Skip to content

New package RPostgreSQL 0.1.0

5 messages · Dirk Eddelbuettel, Brian Ripley, Bo Zhao +1 more

Original
Brian Ripley
# 
Following up the Windows point on a specialized list (R-sig-db), I was 
able to get this to build with

setenv PG_HOME D:/Program Files/PostgreSQL/8.3

and a src/Makevars.win containing

PKG_CPPFLAGS=-I"$(PG_HOME)"/include
PKG_LIBS=-L"$(PG_HOME)"/lib -lpq

The problem is that you do need a fairly complete Windows install of 
PostgreSQL, and I had to go back and add the development files.

Then you will need libpq.dll in your path: that is in (PG_HOME/bin). The 
tests/*.R did not work, but that's a permissions issue (as for me on 
Linux) and I needed to specify user="postgresql" and password="...". 
Then minimal tests worked suitably.

I'll put a Windows binary on CRAN extras, at least for now.
On Tue, 14 Oct 2008, Dirk Eddelbuettel wrote:

            

      
      
      
    

        

      
      
    

        

  
    

      
      
    

  


  


    

    

      
      

        


  

        

      Dirk Eddelbuettel
    

    

      
      #
    

  


  

      
On 17 October 2008 at 12:42, Prof Brian Ripley wrote:

        
| Following up the Windows point on a specialized list (R-sig-db), I was 
| able to get this to build with
| 
| setenv PG_HOME D:/Program Files/PostgreSQL/8.3
| 
| and a src/Makevars.win containing
| 
| PKG_CPPFLAGS=-I"$(PG_HOME)"/include
| PKG_LIBS=-L"$(PG_HOME)"/lib -lpq

Thank you very much! 

I also learned from Jan that the simple (but too simple) configure.in that
avoids pg_config (which somehow wasn't in the path on Sameer's FC machine)
leads to malfunction on OS X, so I have to fix that too.

Do your FC / RH system all have pg_config in the path ?
 
| The problem is that you do need a fairly complete Windows install of 
| PostgreSQL, and I had to go back and add the development files.

Right. I guess I need to document that. Exactly what file did you fetch?

| Then you will need libpq.dll in your path: that is in (PG_HOME/bin). The 
| tests/*.R did not work, but that's a permissions issue (as for me on 
| Linux) and I needed to specify user="postgresql" and password="...". 
| Then minimal tests worked suitably.

Because we cannot assume a Pg backend to be present and to be configured, we
cannot run tests automagically. See inst/devTests in the sources, or
devTests/ in the installed package for some more tests -- not the most
refined set, but it grew as we grew the package.
 
| I'll put a Windows binary on CRAN extras, at least for now.

Many thanks for that, and thanks also for following up on r-sig-db!

Dirk

        
| On Tue, 14 Oct 2008, Dirk Eddelbuettel wrote:

        
| 
| >
| >       RPostgreSQL version 0.1.0
| >
| > We are pround to announce the availability of the RPostgreSQL package on CRAN
| > and its mirrors.  This package provides an a DBI-compliant interface between
| > PostgreSQL and R.
| >
| > RPostgreSQL was developed as part of the Google Summer of Code 2008 program
| > by Sameer Kumar Prayaga.
| >
| > Some highlights:
| >
| >  o  Implemented all the DBI features
| >
| >  o  Added support for transaction management
| >
| >  o  Added type mapping for dates & timestamps
| >
| > Known bugs/deficiencies:
| >
| >  o  Building on Windows unknown/untested. Feedback welcome
| >
| > RPostgreSQL is hosted on Google Code, for more information see
| >
| >   http://rpostgresql.googlecode.com/
| >
| > For any suggestions and queries, please contact:
| >
| >   Sameer Kumar Prayaga <sameer.bits at gmail.com>
| >   Dirk Eddelbuettel <edd at debian.org>
| >
| > -- 
| > Three out of two people have difficulties with fractions.
| >
| > _______________________________________________
| > R-packages mailing list
| > R-packages at r-project.org
| > https://stat.ethz.ch/mailman/listinfo/r-packages
| >
| > ______________________________________________
| > R-help at r-project.org mailing list
| > https://stat.ethz.ch/mailman/listinfo/r-help
| > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
| > and provide commented, minimal, self-contained, reproducible code.
| >
| 
| -- 
| Brian D. Ripley,                  ripley at stats.ox.ac.uk
| Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
| University of Oxford,             Tel:  +44 1865 272861 (self)
| 1 South Parks Road,                     +44 1865 272866 (PA)
| Oxford OX1 3TG, UK                Fax:  +44 1865 272595

  
    

      
      
    

  


  


      

    

    

      
      

        


  

        

      Brian Ripley
    

    

      
      #
    

  


  

      
On Fri, 17 Oct 2008, Dirk Eddelbuettel wrote:

        

            

      
      
      
    

        
Those which have development files on.  (We have some production 
PostgreSQL systems without it, since the CMS for our websites runs on 
PostgreSQL.)

gannet% whereis pg_config
pg_config: /usr/bin/pg_config /usr/include/pg_config.h 
/usr/share/man/man1/pg_config.1.gz
gannet% rpm -q --whatprovides /usr/bin/pg_config
postgresql-devel-8.2.10-1.fc8

so it is part of the -devel RPM as one might expect from RH.

            

      
      
      
    

        
It's part of the standard .msi installer, but you have to select 
non-default options, the Development group (only to do a source build of 
RPostgreSQL, as far as I can tell).  It *should* be easy to go back and 
add those options, but on my box the installation path had been lost so I 
actually had to uninstall and reinstall.

            

      
      
      
    

        
Sure, but even if it is, a standard setup (even on Windows) will require 
both a user and a password (and the only user you can guarantee is present 
is "postgresql").  I was just warning people that failure is expected.

            

      
      
      
    

        

      
      
    

        

  
    

      
      
    

  


  


      

    

    

      
      

        


  

        

      Bo Zhao
    

    

      
      #
    

  


  

      
it is a exciting news that the RPostgreSQL 1.0 released.

On Fri, Oct 17, 2008 at 8:44 AM, Prof Brian Ripley

        
<ripley at stats.ox.ac.uk> wrote:

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

      
      
    

        

  
    

      
      
    

  


  


      

    

    
3 days later