Upload 6294 files

.gitattributes

@@ -33,3 +33,30 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+rdkit/source/Code/GraphMol/SynthonSpaceSearch/data/idorsia_toy_space_a.spc filter=lfs diff=lfs merge=lfs -text
+rdkit/source/Contrib/CalcLigRMSD/figures/1C9.png filter=lfs diff=lfs merge=lfs -text
+rdkit/source/Contrib/CalcLigRMSD/figures/5dpv_complex.png filter=lfs diff=lfs merge=lfs -text
+rdkit/source/Contrib/CalcLigRMSD/figures/6c83_complex.png filter=lfs diff=lfs merge=lfs -text
+rdkit/source/Contrib/CalcLigRMSD/figures/AKI.png filter=lfs diff=lfs merge=lfs -text
+rdkit/source/Contrib/CalcLigRMSD/figures/JVE.png filter=lfs diff=lfs merge=lfs -text
+rdkit/source/Contrib/CalcLigRMSD/figures/N15.png filter=lfs diff=lfs merge=lfs -text
+rdkit/source/Contrib/NIBRSubstructureFilters/FilterSet_NIBR2019_wPubChemExamples.html filter=lfs diff=lfs merge=lfs -text
+rdkit/source/Docs/Analysis/ForceField/ForceFieldTerms.new.odt filter=lfs diff=lfs merge=lfs -text
+rdkit/source/Docs/Book_jp/supple_images/jp_supple_15.png filter=lfs diff=lfs merge=lfs -text
+rdkit/source/Docs/Book_jp/supple_images/jp_supple_24.png filter=lfs diff=lfs merge=lfs -text
+rdkit/source/Docs/Book_jp/supple_images/jp_supple_39.png filter=lfs diff=lfs merge=lfs -text
+rdkit/source/Docs/Book_jp/supple_images/jp_supple_42.png filter=lfs diff=lfs merge=lfs -text
+rdkit/source/Docs/Book_jp/supple_images/jp_supple_47.png filter=lfs diff=lfs merge=lfs -text
+rdkit/source/Docs/Book_jp/supple_images/jp_supple_49.png filter=lfs diff=lfs merge=lfs -text
+rdkit/source/Docs/Book_jp/supple_images/jp_supple_6.png filter=lfs diff=lfs merge=lfs -text
+rdkit/source/Docs/Book/images/picture_10.jpg filter=lfs diff=lfs merge=lfs -text
+rdkit/source/Docs/Book/images/pull_req1.png filter=lfs diff=lfs merge=lfs -text
+rdkit/source/Docs/Book/images/pull_req2.png filter=lfs diff=lfs merge=lfs -text
+rdkit/source/Docs/Book/images/rdkit_fork_clone.png filter=lfs diff=lfs merge=lfs -text
+rdkit/source/Docs/Book/images/similarity_map_rf.png filter=lfs diff=lfs merge=lfs -text
+rdkit/source/Docs/Code/Chem/Pharm2D.signatures.ai filter=lfs diff=lfs merge=lfs -text
+rdkit/source/Docs/Code/Chem/Pharm2D.signatures.jpg filter=lfs diff=lfs merge=lfs -text
+rdkit/source/Docs/Code/Chem/Pharm2D.triangles.ai filter=lfs diff=lfs merge=lfs -text
+rdkit/source/Docs/Code/Chem/Pharm2D.triangles.jpg filter=lfs diff=lfs merge=lfs -text
+rdkit/source/Docs/Images/logo.ai filter=lfs diff=lfs merge=lfs -text
+rdkit/source/rdkit/Chem/Draw/FreeSans.ttf filter=lfs diff=lfs merge=lfs -text

Dockerfile

@@ -0,0 +1,18 @@
+FROM python:3.10
+
+RUN useradd -m -u 1000 user && python -m pip install --upgrade pip
+USER user
+ENV PATH="/home/user/.local/bin:$PATH"
+
+WORKDIR /app
+
+COPY --chown=user ./requirements.txt requirements.txt
+RUN pip install --no-cache-dir --upgrade -r requirements.txt
+
+COPY --chown=user . /app
+ENV MCP_TRANSPORT=http
+ENV MCP_PORT=7860
+
+EXPOSE 7860
+
+CMD ["python", "rdkit/mcp_output/start_mcp.py"]

app.py

@@ -0,0 +1,45 @@
+from fastapi import FastAPI
+import os
+import sys
+
+mcp_plugin_path = os.path.join(os.path.dirname(__file__), "rdkit", "mcp_output", "mcp_plugin")
+sys.path.insert(0, mcp_plugin_path)
+
+app = FastAPI(
+    title="Rdkit MCP Service",
+    description="Auto-generated MCP service for rdkit",
+    version="1.0.0"
+)
+
+@app.get("/")
+def root():
+    return {
+        "service": "Rdkit MCP Service",
+        "version": "1.0.0",
+        "status": "running",
+        "transport": os.environ.get("MCP_TRANSPORT", "http")
+    }
+
+@app.get("/health")
+def health_check():
+    return {"status": "healthy", "service": "rdkit MCP"}
+
+@app.get("/tools")
+def list_tools():
+    try:
+        from mcp_service import create_app
+        mcp_app = create_app()
+        tools = []
+        for tool_name, tool_func in mcp_app.tools.items():
+            tools.append({
+                "name": tool_name,
+                "description": tool_func.__doc__ or "No description available"
+            })
+        return {"tools": tools}
+    except Exception as e:
+        return {"error": f"Failed to load tools: {str(e)}"}
+
+if __name__ == "__main__":
+    import uvicorn
+    port = int(os.environ.get("PORT", 7860))
+    uvicorn.run(app, host="0.0.0.0", port=port)

port.json

@@ -0,0 +1,5 @@
+{
+  "repo": "rdkit",
+  "port": 7914,
+  "timestamp": 1773264143
+}

rdkit/mcp_output/README_MCP.md

@@ -0,0 +1,128 @@
+# RDKit MCP (Model Context Protocol) Service README
+
+## 1) Project Introduction
+
+This project provides an MCP (Model Context Protocol) service wrapper around RDKit for cheminformatics workflows.  
+It is intended for LLM/tooling integrations that need molecule parsing, descriptor calculation, fingerprinting, similarity, substructure matching, and rendering.
+
+Typical use cases:
+- SMILES/InChI/mol block conversion and validation
+- Molecular descriptors (2D/3D, QED, Lipinski, etc.)
+- Fingerprint generation and similarity search
+- Substructure and MCS matching
+- Conformer/geometry operations
+- Optional utilities for database workflows and feature extraction
+
+---
+
+## 2) Installation Method
+
+### System requirements
+- Python 3.10+ recommended
+- RDKit installed with compiled C++ extensions
+- `numpy` (required)
+- Optional: `Pillow`, `pandas`, `scipy`, `matplotlib`, `IPython`, drawing backends (Cairo/Qt), InChI libs
+
+### Recommended install (Conda)
+- conda create -n rdkit-mcp -c conda-forge python=3.11 rdkit numpy
+- conda activate rdkit-mcp
+
+### Optional extras
+- pip install pillow pandas scipy matplotlib ipython
+
+### Verify installation
+- python -c "from rdkit import Chem; print(Chem.MolFromSmiles('CCO') is not None)"
+
+---
+
+## 3) Quick Start
+
+### Basic molecule operations
+- Parse: `Chem.MolFromSmiles(...)`
+- Canonicalize: `Chem.CanonSmiles(...)`
+- Substructure: `mol.HasSubstructMatch(query)`
+- Descriptors: `Descriptors.CalcMolDescriptors(mol)`
+- Fingerprints + similarity: `DataStructs.FingerprintSimilarity(fp1, fp2)`
+- Draw: `Draw.MolToImage(mol)`
+
+### Typical MCP (Model Context Protocol) service flow
+1. Client sends molecule input (SMILES/InChI/mol block).
+2. Service validates and normalizes structure.
+3. Service executes requested operation (descriptor/fingerprint/search/render/etc.).
+4. Service returns structured JSON (result, warnings, errors).
+
+---
+
+## 4) Available Tools and Endpoints List
+
+Suggested MCP (Model Context Protocol) service endpoints/tools:
+
+- `parse_molecule`
+  - Input: SMILES/InChI/mol block
+  - Output: normalized molecule metadata, canonical SMILES, validation status
+
+- `compute_descriptors`
+  - Output: selected descriptors (e.g., MolWt, LogP, TPSA, QED, counts)
+
+- `generate_fingerprint`
+  - Output: RDKit/Morgan/AtomPair/Torsion fingerprints (configurable params)
+
+- `similarity`
+  - Output: Tanimoto/Dice/Cosine similarity between two molecules/fingerprints
+
+- `substructure_search`
+  - Input: target + SMARTS query
+  - Output: match boolean, atom/bond match indices
+
+- `find_mcs`
+  - Output: maximum common substructure (SMARTS, atom/bond counts)
+
+- `enumerate_stereoisomers`
+  - Output: stereoisomer set and count
+
+- `enumerate_heterocycles`
+  - Output: transformed heterocycle variants
+
+- `standardize_molecule`
+  - Output: cleaned/neutralized/tautomer-ordered form (MolStandardize)
+
+- `draw_molecule`
+  - Output: PNG/SVG payload or file reference
+
+- `reaction_enumeration` (optional)
+  - Output: products from reaction + building blocks
+
+- `feature_finder_cli_bridge` (optional)
+  - Bridges RDKit FeatFinderCLI-style functionality
+
+- `db_create` / `db_search` (optional)
+  - Wraps `Projects/DbCLI/CreateDb.py` and `SearchDb.py` workflows
+
+---
+
+## 5) Common Issues and Notes
+
+- RDKit build/runtime mismatch:
+  - Ensure the Python environment uses the same RDKit binary installation.
+- Missing optional drawing deps:
+  - Install Pillow/Cairo/Qt backends for image generation reliability.
+- InChI failures:
+  - Confirm InChI support is present in your RDKit build.
+- Large batch performance:
+  - Prefer batched calls; avoid per-molecule process startup.
+  - Cache parsed molecules/fingerprints when possible.
+- Threading:
+  - Some operations are CPU-heavy; use process pools for throughput-critical services.
+- Error handling:
+  - Return clear parse/validation errors rather than failing silently.
+
+---
+
+## 6) Reference Links or Documentation
+
+- RDKit repository: https://github.com/rdkit/rdkit
+- Main README: https://github.com/rdkit/rdkit/blob/master/README.md
+- RDKit Book / docs: https://github.com/rdkit/rdkit/tree/master/Docs/Book
+- Installation guide: https://github.com/rdkit/rdkit/blob/master/Docs/Book/Install.md
+- C++/Python getting started: https://github.com/rdkit/rdkit/blob/master/Docs/Book/GettingStartedInC%2B%2B.md
+- Contrib utilities: https://github.com/rdkit/rdkit/tree/master/Contrib

rdkit/mcp_output/diff_report.md

@@ -0,0 +1,152 @@
+# Difference Report — `rdkit`
+
+**Generated:** 2026-03-12 05:21:17  
+**Repository:** `rdkit`  
+**Project Type:** Python library  
+**Scope/Intrusiveness:** None  
+**Workflow Status:** ✅ Success  
+**Test Status:** ❌ Failed  
+**Files Changed:** 8 added, 0 modified, 0 deleted
+
+---
+
+## 1) Project Overview
+
+This change set introduces **8 new files** to the `rdkit` repository without modifying existing files.  
+Given the “Basic functionality” feature label and zero intrusive changes, this appears to be an **additive update** intended to extend or scaffold functionality while minimizing regression risk in existing code paths.
+
+However, despite successful workflow execution, the test suite failed, indicating at least one incompatibility, missing dependency, or incomplete integration in the newly added components.
+
+---
+
+## 2) Change Summary
+
+| Metric | Value |
+|---|---|
+| New files | 8 |
+| Modified files | 0 |
+| Deleted files | 0 |
+| Net impact | Additive only |
+| CI/workflow | Passed |
+| Tests | Failed |
+
+### High-level interpretation
+- **Positive:** Existing files remain untouched, reducing direct risk to stable modules.
+- **Concern:** Test failure means the new additions are not yet production-ready as-is.
+
+---
+
+## 3) Difference Analysis
+
+Because no existing files were modified, the differences are likely in one or more of these categories:
+
+1. **New module/package files** (e.g., new APIs, helpers, utilities)
+2. **New test files** that expose existing latent issues or incomplete new behavior
+3. **Configuration/metadata additions** (packaging, CI, docs, examples)
+4. **Data/resource files** needed by functionality but possibly not wired correctly
+
+### Risk profile of additive-only changes
+- **Low risk** of breaking legacy behavior directly.
+- **Medium risk** of integration failure if:
+  - imports are unresolved,
+  - package discovery excludes new modules,
+  - runtime dependencies are not pinned/installed,
+  - tests assume unavailable environment/resources.
+
+---
+
+## 4) Technical Analysis
+
+## CI vs Test Mismatch
+Workflow success with test failure usually indicates:
+- pipeline stages before tests succeeded (lint/build/setup),
+- unit/integration tests failed at runtime or assertion level.
+
+### Likely technical causes
+- **Import path issues** (`__init__.py`, namespace/package registration)
+- **Missing optional dependency** not included in test environment
+- **Inconsistent test assumptions** (OS/path/timezone/random seed)
+- **Uninitialized fixtures/resources** for new functionality
+- **Version skew** (Python, RDKit build variants, third-party libs)
+
+### Validation focus areas
+- Ensure new files are included in package manifest/build system.
+- Confirm test runner discovers and executes intended tests only.
+- Verify deterministic tests (no hidden external state dependencies).
+- Check whether any newly added tests rely on unavailable local artifacts.
+
+---
+
+## 5) Quality and Stability Assessment
+
+**Current readiness:** ⚠️ Not release-ready (due to failed tests)
+
+**Confidence level:** Moderate (structural change is limited, but functional validity is not established)
+
+**Impact on core library:** Likely low direct impact, but unresolved failures block merge confidence and may mask deeper integration problems.
+
+---
+
+## 6) Recommendations & Improvements
+
+## Immediate (P0)
+1. **Triage failed tests from CI logs**
+   - Identify exact failing test files/cases and stack traces.
+2. **Classify failure type**
+   - Environment/setup vs code logic vs test design.
+3. **Apply minimal corrective patch**
+   - Keep change non-intrusive; avoid unrelated refactors.
+4. **Re-run full test matrix**
+   - Especially across supported Python and platform versions.
+
+## Short-term (P1)
+1. Add/adjust **dependency declarations** for new functionality.
+2. Harden tests with deterministic fixtures and explicit mocks.
+3. Add smoke tests for new modules’ importability and basic API behavior.
+4. Ensure docs/examples (if added) match executable behavior.
+
+## Medium-term (P2)
+1. Add CI guardrails:
+   - fail-fast on missing resources,
+   - stricter packaging checks (sdist/wheel contents),
+   - optional-dependency test lanes.
+2. Improve observability in tests:
+   - clearer assertion messages,
+   - structured logs for flaky paths.
+
+---
+
+## 7) Deployment Information
+
+Since tests failed, deployment should be treated as **blocked**.
+
+### Suggested deployment decision
+- **Do not promote** to release branches or package index yet.
+- Permit only **draft/pre-merge validation artifacts** if needed for debugging.
+
+### Pre-deployment checklist
+- [ ] All newly added files are packaged/discoverable
+- [ ] Full test suite green on required matrix
+- [ ] No new warnings elevated to errors in release config
+- [ ] Changelog/release notes prepared for added functionality
+- [ ] Rollback plan documented (if feature flagging is unavailable)
+
+---
+
+## 8) Future Planning
+
+1. **Stabilization sprint for additive features**
+   - Resolve current failures and add regression tests.
+2. **Incremental integration strategy**
+   - Land scaffolding first, behavior next, then performance tuning.
+3. **Compatibility verification**
+   - Validate across key RDKit environments and dependency variants.
+4. **Release governance**
+   - Require green tests + minimal coverage threshold for new modules.
+
+---
+
+## 9) Executive Conclusion
+
+This update is structurally safe (add-only, non-intrusive) but **functionally incomplete** due to test failures.  
+The recommended path is rapid failure triage, targeted fixes, and full matrix revalidation before any release action. Once tests pass, this change set should be low risk to integrate.

