Skip to content

Codebase intelligence for AI and humans: code health scores, auto-generated docs, git analytics, dead code detection, and architectural decisions via MCP.

fix(stats): one symbol count across status, dashboard, workspace and get_overview (#4367)

Raghav Chamadiya04b98f79fd0e
diff --git a/docs/reference/COMPUTED_GLOSSARY.md b/docs/reference/COMPUTED_GLOSSARY.md
index f03a3a4ca..758495701 100644
--- a/docs/reference/COMPUTED_GLOSSARY.md
+++ b/docs/reference/COMPUTED_GLOSSARY.md
@@ -72,6 +72,7 @@ workspace overlays, MCP responses, and CLI output.
| Symbol node | Graph node for an extracted symbol. | `GraphBuilder.add_file()` | `{node_type: "symbol", kind: "function", name: "main"}` |
| External node | Node for third-party or unresolvable dependencies. | Import resolution paths | `external:react` |
| Synthetic module symbol | Symbol node for top-level calls in a file. | `GraphBuilder.add_file()` | `src/app.py::__module__` |
+| Symbol count | The "symbols" figure on `repowise status`, the dashboard, workspace views and `get_overview`: one per distinct symbol id in the symbol index. An overload set counts once, and the synthetic module symbol is not counted. | `count_symbols()` | `76,924 symbols` |
| `defines` edge | File-to-symbol containment. | `GraphBuilder.add_file()` | `src/app.py -> src/app.py::main` |
| `imports` edge | File-to-file import relationship. | `GraphBuilder.build()` | `src/app.py -> src/settings.py` |
| `imported_names` edge payload | Names imported along an import edge. | `GraphBuilder.build()` | `["Settings", "load_config"]` |
diff --git a/packages/cli/src/repowise/cli/commands/status_cmd.py b/packages/cli/src/repowise/cli/commands/status_cmd.py
index 63b6a2c5a..feffdc1ff 100644
--- a/packages/cli/src/repowise/cli/commands/status_cmd.py
+++ b/packages/cli/src/repowise/cli/commands/status_cmd.py
@@ -69,6 +69,7 @@ def _query_repo_counts(repo_path: Path) -> tuple[int, int]:
create_session_factory,
get_session,
)
+ from repowise.core.persistence.crud import count_symbols
from repowise.core.persistence.models import GraphNode, Repository
url = get_db_url_for_repo(repo_path)
@@ -87,7 +88,6 @@ def _query_repo_counts(repo_path: Path) -> tuple[int, int]:
if repo_id is None:
return 0, 0
- # Count file nodes and symbol nodes
file_count_result = await session.execute(
sa_select(sa_func.count())
.select_from(GraphNode)
@@ -96,17 +96,9 @@ def _query_repo_counts(repo_path: Path) -> tuple[int, int]:
GraphNode.node_type == "file",
)
)
- symbol_count_result = await session.execute(
- sa_select(sa_func.count())
- .select_from(GraphNode)
- .where(
- GraphNode.repository_id == repo_id,
- GraphNode.node_type == "symbol",
- )
- )
return (
file_count_result.scalar_one(),
- symbol_count_result.scalar_one(),
+ await count_symbols(session, repo_id),
)
finally:
await engine.dispose()
diff --git a/packages/core/src/repowise/core/ingestion/graph/builder.py b/packages/core/src/repowise/core/ingestion/graph/builder.py
index 71fb2a68d..5ddfb5680 100644
--- a/packages/core/src/repowise/core/ingestion/graph/builder.py
+++ b/packages/core/src/repowise/core/ingestion/graph/builder.py
@@ -278,7 +278,7 @@ class GraphBuilder(MetricsMixin, ResolveMixin, EdgesMixin, SerializeMixin, Rehyd
path,
node_type="file",
language=parsed.file_info.language,
- symbol_count=len(parsed.symbols),
+ symbol_count=parsed.symbol_count,
has_error=bool(parsed.parse_errors),
is_test=parsed.file_info.is_test,
is_entry_point=parsed.file_info.is_entry_point,
diff --git a/packages/core/src/repowise/core/ingestion/models.py b/packages/core/src/repowise/core/ingestion/models.py
index e67bd5e29..e89a97761 100644
--- a/packages/core/src/repowise/core/ingestion/models.py
+++ b/packages/core/src/repowise/core/ingestion/models.py
@@ -752,6 +752,12 @@ class ParsedFile:
# resolution tiers. They become ``references`` edges, not ``calls``.
references: list[CallSite] = field(default_factory=list)
+ @property
+ def symbol_count(self) -> int:
+ """Distinct symbols. An overload set or a re-declaration shares one id,
+ and the graph and ``wiki_symbols`` both store it once."""
