openviking: Add versions 0.4.17, 0.4.17.1, 0.4.18, 0.4.19, 0.4.20, 0.4.21, 0.4.22, 0.4.23, 0.5.0

.github/workflows/build-openviking.yml

@@ -116,6 +116,13 @@ jobs:
           cp third_party/spdlog-1.14.1/LICENSE LICENSE.spdlog
           cp third_party/croaring/LICENSE LICENSE.croaring
           cp third_party/rapidjson/LICENSE LICENSE.rapidjson
+          # From 0.4.18 ragfs's cache feature pulls fred with vendored-openssl, so
+          # openssl-src links OpenSSL statically into the ragfs extension.
+          openssl=$(grep -A1 '^name = "openssl-src"$' Cargo.lock | sed -nE 's/^version = "[0-9.]+\+([0-9.]+)"$/\1/p' || true)
+          if [ -n "${openssl}" ]; then
+            curl -fsSL -o LICENSE.openssl "https://raw.githubusercontent.com/openssl/openssl/openssl-${openssl}/LICENSE.txt"
+            echo "OV_EXTRA_LICENCES=LICENSE.openssl" >> "${GITHUB_ENV}"
+          fi
 
       - name: Write smoke test
         run: |
@@ -186,10 +193,21 @@ jobs:
           CIBW_ARCHS: riscv64
           CIBW_BUILD: "cp310-* cp311-* cp312-* cp313-* cp314-*"
           CIBW_SKIP: "*-musllinux_*"
+          # openssl-src (0.4.18+) runs OpenSSL's Configure, which needs perl modules the
+          # image's minimal perl-interpreter lacks (gotcha 46).
           CIBW_BEFORE_ALL_LINUX: >-
+            dnf -y --disablerepo=extras install perl &&
             curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y --profile minimal
+          # Fat LTO of ragfs-python (workspace [profile.release] lto = true) intermittently
+          # corrupts rustc's heap on the riscv64 runners from 0.4.17.1 on: "double free or
+          # corruption" / "free(): corrupted unsorted chunks" SIGABRTs, or a hang until the
+          # job timeout. Thin LTO avoids the single huge LTO module, but concurrent rustc
+          # codegen threads can still exhaust these runners' memory the same way (gotcha
+          # 228), so also cap cargo's own parallelism.
           CIBW_ENVIRONMENT: >-
             PATH="$PATH:$HOME/.cargo/bin"
+            CARGO_PROFILE_RELEASE_LTO=thin
+            CARGO_BUILD_JOBS=2
             PIP_EXTRA_INDEX_URL=https://pypi.riseproject.dev/simple/
             OV_REQUIRE_RAGFS_BUILD=1
             SETUPTOOLS_SCM_PRETEND_VERSION_FOR_OPENVIKING=${{ env.OPENVIKING_VERSION }}
@@ -201,7 +219,7 @@ jobs:
           CIBW_TEST_SOURCES: tests pyproject.toml smoke_test.py
           CIBW_TEST_COMMAND: >-
             python smoke_test.py &&
-            python -c "import importlib.metadata as m; l = set(p.name for p in m.files('openviking') if '.dist-info/licenses/' in str(p)); assert set(['LICENSE', 'LICENSE.leveldb', 'LICENSE.spdlog', 'LICENSE.croaring', 'LICENSE.rapidjson']) <= l, l" &&
+            python -c "import importlib.metadata as m; l = set(p.name for p in m.files('openviking') if '.dist-info/licenses/' in str(p)); assert set(['LICENSE', 'LICENSE.leveldb', 'LICENSE.spdlog', 'LICENSE.croaring', 'LICENSE.rapidjson'] + '${{ env.OV_EXTRA_LICENCES }}'.split()) <= l, l" &&
             python -m pytest -q -o addopts=''
             tests/vectordb/test_cuvs_config.py
             tests/vectordb/test_cuvs_index.py

docs/packages/openviking.yaml

@@ -10,3 +10,12 @@ versions:
   - filename: openviking-0.4.16-cp310-abi3-manylinux_2_39_riscv64.whl
     sha256: 391ae599f92b7611fe422207f732f80c4315ca0d903cb8f50e5c66fe02c12244
     requires-python: '>=3.10'
+- version: 0.4.17
+- version: 0.4.17.1
+- version: 0.4.18
+- version: 0.4.19
+- version: 0.4.20
+- version: 0.4.21
+- version: 0.4.22
+- version: 0.4.23
+- version: 0.5.0

patches/openviking/0.4.17.1/0001-build-only-require-Development.Module-from-FindPytho.patch

