hadoop-mapreduce-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From abhijeet gaikwad <abygaikwa...@gmail.com>
Subject CompressionCodecFactory in LineRecordReader
Date Thu, 09 Nov 2017 00:31:48 GMT

I see that the LineRecordReader uses CompressionCodecFactory and tries to
guess codec using file name extension. Currently this logic is case
sensitive, for example it will work for "*.gz" but not "*.GZ". Do you see
any challenges if we make this case insensitive - to be precise logic in
getCodec(Path file) method of CompressionCodecFactory?


  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message