docs: align Python platform support

Author: tae898

bindings/python/README.md

@@ -2,7 +2,7 @@
 
 Native Python bindings for ArcadeDB - the multi-model database that supports Graph, Document, Key/Value, Search Engine, Time Series, and Vector models.
 
-**Status**: ✅ Production Ready | **Tests**: 252 Passing | **Platforms**: 6 Supported
+**Status**: ✅ Production Ready | **Tests**: 252 Passing | **Platforms**: 3 Supported
 
 ---
 
@@ -31,10 +31,9 @@ uv pip install arcadedb-embedded
 **Requirements:**
 
 - **Python 3.10–3.14** (packaged/tested on CPython 3.12) - No Java installation required!
-- **Supported Platforms**: Prebuilt wheels for **6 platforms**
+- **Supported Platforms**: Prebuilt wheels for **3 platforms**
   - Linux: x86_64, ARM64
-  - macOS: Intel (x86_64), Apple Silicon (ARM64)
-  - Windows: x86_64, ARM64
+  - macOS: Apple Silicon (ARM64)
 - **Development version**: Use `--pre` flag to install `.devN` versions
 
 !!! tip "Development Releases"
@@ -72,7 +71,7 @@ with arcadedb.create_database("./mydb") as db:
 ## ✨ Features
 
 - ☕ **No Java Installation Required**: Bundled JRE (~249MB uncompressed)
-- 🌍 **6 Platforms Supported**: Linux, macOS, Windows (x86_64 + ARM64)
+- 🌍 **3 Platforms Supported**: Linux (x86_64, ARM64), macOS (ARM64)
 - 🚀 **Embedded Mode**: Direct database access in Python process (no network)
 - 🌐 **Server Mode**: Optional HTTP server with Studio web interface
 - 📦 **Self-contained**: All dependencies bundled (~116MB wheel)
@@ -89,7 +88,7 @@ with arcadedb.create_database("./mydb") as db:
 
 The `arcadedb-embedded` package is platform-specific and self-contained:
 
-**Package Contents (all platforms, ballpark):**
+**Package Contents (all platforms):**
 
 - **Wheel size (compressed)**: ~116MB
 - **ArcadeDB JARs (uncompressed)**: ~32MB
@@ -104,7 +103,7 @@ Import: `import arcadedb_embedded as arcadedb`
 
 ## 🧪 Testing
 
-**Status**: 252 tests + 7 example scripts passing on all 6 platforms
+**Status**: 252 tests + example scripts passing on all 3 platforms
 
 ```bash
 # Run all tests
@@ -123,7 +122,7 @@ See [tests/README.md](tests/README.md) for detailed test documentation.
 **Requirements vary by platform:**
 
 - **Linux**: Docker (handles all dependencies)
-- **macOS/Windows**: Java 25+ JDK with jlink (to build the bundled JRE)
+- **macOS**: Java 25+ JDK with jlink (to build the bundled JRE)
 
 ### Setup Virtual Environment
 
@@ -156,7 +155,6 @@ uv pip install -e ".[test]"
 # Build for specific platform
 ./build.sh linux/amd64    # Requires Docker
 ./build.sh darwin/arm64   # Requires Java JDK (native build)
-./build.sh windows/arm64  # Requires Java JDK (native build)
 # etc.
 ```
 
@@ -181,7 +179,7 @@ Built wheels will be in `dist/`. **[Build instructions](https://humemai.github.i
 - `windows/amd64` (Native build on Windows x64)
 - `windows/arm64` (Native build on Windows ARM64)
 
-> **Developer Note:** See [docs/development/build-architecture.md](docs/development/build-architecture.md) for comprehensive documentation of the multi-platform build architecture, including how we achieve platform-specific JRE bundling across all 6 platforms on GitHub Actions.
+> **Developer Note:** See [docs/development/build-architecture.md](docs/development/build-architecture.md) for comprehensive documentation of the multi-platform build architecture, including how we achieve platform-specific JRE bundling across the supported platforms on GitHub Actions.
 
