jq --arg NAME "John" '.[] | select(.group | contains ($NAME))
this should return only the name John, but it is displaying all values where John is included..
Eg: displays Johnny, Johnie
It's giving these values also as output.
I tried with including * and also tried with single and double quotes. If I use * its giving empty output. Can someone help me on this?
From the manual on contains
The filter contains(b) will produce true if b is completely contained within the input. A string B is contained in a string A if B is a substring of A.
Therefore, contains also matches on substrings. Use == for exact matches, i.e. comparing with the entire string only:
jq --arg NAME "John" '.[] | select(.group == $NAME)'
Related
I am new to jq and facing an issue while parsing my json
I have a json stored in a variable like this
temp='{ "1": { "my_name": "one" }, "2": { "my_name": "two" } }'
Now I need to get the value of my_name for both other entries
I have tried something like this
echo $temp | jq '.1' //out put 0.1
I was assuming to get { "my_name": "one" }
And similarly to get my_name value I did
echo $temp | jq '.1.my_name' // Its output is giving me error
Can any one please help to identify what is wrong with my syntax and how can I correct it.
Just the number gets interpreted as a float. You need to use it in a context where it is unambiguously a key string.
echo "$temp" | jq '.["1"]["my_name"]'
and
echo "$temp" | jq '.["1"]'
to get the containing dict.
With a sufficiently new jq (I think >= 1.4) you can also say
echo "$temp" | jq '."1".my_name'
Whenever you are trying to reference a key that is not a valid identifier, you have to quote it. See the manual for more details.
To select the item under the key 1, you'd do this:
."1"
For your other question on how to obtain the my_name values, you could do this:
to_entries | map(.value.my_name)
How can I titleize text (capitalize each word) with jq?
Expected transformation: "lorum ipsum" → "Lorum Ipsum"
Given a string you consider calling a word, you may uppercase its first character by converting it into an array (./""), modifying the first element (first|=ascii_upcase) and joining all elements back together (add) like so
./"" | first |= ascii_upcase | add
Similarly, given a string of words, split it into an array of words, apply the above to each element using map and join them back together. Depending on what you (or the input data requires you to) consider a word, splitting the string might differ. Given your input example ("lorum ipsum"), using a single space character as word separator will suffice. I'll set it in a variable (--arg ws ' ') and use it for splitting (./$ws) and joining (join($ws)):
echo '"lorum ipsum"' | jq --arg ws ' ' '
./$ws | map(
./"" | first |= ascii_upcase | add
) | join($ws)
'
this may be achieved with gsub, ascii_upcase, and named capture groups:
$ echo '"lorum ipsum"' | jq 'gsub("(?<x>[A-z])(?<y>[A-z]+)"; "\(.x|ascii_upcase)\(.y)")'
"Lorum Ipsum"
You can use nawk for loop, toupper, sub and substr as below:
echo 'lorum ipsum' | nawk '{for(i=1;i<=NF;i++)sub(/./,toupper(substr($i,1,1)),$i)}1'
I have this below YAML input and I am trying to extract shown output using yq. I want to remove pairs where key name (VAR-A) in value {{a.b.VAR-A}} (after a.b.) matches and If I have more than one {{a.b.VAR-A}} in values separated by - , I want to keep them.
VAR-A: '{{a.b.VAR-A}}'
VAR-B: '{{a.b.VAR-B}}'
VAR-C: v0.0
VAR-D: '{{a.b.VAR-D}}-{{a.b.VAR-A}}'
VAR-E: '{{a.b.VAR-C}}-{{a.b.VAR-B}}-{{a.b.VAR-A}}'
VAR-F: True
Expected Output:
VAR-C: v0.0
VAR-D: '{{a.b.VAR-D}}-{{a.b.VAR-A}}'
VAR-E: '{{a.b.VAR-C}}-{{a.b.VAR-B}}-{{a.b.VAR-A}}'
VAR-F: True
This question works if I have all strings, but it fails when I have boolean value in yaml. Extract Key Value pairs which matches the regex in YAML using yq/sed/grep
I get below error:
Error: cannot substitute with !!bool, can only substitute strings. Hint: Most often you'll want to use '|=' over '=' for this operation.
There are at least two very different extant "yq" projects: a Python-based one, which is the focus of Part 1 below, and a Go-based one, which is the focus of Part 2.