+ return len({sym.id for sym in self.symbols})
+
def symbol_id_language(parsed_files: dict[str, ParsedFile], symbol_id: str) -> str | None:
"""Language of the file a symbol ID belongs to, or ``None`` if unparsed.
diff --git a/packages/core/src/repowise/core/persistence/crud/external_systems.py b/packages/core/src/repowise/core/persistence/crud/external_systems.py
index 96f66886d..21f89ad0e 100644
--- a/packages/core/src/repowise/core/persistence/crud/external_systems.py
+++ b/packages/core/src/repowise/core/persistence/crud/external_systems.py
@@ -8,7 +8,7 @@ from __future__ import annotations
from typing import Any
-from sqlalchemy import select
+from sqlalchemy import func, select
from sqlalchemy.ext.asyncio import AsyncSession
from ...analysis.external_systems.links import (
@@ -258,6 +258,28 @@ async def batch_upsert_symbols(
)
+async def count_symbols_by_repo(session: AsyncSession) -> dict[str, int]:
+ """Indexed symbols per repository: the figure every surface calls "symbols".
+
+ One ``wiki_symbols`` row per symbol id, so an overload set counts once, and
+ the graph's per-file ``__module__`` node is not a symbol.
+ """
+ rows = await session.execute(
+ select(WikiSymbol.repository_id, func.count(WikiSymbol.id)).group_by(
+ WikiSymbol.repository_id
+ )
+ )
+ return {repo_id: int(n) for repo_id, n in rows.all()}
+
+
+async def count_symbols(session: AsyncSession, repository_id: str) -> int:
+ """:func:`count_symbols_by_repo` for one repository."""
+ n = await session.scalar(
+ select(func.count(WikiSymbol.id)).where(WikiSymbol.repository_id == repository_id)
+ )
+ return int(n or 0)
+
+
# Chunk size for the scoped existence SELECT — stays under SQLite's
# host-parameter limit when a wide catch-up update touches many files.
_SYMBOL_RECONCILE_CHUNK = 400
diff --git a/packages/core/src/repowise/core/pipeline/_run/index.py b/packages/core/src/repowise/core/pipeline/_run/index.py
index 108cf8073..c4d533828 100644
--- a/packages/core/src/repowise/core/pipeline/_run/index.py
+++ b/packages/core/src/repowise/core/pipeline/_run/index.py
@@ -174,7 +174,7 @@ def _emit_index_summary(run: PipelineRun) -> None:
progress.on_message(
"info",
f"→ {len(run.parsed_files):,} files parsed · "
- f"{sum(len(pf.symbols) for pf in run.parsed_files):,} symbols extracted",
+ f"{sum(pf.symbol_count for pf in run.parsed_files):,} symbols extracted",
)
progress.on_message(
"info",
diff --git a/packages/core/src/repowise/core/pipeline/orchestrator.py b/packages/core/src/repowise/core/pipeline/orchestrator.py
index 25cdd4ff2..a0207cced 100644
--- a/packages/core/src/repowise/core/pipeline/orchestrator.py
+++ b/packages/core/src/repowise/core/pipeline/orchestrator.py
@@ -317,7 +317,7 @@ def _build_result(run: PipelineRun) -> PipelineResult:
file_infos = run.file_infos
parsed_files = run.parsed_files
languages = {fi.language for fi in file_infos if hasattr(fi, "language") and fi.language}
- symbol_count = sum(len(pf.symbols) for pf in parsed_files)
+ symbol_count = sum(pf.symbol_count for pf in parsed_files)
resume_controller = run.resume_controller
return PipelineResult(
diff --git a/packages/core/src/repowise/core/stats_highlights.py b/packages/core/src/repowise/core/stats_highlights.py
index 349cd0d51..a0d582f83 100644
--- a/packages/core/src/repowise/core/stats_highlights.py
+++ b/packages/core/src/repowise/core/stats_highlights.py
@@ -147,20 +147,17 @@ def _ranked_languages(counts: Counter[str]) -> list[dict[str, Any]]:
return [{"language": lang, "file_count": n} for lang, n in ranked]
-def build_scale(file_nodes: Iterable[Row], metrics: Iterable[Row]) -> dict[str, Any]:
+def build_scale(
+ file_nodes: Iterable[Row], metrics: Iterable[Row], symbol_count: int
+) -> dict[str, Any]:
"""Counts and NLOC, plus the :func:`file_mix` languages and module count.
