Grok custom pattern for space delimited file

[atchyuth]

I'm trying to load a file to structured table in Athena. I am using GROK pattern to load it to the table but not able to find the correct pattern. The file format is as below:

L1127 ACTUALS 214171 ON 27649075 -00000000000000000409618.02 601 MBS DAILY VISION - CAN OS

L1127 ACTUALS 412821 ON 27649075 002060 -00000000000000000002657.33 521 MBS DAILY VISION - CAN OS

GROK pattern I'm using : (?.{5})%{SPACE}(?.{7})%{SPACE}(?.{6})%{SPACE}(?.{2})%{SPACE}(?.{8})%{SPACE}(?.{6})%{SPACE}(?.{27})%{SPACE}(?.{3})%{SPACE}(?.{35})

I'm having trouble when the ProductId has no value.

Any help would be appreciated

[Badger]

You cannot use . everywhere since that will always match. If you use NOTSPACE for the fields and make the sixth field (and preceding space) optional by appending ? to the pattern then it will work.

grok { match => { "message" => "^(?<a>[^ ]{5})%{SPACE}(?<b>[^ ]{7})%{SPACE}(?<c>[^ ]{6})%{SPACE}(?<d>[^ ]{2})%{SPACE}(?<e>[^ ]{8})%{SPACE}?(?<f>[^ ]{6})?%{SPACE}(?<g>[^ ]{27})%{SPACE}(?<h>.{3})%{SPACE}(?<i>.+)" } }

Also, you do not have 35 characters at the end, so change the {35} to +