rdkit/mcp_output/mcp_plugin/adapter.py

@@ -0,0 +1,406 @@
+import os
+import sys
+import importlib
+import inspect
+from typing import Any, Dict, Optional, List
+
+source_path = os.path.join(
+    os.path.dirname(os.path.dirname(os.path.dirname(os.path.abspath(__file__)))),
+    "source",
+)
+sys.path.insert(0, source_path)
+
+
+class Adapter:
+    """
+    MCP Import-mode adapter for RDKit repository modules discovered via analysis.
+
+    This adapter attempts direct module imports first ("import" mode). If imports fail,
+    it gracefully falls back to "fallback" mode and returns actionable guidance.
+    All public methods return a unified dictionary payload with a required `status` field.
+    """
+
+    def __init__(self) -> None:
+        self.mode = "import"
+        self._modules: Dict[str, Optional[Any]] = {}
+        self._import_errors: Dict[str, str] = {}
+        self._load_modules()
+
+    # ==========================================================================
+    # Core utilities
+    # ==========================================================================
+
+    def _result(
+        self,
+        status: str,
+        message: str,
+        data: Optional[Any] = None,
+        error: Optional[str] = None,
+        guidance: Optional[str] = None,
+    ) -> Dict[str, Any]:
+        return {
+            "status": status,
+            "mode": self.mode,
+            "message": message,
+            "data": data,
+            "error": error,
+            "guidance": guidance,
+        }
+
+    def _import_module(self, module_path: str) -> Optional[Any]:
+        try:
+            module = importlib.import_module(module_path)
+            self._modules[module_path] = module
+            return module
+        except Exception as exc:
+            self._modules[module_path] = None
+            self._import_errors[module_path] = (
+                f"Failed to import '{module_path}': {exc}"
+            )
+            return None
+
+    def _load_modules(self) -> None:
+        target_modules = [
+            "Regress.Scripts.new_timings",
+            "Web.RDExtras.MolImage",
+            "Web.RDExtras.MolDepict",
+            "Projects.DbCLI.CreateDb",
+            "Projects.DbCLI.SearchDb",
+            "Code.DataManip.MetricMatrixCalc.Wrap.testMatricCalc",
+            "Code.DataStructs.Wrap.testBV",
+            "Code.DataStructs.Wrap.testSparseIntVect",
+            "rdkit.Chem.FeatFinderCLI",
+            "Contrib.MolVS.molvs_cli",
+        ]
+        for module_path in target_modules:
+            self._import_module(module_path)
+
+        if all(self._modules.get(m) is None for m in target_modules):
+            self.mode = "fallback"
+
+    def health(self) -> Dict[str, Any]:
+        """
+        Return adapter import health status.
+
+        Returns:
+            dict: Unified status dict including loaded modules and import errors.
+        """
+        loaded = [k for k, v in self._modules.items() if v is not None]
+        failed = {k: e for k, e in self._import_errors.items()}
+        return self._result(
+            status="ok" if loaded else "degraded",
+            message="Adapter health check completed.",
+            data={"loaded_modules": loaded, "failed_modules": failed},
+            guidance=(
+                "If critical modules failed, ensure C++ extension-backed RDKit build artifacts "
+                "are available under the source tree and required runtime deps are installed."
+            ),
+        )
+
+    def _call_function(
+        self, module_path: str, func_name: str, *args: Any, **kwargs: Any
+    ) -> Dict[str, Any]:
+        module = self._modules.get(module_path)
+        if module is None:
+            return self._result(
+                status="error",
+                message=f"Module '{module_path}' is not available.",
+                error=self._import_errors.get(module_path, "Unknown import failure."),
+                guidance=(
+                    "Verify repository source path injection and local build artifacts. "
+                    "For RDKit modules, ensure compiled extensions and numpy are available."
+                ),
+            )
+        try:
+            func = getattr(module, func_name, None)
+            if func is None or not callable(func):
+                available = [n for n, o in inspect.getmembers(module) if callable(o)]
+                return self._result(
+                    status="error",
+                    message=f"Function '{func_name}' not found in '{module_path}'.",
+                    error="Missing callable in target module.",
+                    data={"available_callables_sample": available[:25]},
+                    guidance="Check function name spelling and module version compatibility.",
+                )
+            result = func(*args, **kwargs)
+            return self._result(
+                status="ok",
+                message=f"Function '{func_name}' executed successfully.",
+                data=result,
+            )
+        except Exception as exc:
+            return self._result(
+                status="error",
+                message=f"Function '{func_name}' execution failed.",
+                error=str(exc),
+                guidance=(
+                    "Validate arguments and required environment. "
+                    "If the callable depends on RDKit backend libs, ensure they are compiled."
+                ),
+            )
+
+    # ==========================================================================
+    # Regress.Scripts.new_timings
+    # ==========================================================================
+
+    def call_new_timings_data(self, *args: Any, **kwargs: Any) -> Dict[str, Any]:
+        """
+        Call Regress.Scripts.new_timings.data.
+
+        Parameters:
+            *args: Positional arguments forwarded to data().
+            **kwargs: Keyword arguments forwarded to data().
+
+        Returns:
+            dict: Unified status response with callable result or actionable error.
+        """
+        return self._call_function("Regress.Scripts.new_timings", "data", *args, **kwargs)
+
+    # ==========================================================================
+    # Web.RDExtras.MolImage
+    # ==========================================================================
+
+    def call_molimage_gif(self, *args: Any, **kwargs: Any) -> Dict[str, Any]:
+        """
+        Call Web.RDExtras.MolImage.gif.
+
+        Parameters:
+            *args: Positional arguments for gif().
+            **kwargs: Keyword arguments for gif().
+
+        Returns:
+            dict: Unified status response.
+        """
+        return self._call_function("Web.RDExtras.MolImage", "gif", *args, **kwargs)
+
+    def call_molimage_svg(self, *args: Any, **kwargs: Any) -> Dict[str, Any]:
+        """
+        Call Web.RDExtras.MolImage.svg.
+
+        Parameters:
+            *args: Positional arguments for svg().
+            **kwargs: Keyword arguments for svg().
+
+        Returns:
+            dict: Unified status response.
+        """
+        return self._call_function("Web.RDExtras.MolImage", "svg", *args, **kwargs)
+
+    # ==========================================================================
+    # Web.RDExtras.MolDepict
+    # ==========================================================================
+
+    def call_moldepict_page(self, *args: Any, **kwargs: Any) -> Dict[str, Any]:
+        """
+        Call Web.RDExtras.MolDepict.page.
+
+        Parameters:
+            *args: Positional arguments for page().
+            **kwargs: Keyword arguments for page().
+
+        Returns:
+            dict: Unified status response.
+        """
+        return self._call_function("Web.RDExtras.MolDepict", "page", *args, **kwargs)
+
+    # ==========================================================================
+    # Projects.DbCLI.CreateDb
+    # ==========================================================================
+
+    def call_createdb_createdb(self, *args: Any, **kwargs: Any) -> Dict[str, Any]:
+        """
+        Call Projects.DbCLI.CreateDb.CreateDb.
+
+        Parameters:
+            *args: Positional arguments for CreateDb().
+            **kwargs: Keyword arguments for CreateDb().
+
+        Returns:
+            dict: Unified status response.
+        """
+        return self._call_function("Projects.DbCLI.CreateDb", "CreateDb", *args, **kwargs)
+
+    def call_createdb_initparser(self, *args: Any, **kwargs: Any) -> Dict[str, Any]:
+        """
+        Call Projects.DbCLI.CreateDb.initParser.
+
+        Parameters:
+            *args: Positional arguments for initParser().
+            **kwargs: Keyword arguments for initParser().
+
+        Returns:
+            dict: Unified status response.
+        """
+        return self._call_function("Projects.DbCLI.CreateDb", "initParser", *args, **kwargs)
+
+    # ==========================================================================
+    # Projects.DbCLI.SearchDb
+    # ==========================================================================
+
+    def call_searchdb_get_mols_from_sdfile(self, *args: Any, **kwargs: Any) -> Dict[str, Any]:
+        """
+        Call Projects.DbCLI.SearchDb.GetMolsFromSDFile.
+
+        Parameters:
+            *args: Positional arguments for GetMolsFromSDFile().
+            **kwargs: Keyword arguments for GetMolsFromSDFile().
+
+        Returns:
+            dict: Unified status response.
+        """
+        return self._call_function("Projects.DbCLI.SearchDb", "GetMolsFromSDFile", *args, **kwargs)
+
+    def call_searchdb_get_mols_from_smiles_file(self, *args: Any, **kwargs: Any) -> Dict[str, Any]:
+        """
+        Call Projects.DbCLI.SearchDb.GetMolsFromSmilesFile.
+
+        Parameters:
+            *args: Positional arguments for GetMolsFromSmilesFile().
+            **kwargs: Keyword arguments for GetMolsFromSmilesFile().
+
+        Returns:
+            dict: Unified status response.
+        """
+        return self._call_function("Projects.DbCLI.SearchDb", "GetMolsFromSmilesFile", *args, **kwargs)
+
+    def call_searchdb_get_neighbor_lists(self, *args: Any, **kwargs: Any) -> Dict[str, Any]:
+        """
+        Call Projects.DbCLI.SearchDb.GetNeighborLists.
+
+        Parameters:
+            *args: Positional arguments for GetNeighborLists().
+            **kwargs: Keyword arguments for GetNeighborLists().
+
+        Returns:
+            dict: Unified status response.
+        """
+        return self._call_function("Projects.DbCLI.SearchDb", "GetNeighborLists", *args, **kwargs)
+
+    # ==========================================================================
+    # Code.DataManip.MetricMatrixCalc.Wrap.testMatricCalc
+    # ==========================================================================
+
+    def call_testmatriccalc_feq(self, *args: Any, **kwargs: Any) -> Dict[str, Any]:
+        """
+        Call Code.DataManip.MetricMatrixCalc.Wrap.testMatricCalc.feq.
+
+        Parameters:
+            *args: Positional arguments for feq().
+            **kwargs: Keyword arguments for feq().
+
+        Returns:
+            dict: Unified status response.
+        """
+        return self._call_function(
+            "Code.DataManip.MetricMatrixCalc.Wrap.testMatricCalc", "feq", *args, **kwargs
+        )
+
+    # ==========================================================================
+    # Code.DataStructs.Wrap.testBV
+    # ==========================================================================
+
+    def call_testbv_feq(self, *args: Any, **kwargs: Any) -> Dict[str, Any]:
+        """
+        Call Code.DataStructs.Wrap.testBV.feq.
+
+        Parameters:
+            *args: Positional arguments for feq().
+            **kwargs: Keyword arguments for feq().
+
+        Returns:
+            dict: Unified status response.
+        """
+        return self._call_function("Code.DataStructs.Wrap.testBV", "feq", *args, **kwargs)
+
+    # ==========================================================================
+    # Code.DataStructs.Wrap.testSparseIntVect
+    # ==========================================================================
+
+    def call_testsparseintvect_feq(self, *args: Any, **kwargs: Any) -> Dict[str, Any]:
+        """
+        Call Code.DataStructs.Wrap.testSparseIntVect.feq.
+
+        Parameters:
+            *args: Positional arguments for feq().
+            **kwargs: Keyword arguments for feq().
+
+        Returns:
+            dict: Unified status response.
+        """
+        return self._call_function("Code.DataStructs.Wrap.testSparseIntVect", "feq", *args, **kwargs)
+
+    # ==========================================================================
+    # CLI-oriented helper wrappers (from analysis)
+    # ==========================================================================
+
+    def call_featfindercli_module(self, argv: Optional[List[str]] = None) -> Dict[str, Any]:
+        """
+        Execute rdkit.Chem.FeatFinderCLI entry-style behavior if available.
+
+        Parameters:
+            argv: Optional argument list. If omitted, module defaults are used.
+
+        Returns:
+            dict: Unified status response.
+        """
+        module_path = "rdkit.Chem.FeatFinderCLI"
+        module = self._modules.get(module_path)
+        if module is None:
+            return self._result(
+                status="error",
+                message="FeatFinderCLI module is unavailable.",
+                error=self._import_errors.get(module_path, "Unknown import failure."),
+                guidance="Confirm RDKit Python modules and compiled backends are importable.",
+            )
+        try:
+            if hasattr(module, "main") and callable(module.main):
+                result = module.main(argv) if argv is not None else module.main()
+                return self._result("ok", "FeatFinderCLI executed.", data=result)
+            return self._result(
+                status="error",
+                message="No callable main() found in FeatFinderCLI.",
+                guidance="Inspect module API and call exported functions directly.",
+            )
+        except Exception as exc:
+            return self._result(
+                status="error",
+                message="FeatFinderCLI execution failed.",
+                error=str(exc),
+                guidance="Check CLI arguments and required feature definition inputs.",
+            )
+
+    def call_molvs_cli_module(self, argv: Optional[List[str]] = None) -> Dict[str, Any]:
+        """
+        Execute Contrib.MolVS.molvs_cli entry-style behavior if available.
+
+        Parameters:
+            argv: Optional argument list for CLI-like execution.
+
+        Returns:
+            dict: Unified status response.
+        """
+        module_path = "Contrib.MolVS.molvs_cli"
+        module = self._modules.get(module_path)
+        if module is None:
+            return self._result(
+                status="error",
+                message="molvs_cli module is unavailable.",
+                error=self._import_errors.get(module_path, "Unknown import failure."),
+                guidance="Ensure Contrib modules exist in source path and dependencies are installed.",
+            )
+        try:
+            if hasattr(module, "main") and callable(module.main):
+                result = module.main(argv) if argv is not None else module.main()
+                return self._result("ok", "molvs_cli executed.", data=result)
+            return self._result(
+                status="error",
+                message="No callable main() found in molvs_cli.",
+                guidance="Inspect module for alternate callable entry points.",
+            )
+        except Exception as exc:
+            return self._result(
+                status="error",
+                message="molvs_cli execution failed.",
+                error=str(exc),
+                guidance="Validate CLI arguments and molecule input format.",
+            )

rdkit/mcp_output/mcp_plugin/main.py

@@ -0,0 +1,13 @@
+"""
+MCP Service Auto-Wrapper - Auto-generated
+"""
+from mcp_service import create_app
+
+def main():
+    """Main entry point"""
+    app = create_app()
+    return app
+
+if __name__ == "__main__":
+    app = main()
+    app.run()

rdkit/mcp_output/mcp_plugin/mcp_service.py

@@ -0,0 +1,149 @@
+import os
+import sys
+
+source_path = os.path.join(os.path.dirname(os.path.dirname(os.path.dirname(os.path.abspath(__file__)))), "source")
+if source_path not in sys.path:
+    sys.path.insert(0, source_path)
+
+from fastmcp import FastMCP
+
+from Regress.Scripts.new_timings import data
+from Web.RDExtras.MolImage import svg, gif
+from Web.RDExtras.MolDepict import page
+from Projects.DbCLI.CreateDb import initParser, CreateDb
+from Projects.DbCLI.SearchDb import GetMolsFromSDFile, GetNeighborLists, GetMolsFromSmilesFile
+from Code.DataManip.MetricMatrixCalc.Wrap.testMatricCalc import feq
+from Code.DataStructs.Wrap.testBV import feq
+from Code.DataStructs.Wrap.testSparseIntVect import feq
+
+mcp = FastMCP("unknown_service")
+
+
+@mcp.tool(name="data", description="Auto-wrapped function data")
+def data(payload: dict):
+    try:
+        if data is None:
+            return {"success": False, "result": None, "error": "Function data is not available"}
+        result = data(**payload)
+        return {"success": True, "result": result, "error": None}
+    except Exception as e:
+        return {"success": False, "result": None, "error": str(e)}
+
+@mcp.tool(name="gif", description="Auto-wrapped function gif")
+def gif(payload: dict):
+    try:
+        if gif is None:
+            return {"success": False, "result": None, "error": "Function gif is not available"}
+        result = gif(**payload)
+        return {"success": True, "result": result, "error": None}
+    except Exception as e:
+        return {"success": False, "result": None, "error": str(e)}
+
+@mcp.tool(name="svg", description="Auto-wrapped function svg")
+def svg(payload: dict):
+    try:
+        if svg is None:
+            return {"success": False, "result": None, "error": "Function svg is not available"}
+        result = svg(**payload)
+        return {"success": True, "result": result, "error": None}
+    except Exception as e:
+        return {"success": False, "result": None, "error": str(e)}
+
+@mcp.tool(name="page", description="Auto-wrapped function page")
+def page(payload: dict):
+    try:
+        if page is None:
+            return {"success": False, "result": None, "error": "Function page is not available"}
+        result = page(**payload)
+        return {"success": True, "result": result, "error": None}
+    except Exception as e:
+        return {"success": False, "result": None, "error": str(e)}
+
+@mcp.tool(name="CreateDb", description="Auto-wrapped function CreateDb")
+def CreateDb(payload: dict):
+    try:
+        if CreateDb is None:
+            return {"success": False, "result": None, "error": "Function CreateDb is not available"}
+        result = CreateDb(**payload)
+        return {"success": True, "result": result, "error": None}
+    except Exception as e:
+        return {"success": False, "result": None, "error": str(e)}
+
+@mcp.tool(name="initParser", description="Auto-wrapped function initParser")
+def initParser(payload: dict):
+    try:
+        if initParser is None:
+            return {"success": False, "result": None, "error": "Function initParser is not available"}
+        result = initParser(**payload)
+        return {"success": True, "result": result, "error": None}
+    except Exception as e:
+        return {"success": False, "result": None, "error": str(e)}
+
+@mcp.tool(name="GetMolsFromSDFile", description="Auto-wrapped function GetMolsFromSDFile")
+def GetMolsFromSDFile(payload: dict):
+    try:
+        if GetMolsFromSDFile is None:
+            return {"success": False, "result": None, "error": "Function GetMolsFromSDFile is not available"}
+        result = GetMolsFromSDFile(**payload)
+        return {"success": True, "result": result, "error": None}
+    except Exception as e:
+        return {"success": False, "result": None, "error": str(e)}
+
+@mcp.tool(name="GetMolsFromSmilesFile", description="Auto-wrapped function GetMolsFromSmilesFile")
+def GetMolsFromSmilesFile(payload: dict):
+    try:
+        if GetMolsFromSmilesFile is None:
+            return {"success": False, "result": None, "error": "Function GetMolsFromSmilesFile is not available"}
+        result = GetMolsFromSmilesFile(**payload)
+        return {"success": True, "result": result, "error": None}
+    except Exception as e:
+        return {"success": False, "result": None, "error": str(e)}
+
+@mcp.tool(name="GetNeighborLists", description="Auto-wrapped function GetNeighborLists")
+def GetNeighborLists(payload: dict):
+    try:
+        if GetNeighborLists is None:
+            return {"success": False, "result": None, "error": "Function GetNeighborLists is not available"}
+        result = GetNeighborLists(**payload)
+        return {"success": True, "result": result, "error": None}
+    except Exception as e:
+        return {"success": False, "result": None, "error": str(e)}
+
+@mcp.tool(name="feq", description="Auto-wrapped function feq")
+def feq(payload: dict):
+    try:
+        if feq is None:
+            return {"success": False, "result": None, "error": "Function feq is not available"}
+        result = feq(**payload)
+        return {"success": True, "result": result, "error": None}
+    except Exception as e:
+        return {"success": False, "result": None, "error": str(e)}
+
+@mcp.tool(name="feq", description="Auto-wrapped function feq")
+def feq(payload: dict):
+    try:
+        if feq is None:
+            return {"success": False, "result": None, "error": "Function feq is not available"}
+        result = feq(**payload)
+        return {"success": True, "result": result, "error": None}
+    except Exception as e:
+        return {"success": False, "result": None, "error": str(e)}
+
+@mcp.tool(name="feq", description="Auto-wrapped function feq")
+def feq(payload: dict):
+    try:
+        if feq is None:
+            return {"success": False, "result": None, "error": "Function feq is not available"}
+        result = feq(**payload)
+        return {"success": True, "result": result, "error": None}
+    except Exception as e:
+        return {"success": False, "result": None, "error": str(e)}
+
+
+
+def create_app():
+    """Create and return FastMCP application instance"""
+    return mcp
+
+if __name__ == "__main__":
+    mcp.run(transport="http", host="0.0.0.0", port=8000)

rdkit/mcp_output/requirements.txt

@@ -0,0 +1,6 @@
+fastmcp
+fastapi
+uvicorn[standard]
+pydantic>=2.0.0
+rdkit compiled Python package (C++ extension modules)
+numpy

rdkit/mcp_output/start_mcp.py

@@ -0,0 +1,30 @@
+
+"""
+MCP Service Startup Entry
+"""
+import sys
+import os
+
+project_root = os.path.dirname(os.path.abspath(__file__))
+mcp_plugin_dir = os.path.join(project_root, "mcp_plugin")
+if mcp_plugin_dir not in sys.path:
+    sys.path.insert(0, mcp_plugin_dir)
+
+from mcp_service import create_app
+
+def main():
+    """Start FastMCP service"""
+    app = create_app()
+    # Use environment variable to configure port, default 8000
+    port = int(os.environ.get("MCP_PORT", "8000"))
+    
+    # Choose transport mode based on environment variable
+    transport = os.environ.get("MCP_TRANSPORT", "stdio")
+    if transport == "http":
+        app.run(transport="http", host="0.0.0.0", port=port)
+    else:
+        # Default to STDIO mode
+        app.run()
+
+if __name__ == "__main__":
+    main()

