feat(rules) add support of proper yaml arrays in rules

Author: llalyeva

pacta/rules/parser.py

@@ -79,23 +79,11 @@ def parse_file(self, path: str | Path) -> RulesDocumentAst:
     def parse_text(self, text: str, *, filename: str | None = None) -> RulesDocumentAst:
         filename = filename or "<rules>"
 
-        blocks = self._split_rule_blocks(text)
-        rules: list[RuleAst] = []
-
-        for i, block in enumerate(blocks, start=1):
-            try:
-                rules.append(self._parse_rule_block(block, filename=filename))
-            except RulesParseError:
-                raise
-            except Exception as e:
-                raise RulesParseError(
-                    message=f"Failed to parse rule block #{i}: {e}",
-                    file=filename,
-                ) from e
-
-        return RulesDocumentAst(rules=tuple(rules), span=SourceSpan(file=filename))
-
-    def _parse_rule_block(self, block: str, *, filename: str) -> RuleAst:
+        parsed_rules = self._parse_rule_block(text,filename=filename)
+
+        return RulesDocumentAst(rules=tuple(parsed_rules), span=SourceSpan(file=filename))
+
+    def _parse_rule_block(self, block: str, *, filename: str) -> list[RuleAst]:
         try:
             import yaml
         except ImportError as e:
@@ -109,13 +97,26 @@ def _parse_rule_block(self, block: str, *, filename: str) -> RuleAst:
         except Exception as e:
             raise RulesParseError(message=f"Invalid YAML in rule block: {e}", file=filename) from e
 
-        if not isinstance(doc, Mapping) or doc.get("rule") is None:
-            raise RulesParseError(message="Rule block must contain top-level 'rule:' mapping", file=filename)
+        if not isinstance(doc, Mapping) or doc.get("rules") is None:
+            raise RulesParseError(message="Rule block must contain top-level 'rules:' list", file=filename)
+
+        rules_list = doc["rules"]
+        if not isinstance(rules_list, list):
+            raise RulesParseError(message="'rules list' must be a list", file=filename)
+
+        parsed_rules = []
 
-        root = doc["rule"]
-        if not isinstance(root, Mapping):
-            raise RulesParseError(message="'rule' must be a mapping", file=filename)
+        for root in rules_list:
+            if not isinstance(root, Mapping):
+                raise RulesParseError(message="'rule' must be a mapping", file=filename)
+            rule_item = self._add_rule_in_list(root, filename = filename)
+            parsed_rules.append(rule_item)
 
+        return parsed_rules    
+
+
+
+    def _add_rule_in_list(self, root: Mapping, *, filename:str) -> RuleAst:
         # Required fields
         rid = self._req_str(root, "id", filename)
         name = self._req_str(root, "name", filename)
@@ -152,10 +153,6 @@ def _parse_rule_block(self, block: str, *, filename: str) -> RuleAst:
             metadata={"parser": "dsl-v0+pyyaml"},
         )
 
-    # Keep your existing _split_rule_blocks, _parse_when_predicate, _parse_pred_item,
-    # _parse_inline_compare, _parse_literal, _req_str, _opt_str as-is (they already handle
-    # strings like "from.layer == domain" and nested any/all/not).
-
     def _split_rule_blocks(self, text: str) -> list[str]:
         lines = [ln.rstrip("\n") for ln in text.splitlines()]
         blocks: list[list[str]] = []
@@ -177,7 +174,7 @@ def flush():
         filtered = []
         for b in blocks:
             first = next((ln for ln in b if ln.strip() and not ln.strip().startswith("#")), "")
-            if first.strip() != "rule:":
+            if first.strip() != "rules:":
                 continue
             filtered.append("\n".join(b))
         return filtered