@@ -0,0 +1,36 @@
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Ludovic Henry <git@ludovic.dev>
+Date: Fri, 25 Sep 2026 00:00:00 +0000
+Subject: [PATCH] build: only require Development.Module from FindPython3
+
+The vector engine is a Python extension module, so it needs the Python
+headers but never links libpython. Asking FindPython3 for the umbrella
+`Development` component also pulls in `Development.Embed`, which requires
+libpython to exist. The manylinux interpreters used to build wheels ship
+no libpython at all, so configuring fails on every manylinux image before
+anything is compiled:
+
+    Could NOT find Python3 (missing: Python3_LIBRARIES Development
+    Development.Embed)
+
+Upstream's own Linux wheels build against a CPython compiled with
+--enable-shared, which is why their CI does not hit it. Reproduces on
+manylinux_2_28_x86_64 as well as manylinux_2_39_riscv64.
+
+Upstream-Status: To upstream [not yet submitted; still present on volcengine/OpenViking main]
+
+Signed-off-by: Ludovic Henry <git@ludovic.dev>
+---
+diff --git a/src/CMakeLists.txt b/src/CMakeLists.txt
+index 1a79b8a4..da2a9495 100644
+--- a/src/CMakeLists.txt
++++ b/src/CMakeLists.txt
+@@ -43,7 +43,7 @@ set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -Wno-error -Wno-deprecated-declarations
+ set(CMAKE_CXX_LINK_EXECUTABLE "${CMAKE_CXX_LINK_EXECUTABLE} -lpthread")
+ set(Python3_ARCH_INCLUDE_DIR "/usr/include/${CMAKE_SYSTEM_PROCESSOR}-linux-gnu/")
+
+-find_package(Python3 COMPONENTS Interpreter Development REQUIRED)
++find_package(Python3 COMPONENTS Interpreter Development.Module REQUIRED)
+ if(WIN32 AND CMAKE_VERSION VERSION_GREATER_EQUAL "3.26")
+     find_package(Python3 COMPONENTS Development.SABIModule QUIET)
+ endif()

patches/openviking/0.4.17/0001-build-only-require-Development.Module-from-FindPytho.patch

@@ -0,0 +1,36 @@
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Ludovic Henry <git@ludovic.dev>
+Date: Fri, 25 Sep 2026 00:00:00 +0000
+Subject: [PATCH] build: only require Development.Module from FindPython3
+
+The vector engine is a Python extension module, so it needs the Python
+headers but never links libpython. Asking FindPython3 for the umbrella
+`Development` component also pulls in `Development.Embed`, which requires
+libpython to exist. The manylinux interpreters used to build wheels ship
+no libpython at all, so configuring fails on every manylinux image before
+anything is compiled:
+
+    Could NOT find Python3 (missing: Python3_LIBRARIES Development
+    Development.Embed)
+
+Upstream's own Linux wheels build against a CPython compiled with
+--enable-shared, which is why their CI does not hit it. Reproduces on
+manylinux_2_28_x86_64 as well as manylinux_2_39_riscv64.
+
+Upstream-Status: To upstream [not yet submitted; still present on volcengine/OpenViking main]
+
+Signed-off-by: Ludovic Henry <git@ludovic.dev>
+---
+diff --git a/src/CMakeLists.txt b/src/CMakeLists.txt
+index 1a79b8a4..da2a9495 100644
+--- a/src/CMakeLists.txt
++++ b/src/CMakeLists.txt
+@@ -43,7 +43,7 @@ set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -Wno-error -Wno-deprecated-declarations
+ set(CMAKE_CXX_LINK_EXECUTABLE "${CMAKE_CXX_LINK_EXECUTABLE} -lpthread")
+ set(Python3_ARCH_INCLUDE_DIR "/usr/include/${CMAKE_SYSTEM_PROCESSOR}-linux-gnu/")
+
+-find_package(Python3 COMPONENTS Interpreter Development REQUIRED)
++find_package(Python3 COMPONENTS Interpreter Development.Module REQUIRED)
+ if(WIN32 AND CMAKE_VERSION VERSION_GREATER_EQUAL "3.26")
+     find_package(Python3 COMPONENTS Development.SABIModule QUIET)
+ endif()

patches/openviking/0.4.18/0001-build-only-require-Development.Module-from-FindPytho.patch