 ## Development
 

bindings/python/build.sh

@@ -43,10 +43,7 @@ print_usage() {
     echo "  Auto-detected if not specified"
     echo "  linux/amd64    Linux x86_64 (Docker build)"
     echo "  linux/arm64    Linux ARM64 (Docker build, native ARM64 runner)"
-    echo "  darwin/amd64   macOS x86_64 (native build on macOS)"
     echo "  darwin/arm64   macOS ARM64 Apple Silicon (native build on macOS)"
-    echo "  windows/amd64  Windows x86_64 (native build on Windows)"
-    echo "  windows/arm64  Windows ARM64 (native build on Windows)"
     echo ""
     echo "PYTHON_VERSION:"
     echo "  Python version for wheel (default: 3.12)"
@@ -57,7 +54,7 @@ print_usage() {
     echo "  If omitted, JARs are pulled from arcadedata/arcadedb:<version>"
     echo ""
     echo "Build Methods:"
-    echo "  Native: macOS and Windows build natively on their platforms"
+    echo "  Native: macOS builds natively on its platform"
     echo "  Docker: Linux uses Docker for manylinux compliance"
     echo ""
     echo "Examples:"
@@ -70,7 +67,7 @@ print_usage() {
     echo "Package features:"
     echo "  ✅ Bundled platform-specific JRE (no Java required)"
     echo "  ✅ Optimized JAR selection (see jar_exclusions.txt)"
-    echo "  ✅ Multi-platform support (6 platforms)"
+    echo "  ✅ Multi-platform support (3 platforms)"
     echo "  📦 Size: ~215MB (compressed), ~289MB (installed)"
     echo ""
 }

bindings/python/docs/development/build-architecture.md

@@ -4,22 +4,19 @@ This document describes the build architecture for creating platform-specific Py
 
 ## Overview
 
-**Goal:** Distribute a single `arcadedb-embedded` package that works on 6 platforms with **zero Java installation required**.
+**Goal:** Distribute a single `arcadedb-embedded` package that works on 3 platforms with **zero Java installation required**.
 
-**Achievement:** 6 platform-specific wheels (~215MB each) with bundled platform-specific JRE, built and tested on GitHub Actions using native runners.
+**Achievement:** 3 platform-specific wheels (~215MB each) with bundled platform-specific JRE, built and tested on GitHub Actions using native runners.
 
 ## Supported Platforms
 
 | Platform | Wheel Size | JRE Size | Runner | Build Method | Notes |
 |----------|-----------|----------|---------|--------------|-------|
 | **linux/amd64** | 215.0M | 62.7M | `ubuntu-24.04` | Docker native | Most common Linux platform |
 | **linux/arm64** | 214.1M | 61.8M | `ubuntu-24.04-arm` | Docker native | ARM64 servers, Raspberry Pi |
-| **darwin/amd64** | 211.9M | 55.3M | `macos-15-intel` | Native build | Intel Macs (2006-2020) |
 | **darwin/arm64** | 210.8M | 53.9M | `macos-15` | Native build | Apple Silicon Macs (2020+) |
-| **windows/amd64** | 211.6M | 51.6M | `windows-2025` | Native build | 64-bit Windows |
-| **windows/arm64** | 209.4M | 47.6M | `windows-11-arm` | Native build | ARM64 Windows (Surface, etc.) |
 
-**All platforms:**
+**All supported platforms:**
 
 - ✅ 252 tests passing
 - ✅ 226.0M JARs (83 files, identical across platforms)
@@ -37,10 +34,10 @@ We use a **hybrid build approach** to create platform-specific wheels:
    - linux/arm64: Native Docker on `ubuntu-24.04-arm` (GitHub ARM64 runner)
    - Builds platform-specific JRE via `jlink`
 