Part 1
python-yq 'del(.[] | select( ( type == "string" and test("^{{a[.]b[.][^}]*}}$" ))))' so-vars.yaml
or
python-yq 'map_values( select( ( type == "string" and test("^{{a[.]b[.][^}]*}}$" )) | not))' so-vars.yaml
Output:
{
"VAR-C": "v0.0",
"VAR-D": "{{a.b.VAR-D}}-{{a.b.VAR-A}}",
"VAR-E": "{{a.b.VAR-C}}-{{a.b.VAR-B}}-{{a.b.VAR-A}}",
"VAR-F": true
}
Part 2
The Go-based version of yq that I have (4.6.3) might not be able to handle your requirements directly, but here's a solution that uses this yq to translate to and from JSON, and jq to do the rest:
yq -j eval . input.yaml |
jq 'del(.[] | select(( type == "string" and test("^{{a[.]b[.][^}]*}}$" ))))' > tmp.json
yq -P eval . tmp.json
The del-free version of the jq program:
map_values( select( type == "string" and test("^{{a[.]b[.][^}]*}}$" | not)
Output:
VAR-C: v0.0
VAR-D: '{{a.b.VAR-D}}-{{a.b.VAR-A}}'
VAR-E: '{{a.b.VAR-C}}-{{a.b.VAR-B}}-{{a.b.VAR-A}}'
VAR-F: true
I'm seeing a is not defined at <top-level> when calling jq like so:
jq ".Changes[0].ResourceRecordSet.Name = word-is-here.domain.com" someFile.json
The error repeats for each word separated by a dash in the second side of the replacement. The full error is like
jq: error: word/0 is not defined at <top-level>, line 1:
.Changes[0].ResourceRecordSet.Name = word-is-here.domain.com
I've tried escaping quotes in many different ways but that didn't help. (what I mean by this is doing "'"'" weird stuff, I'm still learning bash so I'm just trowing stuff at the wall until it sticks)
EDIT:
So I'm trying to run this in a bash script, and both side of the = signs are variables such as jq --arg value "$value" --arg key "$key" '$key = $value' "$path" (what I tried after a suggestion)
and got the error:
Invalid path expression with result ".Changes[0].ResourceRecor...
The json I'm using is as such:
{
"Changes": [
{
"Action": "do something",
"ResourceRecordSet": {
"Name": "some name here to replace",
...
}
}
]
}
jq '.Changes[0].ResourceRecordSet.Name = "word-is-here.domain.com"' file.json
Quote the string you are assigning. Or pass it to jq via an argument:
jq --arg foo 'words-here' '.Changes[0].ResourceRecordSet.Name = $foo' file.json
For passing the path to the key you want as an argument, a suggestion from https://github.com/stedolan/jq/issues/1493 might work:
jq --argjson path '["Changes",0,"ResourceRecordSet","Name"]' \
--arg val 'word-is-here.domain.com' \
'getpath($path) = $val' file.json
The problem (or at least the obvious problem) here is evidently the string: word-is-here.domain.com, since jq is interpreting the dash ("-") as an operation ("minus").
Unfortunately, since you haven't given us many clues, it's not completely clear what specifically needs to be changed, but a reasonable guess is that word-is-here.domain.com is intended as a fixed string. If so, you would have to present it as a JSON string. So in a bash or bash-like environment, you could write:
jq '.Changes[0].ResourceRecordSet.Name = "word-is-here.domain.com"' someFile.json
Specifying the LHS path via a shell variable
If the LHS path must be specified by a shell variable, it should if possible be passed in as a JSON array, e.g. using the --argjson command-line option; one can then use an expression of the form setpath($path; $value) to update the path.
If for some reason a solution allowing the LHS to be specified as a jq path is preferred, then shell string-interpolation could be used, though as with any such interpolation, this should be done with care.
To get what is between "aa=" and either % or empty
string = "aa=value%bb"
string2 = "bb=%aa=value"
The rule must work on both strings to get the value of "aa="
I would like a BASH LANGUAGE solution if possible.
Use this:
result=$(echo "$string" | grep -o 'aa=[^%]*')
result=${result:3} # remove aa=
[^%]* matches any sequence of characters that doesn't contain %, so it will stop when it gets to % or the end of the string. $(result:3} expands to the substring starting from character 3, which removes aa= from the beginning.