Hi all,
i am new to unix,i have never worked on scripting and all,i am learning now and i have to write a script to compare two files.
The requirement is like : in the first file i am searching for a word and after i get that word i have to select everything from the rest of the file and redirect to some output file.the second part is to compare the output file with my second file.the second part is feasible for me to do,but i am facing problem to proceed with the first part,Could anyone help me.
For the first part you can use sed:
sed -n '/pattern/,$p' file
Regards
Please refer the link, this might help you.
Thanks.
Hi all,thanks for ur help but i think i couldn't get what i needed.
Here is the sample file eg:abc.txt
<servergroup id="8">
<application value="OnlineDisputes"></application>
<minheap value="256"></minheap>
<maxheap value="512"></maxheap>
<httpport value="51058"></httpport>
<additionalcmdlineargs value="-DSLConfigFile=app_servicelocator.xml -verbosegc -Xss8m"/>
</servergroup>
<servergroup id="9">
<application value="Transfer"></application>
<minheap value="256"></minheap>
<maxheap value="512"></maxheap>
<httpport value="51059"></httpport>
<additionalcmdlineargs value="-DSLConfigFile=app_servicelocator.xml -verbosegc -Xrunheappro
file"/>
</servergroup>
<servergroup id="10">
<application value="INTLEStatement"></application>
<minheap value="256"></minheap>
<maxheap value="512"></maxheap>
<httpport value="51063"></httpport>
<additionalcmdlineargs value="-DSLConfigFile=app_servicelocator.xml -verbosegc
-XX:MaxPermSize=128m"/>
</servergroup>
</servergroup-info>
<clone-info>
<clone id="1">
<application value="EnterpriseServices"></application>
<cloneindex value="1"></cloneindex>
<node value="spdwa013"></node>
<httpport value="21050"></httpport>
<workingdir value="/tmp"></workingdir>
</clone>
<clone id="2">
<application value="GlobalMR"></application>
<cloneindex value="1"></cloneindex>
<node value="spdwa013"></node>
<httpport value="21051"></httpport>
<workingdir value="/tmp"></workingdir>
</clone>
<clone id="3">
<application value="GlobalUserManagement"></application>
<cloneindex value="1"></cloneindex>
<node value="spdwa013"></node>
<httpport value="21052"></httpport>
<workingdir value="/tmp"></workingdir>
</clone>
<clone id="4">
<application value="USAccountSummary"></application>
<cloneindex value="1"></cloneindex>
<node value="spdwa013"></node>
<httpport value="21053"></httpport>
<workingdir value="/tmp"></workingdir>
</clone>
In the above file i need to select everything below <clone-info>
and redirect it to an output file.
I would be very greatful if anybody can help me in this regard.
You 'think' or are you sure?
What is it that you 'know' that you want as the output?
And how's that compare to the 'sed' solution provided?
Franklin's solution works with slight modification:
sed -n '/<clone-info>/,$p' file_in > file_out
Hey all, Thanks a lot..!!!
Its working now..!!
There are two files file1.txt and file2.txt
and the contents of files are given as follows:
file1.txt
{
EnterpriseServices
GlobalUserManagement
USAccountSummary
USEStatement
MYCAServices
EnterpriseServices
MYCAServices
USEStatement
}
file2.txt
{
USEStatement
MYCAServices
EnterpriseServices
MYCAServices
USEStatement
DocGen
OnlineDisputes
Transfer
INTLEStatement
}
Now the problem is how to compare these two files and redirect to an output file the contents which are present in file1.txt but which are not present in file2.txt
Please help in this regard.
You can use comm -13 or comm -23. Just play with it a little
I am not sure how exactly i can use comm -13 or comm -23 because i need to write some generic script to compare two files and redirect the output to a third file with the contents which are present in first file but not present in second file.
Hi all,
i tried comm command in all the possible ways but could not get the expected result.
I will explain one more time.i have two files containing some words ,it is not necessary that the worrds are in same order in both the files but same words will be there.i want to compare those files and redirect the result to a third file which will contain all the words which are present in first file but not in second.
comm command is not solving the purpose as it compares line by line even if a same word is present in 1st line of file one and 4th line of file two then it is giving the output.
The result should be like : if same words are present in both the files no matter on which line number they are then the final output should be zero.
Please help.
Thanks
Something like this???
grep -v -f file2 file1
The problem in the solution you have given is there is no -f option with grep command?
Could anyone help me out??
Can anyone tell how to use a loop inside awk command so that the words of second file are compared with all the words of first file and those that are present in first file but not in second are given as output??
Thanks in advance
Try this it should work as long as you have saved file1.txt & file2.txt seprately
diff file1.txt file2.txt | grep ">" | sed 's/>//g'
Hope this helps:
ddl10197@GT109867 ~
$ sort foo | uniq > foo.tmp
ddl10197@GT109867 ~
$ sort bar | uniq > bar.tmp
ddl10197@GT109867 ~
$ comm -23 foo.tmp bar.tmp
GlobalUserManagement
USAccountSummary
ddl10197@GT109867 ~
$ grep GlobalUserManagement bar
ddl10197@GT109867 ~
$ grep USAccountSummary bar
Hi All,
Thanks for your response.But the solution provided seems not to be dynamic.
If any one can help running a for loop comparing the strings one by one in both the files i.e(comparing the first string from first file with all the strings in second file if it is not present outputting into third file).
i have got a code chunk,but it is not working,
can anyone help me modify it for the above requirement?
first code chunk i got is:
awk 'FILENAME=="file1" { if( $0 in arr) {continue} else {print $0 } } FILENAME=="file2" {arr[$0]++ }' file2 file1
but i guess there is some syntax problem as it is not working
The other code chunk i got is:
#!/usr/bin/ksh
set -x
> outputfile
awk '
BEGIN {
while ( getline < "file1" ) { arr[$0]=1 }
}
{ if ( arr[$0] != 1 ) print FILENAME":" $0
else delete arr[$0];
}
END {
for( key in arr )
if ( arr[key] == 1 ) print "file1:" key
} ' file2 >> outputfile
Can anyone help me finding what is the problem with the above two code chunks as i am not able to run these,getting some errors
Else,if both the codes doesnt work,could anyone help me providing the code to compare two files containing strings and output the string that are present in first file but not in second in third file.
You can use arrays or any other way,
The files will look like
file1
BasicServicesClone2
BasicServicesClone1
DocGenClone1
USEStatementClone2
USEStatementClone1
file2
BasicServicesClone2
BasicServicesClone1
DocGenClone1
USEStatementClone2
the output should be
USEStatementClone1
as it is present in file 1 but not in file2
angheloko's solution works.
Do you have a grep version without the -f option? With awk you can get the desired output with:
awk 'NR==FNR{a[$0];next}!($1 in a)' file2 file1
Regards
Hi Franklin52,
Thanks for your update but when i am trying the code which you have given,
It is showing as below:
awk 'NR==FNR{a[$0];next}!($1 in a)' file2 file1
awk: syntax error near line 1
awk: bailing out near line 1
angheloko's solution was good,there he was grepping seperately for each string which should not be the case, as i have to compare two files containing strings line by line and it should be generic i.e the strings may change and i have to output the strings which are present in first file and not in second.The file format i have mentione in my previous post.
Could anyone help me writing a generic script to first store all the strings of both the files in arrays and compare those arrays and one by one??
Because the above line of code is not working here,Syntax error?
Could anyone help??
Thanks in advance
Use nawk or /usr/xpg4/bin/awk on Solaris.
Regards