@@ -0,0 +1,36 @@
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Ludovic Henry <git@ludovic.dev>
+Date: Fri, 25 Sep 2026 00:00:00 +0000
+Subject: [PATCH] build: only require Development.Module from FindPython3
+
+The vector engine is a Python extension module, so it needs the Python
+headers but never links libpython. Asking FindPython3 for the umbrella
+`Development` component also pulls in `Development.Embed`, which requires
+libpython to exist. The manylinux interpreters used to build wheels ship
+no libpython at all, so configuring fails on every manylinux image before
+anything is compiled:
+
+    Could NOT find Python3 (missing: Python3_LIBRARIES Development
+    Development.Embed)
+
+Upstream's own Linux wheels build against a CPython compiled with
+--enable-shared, which is why their CI does not hit it. Reproduces on
+manylinux_2_28_x86_64 as well as manylinux_2_39_riscv64.
+
+Upstream-Status: To upstream [not yet submitted; still present on volcengine/OpenViking main]
+
+Signed-off-by: Ludovic Henry <git@ludovic.dev>
+---
+diff --git a/src/CMakeLists.txt b/src/CMakeLists.txt
+index 1a79b8a4..da2a9495 100644
+--- a/src/CMakeLists.txt
++++ b/src/CMakeLists.txt
+@@ -43,7 +43,7 @@ set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -Wno-error -Wno-deprecated-declarations
+ set(CMAKE_CXX_LINK_EXECUTABLE "${CMAKE_CXX_LINK_EXECUTABLE} -lpthread")
+ set(Python3_ARCH_INCLUDE_DIR "/usr/include/${CMAKE_SYSTEM_PROCESSOR}-linux-gnu/")
+
+-find_package(Python3 COMPONENTS Interpreter Development REQUIRED)
++find_package(Python3 COMPONENTS Interpreter Development.Module REQUIRED)
+ if(WIN32 AND CMAKE_VERSION VERSION_GREATER_EQUAL "3.26")
+     find_package(Python3 COMPONENTS Development.SABIModule QUIET)
+ endif()

patches/openviking/0.4.19/0001-build-only-require-Development.Module-from-FindPytho.patch

@@ -0,0 +1,36 @@
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Ludovic Henry <git@ludovic.dev>
+Date: Fri, 25 Sep 2026 00:00:00 +0000
+Subject: [PATCH] build: only require Development.Module from FindPython3
+
+The vector engine is a Python extension module, so it needs the Python
+headers but never links libpython. Asking FindPython3 for the umbrella
+`Development` component also pulls in `Development.Embed`, which requires
+libpython to exist. The manylinux interpreters used to build wheels ship
+no libpython at all, so configuring fails on every manylinux image before
+anything is compiled:
+
+    Could NOT find Python3 (missing: Python3_LIBRARIES Development
+    Development.Embed)
+
+Upstream's own Linux wheels build against a CPython compiled with
+--enable-shared, which is why their CI does not hit it. Reproduces on
+manylinux_2_28_x86_64 as well as manylinux_2_39_riscv64.
+
+Upstream-Status: To upstream [not yet submitted; still present on volcengine/OpenViking main]
+
+Signed-off-by: Ludovic Henry <git@ludovic.dev>
+---
+diff --git a/src/CMakeLists.txt b/src/CMakeLists.txt
+index 1a79b8a4..da2a9495 100644
+--- a/src/CMakeLists.txt
++++ b/src/CMakeLists.txt
+@@ -43,7 +43,7 @@ set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -Wno-error -Wno-deprecated-declarations
+ set(CMAKE_CXX_LINK_EXECUTABLE "${CMAKE_CXX_LINK_EXECUTABLE} -lpthread")
+ set(Python3_ARCH_INCLUDE_DIR "/usr/include/${CMAKE_SYSTEM_PROCESSOR}-linux-gnu/")
+
+-find_package(Python3 COMPONENTS Interpreter Development REQUIRED)
++find_package(Python3 COMPONENTS Interpreter Development.Module REQUIRED)
+ if(WIN32 AND CMAKE_VERSION VERSION_GREATER_EQUAL "3.26")
+     find_package(Python3 COMPONENTS Development.SABIModule QUIET)
+ endif()

patches/openviking/0.4.20/0001-build-only-require-Development.Module-from-FindPytho.patch

