This problem is also fixed on trunk. Thanks for the report. On 4/11/15, Stephen Lee <stephen.lee at agsgeo.eu> wrote: > Hi > > > > I am using V3.8.9 of SQLite3.exe on Windows 7 (64bit). > > > > I have a table in a database which shows the following behaviour. > > The table, 'RawDataSamples', has columns ChannelID, SeqIndex and then > Sample_1 .. Sample_30. > > > > Select count(*) from RawDataSamples order by ChannelID, SeqIndex will > return > 69770. (Correct value) > > > > Select * from RawDataSamples order by ChannelID, SeqIndex will return 83665 > rows. > > > > Select * from RawDataSamples order by SeqIndex,ChannelID will return 69770 > rows. (Again correct) > > > > The ChannelID values are 1 .. 10, while the SeqIndex values are from 0 .. > 6976 for each ChannelID value. > > For ChannelID 1 .. 7, only Sample_0 has data, the other columns are null. > > For ChannelID 8 ..10, all columns have data. > > > > In the second query above, most, but not all, the results for ChannelID 9 > and 10 are duplicated. > > If I delete all other ChannelID values and just leave 9 and 10, then the > select will produce correct single values for ChannelID 9, but most results > for 10 will still be duplicated. > > If I delete ChannelID = 9 values, ChannelID 10 values are no longer > duplicated. > > > > I have other databases with the same structure which have fewer rows per > ChannelID value, and they are ok. It seems to be related to the number of > rows for the first order by column. > > > > The database file is at > https://www.dropbox.com/s/1qo0qdeyn71biq4/land.sldb?dl=0 > > > > I did look through the bug list, and if I missed this one then I apologise. > > > > Regards > > > > Steve > > _______________________________________________ > sqlite-users mailing list > sqlite-users at mailinglists.sqlite.org > http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users >
-- D. Richard Hipp drh at sqlite.org
