I am trying to cat the last 2 columns of multiple text files side by side. The files are in a directory of various types of files. All files have >2 columns, but no guarantee all files have the same number of columns.
For example, if I have:
file1.txt
1 a b J H
2 b c E E
3 c d L L
4 d e L L
5 e f O O
file2.txt
1 a b M B
2 b c O E
3 c d O E
I want:
J H M B
E E O E
L L O E
L L
O O
The closest I've got is:
awk '{print $(NF-1), "\t", $NF}' *.txt
Which is almost what I want.
For the concatenation, I was thinking something like here for concatenation
pr -m -t one.txt two.txt
awk 'NR==FNR{a[NR]=$(NF-1)" "$NF;next}{print $(NF-1),$NF,a[FNR]}' file2.txt file1.txt
Tested:
> cat temp2
1 a b M B
2 b c O E
3 c d O E
> cat temp1
1 a b J H
2 b c E E
3 c d L L
4 d e L L
5 e f O O
> awk 'NR==FNR{a[NR]=$(NF-1)" "$NF;next}{print $(NF-1),$NF,a[FNR]}' temp2 temp1
J H M B
E E O E
L L O E
L L
O O
>
join -a1 -a2 one.txt two.txt | cut -d' ' -f4,5,8,9
Related
here a piece of code :
$> ls
` = _ ; ? ( ] # \ % 1 4 7 a B d E g H J l M o P r S u V x Y
^ > - : ' ) { $ & + 2 5 8 A c D f G i k L n O q R t U w X z
< | , ! " [ } * # 0 3 6 9 b C e F h I K m N p Q s T v W y Z
I'm printing all ASCII character, each element is a folder, and I'm trying to understand the default sorting order of the ls command.
I understand that's there is a case insensitive comparison to sort alphabetic character, with digit coming first.
I've some trouble to understand how special character are sorted, and I'm not able to find something clear. I was thinking it could be related to the ASCII table, but when we see how things are ordered it really make no sens with it... Where is this order coming from ?
Thanks
First, I would like to apologize for my extremely basic knowledge about coding. Then I hope that I will be able to express myself correctly about my issue. Do no hesitate to ask for further clarifications or anything else...
I'm encountering troubles postprocessing data...
My goal is to recombine data which were swapped.
EDIT : here is a .rar folder containing my test example which works and the one that I try to make working... (do not be afraid by the time it requires to process the data)
https://drive.google.com/file/d/1AEPUc8haT5_Z3LR3jnZZlpyfxhdDwwo6/view?usp=sharing
EDIT 2 : Here is what I expect on paper (Its my TestReorder3OK folder in my rar archive)
enter image description here
EDIT 3 : MINIMAL COMPLETE EXAMPLE
Script :
#!/bin/bash
# Definir le nombre de replica
NP=3
NP1=$[NP-1]
rm torder*
for repl in `seq 0 $NP1`
do
echo $repl
# colle la colonne 2 du fichier .lammps dans un fichier rep_0, puis dans la seconde boucle, la colonne 3 dans rep_1, etc.
awk -v rep=$repl '{r2=rep+2;print $r2}' < log.lammps > rep_$repl
i=0
j=0
# cree une boucle dans la boucle
for a in `cat rep_$repl`
do
i=$[i+1]
j=$[j+3]
head -$i screen.$repl.temp | tail -1 >> torder.$a
head -$j ccccd2_H_${repl}_col.bak2 | tail -3 >> ccccd2_H_${a}_temp_col.bak2
done
done
log.lammps file
1 0 1 2
2 1 0 2
3 1 2 0
Starting at column 2, this file contains the number associated to the inputs below. Here is an expanded explanation :
column 2 has three values : 0, 1 and 1 ; the 0 is associated to the first three lines of the file ccccd2_H_0_col.bak2, the next three ones are associated the 1 and the last three ones again to the value 1.
column 3 has also three values : 1, 0 and 2 ; the 1 is associated to the first three lines of the file ccccd2_H_1_col.bak2, the next three ones are associated the 0 and the last three ones again to the value 2.
