As a workaround you can use a predicate to filter out the the headers.

Simply apply a predicate to a column header where none of the column values 
will be the same as the header.

—Andries


On Apr 17, 2015, at 8:43 AM, Abhishek Girish <[email protected]> wrote:

> Hello Charles,
> 
> Support for parsing/skipping headers in CSV files is not available at
> present, but is coming soon. You can watch DRILL-951
> <https://issues.apache.org/jira/browse/DRILL-951> / DRILL-1440
> <https://issues.apache.org/jira/browse/DRILL-1440> for updates.
> 
> And regarding your second question, are you talking about Text files, but
> with custom delimiters (which is supported), or custom data formats (other
> than Text, Parquet, JSON, ...). If the latter, can you please clarify
> further?
> 
> Regards,
> Abhishek
> 
> 
> On Fri, Apr 17, 2015 at 8:27 AM, Charles Givre <[email protected]> wrote:
> 
>> Hello all,
>> I just started using Apache Drill and I'm very impressed so far.  I
>> had 2 questions and I couldn't find any answers online.
>> 
>> 1.  If you have a CSV file in which the first row are field names, is
>> it possible for drill to either read those field names and assign them
>> to the appropriate column, or skip that row entirely?
>> 
>> 2.  I think there would be enormous utility in writing plugins which
>> would enable drill to query log files which have a standard format, bu
>> that isn't one of the supported formats.  Is there any documentation
>> about how to get custom data into Drill?
>> Thanks,
>> -- Charles
>> 

Reply via email to