``file_nodes`` are graph file nodes; external dependency nodes are dropped
- so the file count describes the repo.
+ so the file count describes the repo. ``symbol_count`` is
+ ``crud.count_symbols``, the figure every other surface shows.
"""
file_nodes = list(file_nodes)
- file_count = 0
- symbol_count = 0
- for node in file_nodes:
- if _is_external(node):
- continue
- file_count += 1
- symbol_count += _int(node.get("symbol_count"))
+ file_count = sum(1 for node in file_nodes if not _is_external(node))
total_nloc = 0
test_nloc = 0
diff --git a/packages/core/src/repowise/core/workspace/update/repo.py b/packages/core/src/repowise/core/workspace/update/repo.py
index 067a73709..f93c15def 100644
--- a/packages/core/src/repowise/core/workspace/update/repo.py
+++ b/packages/core/src/repowise/core/workspace/update/repo.py
@@ -248,7 +248,7 @@ async def _incremental_repo_update(
alias=alias,
updated=True,
file_count=file_count,
- symbol_count=sum(len(pf.symbols) for pf in parsed_files),
+ symbol_count=sum(pf.symbol_count for pf in parsed_files),
kg_state=_export_kg_state(repo_path, alias, kg),
working_tree_paths=working_tree_paths,
git_history_refreshed=bool(full_git_summaries) and "Git persist" not in failed_steps,
diff --git a/packages/server/src/repowise/server/mcp_server/tool_overview/workspace.py b/packages/server/src/repowise/server/mcp_server/tool_overview/workspace.py
index 846982467..1a6016fba 100644
--- a/packages/server/src/repowise/server/mcp_server/tool_overview/workspace.py
+++ b/packages/server/src/repowise/server/mcp_server/tool_overview/workspace.py
@@ -8,6 +8,7 @@ from sqlalchemy import func as sa_func
from sqlalchemy import select
from repowise.core.ingestion.languages.registry import REGISTRY
+from repowise.core.persistence.crud import count_symbols
from repowise.core.persistence.crud.analysis.health.metrics import count_health_files
from repowise.core.persistence.database import get_session
from repowise.core.persistence.models import GraphNode
@@ -62,7 +63,7 @@ async def _repo_entry(ctx: Any, registry: Any) -> dict[str, Any]:
"path": str(ctx.path),
"summary": _lead_sentence(ov_content),
"code_files": await _code_files(session, repo_obj.id),
- "symbol_count": await _count_nodes(session, repo_obj.id, "symbol"),
+ "symbol_count": await count_symbols(session, repo_obj.id),
"is_default": registry is not None and ctx.alias == registry.get_default_alias(),
}
@@ -103,18 +104,6 @@ async def _code_files(session: Any, repository_id: str) -> int:
return result.scalar_one()
-async def _count_nodes(session: Any, repository_id: str, node_type: str) -> int:
- result = await session.execute(
- select(sa_func.count())
- .select_from(GraphNode)
- .where(
- GraphNode.repository_id == repository_id,
- GraphNode.node_type == node_type,
- )
- )
- return result.scalar_one()
-
-
def _cross_repo_topology(repos_info: list[dict[str, Any]]) -> dict[str, Any]:
"""Cross-repo summary and contracts; also tags each repo with the repos it depends on."""
