Community
Participate
Working Groups
Build Identifier: 3.7.1 Running a BIRT report using a 36MB FlatFile datasource (500 columns, 10000 rows) takes a very long time - >30minutes. Running SysInternals Process Monitor shows that it is reading one byte from the file every ~0.1ms (it's actually a bit quicker than that, because that would take an hour to read). Looking in the source for org.eclipse.datatools.connectivity.oda.flatfile shows that it is using a bare FileInputStream. Is there a good reason why this couldn't be wrapped in a BufferedInputStream? Reproducible: Always Steps to Reproduce: 1. Create a CSV with 500 columns and 10000 rows. 2. Add that file as a data source to a BIRT report. 3. Run the report, and use Process Monitor to track file accesses.