Skip records in a text file

I am working with large text files (~27 GB per file) containing 570,000 rows (records). I want to locate and access the contents of a particular row as efficiently as possible. The rows all consist of an initial 21 character identifying string, followed by a blank, followed by 8386 "%f" fields separated by blanks, and then ending with a new-line. I know the record number (and identifying string) of the record I would like to read. Unfortunately, I cannot use fseek to position the file since the number of bytes per record is not constant since the "%f" fields have varying sizes.

Is there a capability in MATLAB of skipping N rows (records) of text on read in MATLAB?

Many thanks,

Duane

Answer 1

[per isakson]

"Is there a capability in MATLAB of skipping N rows"

The only one I can think of is the Headerlines of textscan