CamelCase only for some part of the line using shell tool

November 3, 2023

my goal is to have a one liner to generate a list of variables composed from the name of the variable and its value. The value must be kept "as-is" while the name of the variable must be CamelCase, when removing leading slash "/" or dot "." or dash "-"

Sample of input:

auto-scaling/v2/HorizontalPodAutoscaler
v1/Service
apiextensions.k8s.io/v1/CustomResourceDefinition

Expected output:

AutoScalingV2HorizontalPodAutoscaler.$ : "auto-scaling/v2/HorizontalPodAutoscaler"
V1Service.$ : "v1/Service"
ApiextensionsK8sIoV1CustomResourceDefinition.$ : "apiextensions.k8s.io/v1/CustomResourceDefinition"

I found some clue in stackoverflow to transform a full string in CamelCase using awk or sed but those are for the full line while I would only what the part before the .$" (or the ":" as I could add after the ".$ " part)

echo "apiextensions.k8s.io/v1/CustomResourceDefinition" | sed -r 's/(^|[-_/\.]+)([0-9a-z])/\U\2/g'
ApiextensionsK8sIoV1/CustomResourceDefinition

echo "apiextensions.k8s.io/v1/CustomResourceDefinition" | awk 'BEGIN{FS="";RS="[-./]";ORS=""} {$0=toupper(substr($0,1,1)) substr($0,2)} 1'
ApiextensionsK8sIoV1CustomResourceDefinition

>Solution :

You can use the below awk to achieve your purpose:

awk 'BEGIN{FS="";RS="[-./]";ORS=""} {$0=toupper(substr($0,1,1)) substr($0,2)} 1'

Explanation of the above command:

BEGIN{FS="";RS="[-./]";ORS=""} – Set the separators.
- FS="" – Field Separator (FS) is set to make each character behave like a separate field.
- RS="[-./]" – Record Separator (RS) is set to a regex which matches [-./] and separates according to it. You can add or remove them based on your specific purpose.
- ORS="" – Output Record Separator (ORS) is set to avoid printing new lines.
{toupper(substr($0,1,1)) – It will take the first character of record and converts it to uppercase and prepend it to the output.
substr($0,2) – Appends the rest of input to the output record.
1 – shorthand to {print} current record.