UNIX for Dummies Questions & Answers

Hi UNIX Members,

I've been tasked with performing the following:
Extract a block of data in column form
#This data changes each time, therefore automating future procedures

Please Note the following:
line = reading a line from a file_list that leads to the data
The filename is called mol_1($line).opt.out, the block of text is in the opt.out file

So far I have:
Set variables to assign line numbers the text I require lies between ( $ORB and $MULL)
######I would like to extract the text between these two variables#####
I'd like to make a seperate file 'HOMO'(Highest Occupied Molecular Orbital) in which is a temporary data file which I can work with.

Scope of file : $ORB = 18172, $MULL = 18278
Line numbers needed to extract is 18172-18278 (however these change therefore are created as variables)



The current script I have is:

Code:

#!/bin/bash -f
while read line
do

#mkdir HOMO
#cd ./B3LYP_631Gstar2.$line

OBEGIN=$(cat -n $line.opt.out | grep "ORBITAL ENERGIES" | tail -1 | cut -f 1)
OEND=4
ORB=$(( $OBEGIN + $OEND ))
#echo $ORB

MBEGIN=$(cat -n $line.opt.out | grep "MULLIKEN POPULATION ANALYSIS" | tail -1 | cut -f 1)
MEND=3
MULL=$(( $MBEGIN - $MEND ))

###### Works up till here, above is to show what i'm working with####
###Below is the 3 methods i've tried

#sed '$ORB,$MULL!d' $line.opt.out > HOMO
#awk 'NR==$ORB,NR==$MULL' $line.opt.out | cut -f1  > HOMO
#grep -E '($ORB: ){5}$MULL' < $line.opt.out > HOMO

#echo $MULL
done<$1

~

Summary of task: Extracting a block of text from a large file using variables $ORB and $MULL as numbers

#######EXTRA#####
If you know how to extract data that is in columns ( Say I need only column 2 and column 4 out of 6 columns) to be in columns in another file (sort of like an excel file) that's be much appreciated!


Warm Regards, Klor

Hello,

Could you please confirm this is not a homework, if yes then kindly post the query accordingly. If no, then kindly let us know the input and expected output on same.



Thanks,
R. Singh

Hi Mr.Singh
No this isn't homework, it's a way to automate my tasks and learn scripting whilst working on computer modelling. I am a Chemistry student at a Third level in university, working on computational modelling and my Lecturer proposed this idea as an 'extra learning opportunity'

####It is not graded##

The input file is large but the arrangement follows:

Code:

ORBITAL ENERGIES

Title1      Title2      Title3      Title4      Title5
Data1      Data2      Data3      Data4      Data5
Data1.1    Data2.1   Data3.1   Data4.1   Data5.1
Data1.2    Data2.2   Data3.2   Data4.2   Data5.2

MULLIKEN POPULATION ANALYSIS

the expected output is:

Code:

Data2      Data 3    Data 4
Data2.1   Data3.1   Data4.1

Warm Regards,
Jason Deacon

What would be the parameters to apply on the sample input to get this output?

Hello Klor,

Following could help you.

EDIT: This should help in all in all data input file.

Code:

awk '/ORBITAL ENERGIES/ {getline;getline;getline;{a=$2" "$3" "$4};getline;{b=$2" "$3" "$4};getline;{c=$2" "$3" "$4};getline} END{print a ORS b ORS c}' check_parameters

Output will be as follows.

Code:

Data2 Data3 Data4
Data2.1 Data3.1 Data4.1
Data2.2 Data3.2 Data4.2

NOTE: Following code will work only for given input in thread.

Code:

awk 'NR==4 || NR==5 {print $2" "$3" "$4}' check_parameters

Output will be as follows.

Code:

Data2 Data3 Data4
Data2.1 Data3.1 Data4.1

NOTE: Where check_parameters is the Input file name.

Thanks,
R. Singh

Thank you Mr.Singh,
I apologize that I didn't evaluate more.
The input file has many ORBITAL ENERGIES as titles which is why I used tail -1 when cutting the line number, as it is the last one I am interested in.

Also the data line will vary so sometimes i may data.2.5+4.5 ->2.9+4.9 lines, sometimes it may be 2.10+4.10->2.15+4.15 lines so the variables need to change and be integrated into the code as they won't be fixed numbers each time.

@bartus
the parameters are as followed:
#line is defined as mol1 (molecule 1) but it will soon include 3-7 names in columnic order and read the first line then the second as a loop
Anything is # is excluded for now until I can piece together a functioning script.

OBEGIN = Searching to the bottom of the file (text) and extracting the line number from this position (cut -f 1)
OEND = amount of lines until the numeric data starts (important)
therfore ORB = Starting line of important numeric data

MBEGIN = extracting the line number of the next title in the text file
MEND = How many lines until we reach the numeric data
MULL = As it is the next title we must go in reverse and take numbers away therefore MULL is the line number of final numeric data

the #awk #sed #grep afterward are my failed attempts at inputting where fixed numbers would be okay however variables don't work for some reason

the #echos were there to make certain the variables had the correct numbers.

The input file has

~18000lines of text (including 4 other ORBITAL ENERGIES and MULL titles)
ORBITAL
data
MULLIKEN POPULATION ANALYSIS
~500 lines of text

The desired output is

All Datalines,

#######I can manage the below after i extract the data lines####

inwhich I can then do the same to created variables from when
Data = 2.0000
Data = 2.0000
Data = 0.0000
Data = 0.0000
when 2 goes to 0 i will be able to extract those lines using grep

and therefore only have
data = 2.000
data = 0.000

lines in the final file therefore create a graph

Warm Regards, Jason

I think you may be looking for something like this:

Code:

awk '
  $0~ostart{
    c=0
  }
  $0~ostart,$0~mstart{
    A[++c]=$0
  }
  END{
    for(i=oend; i<=c-mend; i++) print A[i]
  }
' ostart="ORBITAL ENERGIES" oend=3 mstart="MULLIKEN POPULATION ANALYSIS" mend=3 file