symgrep-codesearch 0.6.0__tar.gz → 0.6.2__tar.gz

{symgrep_codesearch-0.6.0 → symgrep_codesearch-0.6.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: symgrep-codesearch
-Version: 0.6.0
+Version: 0.6.2
 Summary: Code search with honest recall — structure-aware grep that tells you what it does not know, plus search by meaning
 License-Expression: MIT
 Project-URL: Repository, https://github.com/ashdamle/symgrep
@@ -20,10 +20,12 @@ Requires-Python: >=3.11
 Description-Content-Type: text/markdown
 License-File: LICENSE
 Requires-Dist: tree-sitter-language-pack>=1.0
-Requires-Dist: veccore[fastembed]>=0.2
+Requires-Dist: veccore[fastembed]<1,>=0.2
 Provides-Extra: semantic
 Provides-Extra: mcp
 Requires-Dist: fastmcp>=2.0; extra == "mcp"
+Provides-Extra: rg
+Requires-Dist: ripgrep-bin>=14; extra == "rg"
 Provides-Extra: structural
 Dynamic: license-file
 
@@ -113,8 +115,15 @@ pacman -S ripgrep             # Arch
 # other platforms: https://github.com/BurntSushi/ripgrep#installation
 ```
 
-If `rg` is missing, symgrep fails with a clear, actionable error (never a silent zero-hit). An `rg` installed
-somewhere off `PATH` can be pointed to explicitly with `SYMGREP_RG=/path/to/rg`.
+Or get it from pip, no system package manager needed — the `[rg]` extra bundles the binary (macOS/Linux/Windows):
+
+```bash
+pip install 'symgrep-codesearch[rg]'          # self-contained: ships the rg binary via the ripgrep-bin wheel
+```
+
+This is handy for CI and quick starts; prefer a system `rg` where you have one. If `rg` is missing entirely,
+symgrep fails with a clear, actionable error (never a silent zero-hit). An `rg` installed somewhere off `PATH`
+can be pointed to explicitly with `SYMGREP_RG=/path/to/rg`.
 
 ```bash
 # Distribution name is `symgrep-codesearch` (the bare `symgrep` is parked on PyPI by the Semgrep team);

{symgrep_codesearch-0.6.0 → symgrep_codesearch-0.6.2}/README.md

@@ -84,8 +84,15 @@ pacman -S ripgrep             # Arch
 # other platforms: https://github.com/BurntSushi/ripgrep#installation
 ```
 
-If `rg` is missing, symgrep fails with a clear, actionable error (never a silent zero-hit). An `rg` installed
-somewhere off `PATH` can be pointed to explicitly with `SYMGREP_RG=/path/to/rg`.
+Or get it from pip, no system package manager needed — the `[rg]` extra bundles the binary (macOS/Linux/Windows):
+
+```bash
+pip install 'symgrep-codesearch[rg]'          # self-contained: ships the rg binary via the ripgrep-bin wheel
+```
+
+This is handy for CI and quick starts; prefer a system `rg` where you have one. If `rg` is missing entirely,
+symgrep fails with a clear, actionable error (never a silent zero-hit). An `rg` installed somewhere off `PATH`
+can be pointed to explicitly with `SYMGREP_RG=/path/to/rg`.
 
 ```bash
 # Distribution name is `symgrep-codesearch` (the bare `symgrep` is parked on PyPI by the Semgrep team);

