Implemented a --modules CLI flag to allow runtime selection of the modules to be run

tests/test_new_pipeline.py

@@ -2,11 +2,11 @@ import json
 
 import pytest
 
-from audita.core.config import AuditaConfig
+from audita.core.config import AuditaConfig, ConfigOverrides
 from audita.core.errors import AuditaLLMError
 from audita.core.io import write_report
 from audita.core.schemas import parse_glossary_yaml, parse_source_transcript_json
-from audita.modules import default_module_specs
+from audita.modules import DEFAULT_MODULE_KEYS, default_module_specs, resolve_module_specs
 from audita.pipeline import process_transcript, process_transcript_result
 
 
@@ -73,9 +73,9 @@ def test_process_transcript_runs_noop_framework(tmp_path):
     assert revised[0].text == "Hello. Again."
     assert revised[1].text == "Done."
     assert [call["stage_name"] for call in llm_client.calls] == [
-        "glossary_primary:proposal",
+        "glossary_1:proposal",
         "homophones:proposal",
-        "glossary_secondary:proposal",
+        "glossary_2:proposal",
     ]
 
 
@@ -111,9 +111,9 @@ def test_process_transcript_result_writes_report_and_preserves_skips_per_policy(
 
     assert result.work_dir_retained is True
     assert result.report.pipeline == [
-        "glossary_primary",
+        "glossary_1",
         "homophones",
-        "glossary_secondary",
+        "glossary_2",
         "spoken_word",
         "grammar",
     ]
@@ -142,13 +142,21 @@ def test_external_report_can_be_written(tmp_path):
     write_report(report_path, result.report)
 
     payload = json.loads(report_path.read_text(encoding="utf-8"))
-    assert payload["pipeline"][0] == "glossary_primary"
+    assert payload["pipeline"][0] == "glossary_1"
     assert payload["totals"]["applied_change_count"] == 0
 
 
 def test_default_module_specs_expose_final_validator_order():
     specs = default_module_specs()
 
+    assert DEFAULT_MODULE_KEYS == ("glossary", "homophones", "glossary", "spoken_word", "grammar")
+    assert [spec.instance_name for spec in specs] == [
+        "glossary_1",
+        "homophones",
+        "glossary_2",
+        "spoken_word",
+        "grammar",
+    ]
     assert [validator.name for validator in specs[0].module.validators()] == [
         "proposal_confidence_guard",
         "protected_glossary_guard",
@@ -201,9 +209,9 @@ def test_process_transcript_result_missing_api_key_writes_failed_report(tmp_path
     assert report["status"] == "failed"
     assert report["normalization"]["normalized_segment_count"] == 2
     assert report["pipeline"] == [
-        "glossary_primary",
+        "glossary_1",
         "homophones",
-        "glossary_secondary",
+        "glossary_2",
         "spoken_word",
         "grammar",
     ]
@@ -286,7 +294,7 @@ def test_process_transcript_result_preserves_partial_progress_when_later_module_
 
     assert report["status"] == "failed"
     assert report["normalization"]["normalized_segment_count"] == 1
-    assert [module["instance_name"] for module in report["modules"]] == ["glossary_primary"]
+    assert [module["instance_name"] for module in report["modules"]] == ["glossary_1"]
     assert report["applied_changes"][0]["corrected_text"] == "Jesters"
     assert report["applied_changes"][0]["segment_text_after"] == "There were Jesters at the dam."
     assert report["totals"]["applied_change_count"] == 1
@@ -351,7 +359,7 @@ def test_process_transcript_result_preserves_partial_skips_and_validator_diagnos
 
     run_dir = next((tmp_path / "work").iterdir())
     report = json.loads((run_dir / "report.json").read_text(encoding="utf-8"))
-    validator_dir = run_dir / "glossary_primary"
+    validator_dir = run_dir / "glossary_1"
 
     assert report["status"] == "failed"
     assert report["modules"] == []
@@ -361,3 +369,29 @@ def test_process_transcript_result_preserves_partial_skips_and_validator_diagnos
     assert "unknown correction_index" in report["error"]
     assert (validator_dir / "spoken_form_plausibility_review-prompt-0000.json").exists()
     assert (validator_dir / "spoken_form_plausibility_review-response-0000.json").exists()
+
+
+def test_resolve_module_specs_numbers_repeated_keys():
+    specs = resolve_module_specs(["glossary", "homophones", "glossary"])
+
+    assert [spec.instance_name for spec in specs] == ["glossary_1", "homophones", "glossary_2"]
+    assert [spec.module_key for spec in specs] == ["glossary", "homophones", "glossary"]
+
+
+def test_process_transcript_result_supports_grammar_only_module_override(tmp_path):
+    config = AuditaConfig.from_sources(
+        env={},
+        overrides=ConfigOverrides(work_dir=tmp_path / "work", work_dir_retention="always"),
+    )
+
+    result = process_transcript_result(
+        _transcript(),
+        _glossary(),
+        config,
+        module_keys=["grammar"],
+        llm_client=FakeStructuredLLMClient([]),
+    )
+
+    assert [segment.id for segment in result.transcript] == [1, 2]
+    assert result.report.pipeline == ["grammar"]
+    assert result.report.totals["applied_change_count"] == 0