Visit Our UNIX and Linux User Community


replace string in file.1 with line from file.2


 
Thread Tools Search this Thread
Top Forums Shell Programming and Scripting replace string in file.1 with line from file.2
# 1  
Old 05-11-2009
replace string in file.1 with line from file.2

Hello all, the title makes this sound simple, and maybe it should be.

This is by code:
Code:
#!/bin/sh
cp ch25.txt ch25.fn.tex
n=`grep -c '^\[' ch25_footnotes.txt`

for i in {1..$n}
do
        `grep '^\[$i\]' ch25_footnotes.txt > temp`
        r=`awk -F] '{print $2}' temp`
        `sed 's/\[$i\]/\\footnote{$r}/' ch25.fn.tex`
done

This is what I am trying to do. I have a file like this:
Quote:
First there was nothing. Then there was Calvin[1].
That contains a body of text with footnotes markers, that are of exactly that format [i]. Square brackets are not used otherwise in the text... at least no patters of the form [integer] exist except for footnote markers.

I have a another file of the footnote text, like this
Quote:
[1] A memorable quote from Bill Waterson's Calvin and Hobbes.
What that code I wrote is attempting to do is:
  1. Create a copy of the body text to mess with.
  2. Count the number of footnotes that exist.
  3. Initiate a For loop that will generate an $i for every integer up to $n, the number of footnotes.
  4. grep the line starting with [$i] and save stick it into a temp file.
  5. use awk to grab only the text portion of the footnote (skipping the [$i] number at the beginning of the line) and store that in variable r.
  6. Finally, search the main file for the [$i] marker and add the footnote, alone with \footnote{} markup for latex.
  7. Repeat hundreds of times (-;

I am not all that great with scripting, but this is my thought process and what I have tried. Right now... nothing happens. I execute the script, and it just hangs, and I have to break the script.

In case you are wondering, I am just trying to typeset Marx's Das Kapital for posterity and easier reading. I do not plan on printing or distributing... just taking a freely available reading material and making it pretty.

Thank you!

Last edited by ccox85; 05-11-2009 at 05:28 PM..
# 2  
Old 05-11-2009
PS, I notice that I forgot a semicolon after the for argument... I added that and it still does not run. Thank you!
PPS, I also realized that I should probably have back-ticked the grep and sed commands... but that also did not solve my problem.
PPPS, I am fixing the stupid errors in my code... I hope you don't decline to help me because they existed! lol

Last edited by ccox85; 05-11-2009 at 05:27 PM.. Reason: Updates... I am trying too!
# 3  
Old 05-11-2009
Almost

Alright, so this was an uninteresting problem, I guess... no hits at all except for my own! But, I did solve PART of the problem, so for what its worth I am going to explain that... I am sure I am not the only person to make this mistake.

In my grep command, I was trying to reference a variable integer stored in $i.
Grep was reading the $ independently, as a regular expression meaning "at the end of the line" I think... but at any rate, it was NOT looking for the variable $i and passing that into the argument.

To work around this:
Code:
for i in {1..5}
do 
    echo $i; #to make sure the for loop was working as expected.
    
    str="grep '^\[$i\]' ch25_footnotes.txt"; #I saved the whole thing as a string, which DOES 
                                             #interpret $i as a variable.  If I were to echo that 
                                             #command, it would have a number there 
                                             #rather than a $i.
   
    eval $str; #this actually runs the grep command
done

This works for the grep part... But I guess now the kink is in getting the for command to acknowledge the $n variable... Smilie

Meh... help would be welcome!
# 4  
Old 05-11-2009
Question giving up for now...

Here is what I have managed to come up with:

Code:
#!/bin/sh
cp ch25.txt ch25.fn.tex
n=`grep -c '^\[' ch25_footnotes.txt`
echo $n

#for i in {1..n};  ### This syntax was not working, for some reason

for ((i=1; i <= n ; i++))
do
        str="grep '^\[$i\]' ch25_footnotes.txt > temp"
#      echo $str
        eval $str
        r=`awk -F] '{print $2}' temp`
#      echo $r ## Up to here, things work as expected.
        str2="sed 's/\[$i\]/\\\footnote{$r}/' ch25.fn.tex"
        eval $str2
done

Now, the issue seems to be that it is dumping to standard output, rather than actually editing ch25.fn.tex. I dont really know why thats happening... and I THINK that it printed to stout 140 times, each with the footnote added to only that one number... I THINK that is what happened. To much output to sift through. I feel that I have narrowed down the problem to my sed command.

I have spent too much time playing around already. I hope someone finds this interesting or is sympathetic to my curiosity and helps me out!
# 5  
Old 05-11-2009
Code:
$
$ cat ch25.txt
First there was nothing. Then there was Calvin[1].
blah blah blah
blah blah blah
"On a long enough timeline, the survival rate for everyone drops to zero" [2].
blah blah blah
"Dr. Livingstone, I presume?" [3]
blah blah blah
$
$ cat footnotes.txt
[1] A memorable quote from Bill Waterson's Calvin and Hobbes.
[2] Spoken by Tyler Durden, in Chuck Palahniuk's "Fight Club".
[3] By Henry Morton Stanley, on meeting Dr. David Livingstone near Lake Tanganyika on Nov. 10th, 1871.
$
$
$ perl -ne 'BEGIN {
>   open(FN,"footnotes.txt");
>   while(<FN>) {chomp; $fnote{$1}=$2 if (/(\[\d+\]) (.*)/)}
>   close(FN)}
> chomp;
> if (/(.*?)(\[\d+\])/) {print "$1\\footnote{$fnote{$2}}\n"}
> else {print $_,"\n"}' ch25.txt
First there was nothing. Then there was Calvin\footnote{A memorable quote from Bill Waterson's Calvin and Hobbes.}
blah blah blah
blah blah blah
"On a long enough timeline, the survival rate for everyone drops to zero" \footnote{Spoken by Tyler Durden, in Chuck Palahniuk's "FightClub".}
blah blah blah
"Dr. Livingstone, I presume?" \footnote{By Henry Morton Stanley, on meeting Dr. David Livingstone near Lake Tanganyika on Nov. 10th, 1871.}
blah blah blah
$
$

HTH,
tyler_durden

___________________________________________________________
"It's only after we've lost everything that we're free to do anything."
# 6  
Old 05-12-2009
Thank you!