enricher = _enricher_with_data()
diff --git a/packages/server/src/repowise/server/routers/_repo_summary.py b/packages/server/src/repowise/server/routers/_repo_summary.py
index 9d5158da8..abad6a15b 100644
--- a/packages/server/src/repowise/server/routers/_repo_summary.py
+++ b/packages/server/src/repowise/server/routers/_repo_summary.py
@@ -13,6 +13,7 @@ from sqlalchemy import case, func, select
from sqlalchemy.exc import SQLAlchemyError
from sqlalchemy.ext.asyncio import AsyncSession
+from repowise.core.persistence.crud import count_symbols_by_repo
from repowise.core.persistence.crud.analysis.shown import ungated
from repowise.core.persistence.models import (
DeadCodeFinding,
@@ -39,25 +40,22 @@ async def _file_figures(session: AsyncSession) -> _Figures:
select(
GraphNode.repository_id,
func.count(GraphNode.id),
- func.coalesce(func.sum(GraphNode.symbol_count), 0),
_fresh_case(GraphNode.is_entry_point, True),
)
.where(GraphNode.node_type == "file")
.group_by(GraphNode.repository_id)
)
return [
- (
- repo_id,
- {
- "file_count": int(files or 0),
- "symbol_count": int(symbols or 0),
- "entry_point_count": int(entries or 0),
- },
- )
- for repo_id, files, symbols, entries in result.all()
+ (repo_id, {"file_count": int(files or 0), "entry_point_count": int(entries or 0)})
+ for repo_id, files, entries in result.all()
]
+async def _symbol_figures(session: AsyncSession) -> _Figures:
+ counts = await count_symbols_by_repo(session)
+ return [(repo_id, {"symbol_count": n}) for repo_id, n in counts.items()]
+
+
async def _page_figures(session: AsyncSession) -> _Figures:
# Documentation pages and the fresh subset. Never selects `content`.
result = await session.execute(
@@ -134,6 +132,7 @@ async def _health_figures(session: AsyncSession) -> _Figures:
_SECTIONS = (
_file_figures,
+ _symbol_figures,
_page_figures,
_dead_export_figures,
_hotspot_figures,
diff --git a/packages/server/src/repowise/server/routers/repos.py b/packages/server/src/repowise/server/routers/repos.py
index 408434b96..3ddde678d 100644
--- a/packages/server/src/repowise/server/routers/repos.py
+++ b/packages/server/src/repowise/server/routers/repos.py
@@ -394,13 +394,7 @@ async def get_repo_stats(
),
0,
)
- symbol_count = int(
- await _scalar_or(
- session,
- select(func.sum(GraphNode.symbol_count)).where(GraphNode.repository_id == repo_id),
- 0,
- )
- )
+ symbol_count = await crud.count_symbols(session, repo_id)
entry_point_count = await _scalar_or(
session,
select(func.count(GraphNode.id)).where(
diff --git a/packages/server/src/repowise/server/routers/stats.py b/packages/server/src/repowise/server/routers/stats.py
index 649dac06b..7b9b5689d 100644
--- a/packages/server/src/repowise/server/routers/stats.py
+++ b/packages/server/src/repowise/server/routers/stats.py
@@ -179,7 +179,6 @@ async def stats_highlights(
select(
GraphNode.node_id,
GraphNode.language,
- GraphNode.symbol_count,
GraphNode.external_system_id,
).where(GraphNode.repository_id == repo_id, GraphNode.node_type == "file"),
)
@@ -247,7 +246,7 @@ async def stats_highlights(
index_size = await index_storage_bytes(repo.local_path)
return {
"repo": {"id": repo.id, "name": repo.name},
- "scale": sh.build_scale(file_nodes, metrics),
+ "scale": sh.build_scale(file_nodes, metrics, await crud.count_symbols(session, repo_id)),
"origin": activity["origin"],
"churn": sh.build_churn(totals),
"rhythm": rhythm,
diff --git a/packages/server/src/repowise/server/routers/workspace.py b/packages/server/src/repowise/server/routers/workspace.py
index 27c1fa782..03fd1b67a 100644
--- a/packages/server/src/repowise/server/routers/workspace.py
+++ b/packages/server/src/repowise/server/routers/workspace.py
@@ -162,8 +162,8 @@ def _query_repo_stats(db_path: Path) -> dict:
row = c.execute("SELECT COUNT(*) FROM graph_nodes WHERE node_type = 'file'").fetchone()
result["file_count"] = row[0] if row else 0
- # symbol count
- row = c.execute("SELECT COALESCE(SUM(symbol_count), 0) FROM graph_nodes").fetchone()
+ # The raw-sqlite form of crud.count_symbols: one store per repo here.
