findMatch() reads too much bytes about simplemagic

[j256]

1. I think the answer is the "most files". Some need more bytes and I figured that 10k was pretty small IO wise these days.
2. When you are worried about it reading 10k. :-)
3. I've added code to pay attention to fis.read() return. Thanks.

[stokito]

ok, then maybe we can make some kind of stream processing and if few bytes was enough then terminate the process.
Also, I'm not sure, but I guess that the size should be rounded to file system page size i.e. 4K
https://en.wikipedia.org/wiki/Page_(computer_memory)

[j256]

You linked to a memory page but we are talking about IO. Most disk subsystems have page sizes a lot less than 4k – it depends highly on the underlying filesystem and IO layers. 512 bytes is often cited. I don't think it makes that much difference but I'll consider changing it to 8k.

[j256]

Oh and streaming would be a big change in the code. Not a bad idea but non trivial change.