When working with Ansible's regex_search filter, many developers encounter situations where they need to match patterns dynamically using variables rather than hardcoded strings. The original example shows a common pitfall where the variable name gets interpreted literally instead of being evaluated.

# Problematic example (variable treated as literal)
when: item.key | regex_search('^(vcsourcekit)')

To properly use variables within regex patterns, you need to construct the regex string using string concatenation or formatting:

# Correct solution 1: String concatenation
when: item.key | regex_search('^(' + vcsourcekit + ')')

# Correct solution 2: Using string formatting
when: item.key | regex_search('^%s' % vcsourcekit)

# Correct solution 3: Modern format strings (Python 3.6+)
when: item.key | regex_search(f'^{vcsourcekit}')

Here's a full playbook example demonstrating dynamic pattern matching:

- name: Match VMs with specific source kit version
  hosts: localhost
  vars:
    vcsourcekit: 10
    vmfacts:
      virtual_machines:
        vm1:
          key: "10.0.1"
        vm2:
          key: "9.5.3"

  tasks:
    - name: Process matching VMs
      debug:
        msg: "Found matching VM {{ item.key }}"
      when: item.key | regex_search('^' + vcsourcekit)
      with_dict: "{{ vmfacts.virtual_machines }}"

For more complex scenarios, you can combine multiple variables in your regex patterns:

- name: Match with multiple dynamic components
  vars:
    version_prefix: "10"
    os_type: "linux"
  tasks:
    - debug:
        msg: "Complex pattern match successful"
      when: item.key | regex_search(f'^{version_prefix}.*{os_type}')

When dealing with large datasets, consider these optimizations:

1. Pre-compile complex patterns using regex_escape when variables contain special regex characters
2. Use more specific anchors (^, $) to improve matching speed
3. Cache frequently used patterns when possible

If your patterns aren't matching as expected:

# Debug the actual pattern being used
- debug:
    var: '^' + vcsourcekit

# Verify the variable contents
- debug:
    var: vcsourcekit

# Test with literal values first
- debug:
    msg: "Test match"
  when: "10.0.1" | regex_search('^10')

When working with Ansible's regex_search filter, a common stumbling block is attempting to match against dynamic variables rather than static patterns. The issue arises because Jinja2 templates evaluate variables before regex processing occurs.

In your specific case where vcsourcekit = 10, you're trying to match the pattern ^10 but the filter interprets '^(vcsourcekit)' literally rather than evaluating the variable first. Here's why this happens:

- name: Do something awesome
  vmware_guest:
    hostname: "{{ vcenterhostname }}"
    ...
  when:
      - item.key | regex_search('^(vcsourcekit)')  # Problematic line
  with_dict: "{{ vmfacts.virtual_machines }}"

To properly match against a variable's value, you need to construct the regex pattern using string concatenation:

- name: Correct variable matching
  vmware_guest:
    hostname: "{{ vcenterhostname }}"
    ...
  when:
      - item.key | regex_search('^' ~ vcsourcekit)
  with_dict: "{{ vmfacts.virtual_machines }}"

For more complex scenarios, you might need:

- name: Match with multiple variables
  debug:
    msg: "Pattern matched"
  when:
    - item.key | regex_search('^' ~ var1 ~ '.*' ~ var2)

Or using the regex_replace filter to create patterns:

- name: Dynamic pattern construction
  debug:
    msg: "{{ '^prefix_' ~ dynamic_var ~ '_suffix$' | regex_replace('_', '\-') }}"

Here's how we implemented version matching in a real-world scenario:

- name: Validate software versions
  fail:
    msg: "Incompatible version detected"
  when:
    - current_version | regex_search('^' ~ min_required_version)