MySQL Bugs: #31897: tunneling a mysql connection through http(s)

Bug #31897	tunneling a mysql connection through http(s)
Submitted:	27 Oct 2007 23:49	Modified:	5 Nov 1:46
Reporter:	Don Cohen	Email Updates:
Status:	Won't fix	Impact on me:	None
Category:	Connector / J	Severity:	S4 (Feature request)
Version:		OS:	Any
Assigned to:	Assigned Account	CPU Architecture:	Any

Description:
I'm trying to run an applet inside a network that does not allow direct connections to my server on the internet.  It looks like there is no general proxy available, but there is a proxy for http/https.  I imagine I'm not the first to want to do this, so I thought I'd start by asking whether this is, by any chance, already supported in the connector.
If not, has it been requested before?  Is it in development, or in the queue?
If none of the above, perhaps you could give me some advice about how to do it on my own?  How long would you expect it to take?

Note that I'm trying to live within the restrictions of an applet, so I don't want to run any other applications on the client.

I see at http://idssoftware.com/jdbchttps.html something that looks similar to what I want, but I assume this is a totally different driver, and I expect that changing drivers would be a huge deal - almost like changing databases - in terms of having to change the software because of differences in the drivers.  Do you have any idea how true or false this is?  I.e, are all jdbc drivers really the same or would you expect things to work differently on different ones?

How to repeat:
since this is a feature request there are no instructions for repeating it

Hi Don and thanks for your report.

Taking a quick glance over the latest code I do not find this implemented thus verified as described.

Mark?

I played around this a long time ago, and it's very, very slow, because of the nature of the mysql protocol (or any thing that isn't batch-y enough).

It seems to make more sense to create a RESTful web service where the granularity as at the domain object level and push the persistence concerns to your server, and let the applet interact via HTTP with the domain object model, or to pull the granularity up higher and build actual services?

It's interesting to see a reply out of the blue after almost 2 years.
I did end up using ids software and yes it was a major problem to convert.
It's slower but not enormously so.

I don't understand the second reply.  What's a RESTful web service?
What do you mean by granularity at domain object level?  What's an actual
service?  How are these things supposed to help?

Also, SQL seems pretty batch-y to me.

Feature not in the Connector/J development plan.