dirsql-plugin-embeddings 0.1.13__tar.gz → 0.1.15__tar.gz

{dirsql_plugin_embeddings-0.1.13 → dirsql_plugin_embeddings-0.1.15}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.5
 Name: dirsql-plugin-embeddings
-Version: 0.1.13
+Version: 0.1.15
 Summary: First-party dirsql plugin: semantic search over files.
 Requires-Python: >=3.10
 Requires-Dist: cachetta>=0.7

dirsql_plugin_embeddings-0.1.15/changelog.d/2026-08-12-empty-search-reports-why.md

@@ -0,0 +1,3 @@
+**Fixed**
+
+- **A search that ranks nothing now says why and exits nonzero, instead of printing nothing and exiting 0.** Silent empty output was indistinguishable from "searched and found nothing relevant" — the reported case was running `dirsql-plugin-embeddings 'docs/**/*.md' <query>` from inside `docs/`, where the glob matches nothing. The message names the glob and the directory it searched from (`no files matched 'docs/**/*.md' (searched from /current/dir)`), and distinguishes that from a corpus whose files all matched but had no embeddable text, so it never claims the wrong reason. (#816)

dirsql_plugin_embeddings-0.1.15/changelog.d/2026-08-12-no-per-call-progress-bar.md

@@ -0,0 +1,3 @@
+**Fixed**
+
+- **`embed()` no longer renders a progress bar for every call.** The worker embeds exactly one value per protocol round trip, so the per-call tqdm bar could only ever say `1/1` — on a TTY a 20-file corpus query printed ~21 of them to stderr. The bar is gone; the Hugging Face model-*download* bar, which measures something worth watching, is unchanged and still TTY-gated. (#814)

dirsql_plugin_embeddings-0.1.15/e2e-attestations/claude-814-suppress-per-call-progress.json

@@ -0,0 +1,7 @@
+{
+  "command": "uv run --with-editable ../../packages/python python -m pytest tests/e2e/ -x -q",
+  "ran_at": 1786552300,
+  "exit_code": 0,
+  "commit": "b1fc5a6a475a8f93eab7a3a652b5c982269768a8",
+  "branch": "claude/814-suppress-per-call-progress"
+}

dirsql_plugin_embeddings-0.1.15/e2e-attestations/claude-816-empty-corpus-error.json

@@ -0,0 +1,7 @@
+{
+  "command": "uv run --with-editable ../../packages/python python -m pytest tests/e2e/ -x -q",
+  "ran_at": 1786553959,
+  "exit_code": 0,
+  "commit": "c349f1ed7b83b0817d68019550264edb0a255ddd",
+  "branch": "claude/816-empty-corpus-error"
+}

{dirsql_plugin_embeddings-0.1.13 → dirsql_plugin_embeddings-0.1.15}/src/dirsql_plugin_embeddings/cli/search.py

@@ -1,6 +1,6 @@
 import click
 
-from ..search.run import run_search
+from ..search.run import NothingToRank, run_search
 
 
 @click.command(hidden=True)
@@ -22,5 +22,10 @@ from ..search.run import run_search
 )
 def search(glob, query, limit, model):
     """Rank files matching GLOB by semantic similarity to QUERY."""
-    for line in run_search(glob, query, limit, model):
+    try:
+        lines = run_search(glob, query, limit, model)
+    except NothingToRank as error:
+        click.echo(f"dirsql-plugin-embeddings: {error}", err=True)
+        raise SystemExit(1) from None
+    for line in lines:
         click.echo(line)

{dirsql_plugin_embeddings-0.1.13 → dirsql_plugin_embeddings-0.1.15}/src/dirsql_plugin_embeddings/embedding/worker.py

@@ -3,7 +3,6 @@ from hashlib import sha256
 
 from . import model
 from .cache import make_cache
-from .progress import stderr_is_tty
 from .values import ProtocolError, decode_value
 
 MALFORMED_SHAPE = (
@@ -27,7 +26,10 @@ class Worker:
 
     def _compute(self, digest, identifier):
         text, loaded = self._pending
-        (vector,) = loaded.encode([text], show_progress_bar=stderr_is_tty())
+        # No progress bar: the protocol embeds one value per round trip, so a
+        # per-call bar can only ever say 1/1. The model *download* bar, which
+        # measures something, is TTY-gated in `progress.configure`.
+        (vector,) = loaded.encode([text], show_progress_bar=False)
         return [float(component) for component in vector]
 
     def embed(self, text, model_id):

dirsql_plugin_embeddings-0.1.15/src/dirsql_plugin_embeddings/search/run.py

@@ -0,0 +1,45 @@
+import asyncio
+import os
+from importlib import resources
+
+import dirsql
+
+from .output import format_rows
+from .sql import build_search_sql, count_corpus_sql
+
+
+class NothingToRank(Exception):
+    """The search produced no rows, and why -- so the CLI can say which."""
+
+
+def config_fragment():
+    return str(resources.files("dirsql_plugin_embeddings").joinpath("dirsql.toml"))
+
+
+def no_rows_message(glob, matched, root):
+    if not matched:
+        return f"no files matched {glob!r} (searched from {root})"
+    return (
+        f"{glob!r} matched {matched} file(s), but none had text content to"
+        f" embed -- unreadable or not valid UTF-8 (searched from {root})"
+    )
+
+
+async def _search(glob, query, limit, model):
+    db = dirsql.DirSQL(config=config_fragment())
+    rows = await db.query(build_search_sql(glob, query, limit, model))
+    if rows:
+        return rows, None
+    # Only now, on the error path, is the second scan worth its cost: it is
+    # what tells "no files matched" apart from "matched, none embeddable".
+    # One row, always: `SELECT COUNT(*)` cannot return anything else, and
+    # unpacking says so rather than trusting an index.
+    (counted,) = await db.query(count_corpus_sql(glob))
+    return rows, counted["n"]
+
+
+def run_search(glob, query, limit, model=None):
+    rows, matched = asyncio.run(_search(glob, query, limit, model))
+    if not rows:
+        raise NothingToRank(no_rows_message(glob, matched, os.getcwd()))
+    return format_rows(rows)

{dirsql_plugin_embeddings-0.1.13 → dirsql_plugin_embeddings-0.1.15}/src/dirsql_plugin_embeddings/search/sql.py

@@ -30,3 +30,7 @@ def build_search_sql(glob, query, limit, model=None):
         f" FROM (SELECT path, {inner} AS emb FROM {quote(normalize_glob(glob))})"
         f" ORDER BY distance LIMIT {int(limit):d}"
     )
+
+
+def count_corpus_sql(glob):
+    return f"SELECT COUNT(*) AS n FROM {quote(normalize_glob(glob))}"

dirsql_plugin_embeddings-0.1.13/src/dirsql_plugin_embeddings/search/run.py

@@ -1,20 +0,0 @@
-import asyncio
-from importlib import resources
-
-import dirsql
-
-from .output import format_rows
-from .sql import build_search_sql
-
-
-def config_fragment():
-    return str(resources.files("dirsql_plugin_embeddings").joinpath("dirsql.toml"))
-
-
-async def _query(sql):
-    return await dirsql.DirSQL(config=config_fragment()).query(sql)
-
-
-def run_search(glob, query, limit, model=None):
-    rows = asyncio.run(_query(build_search_sql(glob, query, limit, model)))
-    return format_rows(rows)