Struggling with dynamic environment variable substitution in a shell script for feature flag management
I'm working on a personal project and I'm working on a project and hit a roadblock. Currently developing a prototype for a new feature that includes a feature flag system. In my shell script, I need to dynamically set environment variables based on user input, which should then determine if the feature flag is enabled or disabled. The challenge arises when dealing with variable substitution. I've tried using the basic syntax `${VAR}` for accessing environment variables, but it seems to be failing in the command context where these variables are set. For example: ```bash #!/bin/bash FEATURE_FLAG="true" if [[ ${FEATURE_FLAG} == "true" ]]; then echo "Feature is enabled" else echo "Feature is disabled" fi ``` This behaves as expected until I try to set `FEATURE_FLAG` from another source, like a config file or a user input via command line. I attempted to source the config file using `source config.sh`, but it doesn’t seem to recognize the changes in the current script. Instead, I’ve ended up with output that’s not reflecting the latest value. Also, I’ve experimented with using `export` to set the variable globally, but there are complications when this script is run in different subshells. For instance: ```bash #!/bin/bash read -p "Enable feature flag (true/false): " user_input export FEATURE_FLAG="${user_input}" ./another_script.sh # This script should read the updated value ``` In `another_script.sh`, if I print the `FEATURE_FLAG`, it often shows the old value or nothing at all. How can I ensure that changes to the environment variables are reflected across different scripts? Is there a better pattern for managing such dynamic configurations within shell scripts? I’m looking for best practices that can seamlessly handle this situation without introducing complex dependencies. For context: I'm using Bash on Linux. Is there a better approach? What's the best practice here?