+ row = c.execute("SELECT COUNT(*) FROM wiki_symbols").fetchone()
result["symbol_count"] = int(row[0]) if row else 0
# page count
diff --git a/packages/server/src/repowise/server/services/overview_payload.py b/packages/server/src/repowise/server/services/overview_payload.py
index 3be91fcf4..0cc49acab 100644
--- a/packages/server/src/repowise/server/services/overview_payload.py
+++ b/packages/server/src/repowise/server/services/overview_payload.py
@@ -95,12 +95,7 @@ async def stat_strip(session: AsyncSession, repo_id: str) -> dict:
GraphNode.repository_id == repo_id, GraphNode.node_type == "file"
),
)
- symbol_count = int(
- await session.scalar(
- select(func.sum(GraphNode.symbol_count)).where(GraphNode.repository_id == repo_id)
- )
- or 0
- )
+ symbol_count = await crud.count_symbols(session, repo_id)
entry_point_count = await _count(
session,
select(func.count(GraphNode.id)).where(
diff --git a/tests/unit/ingestion/test_graph.py b/tests/unit/ingestion/test_graph.py
index d8f3a93e0..a0a961598 100644
--- a/tests/unit/ingestion/test_graph.py
+++ b/tests/unit/ingestion/test_graph.py
@@ -118,6 +118,17 @@ class TestAddFile:
# 2 file nodes + 2 synthetic __module__ symbol nodes
assert g.number_of_nodes() == 4
+ def test_symbol_count_counts_an_overload_set_once(self) -> None:
+ """Two overload signatures and the implementation share one id, one node."""
+ b = GraphBuilder()
+ parsed = _parsed("src/pages.ts", "typescript")
+ parsed.symbols = [_sym("src/pages.ts", "get") for _ in range(3)]
+ parsed.symbols.append(_sym("src/pages.ts", "put"))
+ b.add_file(parsed)
+ g = b.graph()
+ assert g.nodes["src/pages.ts"]["symbol_count"] == 2
+ assert parsed.symbol_count == 2
+
# ---------------------------------------------------------------------------
# Docstring propagation
diff --git a/tests/unit/persistence/helpers.py b/tests/unit/persistence/helpers.py
index 12515f6bc..513434b76 100644
--- a/tests/unit/persistence/helpers.py
+++ b/tests/unit/persistence/helpers.py
@@ -6,7 +6,20 @@ from any test file without conftest import restrictions.
from __future__ import annotations
-from repowise.core.persistence.models import Repository
+from repowise.core.persistence.models import Repository, WikiSymbol
+
+
+def make_wiki_symbol(repo_id: str, file_path: str, name: str, **overrides) -> WikiSymbol:
+ """One symbol index row, the rows ``crud.count_symbols`` counts."""
+ return WikiSymbol(
+ repository_id=repo_id,
+ file_path=file_path,
+ symbol_id=f"{file_path}::{name}",
+ name=name,
+ qualified_name=name,
+ kind="function",
+ **overrides,
+ )
def make_repo_kwargs(**overrides) -> dict:
diff --git a/tests/unit/server/test_overview_summary_golden.py b/tests/unit/server/test_overview_summary_golden.py
index 62f7e3d10..077755a29 100644
--- a/tests/unit/server/test_overview_summary_golden.py
+++ b/tests/unit/server/test_overview_summary_golden.py
@@ -23,6 +23,7 @@ from repowise.core.persistence.models import (
GraphNode,
HealthFileMetric,
Page,
+ WikiSymbol,
)
from tests.unit.server.conftest import (
assert_payload_golden,
@@ -60,6 +61,21 @@ async def _seed_index(session_factory, repo_id: str) -> None:
created_at=_CLOCK,
)
)
+ # The symbol index the stat strip counts, matching each file's count.
+ for j in range(i + 2):
+ session.add(
+ WikiSymbol(
+ id=f"ws{i}_{j}",
+ repository_id=repo_id,
+ file_path=path,
+ symbol_id=f"{path}::sym{j}",
+ name=f"sym{j}",
+ qualified_name=f"sym{j}",
+ kind="function",
+ created_at=_CLOCK,
+ updated_at=_CLOCK,
+ )
+ )
for i, (path, owner, hotspot, churn) in enumerate(
[
("src/core.py", "Alice", True, 0.95),
diff --git a/tests/unit/server/test_repos.py b/tests/unit/server/test_repos.py
index 63ddaf4b9..90041cd0c 100644
--- a/tests/unit/server/test_repos.py
+++ b/tests/unit/server/test_repos.py
@@ -213,28 +213,37 @@ async def test_stats_file_count_excludes_symbol_nodes(client: AsyncClient, sessi
"""
from repowise.core.persistence.crud import upsert_repository
from repowise.core.persistence.models import GraphNode
+ from tests.unit.persistence.helpers import make_wiki_symbol
repo = await upsert_repository(session, name="counts", local_path="/tmp/counts")
for i in range(2):
+ # The parser emitted an overload pair: 6 emissions, 5 distinct symbols.
