Re: Backup DB ontape, fgl_xxx list and sqlcmd From Sam Leffler.
This message is in MIME format. The first part should be readable text,
while the remaining parts are likely unreadable without MIME-aware tools.
Send mail to mime@docserver.cac.washington.edu for more info.
---559023410-851401618-879353231=:11347
Content-Type: TEXT/PLAIN; charset=US-ASCII
On Wed, 12 Nov 1997, Christian Hernoux wrote:
> About Q2:
> >No. But I attach some code and a manual page that gives you a C
> >implementation that could be given an I4GL-callable interface
> >with a minimum of difficulty.
> I haven't no files attached to your E-MAil. Problems ?
Oops! Forgetfulness, I think...
Sorry.
Yours,
Jonathan Leffler (johnl@informix.com) #include <witticism.h>
> -----Message d'origine-----
> De : Jonathan Leffler <johnl@informix.com>
> From : Christian Hernoux <cnhx@polisoft.fr>
> Cc : Informix List <informix-list@rmy.emory.edu>
> Date : mardi 11 novembre 1997 23:37
> Objet : Re: Backup DB ontape, fgl_xxx list and sqlcmd From Sam Leffler.
>
> >On Tue, 11 Nov 1997, Christian Hernoux wrote:
> >
> >> Q1 :
> >> ====
> >> In a dbspace I have 3 databases.
> >> When I backup a dbspace, I save these databases.
> >> If I have a problem on a database is it possible to backup this database
> and
> >> restore it ?
> >> (dbunload or dbexport on tape drive ? some examples ?)
> >> I work with Informix ON-Line 5.01 on SCO with a 2GB tape drive.
> >
> >OnLine 5.x does not support dbspace-level restores, let alone
> >database-within-dbspace restores. You would have to upgrade to
> >7.2x, I think, to get that functionality.
> >
> >> Q2 :
> >> ====
> >> With the fgl_getenv("xxx") I get back an environment variable.
> >> does it exist a similar function to set an environment variable ?
> >
> >No. But I attach some code and a manual page that gives you a C
> >implementation that could be given an I4GL-callable interface
> >with a minimum of difficulty.
> >
> >> How can I get the list of fgl_xxx function and their description ?
> >
> >Manuals? I've not checked, but there aren't all that many.
> >
> >> Q3 :
> >> ====
> >> Where i can obtain the software : sqlcmd
> >> (a freeware SQL command interpreter from Sam Leffler)
> >
> >Maybe I should change my name by deed poll.
> >
> >http://www.iiug.org/software/
> >
> >Yours,
> >Jonathan Leffler (johnl@informix.com) #include <witticism.h>
> >
> >
>
>
>
---559023410-851401618-879353231=:11347
Content-Type: TEXT/PLAIN; charset=US-ASCII; name="setenv.c"
Content-Transfer-Encoding: BASE64
Content-ID: <Pine.GSO.3.96.971112084711.11347B@anubis.mp.informix.com>
Content-Description: C Source Code
/*
@(#)File: $RCSfile: setenv.c,v $
@(#)Version: $Revision: 1.5 $
@(#)Last changed: $Date: 1997/06/02 16:41:57 $
@(#)Purpose: setenv(3) -- an alternative to putenv(3)
@(#)Author: J Leffler
@(#)Copyright: (C) JLSS 1995,1997
@(#)Product: :PRODUCT:
*/
/*TABSTOP=4*/
/*
** SETENV(3) -- an alternative to putenv(3)
**
** The basic problems with putenv(3) are manifest in its definition.
** It is given a string which it makes into part of the environment.
** If you overwrite that string subsequently, you modify the environment.
** If the string was allocated and you subsequently reset the environment
** variable without modifying the previous string, you have a memory leak.
**
** setenv() deals with these problems in the following ways:
** 1. It makes a copy of the string it is passed and places the copy in
** the environment.
** 2. It tracks the environment variables it has set and releases
** previously allocated versions of the variable.
**
** setenv() is implemented in two flavours. The default flavour is to
** completely replace putenv() by providing a putenv() entry point which is
** actually the same as setenv(). The alternative flavour continues to use
** putenv() to manage the environment but keeps separate records of what it
** has allocated.
*/
/* -- Include Files */
#include <string.h>
#include <stdlib.h>
#include <assert.h>
#include "setenv.h"
/* -- Constant Definitions */
#define SETENV_FAILURE -1
#define SETENV_SUCCESS 0
#define INCREMENT_SIZE 16
/* -- Configuration Definitions */
/* If defined conservative, override radical; else define radical */
#ifdef CONSERVATIVE_SETENV
#undef RADICAL_SETENV
#else
#define RADICAL_SETENV
#endif /* CONSERVATIVE_SETENV */
/* -- Macro Definitions */
#ifdef RADICAL_SETENV
#define PUTENV(val) fix_env(val)
#else
#define PUTENV(val) putenv(val)
#endif /* RADICAL_SETENV */
#ifdef NDEBUG
#define ENV_CHECK() ((void)0)
#else
#define ENV_CHECK() env_check()
#endif /* NDEBUG */
#ifdef PROFILING
#define NO_EXECUTE
#define PRINTF(x) ((void)0)
#else
#define PRINTF(x) printf x
#endif /* PROFILING */
/* -- Declarations */
extern char **environ;
#ifdef RADICAL_SETENV
static char **env_base = (char **)0;
static int env_size = 0;
static int env_used = 0;
#endif /* RADICAL_SETENV */
static char **alloc_base = (char **)0;
static int alloc_size = 0;
static int alloc_used = 0;
#ifndef lint
static const char rcs[] = "@(#)$Id: setenv.c,v 1.5 1997/06/02 16:41:57 johnl Exp $";
#endif
#ifdef RADICAL_SETENV
/* Environment comparison -- compare two environment entries by name */
static int env_compare(const char *s1, const char *s2)
{
char *eq;
const char *ref;
size_t len;
int rc;
/* One of the strings comes from the environment and contains an '=' */
ref = s1;
eq = strchr(ref, '=');
if (eq == (char *)0)
{
ref = s2;
eq = strchr(ref, '=');
}
assert(eq != (char *)0);
len = eq - ref;
/* Compare up to but excluding equals sign */
rc = strncmp(s1, s2, len);
if (rc == 0)
{
/* Check that we are dealing with distinct variable names */
if (s1[len] != '\\0' && s1[len] != '=')