-2. **macOS/Windows platforms:** Native builds
-   - Uses platform-specific GitHub Actions runners
-   - Native `jlink` creates correct JRE for each platform
-   - Pre-filtered JARs from artifact (eliminates glob issues)
+2. **macOS platform:** Native builds
+  - Uses platform-specific GitHub Actions runner
+  - Native `jlink` creates correct JRE for the platform
+  - Pre-filtered JARs from artifact (eliminates glob issues)
 
 **Critical:** All wheels are **platform-specific** (not `py3-none-any`). This is achieved by:
 
@@ -318,8 +315,8 @@ bindings/python/
    - Uploads artifact for native builds
 
 2. **test job matrix** (lines 91-364)
-   - Builds all 6 platforms
-   - Platform-specific steps (native runners, artifact download, tests)
+  - Builds 3 platforms
+  - Platform-specific steps (native runners, artifact download, tests)
 
 3. **Test parsing** (lines 200-237)
    - JUnit XML generation and parsing

bindings/python/docs/development/ci-setup.md

@@ -2,7 +2,7 @@
 
 ## Overview
 
-The CI/CD workflows now support building and releasing across **6 platforms** for a total of **6 wheel packages** per release, all under the single `arcadedb-embedded` package.
+The CI/CD workflows now support building and releasing across **3 platforms** for a total of **15 wheel packages** per release (3 platforms × 5 Python versions), all under the single `arcadedb-embedded` package.
 
 ## Build Matrix
 
@@ -14,38 +14,29 @@ The CI/CD workflows now support building and releasing across **6 platforms** fo
 
 - **linux/amd64**: Linux on x86_64 (ubuntu-24.04, Docker build)
 - **linux/arm64**: Linux on ARM64 (ubuntu-24.04-arm, Docker build)
-- **darwin/amd64**: macOS on Intel (macos-15-intel, native build)
 - **darwin/arm64**: macOS on Apple Silicon M1/M2/M3/M4 (macos-15, native build)
-- **windows/amd64**: Windows on x86_64 (windows-2025, native build)
-- **windows/arm64**: Windows on ARM64 (windows-11-arm, native build)
 
 ### Total Artifacts
 
-**6 wheels per release**: 1 package × 6 platforms = 6 wheels
+**15 wheels per release**: 1 package × 3 platforms × 5 Python versions = 15 wheels
 
 ## Workflow Changes
 
 ### `test-python-bindings.yml`
 
-- **Matrix**: `platform: [linux/amd64, linux/arm64, darwin/amd64, darwin/arm64, windows/amd64, windows/arm64]`
+- **Matrix**: `platform: [linux/amd64, linux/arm64, darwin/arm64]` and `python-version: [3.10, 3.11, 3.12, 3.13, 3.14]`
 - **Runners**: All native (no QEMU emulation)
   - ubuntu-24.04 (Linux x64)
   - ubuntu-24.04-arm (Linux ARM64)
-  - macos-15-intel (macOS Intel)
   - macos-15 (macOS Apple Silicon)
-  - windows-2025 (Windows x64)
-  - windows-11-arm (Windows ARM64)
-- **Jobs**: 6 total (one per platform for testing)
-- **Artifacts**: `wheel-{platform}-test` (6 artifacts)
+- **Jobs**: 15 total (3 platforms × 5 Python versions)
+- **Artifacts**: `wheel-{platform}-py{version}` (15 artifacts)
 
 ### `release-python-packages.yml`
 
-- **Matrix**: `platform: [linux/amd64, linux/arm64, darwin/amd64, darwin/arm64, windows/amd64, windows/arm64]`
-- **Runners**: All native (pinned versions for reproducibility)
-- **Jobs**: 6 build jobs + 1 publish job (7 total)
-- **Artifacts**: `wheel-{platform}` (6 artifacts)
+- **Artifacts**: `wheel-{platform}-py{version}` (15 artifacts)
 - **Publish Job**:
-  - `publish-pypi`: Collects all 6 wheels and publishes to `arcadedb-embedded`
+  - `publish-pypi`: Collects all 15 wheels and publishes to `arcadedb-embedded`
 
