On Wed, Dec 9, 2015 at 4:57 PM, Kim-Ee Yeoh <ky3@atamo.com> wrote:

On Thu, Dec 10, 2015 at 4:19 AM, Dan Stromberg <strombrg@gmail.com> wrote:

What if I want to be able to deal gracefully with files that aren't readable, whether due to permissions issues or I/O errors? I agree that zip'ing is easier, but is it as robust?

Making sense of this question requires an apples-to-apples comparison. Observe that the original code doesn't deal with read errors either.

In fact, the replies in this thread have done only two things.

They've fixed the typecheck error.

And they've offered idiomatic -- but semantically identical -- rewritings that read-fault in the exact same way as the original code.

Yes, sure.

My thought was that going over the list of filenames+sizes and adding prefix hashes where available, would be easier to make robust, than attempting to get prefix hashes for all and zipping the results.

Is that not correct?

Should I use a Maybe to deal with files that don't hash, so there will always be a one-to-one correspondence, allowing a zip?

Thanks.

Dan Stromberg