Обсуждение: COPY error handling
Hi, I'm not sure if this is the correct group for this question, but I'll post it hoping that it is. I'm loading several ~15 million row files into a table using the COPY command. Apparently one of the rows, about 6 million in, has an invalid entry. This is causing the COPY command to fail, so my question is this: Is there any way to skip invalid rows? Or send them to a separate log file etc to go through later? I've gone through the docs, but I didn't see anything specific to this. Any information, links, or hints are greatly appreciated. Cheers! Bricklen
On Fri, Jun 04, 2004 at 14:11:19 +0000, Bricklen <bricklen-rem@yahoo.comz> wrote: > Hi, I'm not sure if this is the correct group for this question, but > I'll post it hoping that it is. > I'm loading several ~15 million row files into a table using the COPY > command. Apparently one of the rows, about 6 million in, has an invalid > entry. This is causing the COPY command to fail, so my question is this: > Is there any way to skip invalid rows? Or send them to a separate log > file etc to go through later? > I've gone through the docs, but I didn't see anything specific to this. > Any information, links, or hints are greatly appreciated. Currently there isn't a builtin way to do this. You can pass the data through a filter script that removes rows that are not in a proper format.
Bruno Wolff III wrote: > On Fri, Jun 04, 2004 at 14:11:19 +0000, > Bricklen <bricklen-rem@yahoo.comz> wrote: > >>Hi, I'm not sure if this is the correct group for this question, but >>I'll post it hoping that it is. >>I'm loading several ~15 million row files into a table using the COPY >>command. Apparently one of the rows, about 6 million in, has an invalid >>entry. This is causing the COPY command to fail, so my question is this: >>Is there any way to skip invalid rows? Or send them to a separate log >>file etc to go through later? >>I've gone through the docs, but I didn't see anything specific to this. >>Any information, links, or hints are greatly appreciated. > > > Currently there isn't a builtin way to do this. You can pass the data > through a filter script that removes rows that are not in a proper format. > Thanks for the clarification! Cheers, Bricklen