From owner-chemistry@ccl.net Wed Nov 24 11:46:00 2010
From: "FyD fyd%%q4md-forcefieldtools.org" <owner-chemistry a server.ccl.net>
To: CCL
Subject: CCL:G: Gaussian output problem
Message-Id: <-43215-101124113229-24546-gNptjfO0p7SLTJECOEeF/Q a server.ccl.net>
X-Original-From: FyD <fyd-*-q4md-forcefieldtools.org>
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
Content-Type: text/plain;
	charset=ISO-8859-1;
	DelSp="Yes";
	format="flowed"
Date: Wed, 24 Nov 2010 17:31:53 +0100
MIME-Version: 1.0


Sent to CCL by: FyD [fyd()q4md-forcefieldtools.org]
Dear Maciek,

I do not think **** is a bug; simply in the format you have the room  
for only 4 characters...

We encountered the same "problem" (not really one) in the Gaussian  
outputs related to molecular electrostatic potential computation.  
Using Perl, R.E.D. http://q4md-forcefieldtools.org/RED/ (you can  
download the code it is under the GNU General Public License) handles  
that easily using the code below (the code handles $NM molecules; $NC  
conformation & $w reorientations)

- we first calculate the number of MEP points to define the max.  
number of points
    if(/ESP FIT Center/ig){ $nbpoint[$NM]++;}
- then using this max. number of points you can do whatever you need;  
either save in another file with another format or re-build another  
Gaussian output...

I am sure you could follow the same idea: you simply need to find the  
correct filter...

regards, Francois


for($NC=1;$NC <=$nbconf[$NM];$NC++){
   for($w=1;$w<=$nbmod[$NM];$w++){
     $nbpoint[$NM]=0;
     open(JOB2LOG,"<JOB2-gau_m$NM-$NC-$w.log");
     foreach (<JOB2LOG>){ if(/ESP FIT Center/ig){ $nbpoint[$NM]++;} }
     close(JOB2LOG);
     format_name ESPO "HEADER";
     open(ESPO,">espot_m$NM-$NC-$w");
     write ESPO;
     $flag1=$flag2=$i=0;
     open(JOB2LOG,"<JOB2-gau_m$NM-$NC-$w.log");
     foreach (<JOB2LOG>){
        if((/Electrostatic Properties /ig)&&(/Atomic Units/ig)){ $flag1=1; }
        if($flag1 == 1){ $flag2++; }
        if(($flag2 > $nbatoms[$NM]+6)&&($flag2 <  
$nbpoint[$NM]+$nbatoms[$NM]+7)){
           ($charge[$i][$NM])=(split(' '))[2]; $i++;
        }
     }
     close(JOB2LOG);
     $i=0;
     open(JOB2LOG,"<JOB2-gau_m$NM-$NC-$w.log");
     foreach (<JOB2LOG>){
        if(/^       Atomic Center/ig) {
           ($x,$y,$z) = unpack("x32 A10 A10 A10",$_);
           $x=$x/0.529177249;
           $y=$y/0.529177249;
           $z=$z/0.529177249;
           printf ESPO ("                %16.7E%16.7E%16.7E\n",$x,$y,$z); }
           if(/^      ESP Fit Center/ig) {
              ($x,$y,$z) = unpack("x32 A10 A10 A10",$_);
              $x=$x/0.529177249;
              $y=$y/0.529177249;
              $z=$z/0.529177249;
              printf ESPO  
("%16.7E%16.7E%16.7E%16.7E\n",$charge[$i][$NM],$x,$y,$z);
              $i++; }
           }
           close(JOB2LOG);
           close(ESPO);
     }
}

> could you help me with one problem, please? Every time I'm running   
> G03 job I got problem with outputs. The problem is that I cannot   
> extract structures because there are stars **** instead of atom   
> numbers higher than 9999. Have you ever had this problem in your   
> calculation if you had more than 9999 atoms in the system?
>
> It looks like this in output:
>  9995          1      20001030      -12.655571   46.600810   32.540627
>  9996          1      20001030      -12.533484   47.134581   33.952118
>  9997          8      20001021        2.441067  -19.529269   31.331119
>  9998          1      20001030        2.813824  -19.307495   32.184450
>  9999          1      20001030        1.933916  -20.325103   31.491166
>  ****          8      20001021       37.587292  -11.572024   28.812192
>  ****          1      20001030       37.333847  -11.927328   27.960694
>  ****          1      20001030       38.095189  -12.272395   29.221488
>  ****          8      20001021       23.735956  -43.072924   13.135294
>  ****          1      20001030       23.388276  -42.182008   13.115359
>  ****          1      20001030       22.974070  -43.624340   13.312930
>
> Do you have any suggestions?
> Thanks in advance