text-1.2.3.1

* Make `decodeUtf8With` fail explicitly for unsupported non-BMP
  replacement characters instead silent undefined behaviour (haskellgh-213)

* Fix termination condition for file reads via `Data.Text.IO`
  operations (haskellgh-223)

* A serious correctness issue affecting uses of `take` and `drop` with
  negative counts has been fixed (haskellgh-227)

* A bug in the case-mapping functions resulting in unreasonably large
  allocations with large arguments has been fixed (haskellgh-221)

text-1.2.3.0

* Spec compliance: `toCaseFold` now follows the Unicode 9.0 spec (updated from 8.0).

* Bug fix: the lazy `takeWhileEnd` function violated the
  [lazy text invariant](https://github.com/bos/text/blob/1.2.3.0/Data/Text/Internal/Lazy.hs#L51)
  (haskellgh-184).

* Bug fix: Fixed usage of size hints causing incorrect behavior (haskellgh-197).

* New function: `unsnoc` (haskellgh-173).

* Reduce memory overhead in `encodeUTF8` (haskellgh-194).

* Improve UTF-8 decoder error-recovery (haskellgh-182).

* Minor documentation improvements (`@since` annotations, more examples, clarifications).

Update changelog

Bump version to 1.2.2.1

Default integer-pure build flag to False

Added tag 1.2.1.2 for changeset 40f268e67357

Bump version to 1.2.1.2

Bump version to 1.2.1.1

Broke my brain :-(

Update changelog