rdkit/mcp_output/workflow_summary.json

@@ -0,0 +1,230 @@
+{
+  "repository": {
+    "name": "rdkit",
+    "url": "https://github.com/rdkit/rdkit",
+    "local_path": "/Users/ghh/Documents/Code/Code2MCP-private/workspace/rdkit",
+    "description": "Python library",
+    "features": "Basic functionality",
+    "tech_stack": "Python",
+    "stars": 0,
+    "forks": 0,
+    "language": "Python",
+    "last_updated": "",
+    "complexity": "complex",
+    "intrusiveness_risk": "medium"
+  },
+  "execution": {
+    "start_time": 1773263771.541916,
+    "end_time": 1773263949.6153212,
+    "duration": 178.0734052658081,
+    "status": "success",
+    "workflow_status": "success",
+    "nodes_executed": [
+      "download",
+      "analysis",
+      "env",
+      "generate",
+      "run",
+      "review",
+      "finalize"
+    ],
+    "total_files_processed": 18,
+    "environment_type": "unknown",
+    "llm_calls": 0,
+    "deepwiki_calls": 0
+  },
+  "tests": {
+    "original_project": {
+      "passed": false,
+      "details": {},
+      "test_coverage": "100%",
+      "execution_time": 0,
+      "test_files": []
+    },
+    "mcp_plugin": {
+      "passed": true,
+      "details": {},
+      "service_health": "healthy",
+      "startup_time": 0,
+      "transport_mode": "stdio",
+      "fastmcp_version": "unknown",
+      "mcp_version": "unknown"
+    }
+  },
+  "analysis": {
+    "structure": {
+      "packages": [
+        "source.Contrib.RxnRoleAssignment",
+        "source.Scripts.gen_rdkit_stubs",
+        "source.Scripts.patch_rdkit_docstrings",
+        "source.rdkit",
+        "source.rdkit.Avalon",
+        "source.rdkit.Chem",
+        "source.rdkit.DataManip",
+        "source.rdkit.DataStructs",
+        "source.rdkit.Dbase",
+        "source.rdkit.DistanceGeometry",
+        "source.rdkit.ForceField",
+        "source.rdkit.Geometry",
+        "source.rdkit.ML",
+        "source.rdkit.Numerics",
+        "source.rdkit.SimDivFilters",
+        "source.rdkit.VLib",
+        "source.rdkit.sping",
+        "source.rdkit.utils"
+      ]
+    },
+    "dependencies": {
+      "has_environment_yml": false,
+      "has_requirements_txt": false,
+      "pyproject": false,
+      "setup_cfg": true,
+      "setup_py": false
+    },
+    "entry_points": {
+      "imports": [],
+      "cli": [],
+      "modules": []
+    },
+    "risk_assessment": {
+      "import_feasibility": 0.72,
+      "intrusiveness_risk": "medium",
+      "complexity": "complex"
+    },
+    "deepwiki_analysis": {
+      "repo_url": "https://github.com/rdkit/rdkit",
+      "repo_name": "rdkit",
+      "error": "DeepWiki analysis failed",
+      "model": "gpt-5.3-codex",
+      "source": "llm_direct_analysis",
+      "success": false
+    },
+    "code_complexity": {
+      "cyclomatic_complexity": "medium",
+      "cognitive_complexity": "medium",
+      "maintainability_index": 75
+    },
+    "security_analysis": {
+      "vulnerabilities_found": 0,
+      "security_score": 85,
+      "recommendations": []
+    }
+  },
+  "plugin_generation": {
+    "files_created": [
+      "mcp_output/start_mcp.py",
+      "mcp_output/mcp_plugin/__init__.py",
+      "mcp_output/mcp_plugin/mcp_service.py",
+      "mcp_output/mcp_plugin/adapter.py",
+      "mcp_output/mcp_plugin/main.py",
+      "mcp_output/requirements.txt",
+      "mcp_output/README_MCP.md"
+    ],
+    "main_entry": "start_mcp.py",
+    "requirements": [
+      "fastmcp>=0.1.0",
+      "pydantic>=2.0.0"
+    ],
+    "readme_path": "/Users/ghh/Documents/Code/Code2MCP-private/workspace/rdkit/mcp_output/README_MCP.md",
+    "adapter_mode": "import",
+    "total_lines_of_code": 0,
+    "generated_files_size": 0,
+    "tool_endpoints": 0,
+    "supported_features": [
+      "Basic functionality"
+    ],
+    "generated_tools": [
+      "Basic tools",
+      "Health check tools",
+      "Version info tools"
+    ]
+  },
+  "code_review": {},
+  "errors": [],
+  "warnings": [],
+  "recommendations": [
+    "Add a minimal but representative CI smoke-test matrix (Linux/macOS/Windows × Python 3.10–3.12) that imports key compiled modules before running full suites",
+    "split MCP-exposed endpoints into “public API” vs “internal/test-only” and remove autogenerated exposure of `testcase/feq/load_tests` symbols",
+    "introduce an explicit endpoint allowlist with naming normalization to avoid duplicate/conflicting commands (e.g.",
+    "repeated `initParser`)",
+    "add `pyproject.toml` (while keeping `setup.cfg`) to centralize build/test/tool config and declare optional dependency groups",
+    "create reproducible dev environments (`requirements-dev.txt` or `environment.yml`) since none are currently present",
+    "add contract tests for generated MCP adapter signatures and argument schemas to catch AST-discovery drift",
+    "gate heavy/legacy test files (`rough_test.py`",
+    "very large test modules) behind markers and run them separately to improve CI signal",
+    "enforce lint/type checks on Python-only surfaces (ruff + mypy/pyright on scripts/contrib/MCP plugin) with gradual adoption",
+    "add coverage reporting focused on MCP plugin and CLI modules plus per-package thresholds",
+    "harden plugin reliability with structured error handling/timeouts and clear user-facing errors when compiled RDKit components are unavailable",
+    "add security/dependency automation (Dependabot/Renovate + pip-audit) for `mcp_output/requirements.txt` and GitHub workflows",
+    "reduce CI duplication by templating/shared job logic across Azure and GitHub workflows",
+    "document supported/stable MCP endpoints and deprecation policy in `README_MCP.md`",
+    "add basic performance benchmarks for high-use endpoints (`svg/gif`",
+    "fingerprint/search flows) and track regressions in CI",
+    "introduce golden-file tests for rendering outputs (SVG/PNG hash or perceptual checks) to prevent silent depiction changes",
+    "add ownership/CODEOWNERS for core areas (GraphMol",
+    "Draw",
+    "MCP plugin",
+    "Contrib) to speed reviews"
+  ],
+  "performance_metrics": {
+    "memory_usage_mb": 0,
+    "cpu_usage_percent": 0,
+    "response_time_ms": 0,
+    "throughput_requests_per_second": 0
+  },
+  "deployment_info": {
+    "supported_platforms": [
+      "Linux",
+      "Windows",
+      "macOS"
+    ],
+    "python_versions": [
+      "3.8",
+      "3.9",
+      "3.10",
+      "3.11",
+      "3.12"
+    ],
+    "deployment_methods": [
+      "Docker",
+      "pip",
+      "conda"
+    ],
+    "monitoring_support": true,
+    "logging_configuration": "structured"
+  },
+  "execution_analysis": {
+    "success_factors": [
+      "Workflow status is success with all planned nodes executed (download → finalize) and no runtime errors/warnings recorded.",
+      "MCP plugin boot/health checks passed (healthy stdio service), confirming deployable baseline service generation.",
+      "Import-based adapter strategy fits RDKit’s Python-first surfaces and avoided hard dependency on CLI entry points.",
+      "Repository analysis was comprehensive enough (890 files discovered) to generate broad endpoint coverage."
+    ],
+    "failure_reasons": [
+      "No hard workflow failure occurred.",
+      "Quality gap: original project tests are marked not passed/unknown despite reported 100% coverage metadata (likely instrumentation inconsistency).",
+      "DeepWiki analysis failed, reducing architecture/context confidence and forcing fallback analysis.",
+      "Over-exposure from AST discovery (test symbols/classes duplicated as endpoints) is a functional correctness risk, not execution failure."
+    ],
+    "overall_assessment": "good",
+    "node_performance": {
+      "download_time": "Completed successfully; repository imported via zip fallback, indicating resilient acquisition path.",
+      "analysis_time": "Completed with large-scope AST discovery; high breadth but noisy endpoint extraction.",
+      "generation_time": "Completed and produced expected MCP scaffold files; however generated LOC/size metrics are zero, suggesting telemetry/reporting defect.",
+      "test_time": "Plugin health test passed quickly; original project test signal is weak/ambiguous and needs normalization."
+    },
+    "resource_usage": {
+      "memory_efficiency": "Not measurable from provided metrics (reported as 0); add actual runtime telemetry.",
+      "cpu_efficiency": "Not measurable from provided metrics (reported as 0); profiling hooks are missing.",
+      "disk_usage": "Generated artifact set is small and efficient, but file size telemetry is incorrectly reported as 0."
+    }
+  },
+  "technical_quality": {
+    "code_quality_score": 72,
+    "architecture_score": 70,
+    "performance_score": 61,
+    "maintainability_score": 68,
+    "security_score": 85,
+    "scalability_score": 66
+  }
+}

rdkit/source/.azure-pipelines/linux_build.yml

@@ -0,0 +1,88 @@
+steps:
+- bash: |
+    sudo apt-get install g++ wget make libgl1-mesa-dev mesa-common-dev
+    source ${CONDA}/etc/profile.d/conda.sh
+    sudo chown -R ${USER} ${CONDA}
+    conda config --set always_yes yes --set changeps1 no
+    conda update -q conda
+    conda config --set channel_priority strict
+    conda config --add channels conda-forge
+    conda config --set solver libmamba
+    conda info -a
+    conda create --name rdkit_build -c conda-forge --override-channels  $(python) cmake \
+        boost-cpp=$(boost_version) \
+        boost=$(boost_version) \
+        numpy pillow eigen pandas=2.0 matplotlib-base=3.8 \
+        qt=5 cairo
+    conda activate rdkit_build
+    conda config --env --add channels conda-forge
+    conda config --env --set channel_priority strict
+    conda install -c conda-forge --override-channels  pytest nbval ipykernel>=6.0
+  displayName: Setup build environment
+- bash: |
+    source ${CONDA}/etc/profile.d/conda.sh
+    conda activate rdkit_build
+    export CXXFLAGS="${CXXFLAGS} -Wall -Werror"
+    mkdir build && cd build && \
+    cmake .. \
+    -DCMAKE_BUILD_TYPE=Release \
+    -DRDK_INSTALL_INTREE=ON \
+    -DRDK_INSTALL_STATIC_LIBS=OFF \
+    -DRDK_BUILD_CPP_TESTS=ON \
+    -DRDK_BUILD_PYTHON_WRAPPERS=ON \
+    -DRDK_BUILD_COORDGEN_SUPPORT=ON \
+    -DRDK_BUILD_MAEPARSER_SUPPORT=ON \
+    -DRDK_OPTIMIZE_POPCNT=ON \
+    -DRDK_BUILD_TEST_GZIP=ON \
+    -DRDK_BUILD_FREESASA_SUPPORT=ON \
+    -DRDK_BUILD_AVALON_SUPPORT=ON \
+    -DRDK_BUILD_INCHI_SUPPORT=ON \
+    -DRDK_BUILD_YAEHMOP_SUPPORT=ON \
+    -DRDK_BUILD_XYZ2MOL_SUPPORT=ON \
+    -DRDK_BUILD_CAIRO_SUPPORT=ON \
+    -DRDK_BUILD_QT_SUPPORT=ON \
+    -DQt5_DIR=/usr/lib/x86_64-linux-gnu/cmake/Qt5 \
+    -DRDK_BUILD_SWIG_WRAPPERS=OFF \
+    -DRDK_SWIG_STATIC=OFF \
+    -DRDK_BUILD_THREADSAFE_SSS=ON \
+    -DRDK_TEST_MULTITHREADED=ON \
+    -DRDK_BUILD_CFFI_LIB=ON \
+    -DBoost_NO_SYSTEM_PATHS=ON \
+    -DRDK_BOOST_PYTHON3_NAME=$(python_name) \
+    -DPYTHON_EXECUTABLE=${CONDA_PREFIX}/bin/python3 \
+    -DCMAKE_INCLUDE_PATH="${CONDA_PREFIX}/include" \
+    -DCMAKE_LIBRARY_PATH="${CONDA_PREFIX}/lib"
+  displayName: Configure build (Run CMake)
+- bash: |
+    source ${CONDA}/etc/profile.d/conda.sh
+    conda activate rdkit_build
+    cd build
+    make -j $( $(number_of_cores) ) install
+  displayName: Build
+- bash: |
+    source ${CONDA}/etc/profile.d/conda.sh
+    conda activate rdkit_build
+    export RDBASE=`pwd`
+    export PYTHONPATH=${RDBASE}:${PYTHONPATH}
+    export LD_LIBRARY_PATH=${RDBASE}/lib:${CONDA_PREFIX}/lib:${LD_LIBRARY_PATH}
+    echo "LD_LIBRARY_PATH: " $LD_LIBRARY_PATH
+    export QT_QPA_PLATFORM='offscreen'
+    cd build
+    ctest -j $( $(number_of_cores) ) --output-on-failure -T Test
+  displayName: Run tests
+- bash: |
+    source ${CONDA}/etc/profile.d/conda.sh
+    conda activate rdkit_build
+    conda install -c conda-forge --override-channels  ipython=8.12 sphinx myst-parser
+    export RDBASE=`pwd`
+    export PYTHONPATH=${RDBASE}:${PYTHONPATH}
+    export LD_LIBRARY_PATH=${RDBASE}/lib:${LD_LIBRARY_PATH}
+    export QT_QPA_PLATFORM='offscreen'
+    cd Docs/Book
+    make doctest
+  displayName: Run documentation tests
+- task: PublishTestResults@2
+  inputs:
+    testResultsFormat: 'CTest'
+    testResultsFiles: 'build/Testing/*/Test.xml'
+    testRunTitle: $(system.phasedisplayname) CTest Test Run

rdkit/source/.azure-pipelines/linux_build_cartridge.yml

@@ -0,0 +1,60 @@
+steps:
+- bash: |
+    sudo apt-get update
+    sudo apt-get install -y postgresql-14 postgresql-client-14 postgresql-server-dev-14 postgresql-common
+    echo "start postgres server"
+    sudo systemctl start postgresql.service
+    echo "add postgres user"
+    sudo -u postgres createuser -s `whoami`
+    source ${CONDA}/etc/profile.d/conda.sh
+    conda config --set always_yes yes --set changeps1 no
+    conda update -q conda
+    conda info -a
+    conda create --name rdkit_build -c conda-forge --override-channels  cmake \
+        boost-cpp=$(boost_version) \
+        boost=$(boost_version) \
+        eigen \
+        cairo
+    conda activate rdkit_build
+    conda config --env --add channels conda-forge
+    conda config --env --set channel_priority strict
+  displayName: Setup build environment
+- bash: |
+    source ${CONDA}/etc/profile.d/conda.sh
+    conda activate rdkit_build
+    mkdir build && cd build && \
+    cmake .. \
+    -DCMAKE_BUILD_TYPE=Release \
+    -DRDK_INSTALL_INTREE=ON \
+    -DRDK_BUILD_CPP_TESTS=OFF \
+    -DRDK_BUILD_PYTHON_WRAPPERS=OFF \
+    -DRDK_BUILD_COORDGEN_SUPPORT=ON \
+    -DRDK_BUILD_MAEPARSER_SUPPORT=ON \
+    -DRDK_OPTIMIZE_POPCNT=ON \
+    -DRDK_BUILD_TEST_GZIP=ON \
+    -DRDK_BUILD_AVALON_SUPPORT=ON \
+    -DRDK_BUILD_INCHI_SUPPORT=ON \
+    -DRDK_BUILD_SWIG_WRAPPERS=OFF \
+    -DRDK_BUILD_THREADSAFE_SSS=ON \
+    -DRDK_TEST_MULTITHREADED=ON \
+    -DRDK_BUILD_PGSQL=ON \
+    -DPostgreSQL_CONFIG=/usr/bin/pg_config
+  displayName: Configure build (Run CMake)
+- bash: |
+    source ${CONDA}/etc/profile.d/conda.sh
+    conda activate rdkit_build
+    cd build
+    make -j $( $(number_of_cores) ) install
+    sudo sh ./Code/PgSQL/rdkit/pgsql_install.sh
+  displayName: Build
+- bash: |
+    source ${CONDA}/etc/profile.d/conda.sh
+    conda activate rdkit_build
+    cd build
+    sh ./Code/PgSQL/rdkit/pgsql_regress.sh
+    if [ -e ./Code/PgSQL/rdkit/regression.diffs ]
+    then
+      cat ./Code/PgSQL/rdkit/regression.diffs
+      exit 1
+    fi
+  displayName: Run tests

rdkit/source/.azure-pipelines/linux_build_fuzzer.yml

@@ -0,0 +1,57 @@
+steps:
+- bash: |
+    source ${CONDA}/etc/profile.d/conda.sh
+    sudo apt-get install -y clang-9
+    sudo chown -R ${USER} ${CONDA}
+    conda config --set always_yes yes --set changeps1 no
+    conda update -q conda
+    conda info -a
+    conda create --name rdkit_build cmake -c conda-forge --override-channels \
+        boost-cpp=$(boost_version) boost=$(boost_version) \
+        py-boost=$(boost_version) libboost=$(boost_version) \
+        cairo eigen
+    conda activate rdkit_build
+    conda config --env --add channels conda-forge
+    conda config --env --set channel_priority strict	
+  displayName: Setup build environment
+- bash: |
+    source ${CONDA}/etc/profile.d/conda.sh
+    conda activate rdkit_build
+    export CC="clang-9"
+    export CXX="clang++-9"
+    export SANITIZER_FLAGS_address="-fsanitize=address -fsanitize-address-use-after-scope"
+    export COVERAGE_FLAGS="-fsanitize=fuzzer-no-link"
+    export CFLAGS="-O1 -fno-omit-frame-pointer -gline-tables-only -DFUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION $COVERAGE_FLAGS $SANITIZER_FLAGS_address"
+    export CXXFLAGS="$CFLAGS"
+    export LIB_FUZZING_ENGINE="-fsanitize=fuzzer"
+    mkdir build && cd build && \
+    cmake .. \
+    -DCMAKE_BUILD_TYPE=Release \
+    -DRDK_INSTALL_INTREE=ON \
+    -DRDK_BUILD_PYTHON_WRAPPERS=OFF \
+    -DLIB_FUZZING_ENGINE=${LIB_FUZZING_ENGINE} \
+    -DRDK_BUILD_FUZZ_TARGETS=ON \
+    -DRDK_INSTALL_STATIC_LIBS=ON \
+    -DBoost_USE_STATIC_LIBS=ON \
+    -DRDK_BUILD_CPP_TESTS=OFF \
+    -DBoost_NO_SYSTEM_PATHS=ON \
+    -DCMAKE_INCLUDE_PATH="${CONDA_PREFIX}/include" \
+    -DCMAKE_LIBRARY_PATH="${CONDA_PREFIX}/lib" \
+    -DCMAKE_PREFIX_PATH=$CONDA_PREFIX
+  displayName: Configure build (Run CMake)
+- bash: |
+    source ${CONDA}/etc/profile.d/conda.sh
+    conda activate rdkit_build
+    cd build
+    make -j $( $(number_of_cores) ) install
+  displayName: Build
+- bash: |
+    source ${CONDA}/etc/profile.d/conda.sh
+    conda activate rdkit_build
+    corpora=$(find . -type d -name "*_fuzzer")
+    for corpus in $corpora; do
+      corpus_basename=$(basename $corpus)
+      fuzzer_target=$(find . -type f -name "${corpus_basename}" -print -quit)
+      ${fuzzer_target} ${corpus}/*
+    done
+  displayName: Run fuzzer on existing corpora

rdkit/source/.azure-pipelines/linux_build_java.yml

@@ -0,0 +1,62 @@
+steps:
+- bash: |
+    sudo apt-get install g++ wget make libgl1-mesa-dev mesa-common-dev
+    source ${CONDA}/etc/profile.d/conda.sh
+    sudo chown -R ${USER} ${CONDA}
+    conda config --set always_yes yes --set changeps1 no
+    conda update -q conda
+    conda info -a
+    conda create --name rdkit_build -c conda-forge --override-channels  cmake \
+        libboost=$(boost_version) \
+        libboost-devel=$(boost_version) \
+        swig=4.2
+  displayName: Setup build environment
+- bash: |
+    source ${CONDA}/etc/profile.d/conda.sh
+    conda activate rdkit_build
+    conda config --env --add channels conda-forge
+    conda config --env --set channel_priority strict
+    mkdir build && cd build && \
+    cmake .. \
+    -DSWIG_EXECUTABLE=${CONDA_PREFIX}/bin/swig \
+    -DCMAKE_BUILD_TYPE=Release \
+    -DRDK_INSTALL_INTREE=ON \
+    -DRDK_INSTALL_STATIC_LIBS=ON \
+    -DRDK_BUILD_CPP_TESTS=OFF \
+    -DRDK_BUILD_PYTHON_WRAPPERS=OFF \
+    -DRDK_BUILD_COORDGEN_SUPPORT=ON \
+    -DRDK_BUILD_MAEPARSER_SUPPORT=ON \
+    -DRDK_OPTIMIZE_POPCNT=ON \
+    -DRDK_BUILD_TEST_GZIP=OFF \
+    -DRDK_BUILD_FREESASA_SUPPORT=ON \
+    -DRDK_BUILD_AVALON_SUPPORT=ON \
+    -DRDK_BUILD_INCHI_SUPPORT=ON \
+    -DRDK_BUILD_YAEHMOP_SUPPORT=ON \
+    -DRDK_BUILD_XYZ2MOL_SUPPORT=ON \
+    -DRDK_BUILD_SWIG_WRAPPERS=ON \
+    -DRDK_SWIG_STATIC=ON \
+    -DRDK_BUILD_THREADSAFE_SSS=ON \
+    -DRDK_TEST_MULTITHREADED=ON
+  displayName: Configure build (Run CMake)
+- bash: |
+    source ${CONDA}/etc/profile.d/conda.sh
+    conda activate rdkit_build
+    cd build
+    make -j $( $(number_of_cores) ) install
+  displayName: Build
+- bash: |
+    source ${CONDA}/etc/profile.d/conda.sh
+    conda activate rdkit_build
+    export RDBASE=`pwd`
+    export PYTHONPATH=${RDBASE}:${PYTHONPATH}
+    export LD_LIBRARY_PATH=${RDBASE}/lib:${CONDA_PREFIX}/lib:${LD_LIBRARY_PATH}
+    echo "LD_LIBRARY_PATH: " $LD_LIBRARY_PATH
+    export QT_QPA_PLATFORM='offscreen'
+    cd build
+    ctest -j $( $(number_of_cores) ) --output-on-failure -T Test
+  displayName: Run tests
+- task: PublishTestResults@2
+  inputs:
+    testResultsFormat: 'CTest'
+    testResultsFiles: 'build/Testing/*/Test.xml'
+    testRunTitle: $(system.phasedisplayname) CTest Test Run

