Re: PLV8JS - Mailing list pgsql-hackers

From Daniel Farina
Subject Re: PLV8JS
Date
Msg-id CAAZKuFYELQRG-CBorTH2_8NrgTFgOj52ycQ1eix_XhntPaKD4g@mail.gmail.com
Whole thread Raw
In response to PLV8JS  (Milton Labanda <1000ton.lab@gmail.com>)
List pgsql-hackers
On Fri, Sep 21, 2012 at 3:14 PM, Milton Labanda <1000ton.lab@gmail.com> wrote:
> Hi friends, wich context is the apropiate to install this plugin?
> I have
>
> Ubuntu 11
> x64 architecture
> postgresql 9,2
> libv8-3.1  (system package)
>
> but not get install it. Some ideas?

Well, I'm not sure if pgsql-hackers is quite the right mailing list
for this (there's a plv8 mailing list), but because plv8 is probably
going to be so instrumental to many profitable uses I can help you
address it quickly, and maybe some other denizens of -hackers will
have some opinions, especially because this general problem is going
to surface more frequently as number of extensions mushroom.

Until a distribution/operating system integrator commits to packaging
some version of plv8 and/or v8, I recommend employing static linking
of plv8 to a self-downloaded libv8.   This lets you handle multiple
versions of plv8 and libv8 on the same system (but  different database
installs) and lets you get the latest and greatest v8s at-will.

Alternatively, you may be determined to get the plv8 build to work
with your OS of choice, even if they do not have packages.  I think
that's a longer discussion that'd need to happen at a plv8 mailing
list.

In general, I think if there is to be a guideline for Postgres
extensions it should be to enable static linking of dependencies or
perhaps a more detailed guideline involving enabling self-contained
concurrent installs of dynamically linked extension dependencies.  I
am currently in the depths of dependency hell involving some of the
GIS libraries and PostGIS and really, really wish I could do either of
these.

There is a relevant discussion here:

https://code.google.com/p/plv8js/issues/detail?id=36#makechanges

-- 
fdr



pgsql-hackers by date:

Previous
From: Fabrízio de Royes Mello
Date:
Subject: Re: CREATE SCHEMA IF NOT EXISTS
Next
From: Gibheer
Date:
Subject: Re: 9.2 Cascading replication after slave promotion