Wow, that worked like a charm and was less elaborate than my scheme. It'll take me a bit to understand what you did, I don't know perl at all, but it certainly got the job done. Thank you!

Chris
# 7  
Old 05-12-2009
Code:
awk 'FNR==NR{id[$1]="{"substr($0,length($1)+2,length($0))"}";next}
{a=$0;b=match($0,/\[[0-9]+\]/);
if(b>0){
c=substr($0,RSTART,RLENGTH);o=substr($0,1,b-1);print o"\\footnote"id[c]
}
else print
}' footnotes.txt ch25.txt

cheers,
Devaraj Takhellambam

Previous Thread | Next Thread
Test Your Knowledge in Computers #962
Difficulty: Medium
ILOVEYOU, sometimes referred to as Love Bug or Love Letter, was a computer worm that attacked tens of millions of Windows PCs on and after 5 May 2000 local time in the Philippines.
True or False?

10 More Discussions You Might Find Interesting

1. UNIX for Dummies Questions & Answers

Search for a string,delete the line and replace with new string in a file

Hi Everyone, I have a requirement in ksh where i have a set of files in a directory. I need to search each and every file if a particular string is present in the file, delete that line and replace that line with another string expression in the same file. I am very new to unix. Kindly help... (10 Replies)
Discussion started by: Pradhikshan
10 Replies

2. Shell Programming and Scripting

Replace line in file with line in another file based on matching string

HI Can any one guide me how to achieve this task. I have 2 files env.txt #Configuration.Properties values identity_server_url = http://identity.test-hit.com:9783/identity/service/user/register randon_password_length = 6 attachment_file_path = /pass/temp/attachments/... (1 Reply)
Discussion started by: nikilbr86
1 Replies

3. Shell Programming and Scripting

Replace a string with each line from another file repeatedly

I don't know if it's been asked before but seems i gave up seeking. i have 2 files : file1.txt Monday XXXX Tuesday XXXX XXXX Wednesday Thursday XXXX XXXX is in every lines of file1.txt and i want to replace them with each line in file2.txt: home school cinema so output file is: ... (19 Replies)
Discussion started by: perseous
19 Replies

4. Shell Programming and Scripting

Replace and add line in file with line in another file based on matching string

Hi, I want to achieve something similar to what described in another post: The difference is I want to add the line if the pattern is not found. File 1: A123, valueA, valueB B234, valueA, valueB C345, valueA, valueB D456, valueA, valueB E567, valueA, valueB F678, valueA, valueB ... (11 Replies)
Discussion started by: jyu3
11 Replies

5. Shell Programming and Scripting

sed or awk to replace a value in a certain line from another file containing a string

Hi experts, In my text file I have the following alot of lines like below. input.k is as follows. 2684717 -194.7050476 64.2345581 150.6500092 0 0 2684718 -213.1575623 62.7032242 150.6500092 0 0 *INCLUDE $# filename... (3 Replies)
Discussion started by: hamnsan
3 Replies

6. Shell Programming and Scripting

Replace line in file with line in another file based on matching string

Hi I am not the best scripter in the world and have run into a issue which you might be able to guide me on... I have two files. File1 : A123, valueA, valueB B234, valueA, valueB C345, valueA, valueB D456, valueA, valueB E567, valueA, valueB F678, valueA, valueB File2: C345,... (5 Replies)
Discussion started by: luckycharm
5 Replies

7. Shell Programming and Scripting

replace (sed?) a single line/string in file with multiple lines (string) from another file??

Can someone tell me how I can do this? e.g: Say file1.txt contains: today is monday the 22 of NOVEMBER 2010 and file2.txt contains: the 11th month of How do i replace the word NOVEMBER with (5 Replies)
Discussion started by: tuathan
5 Replies

8. Shell Programming and Scripting

To trim Certain field in a line of a file and replace the new string in that position

To trim 3rd field in for all the lines of a file and replace the modified string in that particular field. For example i have a file called Temp.txt having content Temp.txt ----------------- 100,234,M1234 400,234,K1734 300,345,T3456 ---------------- So the modified file output should... (4 Replies)
Discussion started by: rpadhi
4 Replies

9. Shell Programming and Scripting

Replace string in a file within a range of line

Hi, I want to replace the srting '; with ABCD'; in a file from line 1 to line 65. Is there any single command to do it without using awk Thanks for quick reply https://www.unix.com/images/misc/progress.gif (3 Replies)
Discussion started by: tosattam
3 Replies

10. UNIX for Dummies Questions & Answers

how can search a String in one text file and replace the whole line in another file

i am very new to UNIX plz help me in this scenario i have two text files as below file1.txt name=Rajakumar. Discipline=Electronics and communication. Designation=software Engineer. file2.txt name=Kannan. Discipline=Mechanical. Designation=CADD Design Engineer. ... (6 Replies)
Discussion started by: kkraja
6 Replies

Featured Tech Videos