rdkit/source/.azure-pipelines/linux_build_limitexternal.yml

@@ -0,0 +1,70 @@
+steps:
+- bash: |
+    sudo apt-get install g++ wget make libgl1-mesa-dev mesa-common-dev
+    source ${CONDA}/etc/profile.d/conda.sh
+    sudo chown -R ${USER} ${CONDA}
+    conda config --set always_yes yes --set changeps1 no
+    conda update -q conda
+    conda config --set channel_priority strict
+    conda config --add channels conda-forge
+    conda config --set solver libmamba
+    conda info -a
+    conda create --name rdkit_build -c conda-forge --override-channels $(python) cmake \
+        libboost=$(boost_version) libboost-devel=$(boost_version) \
+        libboost-python=$(boost_version) libboost-python-devel=$(boost_version) \
+        numpy pillow eigen pandas=2.1 \
+        qt pytest
+    conda activate rdkit_build
+    conda config --env --add channels conda-forge
+    conda config --env --set channel_priority strict
+  displayName: Setup build environment
+- bash: |
+    source ${CONDA}/etc/profile.d/conda.sh
+    conda activate rdkit_build
+    mkdir build && cd build && \
+    cmake .. \
+    -DCMAKE_BUILD_TYPE=Release \
+    -DRDK_INSTALL_INTREE=ON \
+    -DRDK_INSTALL_STATIC_LIBS=OFF \
+    -DRDK_BUILD_CPP_TESTS=ON \
+    -DRDK_BUILD_PYTHON_WRAPPERS=ON \
+    -DRDK_BUILD_CHEMDRAW_SUPPORT=OFF \
+    -DRDK_BUILD_COORDGEN_SUPPORT=OFF \
+    -DRDK_BUILD_MAEPARSER_SUPPORT=OFF \
+    -DRDK_OPTIMIZE_POPCNT=ON \
+    -DRDK_BUILD_TEST_GZIP=OFF \
+    -DRDK_BUILD_FREESASA_SUPPORT=OFF \
+    -DRDK_BUILD_FREETYPE_SUPPORT=OFF \
+    -DRDK_BUILD_AVALON_SUPPORT=OFF \
+    -DRDK_BUILD_INCHI_SUPPORT=OFF \
+    -DRDK_BUILD_CAIRO_SUPPORT=OFF \
+    -DRDK_BUILD_QT_SUPPORT=OFF \
+    -DRDK_BUILD_XYZ2MOL_SUPPORT=ON \
+    -DRDK_BUILD_PUBCHEMSHAPE_SUPPORT=OFF \
+    -DRDK_BUILD_SWIG_WRAPPERS=OFF \
+    -DRDK_SWIG_STATIC=OFF \
+    -DRDK_BUILD_THREADSAFE_SSS=OFF \
+    -DRDK_TEST_MULTITHREADED=OFF
+  displayName: Configure build (Run CMake)
+- bash: |
+    source ${CONDA}/etc/profile.d/conda.sh
+    conda activate rdkit_build
+    cd build
+    make -j $( $(number_of_cores) ) install
+  displayName: Build
+- bash: |
+    source ${CONDA}/etc/profile.d/conda.sh
+    conda activate rdkit_build
+    export RDBASE=`pwd`
+    export PYTHONPATH=${RDBASE}:${PYTHONPATH}
+    export LD_LIBRARY_PATH=${RDBASE}/lib:${CONDA_PREFIX}/lib:${LD_LIBRARY_PATH}
+    echo "LD_LIBRARY_PATH: " $LD_LIBRARY_PATH
+    export QT_QPA_PLATFORM='offscreen'
+    cd build
+    ctest -j $( $(number_of_cores) ) --output-on-failure -T Test
+  displayName: Run tests
+- task: PublishTestResults@2
+  inputs:
+    testResultsFormat: 'CTest'
+    testResultsFiles: 'build/Testing/*/Test.xml'
+    testRunTitle: $(system.phasedisplayname) CTest Test Run

rdkit/source/.azure-pipelines/linux_build_py312.yml

@@ -0,0 +1,98 @@
+steps:
+- bash: |
+    sudo apt-get install g++ wget make libgl1-mesa-dev mesa-common-dev
+    source ${CONDA}/etc/profile.d/conda.sh
+    sudo chown -R ${USER} ${CONDA}
+    conda config --set always_yes yes --set changeps1 no
+    conda update -q conda
+    conda info -a
+    conda create --name rdkit_build -c conda-forge --override-channels  $(python) cmake \
+        libboost-python-devel=$(boost_version) \
+        libboost-python=$(boost_version) \
+        libboost-devel=$(boost_version) \
+        libboost=$(boost_version) \
+        numpy=2.4 pillow eigen pandas=3 matplotlib-base=3.8 \
+        cairo
+    conda activate rdkit_build
+    conda config --env --add channels conda-forge
+    conda config --env --set channel_priority strict
+    conda install -c conda-forge --override-channels  ipython=8.20 jupyter nbval ipykernel>=6.0
+  displayName: Setup build environment (no Qt due to versioning problems)
+- bash: |
+    source ${CONDA}/etc/profile.d/conda.sh
+    conda activate rdkit_build
+    export CXXFLAGS="${CXXFLAGS} -Wall -Werror -Wextra"
+    mkdir build && cd build && \
+    cmake .. \
+    -DCMAKE_BUILD_TYPE=Release \
+    -DRDK_INSTALL_INTREE=ON \
+    -DRDK_INSTALL_STATIC_LIBS=OFF \
+    -DRDK_BUILD_CPP_TESTS=ON \
+    -DRDK_BUILD_PYTHON_WRAPPERS=ON \
+    -DRDK_BUILD_COORDGEN_SUPPORT=ON \
+    -DRDK_BUILD_MAEPARSER_SUPPORT=ON \
+    -DRDK_OPTIMIZE_POPCNT=ON \
+    -DRDK_BUILD_TEST_GZIP=ON \
+    -DRDK_BUILD_FREESASA_SUPPORT=ON \
+    -DRDK_BUILD_AVALON_SUPPORT=ON \
+    -DRDK_BUILD_INCHI_SUPPORT=ON \
+    -DRDK_BUILD_YAEHMOP_SUPPORT=ON \
+    -DRDK_BUILD_XYZ2MOL_SUPPORT=ON \
+    -DRDK_BUILD_CAIRO_SUPPORT=ON \
+    -DRDK_BUILD_SWIG_WRAPPERS=OFF \
+    -DRDK_SWIG_STATIC=OFF \
+    -DRDK_BUILD_THREADSAFE_SSS=ON \
+    -DRDK_TEST_MULTITHREADED=ON \
+    -DRDK_BUILD_CFFI_LIB=ON \
+    -DBoost_NO_SYSTEM_PATHS=ON \
+    -DBoost_NO_BOOST_CMAKE=TRUE \
+    -DRDK_BOOST_PYTHON3_NAME=$(python_name) \
+    -DPYTHON_EXECUTABLE=${CONDA_PREFIX}/bin/python3 \
+    -DCMAKE_INCLUDE_PATH="${CONDA_PREFIX}/include" \
+    -DCMAKE_LIBRARY_PATH="${CONDA_PREFIX}/lib"
+  displayName: Configure build (Run CMake)
+- bash: |
+    source ${CONDA}/etc/profile.d/conda.sh
+    conda activate rdkit_build
+    cd build
+    make -j $( $(number_of_cores) ) install
+  displayName: Build
+- bash: |
+    source ${CONDA}/etc/profile.d/conda.sh
+    conda activate rdkit_build
+    export RDBASE=`pwd`
+    export PYTHONPATH=${RDBASE}:${PYTHONPATH}
+    export LD_LIBRARY_PATH=${RDBASE}/lib:${CONDA_PREFIX}/lib:${LD_LIBRARY_PATH}
+    echo "LD_LIBRARY_PATH: " $LD_LIBRARY_PATH
+    export QT_QPA_PLATFORM='offscreen'
+    cd build
+    ctest -j $( $(number_of_cores) ) --output-on-failure -T Test
+  displayName: Run tests
+- bash: |
+    source ${CONDA}/etc/profile.d/conda.sh
+    conda activate rdkit_build
+    conda install -c conda-forge --override-channels numpy=1.24 pandas=2.2
+    export RDBASE=`pwd`
+    export PYTHONPATH=${RDBASE}:${PYTHONPATH}
+    export LD_LIBRARY_PATH=${RDBASE}/lib:${CONDA_PREFIX}/lib:${LD_LIBRARY_PATH}
+    echo "LD_LIBRARY_PATH: " $LD_LIBRARY_PATH
+    export QT_QPA_PLATFORM='offscreen'
+    cd build
+    ctest -j $( $(number_of_cores) ) --output-on-failure -T Test
+  displayName: Run tests with older numpy and pandas
+- bash: |
+    source ${CONDA}/etc/profile.d/conda.sh
+    conda activate rdkit_build
+    conda install -c conda-forge --override-channels  sphinx myst-parser
+    export RDBASE=`pwd`
+    export PYTHONPATH=${RDBASE}:${PYTHONPATH}
+    export LD_LIBRARY_PATH=${RDBASE}/lib:${LD_LIBRARY_PATH}
+    export QT_QPA_PLATFORM='offscreen'
+    cd Docs/Book
+    make doctest
+  displayName: Run documentation tests
+- task: PublishTestResults@2
+  inputs:
+    testResultsFormat: 'CTest'
+    testResultsFiles: 'build/Testing/*/Test.xml'
+    testRunTitle: $(system.phasedisplayname) CTest Test Run

rdkit/source/.azure-pipelines/mac_build.yml

@@ -0,0 +1,95 @@
+steps:
+- bash: |
+    export CONDA=/tmp/miniforge
+    wget https://github.com/conda-forge/miniforge/releases/download/24.9.0-0/Miniforge3-24.9.0-0-MacOSX-x86_64.sh
+    bash Miniforge3-24.9.0-0-MacOSX-x86_64.sh -b -p ${CONDA}
+  displayName: install conda
+- script: |
+    echo "Removing homebrew from Azure to avoid conflicts."
+    curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/uninstall > ~/uninstall_homebrew
+    chmod +x ~/uninstall_homebrew
+    ~/uninstall_homebrew -fq
+    rm ~/uninstall_homebrew
+  displayName: Remove homebrew
+- bash: |
+    export CONDA=/tmp/miniforge
+    source ${CONDA}/etc/profile.d/conda.sh
+    sudo chown -R ${USER} ${CONDA}
+    mkdir -p $HOME/.matplotlib
+    echo -e "backend: TkAgg\n" > $HOME/.matplotlib/matplotlibrc
+    conda config --set always_yes yes --set changeps1 no
+    conda update -q conda
+    conda config --set solver libmamba
+    conda config --set channel_priority strict
+    conda config --add channels conda-forge
+    conda create --name rdkit_build $(python) $(compiler) libcxx cmake \
+        libboost=$(boost_version) libboost-devel=$(boost_version) \
+        libboost-python=$(boost_version) libboost-python-devel=$(boost_version) \
+        qt \
+        numpy matplotlib=3.8 cairo pillow eigen pandas=2.1  \
+        jupyter=1.0 ipython=8.20 sphinx myst-parser pytest nbval make
+  displayName: Setup build environment
+- bash: |
+    export CONDA=/tmp/miniforge
+    source ${CONDA}/etc/profile.d/conda.sh
+    conda activate rdkit_build
+    ls -l /Library
+    ls -l /Library/Developer
+    ls -l /Library/Developer/CommandLineTools
+    ls -l /Library/Developer/CommandLineTools/SDKs
+    export SDKROOT=/Library/Developer/CommandLineTools/SDKs/MacOSX$(target_platform).sdk
+    export CONDA_BUILD_SYSROOT=${SDKROOT}
+    export CXXFLAGS="${CXXFLAGS} -Wall -Wextra -Werror"
+    mkdir build && cd build && \
+    cmake .. \
+    -DCMAKE_BUILD_TYPE=Release \
+    -DRDK_INSTALL_INTREE=ON \
+    -DRDK_INSTALL_STATIC_LIBS=OFF \
+    -DRDK_BUILD_CPP_TESTS=ON \
+    -DRDK_BUILD_PYTHON_WRAPPERS=ON \
+    -DRDK_BUILD_COORDGEN_SUPPORT=ON \
+    -DRDK_BUILD_MAEPARSER_SUPPORT=ON \
+    -DRDK_OPTIMIZE_POPCNT=ON \
+    -DRDK_BUILD_TEST_GZIP=ON \
+    -DRDK_BUILD_FREESASA_SUPPORT=ON \
+    -DRDK_BUILD_AVALON_SUPPORT=ON \
+    -DRDK_BUILD_INCHI_SUPPORT=ON \
+    -DRDK_BUILD_YAEHMOP_SUPPORT=ON \
+    -DRDK_BUILD_XYZ2MOL_SUPPORT=ON \
+    -DRDK_BUILD_CAIRO_SUPPORT=ON \
+    -DRDK_BUILD_QT_SUPPORT=ON \
+    -DRDK_BUILD_SWIG_WRAPPERS=OFF \
+    -DRDK_SWIG_STATIC=OFF \
+    -DRDK_BUILD_THREADSAFE_SSS=ON \
+    -DRDK_TEST_MULTITHREADED=ON \
+    -DRDK_BUILD_CFFI_LIB=ON \
+    -DCMAKE_OSX_SYSROOT=${SDKROOT} \
+    -DCMAKE_OSX_DEPLOYMENT_TARGET=$(target_platform)
+  displayName: Configure build (Run CMake)
+- bash: |
+    export CONDA=/tmp/miniforge
+    source ${CONDA}/etc/profile.d/conda.sh
+    conda activate rdkit_build
+    export SDKROOT=/Library/Developer/CommandLineTools/SDKs/MacOSX$(target_platform).sdk
+    export CONDA_BUILD_SYSROOT=${SDKROOT}
+    cd build
+    make -j $( $(number_of_cores) ) install
+  displayName: Build
+- bash: |
+    export CONDA=/tmp/miniforge
+    source ${CONDA}/etc/profile.d/conda.sh
+    conda activate rdkit_build
+    export RDBASE=`pwd`
+    export PYTHONPATH=${RDBASE}:${PYTHONPATH}
+    export DYLD_FALLBACK_LIBRARY_PATH=${RDBASE}/lib:${RDBASE}/rdkit:${CONDA_PREFIX}/lib:${DYLD_FALLBACK_LIBRARY_PATH}
+    export SDKROOT=/Library/Developer/CommandLineTools/SDKs/MacOSX$(target_platform).sdk
+    export CONDA_BUILD_SYSROOT=${SDKROOT}
+    export QT_QPA_PLATFORM='offscreen'
+    cd build
+    ctest -j $( $(number_of_cores) ) --output-on-failure -T Test
+  displayName: Run tests
+- task: PublishTestResults@2
+  inputs:
+    testResultsFormat: 'CTest'
+    testResultsFiles: 'build/Testing/*/Test.xml'
+    testRunTitle: $(system.phasedisplayname) CTest Test Run

rdkit/source/.azure-pipelines/mac_build_java.yml

@@ -0,0 +1,88 @@
+steps:
+- bash: |
+    export CONDA=/tmp/miniforge
+    wget https://github.com/conda-forge/miniforge/releases/download/24.9.0-0/Miniforge3-24.9.0-0-MacOSX-x86_64.sh
+    bash Miniforge3-24.9.0-0-MacOSX-x86_64.sh -b -p ${CONDA}
+  displayName: install conda
+- script: |
+    echo "Removing homebrew from Azure to avoid conflicts."
+    curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/uninstall > ~/uninstall_homebrew
+    chmod +x ~/uninstall_homebrew
+    ~/uninstall_homebrew -fq
+    rm ~/uninstall_homebrew
+  displayName: Remove homebrew
+- bash: |
+    export CONDA=/tmp/miniforge
+    source ${CONDA}/etc/profile.d/conda.sh
+    sudo chown -R ${USER} ${CONDA}
+    mkdir -p $HOME/.matplotlib
+    echo -e "backend: TkAgg\n" > $HOME/.matplotlib/matplotlibrc
+    conda config --set always_yes yes --set changeps1 no
+    conda update -q conda
+    conda create --name rdkit_build  -c conda-forge --override-channels  $(compiler) \
+        libcxx=$(compiler_version) cmake=3.26 \
+        libboost=$(boost_version) \
+        libboost-devel=$(boost_version) \
+        cairo eigen swig=4.2 make
+    conda activate rdkit_build
+    conda config --env --add channels conda-forge
+    conda config --env --set channel_priority strict
+  displayName: Setup build environment
+- bash: |
+    export CONDA=/tmp/miniforge
+    source ${CONDA}/etc/profile.d/conda.sh
+    conda activate rdkit_build
+    ls -l /Library
+    ls -l /Library/Developer
+    ls -l /Library/Developer/CommandLineTools
+    ls -l /Library/Developer/CommandLineTools/SDKs
+    export SDKROOT=/Library/Developer/CommandLineTools/SDKs/MacOSX$(target_platform).sdk
+    export CONDA_BUILD_SYSROOT=${SDKROOT}
+    mkdir build && cd build && \
+    cmake .. \
+    -DCMAKE_BUILD_TYPE=Release \
+    -DRDK_INSTALL_INTREE=ON \
+    -DRDK_INSTALL_STATIC_LIBS=ON \
+    -DRDK_BUILD_CPP_TESTS=OFF \
+    -DRDK_BUILD_PYTHON_WRAPPERS=OFF \
+    -DRDK_BUILD_COORDGEN_SUPPORT=ON \
+    -DRDK_BUILD_MAEPARSER_SUPPORT=ON \
+    -DRDK_OPTIMIZE_POPCNT=ON \
+    -DRDK_BUILD_TEST_GZIP=OFF \
+    -DRDK_BUILD_FREESASA_SUPPORT=ON \
+    -DRDK_BUILD_AVALON_SUPPORT=ON \
+    -DRDK_BUILD_INCHI_SUPPORT=ON \
+    -DRDK_BUILD_YAEHMOP_SUPPORT=ON \
+    -DRDK_BUILD_XYZ2MOL_SUPPORT=ON \
+    -DRDK_BUILD_SWIG_WRAPPERS=ON \
+    -DRDK_SWIG_STATIC=ON \
+    -DRDK_BUILD_THREADSAFE_SSS=ON \
+    -DRDK_TEST_MULTITHREADED=ON
+  displayName: Configure build (Run CMake)
+- bash: |
+    export CONDA=/tmp/miniforge
+    source ${CONDA}/etc/profile.d/conda.sh
+    conda activate rdkit_build
+    export SDKROOT=/Library/Developer/CommandLineTools/SDKs/MacOSX$(target_platform).sdk
+    export CONDA_BUILD_SYSROOT=${SDKROOT}
+    cd build
+    make -j $( $(number_of_cores) ) install
+  displayName: Build
+- bash: |
+    export CONDA=/tmp/miniforge
+    source ${CONDA}/etc/profile.d/conda.sh
+    conda activate rdkit_build
+    export RDBASE=`pwd`
+    export PYTHONPATH=${RDBASE}:${PYTHONPATH}
+    export DYLD_FALLBACK_LIBRARY_PATH=${RDBASE}/lib:${RDBASE}/rdkit:${CONDA_PREFIX}/lib:${DYLD_FALLBACK_LIBRARY_PATH}
+    export SDKROOT=/Library/Developer/CommandLineTools/SDKs/MacOSX$(target_platform).sdk
+    export CONDA_BUILD_SYSROOT=${SDKROOT}
+    export QT_QPA_PLATFORM='offscreen'
+    cd build
+    ctest -j $( $(number_of_cores) ) --output-on-failure -T Test
+  displayName: Run tests
+- task: PublishTestResults@2
+  inputs:
+    testResultsFormat: 'CTest'
+    testResultsFiles: 'build/Testing/*/Test.xml'
+    testRunTitle: $(system.phasedisplayname) CTest Test Run