{symgrep_codesearch-0.6.0 → symgrep_codesearch-0.6.2}/pyproject.toml

@@ -7,12 +7,17 @@ build-backend = "setuptools.build_meta"
 # Semgrep team. The IMPORT name is unchanged — `import symgrep` — so consumers (honestreview, spendguard,
 # 7thsense, warden) are unaffected by the distribution rename. The package dir stays src/symgrep.
 name = "symgrep-codesearch"
+# 0.6.2: ripgrep is a system binary and cannot be a normal pip dependency, but the new `[rg]` extra pulls
+# `ripgrep-bin` (a wheel bundling the rg binary) for a self-contained install; rg_binary() now also finds an
+# rg next to the interpreter (venv bin), so the pip-only path works even under `.venv/bin/python -m …`.
+# 0.6.1: cap the veccore dependency at >=0.2,<1 so a future breaking veccore 1.0 is not pulled in unguarded
+# (verified against veccore 0.3.0).
 # 0.6.0: search-by-meaning is now STANDARD — veccore[fastembed] moved from the `[semantic]` extra into core
 # `dependencies` (the extra is now a no-op back-compat alias). `pip install symgrep-codesearch` now ships
 # semantic search. Minor bump (dependency surface changed; no breaking import/CLI/MCP change).
 # 0.5.0: added the public `file_structure` library API (+ its MCP tool and `symgrep structure` CLI), import
 # module-name normalization, and the FileStructure `connects` list (structure schema v2).
-version = "0.6.0"
+version = "0.6.2"
 description = "Code search with honest recall — structure-aware grep that tells you what it does not know, plus search by meaning"
 readme = "README.md"
 # 3.11+ : the maintained grammar pack (tree-sitter-language-pack, a CORE dep below) ships abi3 wheels for
@@ -42,7 +47,7 @@ dependencies = [
     # pulls fastembed's ONNX runtime, so the base install is heavier — a deliberate trade for semantic working
     # out of the box everywhere symgrep is installed. semantic.py still degrades honestly (UNAVAILABLE) if the
     # stack is somehow absent, but that is now the exception, not the norm.
-    "veccore[fastembed]>=0.2",
+    "veccore[fastembed]>=0.2,<1",     # cap below a future breaking 1.0; verified against 0.3.0
 ]
 # No "License :: ..." trove classifier: PEP 639 carries the license in the SPDX `license` field above, and
 # setuptools>=77 rejects a redundant license classifier alongside it. rg is a system prerequisite, not declared here.
@@ -75,6 +80,11 @@ Repository = "https://github.com/ashdamle/symgrep"
 # alias so `pip install symgrep-codesearch[semantic]` (docs, muscle memory, existing consumers) still resolves.
 semantic = []
 mcp = ["fastmcp>=2.0"]
+# rg is a SYSTEM binary (ripgrep), not normally a pip dependency. This OPTIONAL extra pulls `ripgrep-bin`, a
+# wheel that bundles the rg binary (macOS/Linux/Windows), for a self-contained `pip install` with no system
+# package manager — handy for CI and quick starts. rg_binary() finds the binary it lands in the interpreter's
+# bin dir even when that dir is not on PATH. Prefer a system rg where you have one; this is the pip fallback.
+rg = ["ripgrep-bin>=14"]
 # The grammar deps moved to CORE `dependencies` (grammar-aware symbols must work by default, not opt-in).
 # This extra is kept as a NO-OP back-compat alias so `pip install symgrep[structural]` (CI, docs, muscle
 # memory) still resolves — the grammars it used to add are now unconditional.

