How to replace a comma to a semicolon in a textfile? 2

[zanzemaj]

Hi, I'm a UNIX scripting beginner and need some help. I now have a textfile wherein the contents come from SQL query. The number of lines will be dynamic. Here is a sample contents of the textfile named temp1.txt:
(int.sars),
(int.prev),
(int.batchg),
PROCESS DATA TABLES
(int.keelj),
(int),

I need to transform temp1.txt so that the last line before the line PROCESS DATA TABLES will contain a semicolon instead of comma. The desired output is:
(int.sars),
(int.prev),
(int.batchg);
PROCESS DATA TABLES
(int.keelj),
(int),

Any help is greatly appreciated. Thank you very much.

 

[khalidaaa]

Easy approach is to go directly to that line and change it!

vi temp1.txt

For more information about the vi look into this link:

http://www.apl.jhu.edu/Misc/Unix-info/editors/vi/vi.html

(you will just have to go to that line using arrow keys and then go to the comma then use the combination of ESC+r then just hit the semi-colon key to replace the comma!)

Regards,
Khalid

 

[feherke]

Hi

awk 'NR>1{if(/PROCESS DATA/)sub(/,$/,";",p);print p}{p=$0}END{print p}' /input/file

Feherke.

http://rootshell.be/~feherke/

 

[zanzemaj]

Thank you Khalid. But I need to change this inside a UNIX script. I'm not sure what to use, will I use grep, sed, awk or any other function available.

Here's a part of my script:

echo "NEW FILE" > new_file.txt
### get output of SQLS
FSQLSet 
cat temp1.txt >> new_file.txt

FSQLSet is my function that writes the contents from SQL query.

Here is a sample output file (temp1.txt) produced by FSQLSet

(int.sars),
(int.prev),
(int.batchg),
PROCESS DATA TABLES
(int.keelj),
(int),

Any help is greatly appreciated. Thank you very much.
zanzemaj

 

[zanzemaj]

I tried the solution given by feherke and it gives me the desired output.
Thank you feherke for this valuable post.

 

[zanzemaj]

I have a follow-up question. How do I use cut/paste to replace spaces with delimiter? I tried to cut the first field but the output is wrong. I am not sure how to proceed.

Sample record from a file named temp2.txt:

50fkknm0998 test_table 2007-05-21 12:26:32 9

I tried this:

cat temp2.txt | cut -f 1 temp2.txt

Result is:

50fkknm0998 test_table 2007-05-21 12:26:32 9

Or please suggest another way to replace the spaces with delimiter.

 

[PHV]

Which delimiter ? Say a semicolon ?
One way:

sed 's! !;!g' temp2.txt > temp3.txt

Another (faster) way:

tr ' ' ';' < temp2.txt > temp3.txt

Hope This Helps, PH.
FAQ219-2884
FAQ181-2886

 

[zanzemaj]

Thank you PHV for the quick response. I tried both and it worked. However, I want to cut the first field, second field up to the last field and save them in a variable.

I tried

sed 's! !;!g' temp2.txt > temp3.txt

Result is

50fkknm0998;;test_table;;;;;;;;;;;;;2007-05-21;12:26:32;;;;;;9

when I do a cut

cut -f 1 temp3.txt > temp4.txt

Result is

50fkknm0998;;test_table;;;;;;;;;;;;;2007-05-21;12:26:32;;;;;;9

How will I get the first field, second field and up to the last field? I want to get 50fkknm0998 and save it to a variable. I want to do this to the second field and up to the last field.

Really appreciate your help. Thanks.
zanzemaj

 

[PHV]

I want to cut the first field, second field up to the last field and save them in a variable

eval `awk -F';' '{for(i=1;i<=NF;++i)print "var"i"="$i}' temp3.txt`
echo "var1='$var1' var2='$var2' var3='$var3' ..."

Hope This Helps, PH.
FAQ219-2884
FAQ181-2886

 

[zanzemaj]

Thank you for all the replies. I tried it and everything works as expected.