
I've heard that there is a mysql backend in development for persistent. what is its current status? before you start telling me how postgres is better, let me say that I know it is. but, the AWS RDS implementation of MySQL is very compelling and kind of makes up for the differences.. Max

it doesn't compile. Perhaps a day's work to get it going. Postgresql and
sqlite both use a GenericSql module that MySQL can also use. Either way,
there are already example SQL backends. You are the first haskeller I have
seen show interest in MySQL for a new project, which is the main reasons why
that backend is bit-rotting.
On Tue, Apr 26, 2011 at 1:17 AM, Max Cantor
I've heard that there is a mysql backend in development for persistent. what is its current status? before you start telling me how postgres is better, let me say that I know it is. but, the AWS RDS implementation of MySQL is very compelling and kind of makes up for the differences..
Max _______________________________________________ web-devel mailing list web-devel@haskell.org http://www.haskell.org/mailman/listinfo/web-devel

I vote for ODBC support with GenericSql. ODBC will allow many other databases support. 26.04.2011 19:02, Greg Weber wrote:
it doesn't compile. Perhaps a day's work to get it going. Postgresql and sqlite both use a GenericSql module that MySQL can also use. Either way, there are already example SQL backends. You are the first haskeller I have seen show interest in MySQL for a new project, which is the main reasons why that backend is bit-rotting.
On Tue, Apr 26, 2011 at 1:17 AM, Max Cantor
mailto:mxcantor@gmail.com> wrote: I've heard that there is a mysql backend in development for persistent. what is its current status? before you start telling me how postgres is better, let me say that I know it is. but, the AWS RDS implementation of MySQL is very compelling and kind of makes up for the differences..

Let me clarify three things:
1) I started on a MySQL backend a while ago, and didn't do much work on it.
2) Someone else (I won't point fingers, he can stand up himself if he
chooses) is working on a MySQL backend.
3) Adding arbitrary backends is very easy, *except* for migration
support. So it's within the realm of reason to create a HDBC-generic
backend, sans migration support.
In fact, the PostgreSQL backend is basically an HDBC backend already;
if this is a feature people will actually use, let's do it.
Michael
On Tue, Apr 26, 2011 at 7:33 PM, Nikitiskiy Dmitriy
I vote for ODBC support with GenericSql. ODBC will allow many other databases support.
26.04.2011 19:02, Greg Weber wrote:
it doesn't compile. Perhaps a day's work to get it going. Postgresql and sqlite both use a GenericSql module that MySQL can also use. Either way, there are already example SQL backends. You are the first haskeller I have seen show interest in MySQL for a new project, which is the main reasons why that backend is bit-rotting.
On Tue, Apr 26, 2011 at 1:17 AM, Max Cantor
mailto:mxcantor@gmail.com> wrote: I've heard that there is a mysql backend in development for persistent. what is its current status? before you start telling me how postgres is better, let me say that I know it is. but, the AWS RDS implementation of MySQL is very compelling and kind of makes up for the differences..
_______________________________________________ web-devel mailing list web-devel@haskell.org http://www.haskell.org/mailman/listinfo/web-devel

Hello, web-devel :) On Tue, 2011-04-26 at 19:50 +0300, Michael Snoyman wrote:
Let me clarify three things:
1) I started on a MySQL backend a while ago, and didn't do much work on it. 2) Someone else (I won't point fingers, he can stand up himself if he chooses) is working on a MySQL backend.
Standing up :P I'm going to get on with it this week; hopefully I'll have something to show for it in a few days.
3) Adding arbitrary backends is very easy, *except* for migration support. So it's within the realm of reason to create a HDBC-generic backend, sans migration support.
Indeed, migration is the tedious bit. Migration in Postgre is easier than MySQL.
In fact, the PostgreSQL backend is basically an HDBC backend already; if this is a feature people will actually use, let's do it.
Are you still preferring the native binding route (ala persistent-sqlite)? This is the route I have gone for the MySQL binding.
Michael
On Tue, Apr 26, 2011 at 7:33 PM, Nikitiskiy Dmitriy
wrote: I vote for ODBC support with GenericSql. ODBC will allow many other databases support.
26.04.2011 19:02, Greg Weber wrote:
it doesn't compile. Perhaps a day's work to get it going. Postgresql and sqlite both use a GenericSql module that MySQL can also use. Either way, there are already example SQL backends. You are the first haskeller I have seen show interest in MySQL for a new project, which is the main reasons why that backend is bit-rotting.
On Tue, Apr 26, 2011 at 1:17 AM, Max Cantor
mailto:mxcantor@gmail.com> wrote: I've heard that there is a mysql backend in development for persistent. what is its current status? before you start telling me how postgres is better, let me say that I know it is. but, the AWS RDS implementation of MySQL is very compelling and kind of makes up for the differences..
_______________________________________________ web-devel mailing list web-devel@haskell.org http://www.haskell.org/mailman/listinfo/web-devel
_______________________________________________ web-devel mailing list web-devel@haskell.org http://www.haskell.org/mailman/listinfo/web-devel

