Finished Iframe Implementation

filter_plugins/csp_hashes.py

@@ -0,0 +1,30 @@
+from ansible.errors import AnsibleFilterError
+import copy
+
+def append_csp_hash(applications, application_id, code_one_liner):
+    """
+    Ensures that applications[application_id].csp.hashes['script-src-elem']
+    exists and appends the given one-liner (if not already present).
+    """
+    if not isinstance(applications, dict):
+        raise AnsibleFilterError("`applications` must be a dict")
+    if application_id not in applications:
+        raise AnsibleFilterError(f"Unknown application_id: {application_id}")
+
+    apps = copy.deepcopy(applications)
+    app = apps[application_id]
+    csp = app.setdefault('csp', {})
+    hashes = csp.setdefault('hashes', {})
+
+    existing = hashes.get('script-src-elem', [])
+    if code_one_liner not in existing:
+        existing.append(code_one_liner)
+    hashes['script-src-elem'] = existing
+
+    return apps
+
+class FilterModule(object):
+    def filters(self):
+        return {
+            'append_csp_hash': append_csp_hash
+        }

filter_plugins/text_filters.py

@@ -0,0 +1,44 @@
+# filter_plugins/text_filters.py
+
+from ansible.errors import AnsibleFilterError
+import re
+
+def to_one_liner(s):
+    """
+    Collapse any multi-line string into a single line,
+    trim extra whitespace, and remove JavaScript comments.
+    Supports removal of both '//' line comments and '/*...*/' block comments,
+    but preserves '//' inside string literals and templating expressions.
+    """
+    if not isinstance(s, str):
+        raise AnsibleFilterError("to_one_liner() expects a string")
+
+    # 1) Remove block comments /* ... */
+    no_block_comments = re.sub(r'/\*.*?\*/', '', s, flags=re.DOTALL)
+
+    # 2) Extract string literals to protect them from comment removal
+    string_pattern = re.compile(r"'(?:\\.|[^'\\])*'|\"(?:\\.|[^\"\\])*\"")
+    literals = []
+    def _extract(match):
+        idx = len(literals)
+        literals.append(match.group(0))
+        return f"__STR{idx}__"
+    temp = string_pattern.sub(_extract, no_block_comments)
+
+    # 3) Remove line comments // ...
+    temp = re.sub(r'//.*$', '', temp, flags=re.MULTILINE)
+
+    # 4) Restore string literals
+    for idx, lit in enumerate(literals):
+        temp = temp.replace(f"__STR{idx}__", lit)
+
+    # 5) Collapse all whitespace
+    one_liner = re.sub(r'\s+', ' ', temp).strip()
+
+    return one_liner
+
+class FilterModule(object):
+    def filters(self):
+        return {
+            'to_one_liner': to_one_liner,
+        }