Mike Robbins

For large documents (tens of MB), performance is quadratically slow because each small chunk of data triggers a `split`. This small fix dramatically improves performance for large documents.

Frozen / black screen on launch. "memory allocation bug: object ... has never been allocated"

1

#### Your system information * System information from steam (`Steam` -> `Help` -> `System Information`) in a [gist](https://gist.github.com/): https://gist.github.com/compumike/8628c19caab3490a91dadb750d4da7d1 * Have you checked for system updates?: Yes #### Please describe...

Exception sending query with bytea[] binary array type?

2

I ran into an `Unhandled exception: invalid byte sequence for encoding "UTF8": 0x00 (PQ::PQError)` when trying to do a query that involves the postgres `BYTEA` binary type. Any idea why...

Bug fix for: ``` In lib/awscr-s3/src/awscr-s3/responses/head_object_output.cr:43:7 43 | response.headers.each do |k, v| ^------- Error: undefined local variable or method 'response' for Awscr::S3::Response::HeadObjectOutput ``` Thanks! :heart:

[ENH] Add an optional RDIFF_BACKUP_BLOCKSIZE environment variable to specify a fixed blocksize

3

## Is your feature request related to a problem? Please describe. Increment file size is ~16X too large for database dump backups, because the calculated blocksize is much much larger...