From: Willy
Subject: pg.lisp compile error on clisp
Date: 
Message-ID: <94059dac.0403110039.30ec61a5@posting.google.com>
Hi,

I have a problem doing (load "pg.lisp") from clisp 2.32. It says,

===
[1]> (load "defpackage.lisp")
;; Loading file defpackage.lisp ...
;; Loaded file defpackage.lisp
T
[2]> (load "sysdep.lisp")
;; Loading file sysdep.lisp ...
;; Loaded file sysdep.lisp
T
[3]> (load "pg.lisp")
;; Loading file pg.lisp ...
*** - invalid byte #xF8 in CHARSET:ASCII conversion
Break 1 [4]>

===

I am not sure what does it mean? can anyone help? I am using pg-0.19.

Thanks in advance.


- wil

From: Eric Marsden
Subject: Re: pg.lisp compile error on clisp
Date: 
Message-ID: <wzir7vz7n3e.fsf@melbourne.laas.fr>
>>>>> "wt" == Willy  <··········@yahoo.com> writes:

  wt> [3]> (load "pg.lisp")
  wt> ;; Loading file pg.lisp ...
  wt> *** - invalid byte #xF8 in CHARSET:ASCII conversion
  wt> Break 1 [4]>

see <URL:http://clisp.sourceforge.net/faq.html#enc-err>. You probably
want to start clisp with a "-E latin1" commandline option.
  
  wt> I am using pg-0.19.

some development news: pg-dot-lisp is now hosted at common-lisp.net.
Peter Van Eynde has contributed an implementation of the new
frontend/backend protocol that is supported by PostgreSQL 7.4. This
provides a number of new features:

  - prepare/bind/execute support at the protocol level: these prebound
    queries with placeholder values improve efficiency for certain
    use-cases 

  - better error reporting which should make it possible to improve
    the mapping of PostgreSQL errors onto Common Lisp conditions

Note that the new protocol has been introduced in a
backward-compatible manner, and v7.4 backends are still able to talk
the old version of the protocol, so pg-dot-lisp v0.19 works fine with
PostgreSQL 7.4, though without profiting from these benefits.

I expect that there will be a new release that supports both protocol
version within the next few weeks.

   <URL:http://www.common-lisp.net/project/pg/>

-- 
Eric Marsden                          <URL:http://www.laas.fr/~emarsden/>
From: Stefan Scholl
Subject: Re: pg.lisp compile error on clisp
Date: 
Message-ID: <c2pnpd$g4$1@news.brave.de>
Eric Marsden wrote:
>   - prepare/bind/execute support at the protocol level: these prebound
>     queries with placeholder values improve efficiency for certain
>     use-cases 

And minimize security problems with so called SQL injections.
From: Peter Van Eynde
Subject: Re: pg.lisp compile error on clisp
Date: 
Message-ID: <c2q375$e2f$1@mustyr.pvaneynd.debian.net>
On 2004-03-11, Stefan Scholl <·············@brave.de> wrote:
> Eric Marsden wrote:
>>   - prepare/bind/execute support at the protocol level: these prebound
>>     queries with placeholder values improve efficiency for certain
>>     use-cases 
>
> And minimize security problems with so called SQL injections.

My main reason for doing it. 

Groetjes, Peter

-- 
It's logic Jim, but not as we know it. | ········@debian.org
"God, root, what is difference?" - Pitr| http://people.debian.org/~pvaneynd/
"God is more forgiving." - Dave Aronson| 
From: Alexander Schreiber
Subject: Re: pg.lisp compile error on clisp
Date: 
Message-ID: <slrnc51k8o.tbr.als@mordor.angband.thangorodrim.de>
Stefan Scholl <·············@brave.de> wrote:
> Eric Marsden wrote:
>>   - prepare/bind/execute support at the protocol level: these prebound
>>     queries with placeholder values improve efficiency for certain
>>     use-cases 
> 
> And minimize security problems with so called SQL injections.

Right now, I'm catching that in my own code above pg.lisp, escaping for
PostgreSQL isn't that difficult.

Regards,
        Alex.
-- 
"Opportunity is missed by most people because it is dressed in overalls and
 looks like work."                                      -- Thomas A. Edison