On 10/21/2016 05:35 AM, Martin Tschierschke wrote:
On Thursday, 20 October 2016 at 21:25:50 UTC, Nick Sabalausky wrote:
Minor update to mysql-native: A client driver for MySQL/MariaDB
written natively in D from scratch via the published protocol specs,
with no dependency on the C MySQL client library. Supports either
Phobos or Vide.d sockets (works with or without Vibe.d).
https://github.com/mysql-d/mysql-native
DUB: http://code.dlang.org/packages/mysql-native
In v0.1.7:
- New: Test suite automatically tests with both Vibe and Phobos
sockets, not just Phobos. (@Abscissa)
- Change: Drop support for DMDFE 2.066.1 and below. Compiles on DMDFE
2.067.1 through 2.072.0.
- Fixed: Fix an import deprecation message for DMD 2.071. (@Abscissa)
- Fixed: #57: Added support for passing null parameters in prepared
statements by using Variant(null) (@machindertech)
- Fixed: #63/#69: Add escape module to package import (@Marenz)
- Fixed: #68: Update alias syntax (@Marenz)
Full changelog:
https://github.com/mysql-d/mysql-native/blob/master/CHANGELOG.md
I know I've said docs and examples are a top priority for
mysql-native, but it's become clear that the API is in serious need of
a refresh, so for the most part, I've held off on the docs this time
to avoid wasted effort documenting to-be-deprecated interfaces.
Unless any pressing issues appear, expect the next release to be an
experimental branch showcasing a beta of a new API. If possible, I'd
like to have at least one release where the old API is still
functional, but deprecated.
Please consider to look at the APIs of the other mysql libs, maybe there
is one
which you can use. I found it easier to use the APIs of mysql-d
(https://github.com/paxa/mysql.d) but mysql-lited
(https://github.com/eBookingServices/mysql-lited) looks more enhanced.
Hmm, actually, there appears to be a lot of similarity with part of what
I had in mind:
https://github.com/mysql-d/mysql-native/issues/83
In particular, I *really* want to pretty much get rid of the Command
struct. Minimizing heap activity is also a goal, albeit separate from
this API refresh.
But here too:
Documentation is not ready yet
I would be very happy, if in the end there would be only one superior
mysql-lib for D, so
all can work together to improve this one.
mysql-native gets very few PRs. :(