+ # The stored per-file figure of a pre-dedupe index still says 6.
session.add(
GraphNode(
repository_id=repo.id,
node_id=f"src/mod{i}.py",
node_type="file",
- symbol_count=5,
+ symbol_count=6,
)
)
- for j in range(5):
+ for name in [f"sym{j}" for j in range(5)] + ["__module__"]:
session.add(
GraphNode(
repository_id=repo.id,
- node_id=f"src/mod{i}.py::sym{j}",
+ node_id=f"src/mod{i}.py::{name}",
node_type="symbol",
)
)
+ session.add_all(make_wiki_symbol(repo.id, f"src/mod{i}.py", f"sym{j}") for j in range(5))
await session.commit()
data = (await client.get(f"/api/repos/{repo.id}/stats")).json()
+ overview = (await client.get(f"/api/repos/{repo.id}/overview-summary")).json()
+ symbols = (await client.get("/api/symbols", params={"repo_id": repo.id})).json()
- assert data["file_count"] == 2 # not 12
+ assert data["file_count"] == 2 # not 14
+ # One figure, the symbol index's row count: not 12 symbol nodes, not 12 emissions.
assert data["symbol_count"] == 10
+ assert overview["stats"]["symbol_count"] == 10
+ assert symbols["total"] == 10
diff --git a/tests/unit/server/test_repos_router_paths.py b/tests/unit/server/test_repos_router_paths.py
index 6c2ab4bf7..69509ee99 100644
--- a/tests/unit/server/test_repos_router_paths.py
+++ b/tests/unit/server/test_repos_router_paths.py
@@ -26,6 +26,7 @@ from repowise.core.persistence.database import get_session, init_db
from repowise.core.persistence.models import GraphNode, HealthFileMetric, Repository
from repowise.core.workspace.config import RepoEntry, WorkspaceConfig
from repowise.server.routers import repos as repos_module
+from tests.unit.persistence.helpers import make_wiki_symbol
from tests.unit.server.conftest import create_test_repo
@@ -406,6 +407,8 @@ async def test_stats_counts(client: AsyncClient, session) -> None:
symbol_count=0,
is_entry_point=True,
),
+ make_wiki_symbol(repo["id"], "a.py", "f"),
+ make_wiki_symbol(repo["id"], "a.py", "g"),
]
)
await session.commit()
diff --git a/tests/unit/server/test_repos_summary.py b/tests/unit/server/test_repos_summary.py
index cae38b66b..74fdf2317 100644
--- a/tests/unit/server/test_repos_summary.py
+++ b/tests/unit/server/test_repos_summary.py
@@ -22,7 +22,7 @@ from repowise.core.persistence.models import (
GraphNode,
HealthSnapshot,
)
-from tests.unit.persistence.helpers import make_page_kwargs
+from tests.unit.persistence.helpers import make_page_kwargs, make_wiki_symbol
async def _register(client: AsyncClient, name: str) -> dict:
@@ -51,16 +51,20 @@ async def _seed(session, repo_id: str, *, files: int, symbols_per_file: int = 3)
)
)
# Symbol nodes share the table with file nodes. They are what makes the
- # unscoped count wrong, so every fixture carries some.
- for j in range(symbols_per_file):
+ # unscoped count wrong, so every fixture carries some, plus the per-file
+ # synthetic `__module__` node, which is not a symbol.