@@ -0,0 +1,36 @@
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Ludovic Henry <git@ludovic.dev>
+Date: Fri, 25 Sep 2026 00:00:00 +0000
+Subject: [PATCH] build: only require Development.Module from FindPython3
+
+The vector engine is a Python extension module, so it needs the Python
+headers but never links libpython. Asking FindPython3 for the umbrella
+`Development` component also pulls in `Development.Embed`, which requires
+libpython to exist. The manylinux interpreters used to build wheels ship
+no libpython at all, so configuring fails on every manylinux image before
+anything is compiled:
+
+    Could NOT find Python3 (missing: Python3_LIBRARIES Development
+    Development.Embed)
+
+Upstream's own Linux wheels build against a CPython compiled with
+--enable-shared, which is why their CI does not hit it. Reproduces on
+manylinux_2_28_x86_64 as well as manylinux_2_39_riscv64.
+
+Upstream-Status: To upstream [not yet submitted; still present on volcengine/OpenViking main]
+
+Signed-off-by: Ludovic Henry <git@ludovic.dev>
+---
+diff --git a/src/CMakeLists.txt b/src/CMakeLists.txt
+index 1a79b8a4..da2a9495 100644
+--- a/src/CMakeLists.txt
++++ b/src/CMakeLists.txt
+@@ -43,7 +43,7 @@ set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -Wno-error -Wno-deprecated-declarations
+ set(CMAKE_CXX_LINK_EXECUTABLE "${CMAKE_CXX_LINK_EXECUTABLE} -lpthread")
+ set(Python3_ARCH_INCLUDE_DIR "/usr/include/${CMAKE_SYSTEM_PROCESSOR}-linux-gnu/")
+
+-find_package(Python3 COMPONENTS Interpreter Development REQUIRED)
++find_package(Python3 COMPONENTS Interpreter Development.Module REQUIRED)
+ if(WIN32 AND CMAKE_VERSION VERSION_GREATER_EQUAL "3.26")
+     find_package(Python3 COMPONENTS Development.SABIModule QUIET)
+ endif()

patches/openviking/0.4.21/0001-build-only-require-Development.Module-from-FindPytho.patch

@@ -0,0 +1,36 @@
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Ludovic Henry <git@ludovic.dev>
+Date: Fri, 25 Sep 2026 00:00:00 +0000
+Subject: [PATCH] build: only require Development.Module from FindPython3
+
+The vector engine is a Python extension module, so it needs the Python
+headers but never links libpython. Asking FindPython3 for the umbrella
+`Development` component also pulls in `Development.Embed`, which requires
+libpython to exist. The manylinux interpreters used to build wheels ship
+no libpython at all, so configuring fails on every manylinux image before
+anything is compiled:
+
+    Could NOT find Python3 (missing: Python3_LIBRARIES Development
+    Development.Embed)
+
+Upstream's own Linux wheels build against a CPython compiled with
+--enable-shared, which is why their CI does not hit it. Reproduces on
+manylinux_2_28_x86_64 as well as manylinux_2_39_riscv64.
+
+Upstream-Status: To upstream [not yet submitted; still present on volcengine/OpenViking main]
+
+Signed-off-by: Ludovic Henry <git@ludovic.dev>
+---
+diff --git a/src/CMakeLists.txt b/src/CMakeLists.txt
+index 1a79b8a4..da2a9495 100644
+--- a/src/CMakeLists.txt
++++ b/src/CMakeLists.txt
+@@ -43,7 +43,7 @@ set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -Wno-error -Wno-deprecated-declarations
+ set(CMAKE_CXX_LINK_EXECUTABLE "${CMAKE_CXX_LINK_EXECUTABLE} -lpthread")
+ set(Python3_ARCH_INCLUDE_DIR "/usr/include/${CMAKE_SYSTEM_PROCESSOR}-linux-gnu/")
+
+-find_package(Python3 COMPONENTS Interpreter Development REQUIRED)
++find_package(Python3 COMPONENTS Interpreter Development.Module REQUIRED)
+ if(WIN32 AND CMAKE_VERSION VERSION_GREATER_EQUAL "3.26")
+     find_package(Python3 COMPONENTS Development.SABIModule QUIET)
+ endif()

patches/openviking/0.4.21/0002-pyproject-relax-the-tree-sitter-pin-so-riscv64-can-.patch

