I want to write a json string to a file as part of my systemd unit file. I tried the below, but it fails:
[Unit]
Description=Write json to file
[Service]
ExecStart=/bin/bash -c cat > /tmp/test.json <<- EOM
{
"Hello": "World",
"Test": [{
"foo": "bar"
}]
}
EOM
I don't want to escape quotes as my original json is little longer. Any better ideas?
It doesn't like '{' on line 5. I think it wants command in a single line and use '\' for line continuation
Related
In my declarative Jenkins file, I'm writing some string into a .ini file that I have in my workspace. When running the pipeline, I'm facing issues with writing that string.
Here's the code from my JenkinsFile:
sh "echo -en \n[azcli_servers] >> ./Inventory/hosts.ini"
When running the pipeline, I get the error above:
[azcli_servers]: not found
script returned exit code 127
How can I write these special characters "[]" as regular chars and pass that string "[azcli_servers]" into the file?
Thanks for the response.
Remember to quote your whole string. Example:
sh """echo -en "\n\\[azcli_servers\\]" >> ./Inventory/hosts.ini"""
i run the following function in (git-)bash under windows:
function config_get_container_values() {
local project_name=$1
local container_name=$2
#local container_name="gitea"
echo "###"
buildcmd="jq -r \".containers[]."
echo "$buildcmd"
buildcmd="${buildcmd}${container_name}"
echo "$buildcmd"
buildcmd="${buildcmd}foobar"
echo "$buildcmd"
echo "###"
}
The output of this is the following. Whyever, after using the variable to extend the string, he starts to overwrite $buildcmd. I tried this also with everything in one line as well with the append command (=+). Everytime the same result.
###
jq -r ".containers[].
jq -r ".containers[].gitea
foobar".containers[].gitea
###
The really strange thing is: When i enable the line local container_name="gitea" everything works as expected. The output is:
###
jq -r ".containers[].
jq -r ".containers[].gitea
jq -r ".containers[].giteafoobar
###
When i put this all into a news file, its also works as expected. So i think something goes wrong in the thousands of line before calling this function. Any idea, what could be cause of this behavior?
Regards
Dave
This is not how you should build up the command, DOS line endings aside. Use --arg to pass the name into the filter as a variable. For example,
config_get_container_values() {
local project_name=$1
local container_name=$2
jq -r --arg n "$container_name " '.containers[][$n+"foobar"]'
}
config_get_container foo gitea < some.json
If the function is invoked with
config_get_container_values proj gitea
it produces the "expected" output. If it is invoked with
config_get_container_values proj $'gitea\r'
it produces output that looks like the first output example. $'gitea\r' expands to a string that consists of 'gitea' followed by a Carriage return (CR) character.
One possible cause of the problem is that the container name (gitea) was read from a file that had Windows/DOS line endings (CR-LF). Problems like that are common. See the first question ("Check whether your script or data has DOS style end-of-line characters") in the "Before asking about problematic code" section of the Stack Overflow 'bash' Info page.
I'm using a bash script to send an argument to a node app like so:
testString="\nhello\nthere"
node ./myNodeScript.js $testString
The trouble comes when I use testString inside the node program after capturing it as process.argv[2] -- rather than expand the \n characters to newlines node prints them literally. I need a way to tell node to convert the argument to a javascript string, respecting the formatting characters. Is there a way to go about this?
Try to avoid confusing literal linefeeds and literal backslash followed by literal n.
If you want the string you pass to have linefeeds, you should ignore JavaScript string literal syntax and just pass the linefeeds as linefeeds:
$ cat myNodeScript.js
console.log("Node was passed this: " + process.argv[2])
$ cat myBashScript
testString='
hello
there'
printf 'Bash passes this: %s\n' "$testString"
node myNodeScript.js "$testString"
$ bash myBashScript
Bash passes this:
hello
there
Node was passed this:
hello
there
Arguments should contain data (linefeed) while script files should contain code (quoted linefeed or expanded \n as appropriate in the language). When you make sure not to confuse code and data, you can trivially handle both backslash-en and linefeeds in the same string with no surprises:
testString='
"\nhello\nthere" is JavaScript syntax for:
hello
there'
There are ways to express this on a single line in bash using \n for linefeeds and \\n for backslash-en, you just need to make sure that it remains as code, and doesn't accidentally make it into the variable as data.
Can you try this:
testString=$( printf "\nhello\nthere")
node ./myNodeScript.js "$testString"
And let me know if it works?
I am trying to create a shell script for logs and trying to append data into a text file. I have write this sample "test.sh" code for testing:
#!/bin/sh -e
touch /home/sample.txt
SPTH = '/home/sample'.txt
echo "MY LOG FILE" >> "$SPTH"
echo "DUMP started at $(date +'%d-%m-%Y %H:%M:%S')" >> /home/sample.txt
echo "DUMP finished at $(date +'%d-%m-%Y %H:%M:%S')" >> /home/sample.txt
but in above code all lines are working correct except one line of code i.e.
echo "MY LOG FILE" >> "$SPTH"
It is giving error:
test.sh: line 6: : No such file or directory
I want to replace this full path of file "/home/sample.txt" to variable "$SPATH".
I am executing my shell script using
sh test.sh
What I am doing wrong.
Variable assignments in bash shell does not allow you to have spaces within. It will be actually interpreted as command with = and the subsequent keywords as arguments to the first word, which is wrong.
Change your code to
SPTH="/home/sample.txt"
That is the reason why SPTH was not assigned to the actual path you intended it to have. And you have no reason to have single-quote here and excluding the extension part. Using it fully within double-quotes is absolutely fine.
The syntax for the command line is that the first token is a command, tokens are separated by whitespace. So:
SPTH = '/home/sample'.txt
Has the command as SPTH, the second token is =, and so on. You might think this is daft, but most shells behave like this for historical reasons.
So you need to remove the whitespace:
SPTH='/home/sample'.txt
What I'm trying to accomplish is having a central configuration file, in bash, that defines some variables that are re-used in different bash files. The example below attempts to generate a file name with the current date included in the file name as well as a variable defined in another shell script. However whenever I try to concatenate this external variable it doesn't work. I can concatenate the variable in any other situation.
Example Code:
../config/vars.sh
#!/bin/bash
mysqlUser="backupuser"
mysqlPwd="fakePwd"
mysqlSocket="/var/run/mysqld/mysqld.sock"
mysqlPort="3306"
serverName="s01.catchyservername.com"
./dbBackup.sh
#!/bin/bash
source ../config/vars.sh
tempName=$(date +"%Y%m%d.sql.gz")
fileName="mysqld_${mysqlPort}_${tempName}"
echo "mysqld_${mysqlPort}"
echo ${tempName}
echo ${fileName}
output of dbBackup.sh
mysqld_3306
20140926.sql.gz
_20140926.sql.gz
As you can see when echoing "mysqld_${mysqlPort}" I get the expected output, but when echoing ${fileName} the entire first half of the string is ignored. What am I misunderstanding?
Your vars.sh file was probably created with a DOS/windows text editor:
$ ./dbBackup.sh
mysqld_3306
20140926.sql.gz
_20140926.sql.gz
$ dos2unix vars.sh
dos2unix: converting file vars.sh to Unix format ...
$
$ ./dbBackup.sh
mysqld_3306
20140926.sql.gz
mysqld_3306_20140926.sql.gz
$
As you can see above, I use the dos2unix utility to convert the line separators to Unix style.