 ## GitHub Repository Setup Required
 
@@ -93,34 +84,25 @@ You need to create one environment in GitHub repository settings:
 4. **Monitor the workflow**:
     - Go to Actions tab
     - Watch `Build and Release Python Packages to PyPI`
-    - Check that 6 wheels are built and publish job succeeds
+    - Check that 15 wheels are built and publish job succeeds
 
 ## Validation
 
 ### Expected Artifacts
 
 After a successful release, you should see:
 
-- **6 wheel files** on PyPI for `arcadedb-embedded` (one per platform):
-  - linux/amd64: ~215MB wheel
-  - linux/arm64: ~215MB wheel
-  - darwin/amd64: ~215MB wheel
-  - darwin/arm64: ~215MB wheel
-  - windows/amd64: ~215MB wheel
-  - windows/arm64: ~215MB wheel
+- **15 wheel files** on PyPI for `arcadedb-embedded` (3 platforms × 5 Python versions)
 
 ### Test Results (CI run #96)
 
-All 6 platforms passing 252 tests and 7 example scripts:
+All 3 platforms passing 252 tests and example scripts:
 
 | Platform | Wheel Size | JRE Size | Tests |
 |----------|-----------|----------|-------|
 | linux/amd64 | 215.0M | 62.7M | 252 passed ✅ |
 | linux/arm64 | 214.1M | 61.8M | 252 passed ✅ |
-| darwin/amd64 | 211.9M | 55.3M | 252 passed ✅ |
 | darwin/arm64 | 210.8M | 53.9M | 252 passed ✅ |
-| windows/amd64 | 211.6M | 51.6M | 252 passed ✅ |
-| windows/arm64 | 209.4M | 47.6M | 252 passed ✅ |
 
 **All platforms include:**
 
@@ -136,10 +118,7 @@ All platforms use native GitHub runners:
 
 - **linux/amd64**: ubuntu-24.04 (Docker build)
 - **linux/arm64**: ubuntu-24.04-arm (Docker build, native ARM64)
-- **darwin/amd64**: macos-15-intel (native build)
 - **darwin/arm64**: macos-15 (native build)
-- **windows/amd64**: windows-2025 (native build)
-- **windows/arm64**: windows-11-arm (native build)
 
 ### Build Time Expectations
 
@@ -164,7 +143,6 @@ cd bindings/python
 # Build for specific platform (requires Docker for Linux builds)
 ./build.sh --platform linux/amd64
 ./build.sh --platform darwin/arm64
-./build.sh --platform windows/amd64
 
 # Check the wheels
 ls -lh dist/
@@ -175,13 +153,13 @@ ls -lh dist/
 ```bash
 cd bindings/python
 
-for platform in linux/amd64 linux/arm64 darwin/amd64 darwin/arm64 windows/amd64 windows/arm64; do
+for platform in linux/amd64 linux/arm64 darwin/arm64; do
   echo "Building $platform..."
   ./build.sh --platform "$platform"
 done
 
-# Should have 6 wheels
-ls -1 dist/*.whl | wc -l  # Should output: 6
+# Should have 3 wheels
+ls -1 dist/*.whl | wc -l  # Should output: 3
 ```
 
 ## Troubleshooting
@@ -202,26 +180,23 @@ All platforms now use platform-specific JVM library paths:
 
 ### Wheel count mismatch
 
-- The publish job validates that exactly 6 wheels exist
+- The publish job validates that exactly 15 wheels exist
 - If validation fails, check the build matrix jobs for failures
-- Ensure all 6 platform builds succeeded
+- Ensure all 3 platform builds succeeded
 
 ### Runner availability
 
 All platforms use pinned runner versions for reproducibility:
 
