How to efficiently handle regex parsing over a large response?

See more:

Java

regular-expression

So, I would be getting an 80MB or more excel/csv file as the response from a server and I need to run a list of regexes to manipulate some of the content within the response before presenting it to the end user. I've tried reading the response chunk by chunk, process the regexes over the chunk before proceeding onto the next chunk. With this approach, there is a chance that pattern matching data can span across different chunks and hence cannot be manipulated. If I process all the regexes over the complete response then it takes a large amount of time to process..

Would anyone suggest an efficient way to handle this and bringing down the processing time?