rdkit/source/.azure-pipelines/vs_build.yml

@@ -0,0 +1,85 @@
+steps:
+- powershell: Write-Host "##vso[task.prependpath]$env:CONDA\Scripts"
+  displayName: Setup build environment
+- script: |
+    conda config --set always_yes yes --set changeps1 no
+    conda info -a
+    conda install -n base conda-libmamba-solver
+    conda config --set solver libmamba
+    conda create --name rdkit_build -c conda-forge --override-channels  $(python) ^
+        libboost-python=(boost_version) ^
+        libboost-python-devel=(boost_version) ^
+        libboost=$(boost_version) ^
+        libboost-devel=$(boost_version) ^
+        numpy matplotlib cairo pillow eigen ^
+        numpy=2.4 pandas=3 ^
+        sphinx myst-parser ipython jupyter pytest nbval cmake 
+    call activate rdkit_build
+    conda config --env --add channels conda-forge 
+    conda config --env --set channel_priority strict
+  displayName: Install dependencies
+- script: |
+    set Boost_ROOT=
+    mkdir build && cd build
+    call activate rdkit_build
+    cmake .. ^
+    -G "Visual Studio 17 2022" ^
+    -DCMAKE_BUILD_TYPE=Release ^
+    -DRDK_INSTALL_INTREE=ON ^
+    -DRDK_INSTALL_STATIC_LIBS=OFF ^
+    -DRDK_BUILD_CPP_TESTS=ON ^
+    -DRDK_BUILD_PYTHON_WRAPPERS=ON ^
+    -DRDK_BUILD_COORDGEN_SUPPORT=ON ^
+    -DRDK_BUILD_MAEPARSER_SUPPORT=ON ^
+    -DRDK_OPTIMIZE_POPCNT=ON ^
+    -DRDK_BUILD_TEST_GZIP=ON ^
+    -DRDK_BUILD_FREESASA_SUPPORT=ON ^
+    -DRDK_BUILD_AVALON_SUPPORT=ON ^
+    -DRDK_BUILD_INCHI_SUPPORT=ON ^
+    -DRDK_BUILD_YAEHMOP_SUPPORT=ON ^
+    -DRDK_BUILD_XYZ2MOL_SUPPORT=ON ^
+    -DRDK_BUILD_CAIRO_SUPPORT=ON ^
+    -DRDK_BUILD_THREADSAFE_SSS=ON ^
+    -DRDK_BUILD_SWIG_WRAPPERS=OFF ^
+    -DRDK_BUILD_CFFI_LIB=ON ^
+    -DRDK_SWIG_STATIC=OFF ^
+    -DRDK_TEST_MULTITHREADED=ON ^
+    -DCMAKE_INCLUDE_PATH=%CONDA_PREFIX%/Library/include ^
+    -DCMAKE_LIBRARY_PATH="%CONDA_PREFIX%/Library/lib
+  displayName: Configure build (Run CMake)
+- script: |
+    call activate rdkit_build
+    cd build
+    cmake --build . --config Release --target install -j $(number_of_cores)
+  displayName: Build
+- script: |
+    call activate rdkit_build
+    set RDBASE=%cd%
+    set PYTHONPATH=%RDBASE%;%PYTHONPATH%
+    set PATH=%RDBASE%\lib;%PATH%
+    cd build
+    ctest -C Release -j $(number_of_cores) --output-on-failure -T Test
+  displayName: Run tests
+- script: |
+    call activate rdkit_build
+    conda install -c conda-forge --override-channels  numpy=1.26 pandas=2.2
+    set RDBASE=%cd%
+    set PYTHONPATH=%RDBASE%;%PYTHONPATH%
+    set PATH=%RDBASE%\lib;%PATH%
+    cd build
+    ctest -C Release -j $(number_of_cores) --output-on-failure -T Test
+  displayName: Run with older numpy and pandas
+- script: |
+    call activate rdkit_build
+    conda install -c conda-forge --override-channels  sphinx myst-parser
+    set RDBASE=%cd%
+    set PYTHONPATH=%RDBASE%;%PYTHONPATH%
+    set PATH=%RDBASE%\lib;%PATH%
+    cd Docs/Book
+    make doctest
+  displayName: Run documentation tests
+- task: PublishTestResults@2
+  inputs:
+    testResultsFormat: 'CTest'
+    testResultsFiles: 'build/Testing/*/Test.xml'
+    testRunTitle: $(system.phasedisplayname)  CTest Test Run

rdkit/source/.azure-pipelines/vs_build_dll.yml

@@ -0,0 +1,65 @@
+steps:
+- powershell: Write-Host "##vso[task.prependpath]$env:CONDA\Scripts"
+  displayName: Setup build environment
+- script: |
+    conda config --set always_yes yes --set changeps1 no
+    conda info -a
+    conda install -n base conda-libmamba-solver
+    conda config --set solver libmamba
+    conda create --name rdkit_build -c conda-forge --override-channels  $(python) ^
+        boost=$(boost_version) boost-cpp=$(boost_version) ^
+        libboost-python-devel=(boost_version) ^
+        libboost=$(boost_version) ^
+        libboost-devel=$(boost_version) ^
+        numpy matplotlib cairo pillow eigen pandas=2.1
+    call activate rdkit_build
+    conda config --env --add channels conda-forge
+    conda config --env --set channel_priority strict
+    conda install -c conda-forge --override-channels  cmake ipython pytest nbval
+  displayName: Install dependencies
+- script: |
+    set Boost_ROOT=
+    mkdir build && cd build
+    call activate rdkit_build
+    cmake .. ^
+    -G "Visual Studio 17 2022" ^
+    -DCMAKE_BUILD_TYPE=Release ^
+    -DRDK_INSTALL_INTREE=ON ^
+    -DRDK_INSTALL_STATIC_LIBS=OFF ^
+    -DRDK_INSTALL_DLLS_MSVC=ON ^
+    -DRDK_BUILD_CPP_TESTS=ON ^
+    -DRDK_BUILD_PYTHON_WRAPPERS=OFF ^
+    -DRDK_BUILD_COORDGEN_SUPPORT=ON ^
+    -DRDK_BUILD_MAEPARSER_SUPPORT=ON ^
+    -DRDK_OPTIMIZE_POPCNT=ON ^
+    -DRDK_BUILD_TEST_GZIP=ON ^
+    -DRDK_BUILD_FREESASA_SUPPORT=ON ^
+    -DRDK_BUILD_AVALON_SUPPORT=ON ^
+    -DRDK_BUILD_INCHI_SUPPORT=ON ^
+    -DRDK_BUILD_YAEHMOP_SUPPORT=ON ^
+    -DRDK_BUILD_XYZ2MOL_SUPPORT=ON ^
+    -DRDK_BUILD_CAIRO_SUPPORT=ON ^
+    -DRDK_BUILD_THREADSAFE_SSS=ON ^
+    -DRDK_BUILD_SWIG_WRAPPERS=OFF ^
+    -DRDK_SWIG_STATIC=OFF ^
+    -DRDK_TEST_MULTITHREADED=ON ^
+    -DCMAKE_INCLUDE_PATH=%CONDA_PREFIX%/Library/include ^
+    -DCMAKE_LIBRARY_PATH="%CONDA_PREFIX%/Library/lib
+  displayName: Configure build (Run CMake)
+- script: |
+    call activate rdkit_build
+    cd build
+    cmake --build . --config Release --target install -j $(number_of_cores)
+  displayName: Build
+- script: |
+    call activate rdkit_build
+    set RDBASE=%cd%
+    set PATH=%RDBASE%\lib;%PATH%
+    cd build
+    ctest -C Release -j $(number_of_cores) --output-on-failure -T Test
+  displayName: Run tests
+- task: PublishTestResults@2
+  inputs:
+    testResultsFormat: 'CTest'
+    testResultsFiles: 'build/Testing/*/Test.xml'
+    testRunTitle: $(system.phasedisplayname)  CTest Test Run

rdkit/source/.azure-pipelines/vs_build_swig.yml

@@ -0,0 +1,69 @@
+steps:
+- task: UseDotNet@2
+  displayName: 'Install .NET Core 8 SDK'
+  inputs:
+    version: 8.x
+    performMultiLevelLookup: true
+- powershell: Write-Host "##vso[task.prependpath]$env:CONDA\Scripts"
+  displayName: Setup build environment
+- script: |
+    conda config --set always_yes yes --set changeps1 no
+    conda info -a
+    conda install -n base conda-libmamba-solver
+    conda config --set solver libmamba
+    conda create --name rdkit_build -c conda-forge --override-channels  ^
+        cmake=3.26 ^
+        libboost=$(boost_version) ^
+        libboost-devel=$(boost_version) ^
+        cairo eigen swig=4.2
+    call activate rdkit_build
+    conda config --env --add channels conda-forge
+    conda config --env --set channel_priority strict
+  displayName: Install dependencies
+- script: |
+    set Boost_ROOT=
+    mkdir build && cd build
+    call activate rdkit_build
+    cmake .. ^
+    -G "Visual Studio 17 2022" ^
+    -DCMAKE_BUILD_TYPE=Release ^
+    -DRDK_INSTALL_INTREE=ON ^
+    -DRDK_INSTALL_STATIC_LIBS=OFF ^
+    -DRDK_BUILD_CPP_TESTS=OFF ^
+    -DRDK_BUILD_PYTHON_WRAPPERS=OFF ^
+    -DRDK_BUILD_COORDGEN_SUPPORT=ON ^
+    -DRDK_BUILD_MAEPARSER_SUPPORT=ON ^
+    -DRDK_OPTIMIZE_POPCNT=ON ^
+    -DRDK_BUILD_TEST_GZIP=OFF ^
+    -DRDK_BUILD_FREESASA_SUPPORT=OFF ^
+    -DRDK_BUILD_AVALON_SUPPORT=ON ^
+    -DRDK_BUILD_INCHI_SUPPORT=ON ^
+    -DRDK_BUILD_YAEHMOP_SUPPORT=ON ^
+    -DRDK_BUILD_XYZ2MOL_SUPPORT=ON ^
+    -DRDK_BUILD_CAIRO_SUPPORT=ON ^
+    -DRDK_BUILD_THREADSAFE_SSS=ON ^
+    -DRDK_BUILD_SWIG_WRAPPERS=ON ^
+    -DRDK_BUILD_SWIG_JAVA_WRAPPER=OFF ^
+    -DRDK_BUILD_SWIG_CSHARP_WRAPPER=ON ^
+    -DRDK_BUILD_CFFI_LIB=OFF ^
+    -DRDK_SWIG_STATIC=ON ^
+    -DRDK_TEST_MULTITHREADED=ON
+  displayName: Configure build (Run CMake)
+- script: |
+    call activate rdkit_build
+    cd build
+    cmake --build . --config Release --target install -j $(number_of_cores)
+  displayName: Build
+- script: |
+    call activate rdkit_build
+    set RDBASE=%cd%
+    set PYTHONPATH=%RDBASE%;%PYTHONPATH%
+    set PATH=%RDBASE%\lib;%PATH%
+    cd build
+    ctest -R CSharpTests -C Release -T test --output-on-failure --verbose
+  displayName: Run tests
+- task: PublishTestResults@2
+  inputs:
+    testResultsFormat: 'CTest'
+    testResultsFiles: 'build/Testing/*/Test.xml'
+    testRunTitle: $(system.phasedisplayname)  CTest Test Run

rdkit/source/.clang-format

@@ -0,0 +1,67 @@
+---
+Language:        Cpp
+# BasedOnStyle:  Google
+AccessModifierOffset: -1
+AlignAfterOpenBracket: true
+AlignEscapedNewlinesLeft: true
+AlignOperands:   true
+AlignTrailingComments: true
+AllowAllParametersOfDeclarationOnNextLine: true
+AllowShortBlocksOnASingleLine: false
+AllowShortCaseLabelsOnASingleLine: false
+AllowShortIfStatementsOnASingleLine: true
+AllowShortLoopsOnASingleLine: true
+AllowShortFunctionsOnASingleLine: All
+AllowShortEnumsOnASingleLine: false
+AlwaysBreakAfterDefinitionReturnType: false
+AlwaysBreakTemplateDeclarations: true
+AlwaysBreakBeforeMultilineStrings: true
+BreakBeforeBinaryOperators: None
+BreakBeforeTernaryOperators: true
+BreakConstructorInitializersBeforeComma: false
+BinPackParameters: true
+BinPackArguments: true
+ColumnLimit:     80
+ConstructorInitializerAllOnOneLineOrOnePerLine: true
+ConstructorInitializerIndentWidth: 4
+DerivePointerAlignment: false
+ExperimentalAutoDetectBinPacking: false
+IndentCaseLabels: true
+IndentWrappedFunctionNames: false
+IndentFunctionDeclarationAfterType: false
+MaxEmptyLinesToKeep: 1
+KeepEmptyLinesAtTheStartOfBlocks: false
+NamespaceIndentation: None
+ObjCBlockIndentWidth: 2
+ObjCSpaceAfterProperty: false
+ObjCSpaceBeforeProtocolList: false
+PenaltyBreakBeforeFirstCallParameter: 1
+PenaltyBreakComment: 300
+PenaltyBreakString: 1000
+PenaltyBreakFirstLessLess: 120
+PenaltyExcessCharacter: 1000000
+PenaltyReturnTypeOnItsOwnLine: 200
+PointerAlignment: Right
+SpacesBeforeTrailingComments: 2
+Cpp11BracedListStyle: true
+Standard:        Cpp11
+IndentWidth:     2
+TabWidth:        8
+UseTab:          Never
+BreakStringLiterals: false
+BreakBeforeBraces: Attach
+SpacesInParentheses: false
+SpacesInSquareBrackets: false
+SpacesInAngles:  false
+SpaceInEmptyParentheses: false
+SpacesInCStyleCastParentheses: false
+SpaceAfterCStyleCast: false
+SpacesInContainerLiterals: true
+SpaceBeforeAssignmentOperators: true
+ContinuationIndentWidth: 4
+CommentPragmas:  '^ IWYU pragma:'
+ForEachMacros:   [ foreach, Q_FOREACH, BOOST_FOREACH ]
+SpaceBeforeParens: ControlStatements
+SortIncludes:   false
+DisableFormat:   false
+...

rdkit/source/.clang-tidy

@@ -0,0 +1,18 @@
+---
+Checks:          '-*,boost-use-to-string,modernize-use-auto,modernize-use-nullptr,modernize-loop-convert,modernize-use-override,modernize-pass-by-value,modernize-shrink-to-fit,readability-braces-around-statements,readability-simplify-boolean-expr'
+HeaderFilterRegex: 'Code/'
+User:            glandrum
+CheckOptions:
+  - key:             modernize-loop-convert.MaxCopySize
+    value:           '16'
+  - key:             modernize-loop-convert.MinConfidence
+    value:           reasonable
+  - key:             modernize-loop-convert.NamingStyle
+    value:           CamelCase
+  - key:             modernize-pass-by-value.IncludeStyle
+    value:           llvm
+  - key:             modernize-replace-auto-ptr.IncludeStyle
+    value:           llvm
+  - key:             modernize-use-nullptr.NullMacros
+    value:           'NULL'
+...

rdkit/source/.travis.yml

@@ -0,0 +1,70 @@
+# adapted from http://conda.pydata.org/docs/travis.html
+language: generic
+
+# build matrices
+matrix:
+  include:
+  - os: linux
+    dist: xenial
+    env: CONDA_PYTHON_VERSION=3.6
+  - os: osx
+    osx_image: xcode8.3
+    env: CONDA_PYTHON_VERSION=3.6
+
+addons:
+      apt:
+        packages:
+         - swig
+
+before_install:
+ # download and install miniconda
+ - echo "*************************************************"
+ - echo "Building for $TRAVIS_OS_NAME"
+ - echo "         for Python $CONDA_PYTHON_VERSION"
+ - export PYMAJOR="3";
+ - if [[ "$TRAVIS_OS_NAME" == "linux" ]]; then wget http://repo.continuum.io/miniconda/Miniconda$PYMAJOR-4.5.11-Linux-x86_64.sh -O miniconda.sh; fi
+ - if [[ "$TRAVIS_OS_NAME" == "osx" ]]; then wget http://repo.continuum.io/miniconda/Miniconda$PYMAJOR-4.5.11-MacOSX-x86_64.sh -O miniconda.sh; fi
+ - bash miniconda.sh -b -p $HOME/conda
+ - export PATH="$HOME/conda/bin:$PATH"
+ - hash -r
+ - conda config --set always_yes yes --set changeps1 no
+ - conda update -q conda
+ # Useful for debugging any issues with conda
+ - conda info -a
+
+install:
+ - echo "INSTALL CONDA"
+ # (don't install conda compiler tools on linux -- this is gcc 4.8 right now)
+ #- if [[ "$TRAVIS_OS_NAME" == "linux" ]]; then conda install gcc_linux-64 gcc_linux-64 ; fi
+ #- if [[ "$TRAVIS_OS_NAME" == "linux" ]]; then conda install -q -c rdkit boost=1.63; fi
+ - if [[ "$TRAVIS_OS_NAME" == "linux" ]]; then conda install -q py-boost=1.67.0 libboost=1.67.0 ; fi
+ - if [[ "$TRAVIS_OS_NAME" == "osx" ]]; then conda install -q py-boost=1.65.1 libboost=1.65.1 ; fi
+ - conda install -q numpy pillow pandas cmake eigen # matplotlib
+
+before_script:
+ # RDKit
+ - echo "BEFORE SCRIPT"
+ - export RDBASE=`pwd`
+ - echo $RDBASE
+ - export PYTHONPATH=${RDBASE}
+ - export LD_LIBRARY_PATH=${RDBASE}/lib
+ - export DYLD_FALLBACK_LIBRARY_PATH=${RDBASE}/lib
+
+ - export PYTHON=`which python`
+ - echo $PYTHON
+ - export PY_PREFIX=`$PYTHON -c "import sys; print(sys.prefix)"`
+ - echo $PY_PREFIX
+ - export PY_SP_DIR=$PY_PREFIX/lib/python$CONDA_PYTHON_VERSION/site-packages
+ - echo $PY_SP_DIR
+
+script:
+  - echo "SCRIPT"
+  - cd $RDBASE
+  - mkdir build
+  - cd build
+  - if [[ "$TRAVIS_OS_NAME" == "linux" ]]; then  export CXX="g++" ; fi
+  - if [[ "$TRAVIS_OS_NAME" == "linux" ]]; then  export CC="gcc" ; fi
+  - cmake $REGEX_EXTRA -D PYTHON_EXECUTABLE=$PYTHON -D BOOST_ROOT=$PY_PREFIX -D Boost_NO_SYSTEM_PATHS=ON -D RDK_BUILD_AVALON_SUPPORT=ON -D RDK_BUILD_INCHI_SUPPORT=ON -DRDK_BUILD_THREADSAFE_SSS=on -DRDK_TEST_MULTITHREADED=on -DRDK_INSTALL_STATIC_LIBS=OFF -DRDK_BUILD_SWIG_WRAPPERS=OFF -DRDK_SWIG_STATIC=OFF -DRDK_BUILD_PYTHON_WRAPPERS=ON -DRDK_BUILD_FREESASA_SUPPORT=ON ..
+  - make -j2 install
+  - ls "$PY_PREFIX/lib"
+  - LD_LIBRARY_PATH="$PY_PREFIX/lib:$PREFIX/lib;$SRC_DIR/lib;$LD_LIBRARY_PATH" ctest -j2 --output-on-failure

