refactor: configurable home dir + unified pathspec ignore

graphify/benchmark.py

@@ -62,19 +62,22 @@ def _query_subgraph_tokens(G: nx.Graph, question: str, depth: int = 3) -> int:
 
 
 def run_benchmark(
-    graph_path: str = "graphify-out/graph.json",
+    graph_path: str | Path | None = None,
     corpus_words: int | None = None,
     questions: list[str] | None = None,
 ) -> dict:
     """Measure token reduction: corpus tokens vs graphify query tokens.
 
     Args:
-        graph_path: path to the built graph
+        graph_path: path to the built graph; defaults to ``$GRAPHIFY_HOME/graph.json``
         corpus_words: total word count from detect() output; if None, estimated from graph
         questions: list of questions to benchmark; defaults to _SAMPLE_QUESTIONS
 
     Returns dict with: corpus_tokens, avg_query_tokens, reduction_ratio, per_question
     """
+    if graph_path is None:
+        from . import paths as _paths
+        graph_path = _paths.graph_path()
     data = json.loads(Path(graph_path).read_text(encoding="utf-8"))
     try:
         G = json_graph.node_link_graph(data, edges="links")

graphify/build.py

@@ -180,16 +180,21 @@ def deduplicate_by_label(nodes: list[dict], edges: list[dict]) -> tuple[list[dic
 
 def build_merge(
     new_chunks: list[dict],
-    graph_path: str | Path = "graphify-out/graph.json",
+    graph_path: str | Path | None = None,
     prune_sources: list[str] | None = None,
     *,
     directed: bool = False,
 ) -> nx.Graph:
     """Load existing graph.json, merge new chunks into it, and save back.
 
+    *graph_path* defaults to ``$GRAPHIFY_HOME/graph.json``.
+
     Never replaces — only grows (or prunes deleted-file nodes via prune_sources).
     Safe to call repeatedly: existing nodes and edges are preserved.
     """
+    if graph_path is None:
+        from . import paths as _paths
+        graph_path = _paths.graph_path()
     from networkx.readwrite import json_graph as _jg
 
     graph_path = Path(graph_path)

graphify/cache.py

@@ -44,17 +44,19 @@ def file_hash(path: Path, root: Path = Path(".")) -> str:
 
 
 def cache_dir(root: Path = Path(".")) -> Path:
-    """Returns graphify-out/cache/ - creates it if needed."""
-    d = Path(root).resolve() / "graphify-out" / "cache"
-    d.mkdir(parents=True, exist_ok=True)
-    return d
+    """Per-file extraction cache directory under *root* — created if missing.
+
+    Resolves to ``$GRAPHIFY_HOME/cache``. See :mod:`graphify.paths`.
+    """
+    from . import paths
+    return paths.cache_dir(root, create=True)
 
 
 def load_cached(path: Path, root: Path = Path(".")) -> dict | None:
     """Return cached extraction for this file if hash matches, else None.
 
     Cache key: SHA256 of file contents.
-    Cache value: stored as graphify-out/cache/{hash}.json
+    Cache value: stored as ``<cache_dir>/{hash}.json``.
     Returns None if no cache entry or file has changed.
     """
     try:
@@ -73,7 +75,7 @@ def load_cached(path: Path, root: Path = Path(".")) -> dict | None:
 def save_cached(path: Path, result: dict, root: Path = Path(".")) -> None:
     """Save extraction result for this file.
 
-    Stores as graphify-out/cache/{hash}.json where hash = SHA256 of current file contents.
+    Stores as ``<cache_dir>/{hash}.json`` where hash = SHA256 of current file contents.
     result should be a dict with 'nodes' and 'edges' lists.
 
     No-ops if `path` is not a regular file. Subagent-produced semantic fragments
@@ -108,7 +110,7 @@ def cached_files(root: Path = Path(".")) -> set[str]:
 
 
 def clear_cache(root: Path = Path(".")) -> None:
-    """Delete all graphify-out/cache/*.json files."""
+    """Delete all cached extraction entries."""
     d = cache_dir(root)
     for f in d.glob("*.json"):
         f.unlink()