+ for name in [f"sym{j}" for j in range(symbols_per_file)] + ["__module__"]:
session.add(
GraphNode(
repository_id=repo_id,
- node_id=f"src/mod{i}.py::sym{j}",
+ node_id=f"src/mod{i}.py::{name}",
node_type="symbol",
symbol_count=0,
)
)
+ session.add_all(
+ make_wiki_symbol(repo_id, f"src/mod{i}.py", f"sym{j}") for j in range(symbols_per_file)
+ )
session.add(
GitMetadata(
repository_id=repo_id,
@@ -133,8 +137,8 @@ async def test_file_count_excludes_symbol_nodes(client: AsyncClient, session) ->
assert resp.status_code == 200
row = resp.json()["repos"][0]
- assert row["file_count"] == 4 # not 16
- assert row["symbol_count"] == 12
+ assert row["file_count"] == 4 # not 20
+ assert row["symbol_count"] == 12 # not 16: `__module__` nodes are not symbols
assert row["entry_point_count"] == 1
@@ -254,10 +258,10 @@ async def test_query_count_does_not_grow_with_repo_count(
f"query count grew from {two_repos} (2 repos) to {six_repos} (6 repos) — "
"something in the route is running per repository again"
)
- # One repo list plus five grouped aggregates. Pinned so "flat" has a value:
+ # One repo list plus six grouped aggregates. Pinned so "flat" has a value:
# a change that merges or splits an aggregate should be a deliberate edit
# here, not a silent drift.
- assert six_repos == 6
+ assert six_repos == 7
async def _count_queries(client: AsyncClient, statements: list[str]) -> list[str]:
diff --git a/tests/unit/server/test_stats_superlatives.py b/tests/unit/server/test_stats_superlatives.py
index a1be968ad..44d8ea42b 100644
--- a/tests/unit/server/test_stats_superlatives.py
+++ b/tests/unit/server/test_stats_superlatives.py
@@ -116,9 +116,9 @@ def test_scale_counts_code_languages_and_skips_external_nodes() -> None:
{"nloc": 100, "module": "a", "is_test": False},
{"nloc": 40, "module": "tests", "is_test": True},
]
- scale = build_scale(nodes, metrics)
+ scale = build_scale(nodes, metrics, symbol_count=4)
assert scale["file_count"] == 3
- assert scale["symbol_count"] == 5
+ assert scale["symbol_count"] == 4 # passed in, not the per-file sum
assert scale["language_count"] == 2
assert {lang["language"] for lang in scale["languages"]} == {"python", "typescript"}
assert scale["total_nloc"] == 140
diff --git a/tests/unit/server/test_workspace_router.py b/tests/unit/server/test_workspace_router.py
index 5361f7c4a..d1f97c884 100644
--- a/tests/unit/server/test_workspace_router.py
+++ b/tests/unit/server/test_workspace_router.py
@@ -247,6 +247,7 @@ def _create_workspace_repo_db(
symbol_count INTEGER DEFAULT 0
);
CREATE TABLE wiki_pages (id TEXT PRIMARY KEY, confidence REAL);
+ CREATE TABLE wiki_symbols (id TEXT PRIMARY KEY);
CREATE TABLE git_metadata (
id TEXT PRIMARY KEY,
churn_percentile REAL,
@@ -265,6 +266,7 @@ def _create_workspace_repo_db(
('src/a.py::Foo', 'symbol', 'python', 0),
('src/a.py::Foo.bar', 'symbol', 'python', 0),
('src/b.py::baz', 'symbol', 'python', 0);
+ INSERT INTO wiki_symbols (id) VALUES ('sym-foo'), ('sym-bar'), ('sym-baz');
INSERT INTO wiki_pages (id, confidence) VALUES
('page-a', 0.8),
('page-b', 0.6);
@@ -945,6 +947,7 @@ class TestQueryRepoStats:
symbol_count INTEGER DEFAULT 0
);
CREATE TABLE wiki_pages (id TEXT PRIMARY KEY, confidence REAL);
+ CREATE TABLE wiki_symbols (id TEXT PRIMARY KEY);
CREATE TABLE git_metadata (
id TEXT PRIMARY KEY,
is_hotspot INTEGER NOT NULL DEFAULT 0,
@@ -982,6 +985,15 @@ class TestQueryRepoStats:
assert stats["hotspot_count"] == 2
+ def test_symbol_count_is_the_symbol_index(self, tmp_path: Path) -> None:
+ """The figure `/api/repos/{id}/stats` reports: symbol index rows, not
+ the per-file emission counts (5 here)."""
+ _create_workspace_repo_db(tmp_path, "backend")
+
+ stats = workspace._query_repo_stats(tmp_path / "backend" / ".repowise" / "wiki.db")
+
+ assert stats["symbol_count"] == 3
+
def test_hotspot_count_zero_when_no_hotspots(self, tmp_path: Path) -> None:
db_path = tmp_path / ".repowise" / "wiki.db"
self._make_wiki_db(db_path, rows=[(0, 0.99), (0, 0.80)])