rdkit/source/CMakeLists.txt

@@ -0,0 +1,713 @@
+cmake_minimum_required(VERSION 3.18)
+
+project (RDKit)
+
+IF(NOT CMAKE_BUILD_TYPE)
+  SET(CMAKE_BUILD_TYPE Release CACHE STRING
+      "Choose the type of build, options are: None Debug Release RelWithDebInfo MinSizeRel."
+      FORCE)
+ENDIF(NOT CMAKE_BUILD_TYPE)
+
+# Enable CTest
+include(CTest)
+
+set(CMAKE_CXX_STANDARD 20)
+set(CMAKE_CXX_STANDARD_REQUIRED ON)
+set(CMAKE_C_STANDARD 99)
+set(CMAKE_C_STANDARD_REQUIRED ON)
+
+set(CMAKE_MODULE_PATH
+    ${CMAKE_MODULE_PATH} "${CMAKE_CURRENT_SOURCE_DIR}/Code/cmake/Modules/")
+
+add_library(rdkit_base INTERFACE)
+
+option(RDK_BUILD_SWIG_WRAPPERS "build the SWIG wrappers" OFF )
+option(RDK_BUILD_PYTHON_WRAPPERS "build the standard python wrappers" ON )
+option(RDK_BUILD_COMPRESSED_SUPPLIERS "build in support for compressed MolSuppliers" OFF )
+option(RDK_BUILD_INCHI_SUPPORT "build the rdkit inchi wrapper" OFF )
+option(RDK_BUILD_AVALON_SUPPORT "install support for the avalon toolkit. Use the variable AVALONTOOLS_DIR to set the location of the source." OFF )
+option(RDK_BUILD_PGSQL "build the PostgreSQL cartridge" OFF )
+option(RDK_BUILD_RPATH_SUPPORT "build shared libraries using rpath" OFF)
+option(RDK_PGSQL_STATIC "statically link rdkit libraries into the PostgreSQL cartridge" ON )
+option(RDK_PGSQL_BFP_GIST_SORTSUPPORT "enable the sortsupport method in the PostgreSQL bfp gist index" OFF )
+option(RDK_PGSQL_MOL_GIST_SORTSUPPORT "enable the sortsupport method in the PostgreSQL mol gist index" ON )
+option(RDK_PGSQL_QMOL_GIST_SORTSUPPORT "enable the sortsupport method in the PostgreSQL qmol gist index" ON )
+option(RDK_BUILD_CONTRIB "build the Contrib directory" OFF )
+option(RDK_INSTALL_INTREE "install the rdkit in the source tree (former behavior)" ON )
+option(RDK_INSTALL_DLLS_MSVC "install the rdkit DLLs when using MSVC" OFF)
+option(RDK_INSTALL_STATIC_LIBS "install the rdkit static libraries" ON )
+option(RDK_INSTALL_PYTHON_TESTS "install the rdkit Python tests with the wrappers" OFF )
+option(RDK_BUILD_THREADSAFE_SSS "enable thread-safe substructure searching" ON )
+option(RDK_BUILD_SLN_SUPPORT "include support for the SLN format" ON )
+option(RDK_TEST_MULTITHREADED "run some tests of multithreading" ON )
+option(RDK_BUILD_SWIG_JAVA_WRAPPER "build the SWIG JAVA wrappers (does nothing if RDK_BUILD_SWIG_WRAPPERS is not set)" ON )
+option(RDK_BUILD_SWIG_CSHARP_WRAPPER "build the experimental SWIG C# wrappers (does nothing if RDK_BUILD_SWIG_WRAPPERS is not set)" OFF )
+option(RDK_BUILD_DOTNET_CSHARP_TESTS "build the DotNet Core tests for C# wrappers on Linux (does nothing if RDK_BUILD_SWIG_CSHARP_WRAPPER is not set and is ignored on Windows)" OFF )
+option(RDK_SWIG_STATIC "statically link rdkit libraries into the SWIG wrappers" ON )
+option(RDK_TEST_MMFF_COMPLIANCE "run MMFF compliance tests (requires tar/gzip)" ON )
+option(RDK_BUILD_CPP_TESTS "build the c++ tests (disabing can speed up builds" ON)
+option(RDK_USE_FLEXBISON "use flex/bison, if available, to build the SMILES/SMARTS/SLN parsers" OFF)
+option(RDK_TEST_COVERAGE "Use G(L)COV to compute test coverage" OFF)
+option(RDK_USE_BOOST_SERIALIZATION "Use the boost serialization library if available" ON)
+option(RDK_USE_BOOST_STACKTRACE "use boost::stacktrace to do more verbose invariant output (linux only)" ON)
+option(RDK_BUILD_TEST_GZIP "Build the gzip'd stream test" OFF)
+option(RDK_OPTIMIZE_POPCNT "Use SSE4.2 popcount instruction while compiling." ON)
+option(RDK_USE_STRICT_ROTOR_DEFINITION "Use the most strict rotatable bond definition" ON)
+option(RDK_BUILD_DESCRIPTORS3D "Build the 3D descriptors calculators, requires Eigen3 to be installed" ON)
+option(RDK_BUILD_CHEMDRAW_SUPPORT "build support for the Revvity ChemDraw document format" ON )
+option(RDK_BUILD_FREESASA_SUPPORT "build the rdkit freesasa wrapper" OFF )
+option(RDK_BUILD_COORDGEN_SUPPORT "build the rdkit coordgen wrapper" ON )
+option(RDK_BUILD_MAEPARSER_SUPPORT "build the rdkit MAE parser wrapper" ON )
+option(RDK_BUILD_MOLINTERCHANGE_SUPPORT "build in support for CommonChem molecule interchange" ON )
+option(RDK_BUILD_YAEHMOP_SUPPORT "build support for the YAeHMOP wrapper" OFF)
+option(RDK_BUILD_XYZ2MOL_SUPPORT "build in support for the RDKit's implementation of xyz2mol (in the DetermineBonds library)" OFF )
+option(RDK_BUILD_STRUCTCHECKER_SUPPORT "build in support for the StructChecker alpha (not recommended, use the MolVS integration instead)" OFF )
+option(RDK_BUILD_PUBCHEMSHAPE_SUPPORT "build the rdkit wrapper around pubchem-align3d" ON )
+option(RDK_USE_URF "Build support for Florian Flachsenberg's URF library" ON)
+option(RDK_INSTALL_DEV_COMPONENT "install libraries and headers" ON)
+option(RDK_USE_BOOST_IOSTREAMS "use boost::iostreams" ON)
+option(RDK_BUILD_MINIMAL_LIB "build the minimal RDKit wrapper (for the JS bindings)" OFF)
+option(RDK_MINIMAL_LIB_SUPPORT_LEGACY_BROWSERS "build the minimal RDKit JS wrapper such that it supports legacy browsers" OFF)
+option(RDK_BUILD_CFFI_LIB "build the CFFI wrapper (for use in other programming languges)" OFF)
+option(RDK_CFFI_STATIC "statically link rdkit libraries into the cffi wrapper" ON)
+option(RDK_BUILD_FUZZ_TARGETS "build the fuzz targets" OFF)
+option(RDK_BUILD_MINIMAL_LIB_RXN "build support for reactions into MinimalLib" ON )
+option(RDK_BUILD_MINIMAL_LIB_SUBSTRUCTLIBRARY "build support for SubstructLibrary into MinimalLib" ON )
+option(RDK_BUILD_MINIMAL_LIB_MCS "build support for MCS into MinimalLib" OFF )
+option(RDK_BUILD_MINIMAL_LIB_MOLZIP "build support for molzip into MinimalLib" OFF )
+option(RDK_BUILD_MINIMAL_LIB_MMPA "build support for MMPA into MinimalLib" OFF )
+option(RDK_BUILD_LONG_RUNNING_TESTS "build longer running tests" OFF )
+
+set(RDK_BOOST_VERSION "1.81.0")
+
+if(NOT MSVC AND RDK_OPTIMIZE_POPCNT)
+  if(CMAKE_SYSTEM_PROCESSOR MATCHES "^(ppc64|powerpc64)")
+    set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -mpopcntd")
+  elseif(CMAKE_SYSTEM_PROCESSOR MATCHES "^(x86_64|i*86)")
+    set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -mpopcnt")
+  else()
+    message(STATUS "POPCNT hardware flag not supported or not required for ${CMAKE_SYSTEM_PROCESSOR} architecture")
+  endif()
+endif()
+
+# Detect clang, which masquerades as gcc.  CMake 2.6 doesn't know how to
+# detect it.
+string(REGEX MATCH "clang" CMAKE_COMPILER_IS_CLANG "${CMAKE_C_COMPILER}")
+# Detect emcc
+string(REGEX MATCH "Emscripten" CMAKE_COMPILER_IS_EMCC "${CMAKE_SYSTEM_NAME}")
+
+if(NOT UNIX OR APPLE OR CYGWIN OR CMAKE_COMPILER_IS_EMCC)
+  if(RDK_USE_BOOST_STACKTRACE)
+    MESSAGE("Disabling boost::stacktrace on non-linux and emscripten platforms")
+    set(RDK_USE_BOOST_STACKTRACE OFF)
+  endif(RDK_USE_BOOST_STACKTRACE)
+endif()
+
+if(NOT RDK_BUILD_SWIG_WRAPPERS)
+  set(RDK_BUILD_SWIG_JAVA_WRAPPER OFF)
+  set(RDK_BUILD_SWIG_CSHARP_WRAPPER OFF)
+endif()
+
+if(RDK_SWIG_STATIC AND RDK_BUILD_SWIG_WRAPPERS)
+  if(NOT MSVC AND NOT RDK_INSTALL_STATIC_LIBS)
+    message("Enabling RDK_INSTALL_STATIC_LIBS because RDK_SWIG_STATIC is set.")
+    set(RDK_INSTALL_STATIC_LIBS ON CACHE BOOL "install the rdkit static libraries" FORCE)
+  endif(NOT MSVC AND NOT RDK_INSTALL_STATIC_LIBS)
+endif()
+if ((MSVC AND (NOT RDK_INSTALL_DLLS_MSVC)) OR (WIN32 AND RDK_INSTALL_STATIC_LIBS) OR RDK_BUILD_MINIMAL_LIB)
+  set(RDK_BUILD_STATIC_LIBS_ONLY ON)
+else()
+  set(RDK_BUILD_STATIC_LIBS_ONLY OFF)
+endif()
+if(NOT RDK_BUILD_STATIC_LIBS_ONLY)
+  add_definitions(-DRDKIT_DYN_LINK)
+endif()
+
+if(RDK_PGSQL_STATIC AND RDK_BUILD_PGSQL)
+  if(NOT MSVC AND NOT RDK_INSTALL_STATIC_LIBS)
+    message("Enabling RDK_INSTALL_STATIC_LIBS because RDK_PGSQL_STATIC is set.")
+    set(RDK_INSTALL_STATIC_LIBS ON CACHE BOOL "install the rdkit static libraries" FORCE)
+  endif(NOT MSVC AND NOT RDK_INSTALL_STATIC_LIBS)
+endif()
+
+if(RDK_CFFI_STATIC AND RDK_BUILD_CFFI_LIB)
+  if(NOT MSVC AND NOT RDK_INSTALL_STATIC_LIBS)
+    message("Enabling RDK_INSTALL_STATIC_LIBS because RDK_CFFI_STATIC is set.")
+    set(RDK_INSTALL_STATIC_LIBS ON CACHE BOOL "install the rdkit static libraries" FORCE)
+  endif(NOT MSVC AND NOT RDK_INSTALL_STATIC_LIBS)
+endif()
+
+
+include(TestBigEndian)
+TEST_BIG_ENDIAN(RDK_BIG_ENDIAN)
+
+# At build time put runtime binaries in the bin subdirectory
+set(RDK_RUNTIME_OUTPUT_DIRECTORY "${CMAKE_CURRENT_BINARY_DIR}/bin")
+set(RDK_LIBRARY_OUTPUT_DIRECTORY "${CMAKE_CURRENT_BINARY_DIR}/lib")
+set(RDK_ARCHIVE_OUTPUT_DIRECTORY "${CMAKE_CURRENT_BINARY_DIR}/lib")
+# Also place the python extension where the source tree would expect them
+set(RDK_PYTHON_OUTPUT_DIRECTORY "${CMAKE_CURRENT_BINARY_DIR}/rdkit")
+
+#-------
+# Config variables:
+set(RDKit_Year "2026")
+set(RDKit_Month "03")
+set(RDKit_Revision "1")
+set(RDKit_RevisionModifier "pre")
+set(RDKit_ABI "1")
+
+# we need an integer version of the month later, so remove the zero padding
+# if there is any
+string(REGEX REPLACE "^0" "" RDKit_intMonth ${RDKit_Month} )
+
+set(RDKit_CodeDir "${CMAKE_CURRENT_SOURCE_DIR}/Code")
+set(RDKit_ExternalDir "${CMAKE_CURRENT_SOURCE_DIR}/External")
+set(RDKit_DataDir "${CMAKE_CURRENT_SOURCE_DIR}/Data")
+
+#include catch
+find_package(Catch2 3 QUIET)
+if(NOT Catch2_FOUND)
+  Include(FetchContent)
+
+  FetchContent_Declare(
+    Catch2
+    GIT_REPOSITORY https://github.com/catchorg/Catch2.git
+    GIT_TAG        v3.4.0 # or a later release
+  )
+
+  FetchContent_MakeAvailable(Catch2)
+endif()
+
+# make sure we have better_enums
+Include(FetchContent)
+
+FetchContent_Declare(
+  better_enums
+  GIT_REPOSITORY https://github.com/aantron/better-enums.git
+  GIT_TAG        c35576bed0295689540b39873126129adfa0b4c8 # 0.11.3
+)
+
+if(RDK_INSTALL_INTREE)
+  set(RDKit_BinDir "${CMAKE_SOURCE_DIR}/bin")
+  set(RDKit_LibDir "${CMAKE_SOURCE_DIR}/lib")
+  set(RDKit_HdrDir "Code") # used in rdkit-config.cmake, path prefix not needed.
+  set(RDKit_ShareDir "${CMAKE_SOURCE_DIR}")
+else(RDK_INSTALL_INTREE)
+  set(RDKit_BinDir "bin")
+  set(RDKit_LibDir "lib${LIB_SUFFIX}")
+  set(RDKit_HdrDir "include/rdkit")
+  set(RDKit_ShareDir "share/RDKit")
+endif(RDK_INSTALL_INTREE)
+
+if(RDK_BUILD_RPATH_SUPPORT)
+  # use, i.e. don't skip the full RPATH for the build tree
+  SET(CMAKE_SKIP_BUILD_RPATH  FALSE)
+
+  # when building, don't use the install RPATH already
+  # (but later on when installing)
+  SET(CMAKE_BUILD_WITH_INSTALL_RPATH FALSE)
+
+  message("CMAKE_INSTALL_RPATH ${CMAKE_INSTALL_PREFIX}/lib")
+  SET(CMAKE_INSTALL_RPATH "${CMAKE_INSTALL_PREFIX}/lib")
+
+  # add the automatically determined parts of the RPATH
+  # which point to directories outside the build tree to the install RPATH
+  SET(CMAKE_INSTALL_RPATH_USE_LINK_PATH TRUE)
+
+  # the RPATH to be used when installing, but only if it's not a system
+  # directory
+  LIST(FIND CMAKE_PLATFORM_IMPLICIT_LINK_DIRECTORIES
+  "${CMAKE_INSTALL_PREFIX}/lib" isSystemDir)
+  IF("${isSystemDir}" STREQUAL "-1")
+     SET(CMAKE_INSTALL_RPATH "${CMAKE_INSTALL_PREFIX}/lib")
+  ENDIF("${isSystemDir}" STREQUAL "-1")
+endif()
+
+
+
+if(RDK_BUILD_SWIG_WRAPPERS)
+  set(RDKit_JavaLibDir "${RDKit_ExternalDir}/java_lib")
+endif(RDK_BUILD_SWIG_WRAPPERS)
+
+
+if(RDK_BUILD_COORDGEN_SUPPORT)
+    add_definitions(-DRDK_BUILD_COORDGEN_SUPPORT)
+    include_directories(${RDKit_ExternalDir})
+endif()
+
+if(RDK_BUILD_MAEPARSER_SUPPORT)
+    add_definitions(-DRDK_BUILD_MAEPARSER_SUPPORT)
+    include_directories(${RDKit_ExternalDir})
+endif()
+
+if(RDK_USE_URF)
+  add_definitions(-DRDK_USE_URF)
+  if((MSVC AND (NOT RDK_INSTALL_DLLS_MSVC)) OR ((NOT MSVC) AND WIN32 AND RDK_INSTALL_STATIC_LIBS))
+    add_definitions(-DRDL_WIN_STATIC)
+  endif()
+  include_directories(${RDKit_ExternalDir}/RingFamilies/RingDecomposerLib/src/RingDecomposerLib)
+endif()
+
+if(RDK_BUILD_XYZ2MOL_SUPPORT)
+    add_definitions(-DRDK_BUILD_XYZ2MOL_SUPPORT)
+    include_directories(${RDKit_ExternalDir})
+endif()
+
+if(RDK_BUILD_YAEHMOP_SUPPORT)
+    add_definitions(-DRDK_BUILD_YAEHMOP_SUPPORT)
+endif()
+
+if(CMAKE_SIZEOF_VOID_P EQUAL 4)
+  target_compile_definitions(rdkit_base INTERFACE "-DRDK_32BIT_BUILD")
+else()
+  target_compile_definitions(rdkit_base INTERFACE "-DRDK_64BIT_BUILD")
+endif()
+
+if(MINGW)
+  target_compile_definitions(rdkit_base INTERFACE "-DBOOST_SYSTEM_NO_DEPRECATED")
+endif(MINGW)
+# fallback on match to "Clang" where CMAKE_C_COMPILER=/usr/bin/cc
+if (CMAKE_COMPILER_IS_CLANG OR CMAKE_CXX_COMPILER_ID MATCHES "Clang" OR CMAKE_COMPILER_IS_EMCC)
+  if (CMAKE_CXX_COMPILER_VERSION VERSION_GREATER_EQUAL 15)
+    target_compile_definitions(rdkit_base INTERFACE -DBOOST_SYSTEM_NO_DEPRECATED -DBOOST_NO_CXX98_FUNCTION_BASE -D_HAS_AUTO_PTR_ETC=0)
+    target_compile_options(rdkit_base INTERFACE -Wno-deprecated-builtins -Wno-deprecated-non-prototype -Wno-unused-parameter)
+  endif()
+endif()
+
+# defines macros: rdkit_python_extension, rdkit_test
+include(RDKitUtils)
+
+install(TARGETS rdkit_base EXPORT ${RDKit_EXPORTED_TARGETS}
+        COMPONENT dev )
+
+# disable some warnings that we don't care about
+if(MSVC)
+  # These should occasionally be enabled to check what's going on
+  # 4267 conversion with possible loss of data.
+  # 4305 truncation from 'double' to 'const float'
+  # 4244 conversion from 'uint64_t' to 'unsigned int', possible loss of data
+  target_compile_options(rdkit_base INTERFACE "/wd4267" "/wd4305" "/wd4244")
+  target_compile_definitions(rdkit_base INTERFACE "_CRT_SECURE_NO_WARNINGS")
+endif(MSVC)
+
+
+# extra boost versions
+if(MSVC)
+  # FIX: do we still need this?
+  target_compile_definitions(rdkit_base INTERFACE "-DBOOST_ALL_NO_LIB")
+  if(RDK_INSTALL_DLLS_MSVC)
+    target_compile_definitions(rdkit_base INTERFACE "-DBOOST_ALL_DYN_LINK")
+  endif(RDK_INSTALL_DLLS_MSVC)
+endif(MSVC)
+
+if(RDK_BUILD_INCHI_SUPPORT)
+  find_package(Inchi)
+endif(RDK_BUILD_INCHI_SUPPORT)
+
+if(RDK_BUILD_PYTHON_WRAPPERS)
+  add_library(rdkit_py_base INTERFACE)
+
+  #-------
+  # pull in python:
+  find_package(Python3 COMPONENTS Interpreter Development.Module NumPy)
+  target_include_directories(rdkit_py_base INTERFACE ${Python3_INCLUDE_DIRS})
+  target_include_directories(rdkit_py_base INTERFACE ${Python3_NumPy_INCLUDE_DIRS})
+
+  # determine linkage of python
+  execute_process(
+      COMMAND
+      ${Python3_EXECUTABLE} -c "import sysconfig; print(sysconfig.get_config_var('Py_ENABLE_SHARED'))"
+      OUTPUT_VARIABLE Py_ENABLE_SHARED
+      OUTPUT_STRIP_TRAILING_WHITESPACE
+  )
+
+  if(WIN32 OR "${Py_ENABLE_SHARED}" STREQUAL "1")
+    target_link_libraries(rdkit_py_base INTERFACE ${Python3_LIBRARIES} )