Same story for column 4.
Now what I want, is that every set of three lines associated to the 0 value go into a single file. Every set of three lines associated to the 1 value go into another single file, and the sets of three lines associated to the 2 value to a last file.
Inputs :
ccccd2_H_0_col.bak2
blank line
N a b c
C d e f
N g h i
C j k l
N m n o
C p q r
ccccd2_H_1_col.bak2
blank line
N s t u
C v w x
N y z a
C b c d
N e f g
C h i j
ccccd2_H_2_col.bak2
blank line
N k l m
C n o p
N q r s
C t u v
N w x y
C z a b
Outputs : These are the desired outputs and the one that I get for simple test files
ccccd2_H_0_temp_col
blank line
N a b c
C d e f
N y z a
C b c d
N w x y
C z a b
ccccd2_H_1_temp_col
blank line
N g h i
C j k l
N m n o
C p q r
N s t u
C v w x
ccccd2_H_2_temp_col
blank line
N e f g
C h i j
N k l m
C n o p
N q r s
C t u v
This works fine on small test files (as shown here), but not on my real system. For my real system, I have the log.lammps file that contains 14 rows and 10,001 lines, and my input files that contain 121,121 lines (so 10,001 * block of 121 lines). It creates files 10 times larger with more data than it should.
Can you enlighten me about my issue ? I think this is linked to the difference of line number from my files containing a single row and the files containing cartesian coordinates, but I really don't understand the link nor the way to solve it...
Thank you in advance...
I think I understand what you're trying do do now and this GNU awk script (for ARGIND, ENDFILE and inbuilt open file management) will do it:
$ cat ../tst.awk
ARGIND == 1 {
for (inFileNr=2; inFileNr<=NF; inFileNr++) {
outFileNrs[inFileNr,NR] = $inFileNr
}
next
}
ENDFILE { RS = "" }
{ print ORS $0 > ("ccccd2_H_" outFileNrs[ARGIND,FNR] "_temp_col") }
Look:
INPUT:
$ ls
ccccd2_H_0_col.bak2 ccccd2_H_1_col.bak2 ccccd2_H_2_col.bak2 log.lammps
$ cat log.lammps
1 0 1 2
2 1 0 2
3 1 2 0
$ paste ccccd2_H_0_col.bak2 ccccd2_H_1_col.bak2 ccccd2_H_2_col.bak2 | sed 's/\t/\t\t/g'
N a b c N s t u N k l m
C d e f C v w x C n o p
N g h i N y z a N q r s
C j k l C b c d C t u v
N m n o N e f g N w x y
C p q r C h i j C z a b
SCRIPT EXECUTION:
$ awk -f ../tst.awk log.lammps ccccd2_H_0_col.bak2 ccccd2_H_1_col.bak2 ccccd2_H_2_col.bak2
OUTPUT:
$ ls
ccccd2_H_0_col.bak2 ccccd2_H_1_col.bak2 ccccd2_H_2_col.bak2 log.lammps
ccccd2_H_0_temp_col ccccd2_H_1_temp_col ccccd2_H_2_temp_col
$ paste ccccd2_H_0_temp_col ccccd2_H_1_temp_col ccccd2_H_2_temp_col | sed 's/\t/\t\t/g'
N a b c N g h i N e f g
C d e f C j k l C h i j
N y z a N m n o N k l m
C b c d C p q r C n o p
N w x y N s t u N q r s
C z a b C v w x C t u v
Hello i have a text containing pipe (|) symbol and i want to replace it with space. This is the text in the column description
|TrueCricketLover|M€$$!|
PTI|Capricorn|No DM|#TeamIK|#shaneRWatson33 ❤
Boom Boom❤
Striving to be a better human!