{symgrep_codesearch-0.6.0 → symgrep_codesearch-0.6.2}/src/symgrep/ripgrep.py

@@ -9,6 +9,7 @@ import json
 import os
 import shutil
 import subprocess
+import sys
 import threading
 from dataclasses import dataclass, field
 from typing import Iterable, List, Optional
@@ -71,16 +72,25 @@ def rg_binary() -> str:
         # quietly using whatever rg is on PATH
         raise RipgrepNotFound(f"{RG_ENV_VAR}={env!r} does not exist")
     found = shutil.which("rg")
-    if not found:
-        # rg is a SYSTEM prerequisite, not a pip dependency — symgrep IS ripgrep underneath. Fail with the
-        # exact fix rather than degrade to a silent zero-hit (the confusion this tool exists to prevent).
-        raise RipgrepNotFound(
-            "ripgrep (rg) is required but was not found on PATH. Install it: "
-            "macOS `brew install ripgrep`; Debian/Ubuntu `apt install ripgrep`; "
-            "Fedora `dnf install ripgrep`; Arch `pacman -S ripgrep`; or see "
-            f"https://github.com/BurntSushi/ripgrep#installation . "
-            f"Already installed elsewhere? Point symgrep at it with {RG_ENV_VAR}=/path/to/rg")
-    return found
+    if found:
+        return found
+    # The `symgrep-codesearch[rg]` extra pulls `ripgrep-bin`, which lands `rg` in the interpreter's OWN bin dir
+    # (the venv's bin/). That dir is NOT on PATH when symgrep runs via `.venv/bin/python -m …` without
+    # activation (the common CI invocation), so shutil.which misses it — look there explicitly, making the
+    # pip-only install self-contained regardless of PATH.
+    exe_dir = os.path.dirname(sys.executable or "")
+    if exe_dir:
+        cand = os.path.join(exe_dir, "rg.exe" if os.name == "nt" else "rg")
+        if os.path.isfile(cand) and os.access(cand, os.X_OK):
+            return cand
+    # rg is a SYSTEM binary, not resolvable as a normal pip dependency — symgrep IS ripgrep underneath. Fail
+    # with the exact fix rather than degrade to a silent zero-hit (the confusion this tool exists to prevent).
+    raise RipgrepNotFound(
+        "ripgrep (rg) is required but was not found. Either install the binary — "
+        "macOS `brew install ripgrep`; Debian/Ubuntu `apt install ripgrep`; Fedora `dnf install ripgrep`; "
+        "Arch `pacman -S ripgrep`; or https://github.com/BurntSushi/ripgrep#installation — OR get it from pip "
+        "with the bundled-binary extra: `pip install 'symgrep-codesearch[rg]'`. "
+        f"Already installed elsewhere? Point symgrep at it with {RG_ENV_VAR}=/path/to/rg")
 
 
 def _rg_cmd(pattern: str, roots: Iterable[str], case_insensitive: bool, fixed_string: bool,

{symgrep_codesearch-0.6.0 → symgrep_codesearch-0.6.2}/src/symgrep_codesearch.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: symgrep-codesearch
-Version: 0.6.0
+Version: 0.6.2
 Summary: Code search with honest recall — structure-aware grep that tells you what it does not know, plus search by meaning
 License-Expression: MIT
 Project-URL: Repository, https://github.com/ashdamle/symgrep
@@ -20,10 +20,12 @@ Requires-Python: >=3.11
 Description-Content-Type: text/markdown
 License-File: LICENSE
 Requires-Dist: tree-sitter-language-pack>=1.0
-Requires-Dist: veccore[fastembed]>=0.2
+Requires-Dist: veccore[fastembed]<1,>=0.2
 Provides-Extra: semantic
 Provides-Extra: mcp
 Requires-Dist: fastmcp>=2.0; extra == "mcp"
+Provides-Extra: rg
+Requires-Dist: ripgrep-bin>=14; extra == "rg"
 Provides-Extra: structural
 Dynamic: license-file
 
@@ -113,8 +115,15 @@ pacman -S ripgrep             # Arch
 # other platforms: https://github.com/BurntSushi/ripgrep#installation
 ```
 
-If `rg` is missing, symgrep fails with a clear, actionable error (never a silent zero-hit). An `rg` installed
-somewhere off `PATH` can be pointed to explicitly with `SYMGREP_RG=/path/to/rg`.
+Or get it from pip, no system package manager needed — the `[rg]` extra bundles the binary (macOS/Linux/Windows):
+
+```bash
+pip install 'symgrep-codesearch[rg]'          # self-contained: ships the rg binary via the ripgrep-bin wheel
+```
+
+This is handy for CI and quick starts; prefer a system `rg` where you have one. If `rg` is missing entirely,
+symgrep fails with a clear, actionable error (never a silent zero-hit). An `rg` installed somewhere off `PATH`
+can be pointed to explicitly with `SYMGREP_RG=/path/to/rg`.
 
 ```bash
 # Distribution name is `symgrep-codesearch` (the bare `symgrep` is parked on PyPI by the Semgrep team);

{symgrep_codesearch-0.6.0 → symgrep_codesearch-0.6.2}/src/symgrep_codesearch.egg-info/requires.txt

@@ -1,9 +1,12 @@
 tree-sitter-language-pack>=1.0
-veccore[fastembed]>=0.2
+veccore[fastembed]<1,>=0.2
 
 [mcp]
 fastmcp>=2.0
 
+[rg]
+ripgrep-bin>=14
+
 [semantic]
 
 [structural]