+  endif()
+
+
+  find_package(Boost ${RDK_BOOST_VERSION} COMPONENTS "python${Python3_VERSION_MAJOR}${Python3_VERSION_MINOR}" "numpy${Python3_VERSION_MAJOR}${Python3_VERSION_MINOR}" REQUIRED CONFIG)
+
+  target_link_libraries(rdkit_py_base INTERFACE "Boost::python${Python3_VERSION_MAJOR}${Python3_VERSION_MINOR}" "Boost::numpy${Python3_VERSION_MAJOR}${Python3_VERSION_MINOR}")
+
+  if(RDK_INSTALL_INTREE)
+    set(RDKit_PythonDir "${CMAKE_SOURCE_DIR}/rdkit")
+  else(RDK_INSTALL_INTREE)
+    if (NOT PYTHON_INSTDIR)
+      # Determine correct installation directory for Python bindings
+      execute_process(
+        COMMAND
+        ${Python3_EXECUTABLE} -c "import sys; import sysconfig; \
+          base_key = 'base' if sys.platform == 'win32' else 'platbase'; \
+          schema = 'nt' if sys.platform == 'win32' else 'posix_prefix'; \
+          print(sysconfig.get_path('platlib', schema, vars={base_key: '${CMAKE_INSTALL_PREFIX}'}))"
+        OUTPUT_VARIABLE PYTHON_INSTDIR
+        OUTPUT_STRIP_TRAILING_WHITESPACE
+      )
+    endif (NOT PYTHON_INSTDIR)
+    string(REGEX REPLACE "\\\\" "/" PYTHON_INSTDIR ${PYTHON_INSTDIR})
+    message("Python Install directory ${PYTHON_INSTDIR}")
+    set(RDKit_PythonDir "${PYTHON_INSTDIR}/rdkit")
+    if(RDK_INSTALL_PYTHON_TESTS)
+      install(DIRECTORY rdkit DESTINATION ${PYTHON_INSTDIR}
+        COMPONENT python
+        PATTERN ".svn" EXCLUDE
+        PATTERN "CMake*" EXCLUDE
+        PATTERN "Basement" EXCLUDE
+      )
+    else(RDK_INSTALL_PYTHON_TESTS)
+      install(DIRECTORY rdkit DESTINATION ${PYTHON_INSTDIR}
+        COMPONENT python
+        PATTERN ".svn" EXCLUDE
+        PATTERN "test_data" EXCLUDE
+        PATTERN "testData" EXCLUDE
+        PATTERN "test_list*" EXCLUDE
+        PATTERN "CMake*" EXCLUDE
+        PATTERN "Basement" EXCLUDE
+        PATTERN "UnitTest*" EXCLUDE
+      )
+    endif(RDK_INSTALL_PYTHON_TESTS)
+
+  endif(RDK_INSTALL_INTREE)
+
+  if(NOT WIN32)
+    # See https://bugs.python.org/msg277944
+    # The "command to create shared modules". Used as variable in the "Makefile (and similar) templates to build python modules"
+    # for both in-python and third party modules. Initialized to hold the value which works for third party modules to link
+    # against the _installed_ python.
+    # We strip off the first word though (which will be the compiler name).
+    execute_process(
+        COMMAND
+        ${Python3_EXECUTABLE} -c "import sysconfig; print(sysconfig.get_config_var('LDSHARED').lstrip().split(' ', 1)[1])"
+        OUTPUT_VARIABLE PYTHON_LDSHARED
+        OUTPUT_STRIP_TRAILING_WHITESPACE
+     )
+  endif()
+
+
+  install(TARGETS rdkit_py_base EXPORT ${RDKitPython_EXPORTED_TARGETS}
+          COMPONENT dev )
+
+  # check to see if we can find nbval,
+  execute_process(
+    COMMAND
+    ${Python3_EXECUTABLE} -c "import nbval"
+    ERROR_VARIABLE nbvalERR
+    OUTPUT_STRIP_TRAILING_WHITESPACE)
+  if(NOT nbvalERR)
+    set(RDK_NBVAL_AVAILABLE ON)
+  else()
+    message("nbval not found, disabling the jupyter tests")
+  endif()
+
+else(RDK_BUILD_PYTHON_WRAPPERS)
+  find_package(Boost ${RDK_BOOST_VERSION} REQUIRED CONFIG)
+endif(RDK_BUILD_PYTHON_WRAPPERS)
+
+find_package(Eigen3)
+if(RDK_BUILD_DESCRIPTORS3D)
+  if(NOT EIGEN3_FOUND)
+    add_subdirectory(External/Eigen)
+  endif()
+  target_compile_definitions(rdkit_base INTERFACE "-DRDK_BUILD_DESCRIPTORS3D")
+endif()
+
+if(TARGET Eigen3::Eigen)
+  target_compile_definitions(rdkit_base INTERFACE "-DRDK_HAS_EIGEN3")
+  target_link_libraries(rdkit_base INTERFACE Eigen3::Eigen)
+  set(RDK_HAS_EIGEN ON)
+endif()
+
+if(RDK_BUILD_THREADSAFE_SSS)
+  find_package (Threads)
+  set(RDKit_THREAD_LIBS Threads::Threads)
+  target_compile_definitions(rdkit_base INTERFACE -DRDK_BUILD_THREADSAFE_SSS)
+  if(RDK_TEST_MULTITHREADED)
+    target_compile_definitions(rdkit_base INTERFACE "-DRDK_TEST_MULTITHREADED")
+  endif()
+  target_link_libraries(rdkit_base INTERFACE Threads::Threads)
+else()
+  # otherwise boost flyweight links against pthreads and causes the
+  # RDK_BUILD_THREADSAFE_SSS=OFF build to fail
+  target_compile_definitions(rdkit_base INTERFACE "-DBOOST_DISABLE_THREADS")
+  if(RDK_TEST_MULTITHREADED)
+    message("RDK_TEST_MULTITHREADED does not make sense without RDK_BUILD_THREADSAFE_SSS, disabling it.")
+    set(RDK_TEST_MULTITHREADED OFF)
+  endif()
+endif()
+
+if(RDK_USE_BOOST_SERIALIZATION)
+    find_package(Boost ${RDK_BOOST_VERSION} COMPONENTS serialization iostreams REQUIRED CONFIG)
+    target_link_libraries(rdkit_base INTERFACE ${Boost_LIBRARIES})
+    target_compile_definitions(rdkit_base INTERFACE -DRDK_USE_BOOST_SERIALIZATION)
+    if(NOT Boost_USE_STATIC_LIBS)
+      target_compile_definitions(rdkit_base INTERFACE -DBOOST_SERIALIZATION_DYN_LINK)
+    endif()
+endif()
+
+if(RDK_USE_BOOST_IOSTREAMS)
+    target_compile_definitions(rdkit_base INTERFACE -DRDK_USE_BOOST_IOSTREAMS)
+    find_package(Boost ${RDK_BOOST_VERSION} COMPONENTS iostreams REQUIRED CONFIG)
+    target_link_libraries(rdkit_base INTERFACE ${Boost_LIBRARIES})
+
+    if (NOT Boost_USE_STATIC_LIBS)
+      target_compile_definitions(rdkit_base INTERFACE -DBOOST_IOSTREAMS_DYN_LINK)
+    endif()
+
+    # deal with zlib
+    if (WIN32)
+      find_package(Boost ${RDK_BOOST_VERSION} COMPONENTS zlib CONFIG)
+      if(Boost_zlib_FOUND)
+        set(zlib_lib Boost::zlib)
+      endif()
+    else()
+      if(Boost_USE_STATIC_LIBS)
+        # when we're doing static linkage of boost
+        # to also link against zlib (due to iostreams)
+        find_package(ZLIB)
+        set(zlib_lib ${ZLIB_LIBRARIES})
+     endif()
+    endif()
+    target_link_libraries(rdkit_base INTERFACE ${zlib_lib})
+endif()
+
+
+# set the boost include directories and linkage:
+target_include_directories(rdkit_base INTERFACE
+                           $<BUILD_INTERFACE:${RDKit_CodeDir}>
+                           $<INSTALL_INTERFACE:${RDKit_HdrDir}>
+                           )
+# linking against Boost::boost will add Boost to the include directories without
+# hard-coding Boost's path into the rdkit-targets.cmake that gets generated
+target_link_libraries(rdkit_base INTERFACE ${BOOST_LIBRARIES} Boost::boost)
+
+
+if(RDK_BUILD_PGSQL)
+  find_package(PostgreSQL REQUIRED)
+endif(RDK_BUILD_PGSQL)
+
+# setup our compiler flags:
+if (RDK_TEST_COVERAGE)
+ if (NOT RDK_USE_FLEXBISON)
+   message(FATAL_ERROR, "Test coverage doesn't current work unless FLEX and BISON are run.")
+ endif(NOT RDK_USE_FLEXBISON)
+
+ message("====== Installing test coverage support ======")
+ message("  To run:")
+ message("   make install")
+ message("   make RDKit_coverage")
+ message("")
+ message(" open <build_dir>/coverage/index.html")
+ message("")
+ message(" If any of the RDKit tests fail, coverage will probably not be generated.")
+ message("====== Installing test coverage support ======")
+ message("")
+
+ INCLUDE(CodeCoverage)
+ SET(CMAKE_CXX_FLAGS "-g -O0 -fprofile-arcs -ftest-coverage")
+ SET(CMAKE_C_FLAGS "-g -O0 -fprofile-arcs -ftest-coverage")
+ message("== setup_target_for_coverage(${PROJECT_NAME}_coverage test coverage)")
+ setup_target_for_coverage(${PROJECT_NAME}_coverage ctest coverage)
+
+else(RDK_TEST_COVERAGE)
+ if(CMAKE_COMPILER_IS_GNUCXX)
+   # We don't on C_FLAGS warnings to keep Avalon compiling quiet
+   SET(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -Wno-deprecated -Wno-unused-function -fno-strict-aliasing")
+   SET(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -Wno-deprecated -Wno-unused-function -fno-strict-aliasing -Wall -Wextra")
+  if (NOT CYGWIN)
+    SET(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -fPIC")
+    SET(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -fPIC")
+  endif()
+ endif()
+ if(CMAKE_CXX_COMPILER_ID MATCHES "Clang") # there's at least Clang and AppleClang
+   # We don't on C_FLAGS warnings to keep Avalon compiling quiet
+   SET(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -Wno-array-bounds -fPIC -Wno-parentheses -Wno-logical-op-parentheses -Wno-format")
+   SET(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -Wall -Wextra -Wno-deprecated -Wno-unused-function -fno-strict-aliasing -Wno-format -Wno-logical-op-parentheses -fPIC")
+   if(APPLE)
+     SET(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -stdlib=libc++")
+   endif()
+ endif()
+endif(RDK_TEST_COVERAGE)
+
+if(NOT RDK_INSTALL_INTREE)
+  install(DIRECTORY Data DESTINATION
+      ${RDKit_ShareDir}
+      COMPONENT data
+      PATTERN ".svn" EXCLUDE
+      )
+  install(DIRECTORY Docs DESTINATION
+      ${RDKit_ShareDir}
+      COMPONENT docs
+      PATTERN ".svn" EXCLUDE
+      PATTERN "_build" EXCLUDE
+      )
+  install(DIRECTORY Contrib DESTINATION
+      ${RDKit_ShareDir}
+      COMPONENT extras
+      PATTERN ".svn" EXCLUDE
+      )
+  if(RDK_BUILD_PYTHON_WRAPPERS)
+    install(DIRECTORY Projects DESTINATION
+        ${RDKit_ShareDir}
+        COMPONENT extras
+        PATTERN ".svn" EXCLUDE
+        PATTERN "test_list*" EXCLUDE
+        PATTERN "CMake*" EXCLUDE
+        )
+    if(RDK_INSTALL_PYTHON_TESTS)
+      install(DIRECTORY Scripts DESTINATION
+        ${RDKit_ShareDir}
+        COMPONENT data
+        PATTERN ".svn" EXCLUDE
+        )
+    else(RDK_INSTALL_PYTHON_TESTS)
+      install(DIRECTORY Scripts DESTINATION
+        ${RDKit_ShareDir}
+        COMPONENT data
+        PATTERN ".svn" EXCLUDE
+        PATTERN "run_python_tests.py" EXCLUDE
+        )
+    endif(RDK_INSTALL_PYTHON_TESTS)
+
+
+  endif(RDK_BUILD_PYTHON_WRAPPERS)
+  install(FILES README.md license.txt
+          DESTINATION ${RDKit_ShareDir}
+          COMPONENT base
+          )
+endif(NOT RDK_INSTALL_INTREE)
+
+if (RDK_SQUASH_MVC_SECURE_WARNINGS)
+  MESSAGE("== Squashing MSVC Security warnings (do not use during development)")
+  target_compile_definitions(rdkit_base INTERFACE "-D_SCL_SECURE_NO_WARNINGS")
+  target_compile_definitions(rdkit_base INTERFACE "/wd4996")
+endif(RDK_SQUASH_MVC_SECURE_WARNINGS)
+
+if(RDK_USE_STRICT_ROTOR_DEFINITION)
+    MESSAGE("== Using strict rotor definition")
+    target_compile_definitions(rdkit_base INTERFACE "-DRDK_USE_STRICT_ROTOR_DEFINITION")
+endif()
+
+if (MSVC)
+  # disable warnings:
+  #  - 4267: conversion from 'size_t' to 'unsigned int', possible loss of data
+  add_definitions( "/wd4267" )
+endif(MSVC)
+
+add_subdirectory(External)
+add_subdirectory(Code)
+
+if(RDK_BUILD_PYTHON_WRAPPERS)
+  add_subdirectory(Projects)
+  add_subdirectory(rdkit)
+  add_subdirectory(rdkit-stubs)
+endif(RDK_BUILD_PYTHON_WRAPPERS)
+
+if(RDK_BUILD_CONTRIB)
+  add_subdirectory(Contrib)
+endif(RDK_BUILD_CONTRIB)
+
+# export the project targets (to be included in the cmake package configuration)
+include(CMakePackageConfigHelpers)
+
+# first manage the targets for the base C++ toolkit
+install(
+  EXPORT ${RDKit_EXPORTED_TARGETS}
+  FILE ${RDKit_EXPORTED_TARGETS}.cmake
+  NAMESPACE RDKit::
+  DESTINATION ${RDKit_LibDir}/cmake/rdkit
+)
+
+# create and install package configuration and version files
+write_basic_package_version_file(
+  "${RDKit_BINARY_DIR}/rdkit-config-version.cmake"
+  VERSION ${RDKit_RELEASENAME}
+  COMPATIBILITY AnyNewerVersion
+)
+
+configure_file (
+    ${RDKit_SOURCE_DIR}/rdkit-config.cmake.in
+    ${RDKit_BINARY_DIR}/rdkit-config.cmake @ONLY)
+
+install(FILES
+    ${RDKit_BINARY_DIR}/rdkit-config.cmake
+    ${RDKit_BINARY_DIR}/rdkit-config-version.cmake
+    DESTINATION ${RDKit_LibDir}/cmake/rdkit)
+
+# then manage the targets for the python bindings
+if(RDK_BUILD_PYTHON_WRAPPERS)
+  install(
+    EXPORT ${RDKitPython_EXPORTED_TARGETS}
+    FILE ${RDKitPython_EXPORTED_TARGETS}.cmake
+    NAMESPACE RDKit::
+    DESTINATION ${RDKit_LibDir}/cmake/rdkitpython
+  )
+
+  write_basic_package_version_file(
+    "${RDKit_BINARY_DIR}/rdkitpython-config-version.cmake"
+    VERSION ${RDKit_RELEASENAME}
+    COMPATIBILITY AnyNewerVersion
+  )
+
+  configure_file (
+    ${RDKit_SOURCE_DIR}/rdkitpython-config.cmake.in
+    ${RDKit_BINARY_DIR}/rdkitpython-config.cmake @ONLY)
+
+  install(FILES
+    ${RDKit_BINARY_DIR}/rdkitpython-config.cmake
+    ${RDKit_BINARY_DIR}/rdkitpython-config-version.cmake
+    DESTINATION ${RDKit_LibDir}/cmake/rdkitpython)
+endif(RDK_BUILD_PYTHON_WRAPPERS)
+
+# Memory testing setup
+FIND_PROGRAM(MEMORYCHECK_COMMAND valgrind)
+CONFIGURE_FILE(CTestCustom.ctest.in ${RDKit_BINARY_DIR}/CTestCustom.ctest)
+
+# Packaging
+SET(CPACK_GENERATOR "TGZ;DEB;RPM")
+SET(CPACK_RPM_COMPONENT_INSTALL ON)
+SET(CPACK_DEB_COMPONENT_INSTALL ON)
+SET(CPACK_MONOLITHIC_INSTALL OFF)
+
+SET(CPACK_COMPONENTS_ALL runtime base data docs dev python extras)
+set(CPACK_COMPONENT_RUNTIME_GROUP "Runtime")
+set(CPACK_COMPONENT_BASE_GROUP "Runtime")
+set(CPACK_COMPONENT_DATA_GROUP "Runtime")
+set(CPACK_COMPONENT_DOCS_GROUP "Runtime")
+set(CPACK_COMPONENT_DEV_GROUP "Development")
+set(CPACK_COMPONENT_PYTHON_GROUP "Python")
+set(CPACK_COMPONENT_EXTRAS_GROUP "Extras")
+if(RDK_BUILD_PGSQL)
+  SET(CPACK_COMPONENTS_ALL ${CPACK_COMPONENTS_ALL} pgsql)
+  set(CPACK_COMPONENT_PGSQL_GROUP "PgSQL")
+endif()
+
+SET(CPACK_PACKAGE_DESCRIPTION_SUMMARY "RDKit")
+SET(CPACK_PACKAGE_VENDOR "rdkit.org")
+SET(CPACK_PACKAGE_CONTACT "greg.landrum@gmail.com")
+SET(CPACK_PACKAGE_DESCRIPTION_FILE "${CMAKE_CURRENT_SOURCE_DIR}/README.md")
+SET(CPACK_RESOURCE_FILE_LICENSE "${CMAKE_CURRENT_SOURCE_DIR}/license.txt")
+SET(CPACK_PACKAGE_VERSION_MAJOR ${RDKit_Year})
+SET(CPACK_PACKAGE_VERSION_MINOR ${RDKit_Month})
+SET(CPACK_PACKAGE_VERSION_PATCH ${RDKit_Revision}${RDKit_RevisionModifier})
+SET(CPACK_PACKAGE_INSTALL_DIRECTORY "CMake ${CMake_VERSION_MAJOR}.${CMake_VERSION_MINOR}")
+#IF(WIN32 AND NOT UNIX)
+#  # There is a bug in NSI that does not handle full unix paths properly. Make
+#  # sure there is at least one set of four (4) backlasshes.
+#  SET(CPACK_PACKAGE_ICON "${CMake_SOURCE_DIR}/Utilities/Release\\\\InstallIcon.bmp")
+#  SET(CPACK_NSIS_INSTALLED_ICON_NAME "bin\\\\MyExecutable.exe")
+#  SET(CPACK_NSIS_DISPLAY_NAME "${CPACK_PACKAGE_INSTALL_DIRECTORY} My Famous Project")
+#  SET(CPACK_NSIS_HELP_LINK "http:\\\\\\\\www.my-project-home-page.org")
+#  SET(CPACK_NSIS_URL_INFO_ABOUT "http:\\\\\\\\www.my-personal-home-page.com")
+#  SET(CPACK_NSIS_CONTACT "me@my-personal-home-page.com")
+#  SET(CPACK_NSIS_MODIFY_PATH ON)
+#ELSE(WIN32 AND NOT UNIX)
+#  SET(CPACK_STRIP_FILES "bin/MyExecutable")
+#  SET(CPACK_SOURCE_STRIP_FILES "")
+#ENDIF(WIN32 AND NOT UNIX)
+#SET(CPACK_PACKAGE_EXECUTABLES "MyExecutable" "My Executable")
+
+
+SET(CPACK_SET_DESTDIR ON)
+
+INCLUDE(CPack)