@@ -0,0 +1,39 @@
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Ludovic Henry <git@ludovic.dev>
+Date: Mon, 28 Sep 2026 00:00:00 +0000
+Subject: [PATCH] pyproject: relax the tree-sitter pin so riscv64 can resolve
+ it
+
+pyproject.toml pins `tree-sitter==0.25.2`, which never published a
+riscv64 wheel -- tree-sitter's own PyPI files only gained a
+`manylinux_riscv64` wheel starting with 0.26.0. With
+`PIP_ONLY_BINARY=:all:` in the wheel's test environment, installing the
+built wheel fails outright:
+
+    ERROR: Could not find a version that satisfies the requirement
+    tree-sitter==0.25.2 (from openviking) (from versions: 0.26.0)
+    ERROR: No matching distribution found for tree-sitter==0.25.2
+
+Relax the pin to a lower bound so the resolver can pick 0.26.0 instead;
+the smoke test and pytest selection this port runs never touch
+tree-sitter's Python API directly, so the exact-pin "avoid unvalidated
+upstream API and grammar changes" concern the comment names does not
+apply to this port's own test coverage.
+
+Upstream-Status: Inappropriate [riscv64-wheel availability is specific to this port; upstream's own CI never resolves against a riscv64 index]
+
+Signed-off-by: Ludovic Henry <git@ludovic.dev>
+---
+diff --git a/pyproject.toml b/pyproject.toml
+index 0000000..0000000 100644
+--- a/pyproject.toml
++++ b/pyproject.toml
+@@ -61,7 +61,7 @@ dependencies = [
+     "litellm>=1.83.7,<1.91.2",
+     "python-multipart>=0.0.31",
+     # Pin the parsing stack to avoid unvalidated upstream API and grammar changes.
+-    "tree-sitter==0.25.2",
++    "tree-sitter>=0.25.2",
+     "tree-sitter-python==0.25.0",
+     "tree-sitter-javascript==0.25.0",
+     "tree-sitter-typescript==0.23.2",

patches/openviking/0.4.21/0003-pyproject-relax-tree-sitter-grammar-pins.patch

@@ -0,0 +1,51 @@
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Ludovic Henry <git@ludovic.dev>
+Date: Sun, 11 Oct 2026 00:00:00 +0000
+Subject: [PATCH] pyproject: relax three more tree-sitter grammar pins for
+ riscv64
+
+Same problem as the tree-sitter pin itself (patch 0002): pyproject.toml
+pins tree-sitter-rust==0.24.0, tree-sitter-c-sharp==0.23.1, and
+tree-sitter-language-pack==1.13.3 exactly, but this repo has only ever
+published riscv64 wheels for tree-sitter-rust 0.24.2, tree-sitter-c-sharp
+0.23.5, and tree-sitter-language-pack 1.15.8+ (docs/packages/*.yaml).
+With PIP_ONLY_BINARY=:all:, installing the built wheel fails:
+
+    ERROR: Could not find a version that satisfies the requirement
+    tree-sitter-rust==0.24.0 (from openviking) (from versions: 0.24.2)
+    ERROR: No matching distribution found for tree-sitter-rust==0.24.0
+
+Relax all three to a lower bound for the same reason patch 0002 relaxes
+tree-sitter itself: this port's own test coverage never touches these
+grammar packages' Python APIs directly, so the exact-pin "avoid
+unvalidated upstream API and grammar changes" concern doesn't apply here.
+
+Upstream-Status: Inappropriate [riscv64-wheel availability is specific to this port; upstream's own CI never resolves against a riscv64 index]
+
+Signed-off-by: Ludovic Henry <git@ludovic.dev>
+---
+diff --git a/pyproject.toml b/pyproject.toml
+index 0000000..0000000 100644
+--- a/pyproject.toml
++++ b/pyproject.toml
+@@ -67,9 +67,9 @@ dependencies = [
+     "tree-sitter-typescript==0.23.2",
+     "tree-sitter-java==0.23.5",
+     "tree-sitter-cpp==0.23.4",
+-    "tree-sitter-rust==0.24.0",
++    "tree-sitter-rust>=0.24.0",
+     "tree-sitter-go==0.25.0",
+-    "tree-sitter-c-sharp==0.23.1",
++    "tree-sitter-c-sharp>=0.23.1",
+     "tree-sitter-php==0.24.1",
+     "tree-sitter-lua==0.5.0",
+     # OpenTelemetry
+@@ -85,7 +85,7 @@ dependencies = [
+     "mcp>=1.27,<2",
+     "pathspec>=1.1.1",
+     "grep-ast==0.9.0",
+-    "tree-sitter-language-pack==1.13.3",
++    "tree-sitter-language-pack>=1.13.3",
+ ]
+
+