6-M2 > Reporter: Andrea Aime > Assignee: Andrea Aime > Fix For: 2.6-M2 > > Attachments: GEOT-2683.patch, GEOT-2683.patch, ReadSpeedTest.java > > > Decoding the bytes of the dbf row into chars is one of the most expensive operations the shapefile reader is doing (when the dbf is opened at all, that is). > This is due to che usage of the Decoder class, which is pretty generic. > A custom code path that just assumes the 1-1 byte-char correspondance is quite a bit faster (15-20% overall while reading statepop.shp) -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://jira.codehaus.org/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira |