Loop through CSV to create XML

Question

Hello all, first and foremost I want to apologize if this has been answered somewhere, but I am looking to do the following:
I have 500 records I need put in XML format. The records are currently in CSV format. I can use curl to upload a record in XML format using our applications API. I was under the impression that I could generate 1 large XML file and upload the entire group.
I have a script that will read in data from a CSV and output it to a file in XML format. I got the script the way I thought I needed it working, which was to create a new xml file, append a line to it, then loop through the csv and echo my string repeatedly inputing the variables from the CSV file. I could then upload the XML file where I needed it.
As it turns out, it seems the API will only allow me to upload 1 record at a time. (ALMOST pointless IMO.) So what I am looking to do is change the way the script loops through the CSV file, and instead of creating 1 large file, I would like the script to loop through the CSV one line at a time, create the XML file of only 1 record, upload it using curl, delete itself (or its contents) and repeat that process throughout the entire CSV file.
I am just getting familiar with loops, so Im not sure how I would even start this process, or even word what I want properly to search google. (Which I have been for the last two days, thats how I frankenstiened what I have below. Any help would be greatly appreciated.
The script is as follows:
#!/bin/bash
echo '<?xml version="1.0" encoding="UTF-8" standalone="no"?>' >> xmldata.xml
#:Read variables from XMLInput.csv:#
while read inputline;do
line0="$(echo $inputline | cut -d, -f1)"
line1="$(echo $inputline | cut -d, -f2)"
#:Set auto increment ID:#
idAdd0=$((idAdd0 +1))
#:removes trailing alpha characters leaving only the IP address:#
cutLine0=${line0//[^0-9.]/}
#:Strips trailing octet and appends .1:#
startIP=`echo $cutLine0 | sed 's/.[0-9]*$/.1/'`
#:Strips trailing octet and appens 254:#
endIP=`echo $cutLine0 | sed 's/.[0-9]*$/.254/'`
#:echos line to txt file as XML:#
echo "<network_segment><id>"$idAdd0"</id><name>$line0</name><starting_address>$startIP</starting_address><ending_address>$endIP</ending_address><distribution_point/><url/><netboot_server/><swu_server/><building>$line1</building><department/><override_buildings>false</override_buildings><override_departments>false</override_departments></network_segment>" >> xmldata.xml
done < XMLInput.csv
exit 0

Answer 1

What I ended up doing was having the script create 500 separate XML files in a folder by adding my auto increment variable to the XML file name,
then curling each file as a separate entry. A litte dirtier than I wanted, but in the end, I didn't have to manually enter 500 subnets
and define there locations in a website. Job Done.
Script if interested:
#!/bin/bash
#:Read variables from XMLInput.csv:#
while read inputline;do
line0="$(echo $inputline | cut -d, -f1)"
line1="$(echo $inputline | cut -d, -f2)"
echo
#:Set auto increment ID:#
idAdd0=$((idAdd0 +1))
#:removes trailing alpha characters leaving only the IP address:#
cutLine0=${line0//[^0-9.]/}
#:Strips trailing octet and appends .1:#
startIP=`echo $cutLine0 | sed 's/.[0-9]*$/.1/'`
#:Strips trailing octet and appens 254:#
endIP=`echo $cutLine0 | sed 's/.[0-9]*$/.254/'`
#:echos line to txt file as XML:#
echo '<?xml version="1.0" encoding="UTF-8" standalone="no"?>'"<network_segment><id>"$idAdd0"</id><name>$line0</name><starting_address>
$startIP</starting_address><ending_address>$endIP</ending_address><distribution_point/><url/><netboot_server/><swu_server/><building>$line1</building>
<department/><override_buildings>false</override_buildings><override_departments>false</override_departments></network_segment>" >> xmldata$idAdd0.xml
#:upload XML file to JSS:#
`curl -k -v -u account:Password https://server address/JSSResource/networksegments/name/SegmentName -T "xmldata$idAdd0.xml" -X POST`
done < XMLInput.csv
exit 0