 - ubuntu-24.04 (guaranteed available)
 - ubuntu-24.04-arm (GitHub-hosted ARM64)
-- macos-15-intel (Intel Mac, pinned version)
 - macos-15 (Apple Silicon, pinned version)
-- windows-2025 (Windows x64, pinned version)
-- windows-11-arm (Windows ARM64, GitHub-hosted)
 
 ## Next Steps
 
 1. **Create GitHub environment** (`pypi`)
 2. **Set up PyPI trusted publisher** for `arcadedb-embedded`
 3. **Test with a dev tag**: `git tag 25.10.1.dev0 && git push origin 25.10.1.dev0`
-4. **Verify 6 wheels are published** to PyPI (one per platform)
-5. **Test installation** on all 6 platforms
+4. **Verify 15 wheels are published** to PyPI (3 platforms × 5 Python versions)
+5. **Test installation** on the supported platforms
 6. **Verify no Java required** on end-user systems (JRE bundled)

bindings/python/docs/getting-started/distributions.md

@@ -23,18 +23,18 @@ The package includes everything you need:
 - **ArcadeDB JARs** (~32MB, uncompressed): Core database with all features
 - **Bundled JRE** (~249MB, uncompressed): Platform-specific Java 25 runtime (via jlink)
 
-**Total:** ~116MB compressed wheel, ~281MB uncompressed installed size (ballpark)
+**Total:** ~116MB compressed wheel, ~281MB uncompressed installed size
 
 ## Platform Support
 
-Pre-built **platform-specific** wheels are available for **6 platforms**. Sizes vary slightly by platform, but the ballpark totals are consistent (see size breakdown below).
+Pre-built **platform-specific** wheels are available for **3 platforms**. Sizes vary slightly by platform, but the ballpark totals are consistent (see size breakdown below).
 
 **Key Features:**
 
 - ✅ All platforms use **platform-specific wheels** (not universal)
 - ✅ uv pip automatically selects the correct wheel for your system
 - ✅ Each platform has its own bundled JRE optimized for that architecture
-- ✅ All 6 platforms tested and verified (252/252 tests passing)
+- ✅ All supported platforms tested and verified (252/252 tests passing)
 - ✅ Built on native runners (no emulation) for optimal performance
 
 ## What's Included
@@ -141,7 +141,6 @@ uv pip automatically selects the correct platform-specific wheel:
 ```bash
 # On Linux x64, installs: arcadedb_embedded-X.Y.Z-py3-none-manylinux_2_17_x86_64.whl
 # On macOS ARM64, installs: arcadedb_embedded-X.Y.Z-py3-none-macosx_11_0_arm64.whl
-# On Windows x64, installs: arcadedb_embedded-X.Y.Z-py3-none-win_amd64.whl
 # etc.
 ```
 

bindings/python/docs/getting-started/installation.md

@@ -11,10 +11,9 @@ uv pip install arcadedb-embedded
 **Requirements:**
 
 - **Python 3.10–3.14** (packaged; primary testing on 3.12) - No Java installation required!
-- **Supported Platforms**: Prebuilt wheels for **6 platforms**
+- **Supported Platforms**: Prebuilt wheels for **3 platforms**
   - Linux: x86_64, ARM64
-  - macOS: Intel (x86_64), Apple Silicon (ARM64)
-  - Windows: x86_64, ARM64
+  - macOS: Apple Silicon (ARM64)
 
 ## What's Included
 

bindings/python/tests/README.md

@@ -98,4 +98,4 @@ db = server.create_database("mydb")
 
 - **Questions?** See the [Testing Guide](https://humemai.github.io/arcadedb-embedded-python/latest/development/testing/)
 - **Found a bug?** [Open an issue](https://github.com/humemai/arcadedb-embedded-python/issues)
-- **Contributing?** Read [Contributing Guide](https://github.com/humemai/arcadedb-embedded-python/blob/python-embedded/CONTRIBUTING.md)
+- **Contributing?** Read [Contributing Guide](https://github.com/humemai/arcadedb-embedded-python/blob/main/CONTRIBUTING.md)