UNIX for Dummies Questions & Answers

Hi,
My script should run through all subdirs recursively in SUBDIR_LIST, run a clearcase command in each subdir, print to the merge_results1.txt and add a new line to the results before going to the next subdir. All the results should append to merge_results1.txt, but are seperated by a line so it's easier to read.


Can someone help me with the foreach statement?


#! /usr/bin/ksh
/usr/atria/bin/cleartool startview a442592_merge_diff_2.0to2.5

cd /view/a442592_merge_diff_2.0to2.5/vobs_jil

SUBDIR_LIST=`ls -d fmr*/`


#for all the subdirs run a print merge command and add a new line to #merge_results1.txt

for each ${SUBDIR_LIST}

/usr/atria/bin/cleartool findmerge . -fver .../strutsplus-2.0_bos/LATEST -print > /home/a442592/merge_results1.txt

#loop ends here

cat /home/a442592/merge_results1.txt | sed 's/Needs Merge//g' | sed 's/"//g' |cut -d[ -f1 > /home/a442592/merge_results2.txt

cat ~/merge_results2.txt | mailx -r <form e-mail>-s 'Sync merge between 2.0 and 2.5' <to e-mail>

Thanks

The syntax for a for loop is:

Code:

for file in `ls -d fmr*/`; do
  #
  # do your stuff here with "$file"
  #
done

Regards

Hi.
That worked - thanks.

My problem now is that everytime it goes through the loop it adds a line, which is ok if there is output. My problem is that i am now getting 3 to 4 blank lines so i now need to put an if statement to say if a blank line already exists don't add another.

Do you know how to do that?

CODE:
SUBDIR_LIST=`ls -d fmr*`
for DIR in `ls -d fmr*`; do
/usr/atria/bin/cleartool findmerge $DIR -fver .../strutsplus-2.0_bos/LATEST -print >> /home/a442592/merge_results1.txt
echo >> /home/a442592/merge_results1.txt "\n"
done
cat /home/a442592/merge_results1.txt | sed 's/Needs Merge//g' | sed 's/"//g' |cut -d[ -f1 > /home/a442592/merge_results2.txt


Thanks
Angela

What does this command means?

Code:

echo >> /home/a442592/merge_results1.txt "\n"

Do you have the desired output if you remove this command?
If not, post the contents of merge_results1.txt and the desired output.

Regards

The loops keeps appending to the same file so the echo command adds a line between each loop. if i remove it i get no line seperators.

This is what i am getting now (using 'echo...'):

___________________________________________

fmr-aspects/docs/pub/AspectJ-Policy-Enforcement-User-Guide.doc
fmr-aspects/jil-usage-policies-test/src/com/fmr/commons/aspectj/policy/FMRCommonsSSMCTest.java
fmr-aspects/jil-usage-policies-test/src/com/fmr/commons/aspectj/policy/TestAll.java


fmr-commons-admin/docs/pub/fmr-admin.doc
fmr-commons-admin/src/java/com/fmr/commons/admin/services/jca/JcaFactorySpec.java


fmr-commons-catalog/docs/pub








fmr-commons-cs203/build.properties
fmr-commons-cs203/old_build.xml
fmr-commons-cs203/src/main/java/com/fmr/commons/cs203/utils/CS203Constants.java
fmr-commons-cs203/src/test/java/com/fmr/commons/cs203/validator/CoreValidatorTest.java


fmr-commons-hacp/docs/TSD-FMR-COMMONS-HACP.doc
fmr-commons-hacp/scripts/hacp-build.xml
fmr-commons-hacp/src/test/config/log4j.xml

__________________________________________

The spaces between should only be one line, but my code adds a line regardless of any output of the loop. So i want to say that if a blank line already exists dont add another.

What it should look like is:

___________________________
fmr-aspects/docs/pub/AspectJ-Policy-Enforcement-User-Guide.doc
fmr-aspects/jil-usage-policies-test/src/com/fmr/commons/aspectj/policy/FMRCommonsSSMCTest.java
fmr-aspects/jil-usage-policies-test/src/com/fmr/commons/aspectj/policy/TestAll.java

fmr-commons-admin/docs/pub/fmr-admin.doc
fmr-commons-admin/src/java/com/fmr/commons/admin/services/jca/JcaFactorySpec.java

fmr-commons-catalog/docs/pub

fmr-commons-cs203/build.properties
fmr-commons-cs203/old_build.xml
fmr-commons-cs203/src/main/java/com/fmr/commons/cs203/utils/CS203Constants.java
fmr-commons-cs203/src/test/java/com/fmr/commons/cs203/validator/CoreValidatorTest.java

fmr-commons-hacp/docs/TSD-FMR-COMMONS-HACP.doc
fmr-commons-hacp/scripts/hacp-build.xml
fmr-commons-hacp/src/test/config/log4j.xml

__________________________

Thanks
Angela

It's difficult if you don't know when your command produces an empty line.
You can delete all the emptylines from your output with:

Code:

sed '/^$/d' file > newfile

Regards

Hi,
Thats not really what i need cause i still need 1 blank line.

I found this - it removes duplicate, consecutive lines
awk 'a !~ $0; {a=$0}'


would you know how to change it to remove only 1 of the duplicate lines?

thanks