using sed command to read between sections in a given file

Question

File content.

[kafka_properties]
listeners=PLAINTEXT://:KAFKA_CLIENT_PORT
default.replication.factor=2
ssl.client.auth=required
ssl.cipher.suites=["TLS_DHE_RSA_WITH_AES_256_GCM_SHA384",
            "TLS_DHE_RSA_WITH_AES_128_GCM_SHA256",
            "TLS_DHE_RSA_WITH_AES_256_CBC_SHA256",
            "TLS_DHE_RSA_WITH_AES_128_CBC_SHA256",]
ssl.enabled.protocols=TLSV1.2
ssl.secure.random.implem=SHA1PRNG
security.inter.broker.protocol=PLAINTEXT
security.protocol=SSL
ssl.endpoint.identification.algorithm=https


[kafka_ports]
KAFKA_CLIENT_PORT=9082

[zookeeper_properties]
clientPort=ZK_CLIENT_PORT
syncLimit=2
initLimit=5
tickTime=2000
autopurge.snapRetainCount=3
autopurge.purgeInterval=1
admin.serverPort=ZK_SERVER_ADMIN_PORT

I am trying to read the values from every section, e.g. [kafka_properties], or [kafka_ports] using this command:

cat file.txt | sed -n '0,/kafka_properties/d;/\[/,$d;/^$/d;p'

And write the values into a different file. It works okay if I don't add the parameter:

ssl.cipher.suites=["TLS_DHE_RSA_WITH_AES_256_GCM_SHA384",
            "TLS_DHE_RSA_WITH_AES_128_GCM_SHA256",
            "TLS_DHE_RSA_WITH_AES_256_CBC_SHA256",
            "TLS_DHE_RSA_WITH_AES_128_CBC_SHA256",]

but after adding the ssl.cipher.suites= parameter to the file.txt sed is not working as expected. Where am I going wrong ?

Please edit your question and explain how sed is "not working as expected". What output were you expecting and what output do you actually get? — terdon
– terdon ♦, Commented Apr 9, 2019 at 10:52

pLumo · Accepted Answer · 2019-04-09 10:55:24Z

2

Make [ only match at the beginning of the line with ^:

sed -n '0,/kafka_properties/d;/^\[/,$d;/^$/d;p' file.txt

answered Apr 9, 2019 at 10:55

pLumo

23.2k2 gold badges43 silver badges70 bronze badges

Add a comment |

terdon · Accepted Answer · 2019-04-09 11:17:16Z

For the general case, writing each section to a different file without needing to know the section names, you could do:

awk '/^\[/{n=$1;gsub(/[][]/,"",n)}{print >> n".txt"}' file

This will create these files from your example:

kafka_ports.txt  kakfa_properties.txt  zookeeper_properties.txt

Explanation

/^\[/{n=$1;gsub(/[][]/,"",n)} : if this line starts with a [, save the 1st field as the variable n and remove all [ or ] from it.
print >> n".txt" : append the curent line to the file n.txt where n is the name of the section.

Note that this assumes you never have whitespace in a section name. If you do, try this instead:

awk '/^\[/{n=$0;gsub(/[][]/,"",n); gsub(/ /,"_",n)}{print >> n".txt"}' file

Stack Exchange Network

using sed command to read between sections in a given file

2 Answers 2

Explanation

You must log in to answer this question.

Hot Network Questions

using sed command to read between sections in a given file

2 Answers 2

Explanation

You must log in to answer this question.

Related

Hot Network Questions