codedd-cli 0.1.6__tar.gz → 0.1.7__tar.gz

{codedd_cli-0.1.6 → codedd_cli-0.1.7}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: codedd-cli
-Version: 0.1.6
+Version: 0.1.7
 Summary: CLI tool for CodeDD — run code audits from your terminal
 License: MIT
 License-File: LICENSE
@@ -25,6 +25,7 @@ Requires-Dist: defusedxml (>=0.7.1)
 Requires-Dist: esprima (>=4.0.0) ; extra == "esprima"
 Requires-Dist: httpx (>=0.27.0)
 Requires-Dist: keyring (>=25.0.0)
+Requires-Dist: langgraph (>=0.2.0)
 Requires-Dist: lizard (>=1.17.0)
 Requires-Dist: radon (>=6.0.0)
 Requires-Dist: rich (>=13.0.0)
@@ -257,7 +258,7 @@ Run: codedd audit start
 | `codedd config show-keys` | List which providers have keys configured (not the keys themselves) |
 | `codedd config remove-key <anthropic\|openai>` | Remove a stored LLM API key from keychain |
 | `codedd config provider [anthropic\|openai\|both]` | Set preferred LLM provider |
-| `codedd config concurrency <n>` | Set max concurrent LLM requests (default 6) |
+| `codedd config concurrency <n>` | Set max concurrent LLM requests (1–75, default 4) |
 
 ---
 

{codedd_cli-0.1.6 → codedd_cli-0.1.7}/README.md

@@ -220,7 +220,7 @@ Run: codedd audit start
 | `codedd config show-keys` | List which providers have keys configured (not the keys themselves) |
 | `codedd config remove-key <anthropic\|openai>` | Remove a stored LLM API key from keychain |
 | `codedd config provider [anthropic\|openai\|both]` | Set preferred LLM provider |
-| `codedd config concurrency <n>` | Set max concurrent LLM requests (default 6) |
+| `codedd config concurrency <n>` | Set max concurrent LLM requests (1–75, default 4) |
 
 ---
 

{codedd_cli-0.1.6 → codedd_cli-0.1.7}/codedd_cli/__init__.py

@@ -1,3 +1,3 @@
 """CodeDD CLI — run code audits from your terminal."""
 
-__version__ = "0.1.6"
+__version__ = "0.1.7"

{codedd_cli-0.1.6 → codedd_cli-0.1.7}/codedd_cli/api/endpoints.py

@@ -19,7 +19,6 @@ class Endpoints:
     # Scope
     REGISTER_SCOPE = "/api/cli/scope/register/"
     SCOPE_FILES = "/api/cli/scope/files/"
-    # Legacy web route reused by the CLI (django_codedd.urls → api/delete_repo_from_scope/).
     DELETE_REPO_FROM_SCOPE = "/api/delete_repo_from_scope/"
 
     # Audit lifecycle
@@ -30,6 +29,7 @@ class Endpoints:
 
     # Local audit execution (CLI-side file auditing)
     AUDIT_PLAN = "/api/cli/audit/plan/"
+    AUDIT_PROGRESS = "/api/cli/audit/progress/"
     AUDIT_RESULTS = "/api/cli/audit/results/"
     AUDIT_COMPLETE = "/api/cli/audit/complete/"
 

{codedd_cli-0.1.6 → codedd_cli-0.1.7}/codedd_cli/auditor/archetype_classifier.py

@@ -16,12 +16,10 @@ identical tier definitions.  The only difference is that it does NOT persist
 results to PostgreSQL and operates without Phase 3 graph nodes/edges — the
 _count_deployable_units method falls back to component-level heuristics when
 graph data is unavailable.
-
-Parity reference: E:/CodeDD/django/auditor/auditing_functions/functions/step_5/
-                  architecture_analysis/archetype_classifier.py
 """
 
 import logging
+import re
 import time
 from typing import Any
 
@@ -133,9 +131,38 @@ class DimensionScorer:
                 patterns.append(val)
         return patterns
 
+    @staticmethod
+    def _normalize_token(value: str) -> str:
+        """Lowercase and collapse separators (``_ - . /`` + whitespace) to spaces.
+
+        This lets a single keyword match canonical tech ids regardless of the
+        separator style used upstream — e.g. ``spring`` matches ``spring_boot``,
+        ``spring-boot`` and ``spring boot`` alike — while still respecting word
+        boundaries so ``spring`` does NOT match ``mainspring``.
+        """
+        return re.sub(r"[\s_\-./]+", " ", value.lower()).strip()
+
+    def _matches(self, item: str, needle: str) -> bool:
+        """Left-boundary match of *needle* within *item* after normalization.
+
+        Replaces the previous raw substring test, which produced mid-word false
+        positives (e.g. ``spring`` in ``offspring``/``mainspring``).  We require
+        the needle to begin at a token boundary but deliberately do NOT require a
+        right boundary, so canonical tech ids that concatenate suffixes still
+        match (``express`` → ``expressjs``, ``next`` → ``nextjs``, ``postgres``
+        → ``postgresql``).  Multi-word needles (``api gateway``, ``kafka
+        streams``) work because both sides are separator-normalized first.
+        """
+        item_norm = self._normalize_token(item)
+        needle_norm = self._normalize_token(needle)
+        if not needle_norm:
+            return False
+        pattern = rf"(?<![a-z0-9]){re.escape(needle_norm)}"
+        return re.search(pattern, item_norm) is not None
+
     def _has_any(self, haystack: list[str], *needles: str) -> bool:
         return any(
-            needle in item
+            self._matches(item, needle)
             for item in haystack
             for needle in needles
         )
@@ -143,7 +170,7 @@ class DimensionScorer:
     def _count_matches(self, haystack: list[str], *needles: str) -> int:
         return sum(
             1 for item in haystack
-            if any(needle in item for needle in needles)
+            if any(self._matches(item, needle) for needle in needles)
         )
 
     def _evidence(