v0.3.1

Switches verbose to batchman.verbose so as to not collide with verbose arguments within actual functions.
Switched to semantic versioning.

v0.3.0

A new method, robust_batch, has been added that can attempt batching with automatic error-handling. Call robust_batch(batched_fn, ...) where batched_fn is the function that you want to use after applying batch to it, and ... are the arguments you wanted to pass to batched_fn. The default amount of max retries is 3, but you can change this by passing batchman.retries.
stop = FALSE now works as originally intended to keep going with batch processing even if an error occurs midstream. Batches that result in an error simply return NA instead.

Setting options('batchman.verbose' = FALSE) will now dynamically turn off batchman's printed output, even for batched functions after they are generated. Note that there is still no way to dynamically turn on the verbosity if verbose = FALSE was originally set when the function was made.
Initial prototypes for covr and travis added.

You can now set options('batchman.verbose' = FALSE) to turn off batchman's print messages. This is potentially useful for tests where digging into batchman to set verbose = FALSE is annoying. Keep in mind that this still won't affect batched functions after they are generated.

Incorporates the progress bar from dplyr https://github.com/hadley/dplyr/blob/master/R/progress.R

The way of getting the pre-batched function has changed from looking at the attr of the batched function to using get_before_fn(batched_fn), which instead relies on looking at the environment of batched_fn instead of the attributes.
Batched functions now are of class batch_function (in addition to being of class function), and printing a batched function will show a list with both the before_fn (the function prior to batching) and the after_fn (the batched funtion). This should make looking into the function more clear, while not sacrificing your ability to see what is going on under the hood.