On Wed, Apr 27, 2011 at 1:55 PM, Blake Rain
Hello, web-devel :)
On Tue, 2011-04-26 at 19:50 +0300, Michael Snoyman wrote:
Let me clarify three things:
1) I started on a MySQL backend a while ago, and didn't do much work on it. 2) Someone else (I won't point fingers, he can stand up himself if he chooses) is working on a MySQL backend.
Standing up :P
I'm going to get on with it this week; hopefully I'll have something to show for it in a few days.
3) Adding arbitrary backends is very easy, *except* for migration support. So it's within the realm of reason to create a HDBC-generic backend, sans migration support.
Indeed, migration is the tedious bit. Migration in Postgre is easier than MySQL.
In fact, the PostgreSQL backend is basically an HDBC backend already; if this is a feature people will actually use, let's do it.
Are you still preferring the native binding route (ala persistent-sqlite)? This is the route I have gone for the MySQL binding.
Yes, that's still the preferred route. I only mention having an HDBC-universal backend because it would be relatively painless to implement. Michael
Michael
On Tue, Apr 26, 2011 at 7:33 PM, Nikitiskiy Dmitriy
wrote: I vote for ODBC support with GenericSql. ODBC will allow many other databases support.
26.04.2011 19:02, Greg Weber wrote:
it doesn't compile. Perhaps a day's work to get it going. Postgresql and sqlite both use a GenericSql module that MySQL can also use. Either way, there are already example SQL backends. You are the first haskeller I have seen show interest in MySQL for a new project, which is the main reasons why that backend is bit-rotting.
On Tue, Apr 26, 2011 at 1:17 AM, Max Cantor
mailto:mxcantor@gmail.com> wrote: I've heard that there is a mysql backend in development for persistent. what is its current status? before you start telling me how postgres is better, let me say that I know it is. but, the AWS RDS implementation of MySQL is very compelling and kind of makes up for the differences..
_______________________________________________ web-devel mailing list web-devel@haskell.org http://www.haskell.org/mailman/listinfo/web-devel
_______________________________________________ web-devel mailing list web-devel@haskell.org http://www.haskell.org/mailman/listinfo/web-devel

AWS RDS is just very nice, otherwise i'd have no interest in mysql. is the most up to date mysql backend on github? max On Apr 26, 2011, at 11:02 PM, Greg Weber wrote:
it doesn't compile. Perhaps a day's work to get it going. Postgresql and sqlite both use a GenericSql module that MySQL can also use. Either way, there are already example SQL backends. You are the first haskeller I have seen show interest in MySQL for a new project, which is the main reasons why that backend is bit-rotting.
On Tue, Apr 26, 2011 at 1:17 AM, Max Cantor
wrote: I've heard that there is a mysql backend in development for persistent. what is its current status? before you start telling me how postgres is better, let me say that I know it is. but, the AWS RDS implementation of MySQL is very compelling and kind of makes up for the differences.. Max _______________________________________________ web-devel mailing list web-devel@haskell.org http://www.haskell.org/mailman/listinfo/web-devel

No work has gone on for the MySQL backend for a while, so yes, you will find
github up to date.
Any idea how RDS fared in the amazon meltdown last week? Anecdotally I heard
it had problems also. It is suspected that RDS is built on top of EBS [1].
EBS appeared to be the main problem and seems to be forsaken by cloud
architects now [2]. Recent events and previous experience seems to indicate
that replication across availability zones without EBS is actually more
durable.
[1]
http://stackoverflow.com/questions/5249842/how-does-amazon-rds-backup-snapsh...
[2] http://status.heroku.com/incident/151
On Tue, Apr 26, 2011 at 8:15 PM, Max Cantor
AWS RDS is just very nice, otherwise i'd have no interest in mysql. is the most up to date mysql backend on github?
max
On Apr 26, 2011, at 11:02 PM, Greg Weber wrote:
it doesn't compile. Perhaps a day's work to get it going. Postgresql and sqlite both use a GenericSql module that MySQL can also use. Either way, there are already example SQL backends. You are the first haskeller I have seen show interest in MySQL for a new project, which is the main reasons why that backend is bit-rotting.
On Tue, Apr 26, 2011 at 1:17 AM, Max Cantor
wrote: I've heard that there is a mysql backend in development for persistent. what is its current status? before you start telling me how postgres is better, let me say that I know it is. but, the AWS RDS implementation of MySQL is very compelling and kind of makes up for the differences.. Max _______________________________________________ web-devel mailing list web-devel@haskell.org http://www.haskell.org/mailman/listinfo/web-devel
participants (5)
-
Blake Rain
-
Greg Weber
-
Max Cantor
-
Michael Snoyman
-
Nikitiskiy Dmitriy