I have tried the regexp_replace function like this
regexp_replace(description,'|',' ')
This command returns this value
| T r u e C r i c k e t L o v e r | M € $ $ ! |
P T I | C a p r i c o r n | N o D M | # T e a m I K | # s h a n e R W a t s o n 3 3 ❤
B o o m B o o m ❤
S t r i v i n g t o b e a b e t t e r h u m a n !
L o v e h i m w h o l e a s t D e s e r v e s I t , T h a t ' s i t ❤
It is not replacing the pipe (|) symbol. Kindly help.
Try this:
select regexp_replace(description,'\\|',' ') from table;
Since a pipe character is an OR operator in regex in must be escaped. In Java flavored regex, two escape characters, back slashes, must be used.
Try this one add \ in your regexp_replace function
insert overwrite table_name select regexp_replace(id,'\\|',' ') from table_name
When following this tutorial on God, I run the command god -c path/to/simple.god -D and instead of getting the output as described, I get following weird output.
0000000 G o d . w a t c h d o | w |
778334023 1668571511 1868832872 2088205344
0000020 \n w . n a m e = " s
538976266 1848538912 543518049 1931616317
0000040 i m p l e " \n w . s t a
1819307369 537535077 1998594080 1635021614
0000060 r t = " r u b y / U s e r
1025537138 1970414112 790657378 1919251285
0000100 s / k a m a l / g o d r b / s i
1634414451 795631981 1919184743 1769156450
0000120 m p l e . r b " \n w . k
1701605485 576877102 538976266 1798207264
0000140 e e p a l i v e \n e n d \n
1634755941 1702259052 1684956426 10
0000155
I have no idea why it doesn't work.
I just ran god --version and the output is
od (GNU coreutils) 8.22
Copyright (C) 2013 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>.
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.
Written by Jim Meyering.
It seems it is some GNU god. I reinstalled god (sudo gem install god) and still I get the above output on running god --version. Any workaround?
Just in case: Using Mac OS X.
It seems like you run the command od instead of god.
$ od -c /etc/lsb-release -D
0000000 D I S T R I B _ I D = U b u n t
1414744388 1598179666 1430078537 1953396066
0000020 u \n D I S T R I B _ R E L E A S
1229195893 1230132307 1163026242 1396786508
0000040 E = 1 4 . 1 0 \n D I S T R I B _
875642181 170930478 1414744388 1598179666
0000060 C O D E N A M E = u t o p i c \n
1162104643 1162690894 1869903165 174287216
0000100 D I S T R I B _ D E S C R I P T
1414744388 1598179666 1129530692 1414547794
0000120 I O N = " U b u n t u 1 4 . 1
1028542281 1969378594 544568430 825111601
0000140 0 " \n
664112
0000143
Fix the typo: god
This question already has answers here:
Deleting columns from a file with awk or from command line on linux
(4 answers)
Closed 8 years ago.
I have a text file that looks like this:
A B C A B C A B C A B
G T C A G T C A G T C
A B C A B C A B C A B
A B C A B C A B C A B
A D E A B D E A B D E
A B C A B C A B C A B
C B D G C B D G C B D
Is there a way to extract only certain columns and leave the other columns intact?
For example removing only columns 2 and 5:
A C A C A B C A B
G C A T C A G T C
A C A C A B C A B
A C A C A B C A B
A E A D E A B D E
A C A C A B C A B
C D G B D G C B D
Thanks in advance.
UPDATE:
Found this answer using awk, but this extract whole "block" of columns and I only want to extract some.
Awk for extracting columns 3 to 5:
awk -F 'FS' 'BEGIN{FS="\t"}{for (i=1; i<=NF-1; i++) if(i<3 || i>5) {printf $i FS};{print $NF}}' input.txt
in your case you could do
cat your_file |cut -d ' ' --complement -s -f2,5
where ' ' is the delimiter(in your case the space)