Currently, weather-sp provides an option to set the filename of the splitted file.
For example, for the input file --input-pattern 'gs://bucket-name/2017-1-sp-10u-10v-2t-2d-tp-ptype.gb', I provided --output-template 'gs://bucket-name/monthly-splitted/{date}{time}/{step}/{date}{time}_FH-{step}.gb' and it created splitted files like gs://bucket-name/monthly-splitted/201201010000/0/201201010000_FH-0.gb.
This input file is for a few variables. Now in the future, when a new raw file for a new variable comes for the same timestamp, I cannot split that file in the monthly-splitted folder, because it will override the files as they just includes date, time and step. What I really want is an option to have a functionality to append the raw data filename to the splitted data filename. Somewhat like ability to provide that in the --output-template argument only like --output-template 'gs://bucket-name/monthly-splitted/{date}{time}/{step}/{filename}_{date}{time}_FH-{step}.gb'
Currently, weather-sp provides an option to set the filename of the splitted file.
For example, for the input file
--input-pattern 'gs://bucket-name/2017-1-sp-10u-10v-2t-2d-tp-ptype.gb', I provided--output-template 'gs://bucket-name/monthly-splitted/{date}{time}/{step}/{date}{time}_FH-{step}.gb'and it created splitted files likegs://bucket-name/monthly-splitted/201201010000/0/201201010000_FH-0.gb.This input file is for a few variables. Now in the future, when a new raw file for a new variable comes for the same timestamp, I cannot split that file in the monthly-splitted folder, because it will override the files as they just includes date, time and step. What I really want is an option to have a functionality to append the raw data filename to the splitted data filename. Somewhat like ability to provide that in the --output-template argument only like
--output-template 'gs://bucket-name/monthly-splitted/{date}{time}/{step}/{filename}_{date}{time}_FH-{step}.gb'