rdkit/source/CONTRIBUTING

@@ -0,0 +1 @@
+See Docs/Book/GettingStartedWithContributing.md

rdkit/source/CTestConfig.cmake

@@ -0,0 +1,2 @@
+# For some reason, MEMORYCHECK_SUPPRESSIONS_FILE is not being caught, so I hardcoded it here
+SET(MEMORYCHECK_COMMAND_OPTIONS "--tool=memcheck --error-exitcode=13 --time-stamp=yes --num-callers=20 --gen-suppressions=all --leak-check=full --show-reachable=no --trace-children=yes --suppressions=${RDKit_SOURCE_DIR}/Code/cmake/rdkit_valgrind.suppressions")

rdkit/source/CTestCustom.ctest.in

@@ -0,0 +1,11 @@
+# Require us to have the $RDBASE env var set to run the tests
+IF(NOT DEFINED ENV{RDBASE})
+  MESSAGE(FATAL_ERROR "\n\nPlease set your RDBASE env variable before running the tests.\n\n")
+ENDIF(NOT DEFINED ENV{RDBASE})
+
+SET(CTEST_CUSTOM_MEMCHECK_IGNORE
+  ${CTEST_CUSTOM_MEMCHECK_IGNORE}
+
+  # python tests are not memchecked: these are slow, and difficult to interpret
+  ${RDKIT_PYTEST_CACHE}
+)

rdkit/source/Code/Bench/CMakeLists.txt

@@ -0,0 +1,38 @@
+add_executable(bench EXCLUDE_FROM_ALL bench_common.cpp
+  descriptors.cpp
+  fingerprint.cpp
+  meta.cpp
+  mol.cpp
+  molops.cpp
+  pickle.cpp
+  smiles.cpp
+  stereo.cpp
+  substruct_match.cpp
+)
+target_link_libraries(bench rdkitCatch
+  CIPLabeler
+  Descriptors
+  Fingerprints
+  SmilesParse
+)
+
+if(RDK_BUILD_INCHI_SUPPORT)
+  target_link_libraries(bench RDInchiLib)
+  target_sources(bench PRIVATE inchi.cpp)
+endif(RDK_BUILD_INCHI_SUPPORT)
+
+if(RDK_BUILD_CPP_TESTS)
+  # add a fast version of the benchmarks to the default unit tests
+  # to protect the benchmarks from bit-rot
+  add_test(
+    NAME quickbench
+    COMMAND bench --benchmark-samples 1 --benchmark-warmup-time 0
+  )
+  
+  # ctest does not automatically build test executables
+  # so, like with other unit tests in this repo, we need to manually
+  # build the target before running ctest. This usually gets done with
+  # an `all` or an `install` build. We re-add bench to he `all` group
+  # to keep it similar to the other unit tests
+  set_target_properties(bench PROPERTIES EXCLUDE_FROM_ALL FALSE)
+endif(RDK_BUILD_CPP_TESTS)

rdkit/source/Code/Bench/README.md

@@ -0,0 +1,13 @@
+# RDKit Benchmarks
+
+To run:
+
+```bash
+mkdir build
+cd build
+cmake ..
+cmake --build . --target bench -j "$(nproc)"
+# see `./Code/Bench/bench --help` for options
+export RDBASE=".."
+./Code/Bench/bench
+```

rdkit/source/Code/Bench/bench_common.cpp

@@ -0,0 +1,20 @@
+#include <catch2/catch_all.hpp>
+
+#include <GraphMol/ROMol.h>
+#include <GraphMol/SmilesParse/SmilesParse.h>
+
+#include "bench_common.hpp"
+
+namespace bench_common {
+
+std::vector<RDKit::ROMol> load_samples() {
+  std::vector<RDKit::ROMol> ret;
+  for (auto smiles : SAMPLES) {
+    auto mol = RDKit::v2::SmilesParse::MolFromSmiles(smiles);
+    REQUIRE(mol);
+    ret.push_back(std::move(*mol));
+  }
+  return ret;
+}
+
+}  // namespace bench_common

rdkit/source/Code/Bench/bench_common.hpp

@@ -0,0 +1,53 @@
+#pragma once
+
+#include <cstdint>
+#include <limits>
+#include <vector>
+
+#include <GraphMol/ROMol.h>
+
+namespace bench_common {
+
+constexpr const char *SAMPLES[] = {
+    // highly fused/bridged ring system
+    "O=C1N2[C@H]3N4CN5C(=O)N6C7C5N(C2)C(=O)N7CN2C(=O)N5CN1[C@@H]3N(CN1C5C2N(C1=O)C6)C4=O",
+
+    // multiple stereo groups
+    "OC[C@]12[C@](O)(CC[C@H]3[C@]4(O)[C@@](C)([C@@H](C5COC(=O)C5)CC4)C[C@@H](O)[C@H]13)C[C@@H](O[C@@H]1O[C@@H](C)[C@H](O)[C@@H](O)[C@H]1O)C[C@H]2O",
+
+    // randomly selected
+    "C[C@H](NS(/C=C/c1ccccc1)(=O)=O)C(OCC(N1CCC(C)CC1)=O)=O",
+    "COc1ccc(-n2cc(CNC(C)c3ncncc3)c(-c3cc(F)ccc3)n2)cc1",
+    "O=C1N=C([n+]2ccccc2)/C(=C/[O-])N1c1ccccc1",
+    "c1csc(CNC(=O)C2CCC(CNS(c3ccccc3)(=O)=O)CC2)c1",
+    "CC1=C2C(c3ccc(C)cc3)C3=C(N=C2NN1)CC(C)(C)CC3=O",
+    "NC(N)=NN/C=C1\\C=CC=C([N+]([O-])=O)C1=O",
+    "CCc1c2c(oc(=O)c1)c(CN1CCN(C)CC1)c(O)c(Cl)c2",
+    "COc1ccc(C2CC(C(F)(F)F)N3NC=C(C(Nc4c(C)n(C)n(-c5ccccc5)c4=O)=O)C3=N2)cc1",
+    "Cc1ccc(CC(=O)O/N=C(\\N)Cc2ccc([N+]([O-])=O)cc2)cc1",
+    "Br.COc1ccc(/N=C/C=C2/OC(C)(C)OC(c3ccccc3)=C2)cc1",
+};
+
+std::vector<RDKit::ROMol> load_samples();
+
+constexpr uint64_t nth_random(uint64_t n) noexcept {
+  // https://xoshiro.di.unimi.it/splitmix64.c
+  // inlined here becuse <boost/random/splitmix64.hpp> is not always available
+  std::uint64_t seed = 1;
+  std::uint64_t z = seed + (n + 1) * 0x9E3779B97F4A7C15;
+  z = (z ^ (z >> 30)) * 0xBF58476D1CE4E5B9;
+  z = (z ^ (z >> 27)) * 0x94D049BB133111EB;
+  return z ^ (z >> 31);
+}
+
+static_assert(0x910a2dec89025cc1 == nth_random(0));
+static_assert(0xbeeb8da1658eec67 == nth_random(1));
+static_assert(0xf893a2eefb32555e == nth_random(2));
+static_assert(0x71c18690ee42c90b == nth_random(3));
+static_assert(0x71bb54d8d101b5b9 == nth_random(4));
+static_assert(0x7760003b54a685ae == nth_random(1000));
+static_assert(0x28a1928f0674d152 == nth_random(1000000000000));
+static_assert(0x5692161d100b05e5 ==
+              nth_random(std::numeric_limits<uint64_t>::max()));
+
+}  // namespace bench_common

rdkit/source/Code/Bench/descriptors.cpp

@@ -0,0 +1,31 @@
+#include <catch2/catch_all.hpp>
+#include <string>
+
+#include "bench_common.hpp"
+
+#include <GraphMol/SmilesParse/SmilesParse.h>
+#include <GraphMol/Descriptors/Lipinski.h>
+
+using namespace RDKit;
+
+TEST_CASE("Descriptors::calcNumSpiroAtoms", "[descriptors]") {
+  auto samples = bench_common::load_samples();
+  BENCHMARK("Descriptors::calcNumSpiroAtoms") {
+    auto sum = 0;
+    for (auto &mol : samples) {
+      sum += Descriptors::calcNumSpiroAtoms(mol);
+    }
+    return sum;
+  };
+}
+
+TEST_CASE("Descriptors::calcNumBridgeheadAtoms", "[descriptors]") {
+  auto samples = bench_common::load_samples();
+  BENCHMARK("Descriptors::calcNumBridgeheadAtoms") {
+    auto sum = 0;
+    for (auto &mol : samples) {
+      sum += Descriptors::calcNumBridgeheadAtoms(mol);
+    }
+    return sum;
+  };
+}

rdkit/source/Code/Bench/fingerprint.cpp

@@ -0,0 +1,25 @@
+#include <catch2/catch_all.hpp>
+#include <string>
+
+#include "bench_common.hpp"
+
+#include <GraphMol/SmilesParse/SmilesParse.h>
+#include <GraphMol/Fingerprints/MorganGenerator.h>
+
+using namespace RDKit;
+
+TEST_CASE("MorganFingerprints::getFingerprint", "[fingerprint]") {
+  auto samples = bench_common::load_samples();
+  const auto radius = 2;
+  std::unique_ptr<FingerprintGenerator<uint64_t>> gen(
+      MorganFingerprint::getMorganGenerator<uint64_t>(radius));
+
+  BENCHMARK("MorganFingerprints::getFingerprint") {
+    auto sum = 0;
+    for (auto &mol : samples) {
+      std::unique_ptr<ExplicitBitVect> fp(gen->getFingerprint(mol));
+      sum += fp->getNumOnBits();
+    }
+    return sum;
+  };
+}

rdkit/source/Code/Bench/inchi.cpp

@@ -0,0 +1,55 @@
+#include <catch2/catch_all.hpp>
+#include <string>
+
+#include "bench_common.hpp"
+
+#include <GraphMol/ROMol.h>
+#include <GraphMol/SmilesParse/SmilesParse.h>
+#include <INCHI-API/inchi.h>
+
+using namespace RDKit;
+
+TEST_CASE("MolToInchi", "[inchi]") {
+  auto samples = bench_common::load_samples();
+  BENCHMARK("MolToInchi") {
+    std::vector<std::string> inchis;
+    for (auto &mol : samples) {
+      ExtraInchiReturnValues rv;
+      inchis.push_back(MolToInchi(mol, rv));
+    }
+    return inchis;
+  };
+}
+
+TEST_CASE("InchiToInchiKey", "[inchi]") {
+  auto samples = bench_common::load_samples();
+  std::vector<std::string> inchis;
+  for (auto &mol : samples) {
+    ExtraInchiReturnValues rv;
+    inchis.push_back(MolToInchi(mol, rv));
+  }
+  BENCHMARK("InchiToInchiKey") {
+    std::vector<std::string> inchikeys;
+    for (auto &inchi : inchis) {
+      inchikeys.push_back(InchiToInchiKey(inchi));
+    }
+    return inchikeys;
+  };
+}
+
+TEST_CASE("InchiToMol", "[inchi]") {
+  auto samples = bench_common::load_samples();
+  std::vector<std::string> inchis;
+  for (auto &mol : samples) {
+    ExtraInchiReturnValues rv;
+    inchis.push_back(MolToInchi(mol, rv));
+  }
+  BENCHMARK("InchiToMol") {
+    std::vector<std::unique_ptr<ROMol>> mols;
+    for (auto &inchi : inchis) {
+      ExtraInchiReturnValues rv_inner;
+      mols.emplace_back(InchiToMol(inchi, rv_inner));
+    }
+    return mols;
+  };
+}

rdkit/source/Code/Bench/meta.cpp

@@ -0,0 +1,12 @@
+// benches about benches
+
+#include <catch2/catch_all.hpp>
+#include <string>
+
+#include "bench_common.hpp"
+
+TEST_CASE("bench_common::nth_random", "[meta]") {
+  BENCHMARK("bench_common::nth_random", i) {
+    return bench_common::nth_random(i);
+  };
+}

rdkit/source/Code/Bench/mol.cpp

@@ -0,0 +1,72 @@
+#include <catch2/catch_all.hpp>
+#include <string>
+
+#include "bench_common.hpp"
+
+#include <GraphMol/ROMol.h>
+#include <GraphMol/SmilesParse/SmilesParse.h>
+
+using namespace RDKit;
+
+TEST_CASE("ROMol copy constructor", "[mol]") {
+  auto samples = bench_common::load_samples();
+  BENCHMARK_ADVANCED("ROMol copy constructor")(
+      Catch::Benchmark::Chronometer meter) {
+    std::vector<Catch::Benchmark::storage_for<ROMol>> storage(meter.runs() *
+                                                              samples.size());
+    meter.measure([&](int i) {
+      for (size_t sample = 0; sample < samples.size(); ++sample) {
+        storage[i * samples.size() + sample].construct(samples[sample]);
+      }
+    });
+  };
+}
+
+TEST_CASE("ROMol destructor", "[mol]") {
+  auto samples = bench_common::load_samples();
+  BENCHMARK_ADVANCED("ROMol destructor")(Catch::Benchmark::Chronometer meter) {
+    std::vector<Catch::Benchmark::destructable_object<ROMol>> storage(
+        meter.runs() * samples.size());
+    for (size_t i = 0; i < storage.size(); ++i) {
+      storage[i].construct(samples[i % samples.size()]);
+    }
+    meter.measure([&](int i) {
+      for (size_t sample = 0; sample < samples.size(); ++sample) {
+        storage[i * samples.size() + sample].destruct();
+      }
+    });
+  };
+}
+
+TEST_CASE("memory pressure test", "[mol][size]") {
+  auto cases = bench_common::load_samples();
+  REQUIRE(!cases.empty());
+
+  const size_t N = 10400;
+  std::vector<ROMol> mols;
+  mols.reserve(N);
+  for (size_t i = 0; i < N; ++i) {
+    mols.emplace_back(cases[i % cases.size()]);
+  }
+  REQUIRE(mols.size() == N);
+
+  BENCHMARK("memory pressure test", i) {
+    // copy from one random location to another
+    auto a = bench_common::nth_random(i);
+    auto src_idx = a % mols.size();
+    auto dst_idx = (a / mols.size()) % mols.size();
+    ROMol temp(mols[src_idx]);
+    mols[dst_idx] = std::move(temp);
+  };
+}
+
+TEST_CASE("ROMol::getNumHeavyAtoms", "[mol]") {
+  auto samples = bench_common::load_samples();
+  BENCHMARK("ROMol::getNumHeavyAtoms") {
+    auto sum = 0;
+    for (auto &mol : samples) {
+      sum += mol.getNumHeavyAtoms();
+    }
+    return sum;
+  };
+}

rdkit/source/Code/Bench/molops.cpp

@@ -0,0 +1,48 @@
+#include <catch2/catch_all.hpp>
+#include <string>
+
+#include "bench_common.hpp"
+
+#include <GraphMol/SmilesParse/SmilesParse.h>
+#include <GraphMol/MolOps.h>
+
+using namespace RDKit;
+
+TEST_CASE("MolOps::addHs", "[molops]") {
+  auto samples = bench_common::load_samples();
+  BENCHMARK("MolOps::addHs") {
+    auto total_atoms = 0;
+    for (auto &mol : samples) {
+      RWMol mol_copy(mol);
+      MolOps::addHs(mol_copy);
+      total_atoms += mol_copy.getNumAtoms();
+    }
+    return total_atoms;
+  };
+}
+
+TEST_CASE("MolOps::FindSSR", "[molops]") {
+  auto samples = bench_common::load_samples();
+  BENCHMARK("MolOps::FindSSR") {
+    auto total = 0;
+    for (auto &mol : samples) {
+      total += MolOps::findSSSR(mol);
+    }
+    return total;
+  };
+}
+
+TEST_CASE("MolOps::getMolFrags", "[molops]") {
+  auto samples = bench_common::load_samples();
+  BENCHMARK("MolOps::getMolFrags") {
+    auto total = 0;
+    for (auto &mol : samples) {
+      std::vector<std::unique_ptr<ROMol>> frags;
+      MolOps::getMolFrags(mol, frags);
+      for (auto &frag : frags) {
+        total += frag->getNumAtoms();
+      }
+    }
+    return total;
+  };
+}

rdkit/source/Code/Bench/pickle.cpp

@@ -0,0 +1,41 @@
+#include <catch2/catch_all.hpp>
+#include <string>
+#include <sstream>
+
+#include "bench_common.hpp"
+
+#include <GraphMol/MolPickler.h>
+#include <GraphMol/SmilesParse/SmilesParse.h>
+
+using namespace RDKit;
+
+TEST_CASE("MolPickler::pickleMol", "[pickle]") {
+  auto samples = bench_common::load_samples();
+  BENCHMARK("MolPickler::pickleMol") {
+    std::stringstream buf;
+    for (auto &mol : samples) {
+      MolPickler::pickleMol(mol, buf);
+    }
+    return buf.str().size();
+  };
+}
+
+TEST_CASE("MolPickler::molFromPickle", "[pickle]") {
+  auto samples = bench_common::load_samples();
+  std::vector<std::string> pickles;
+  pickles.reserve(samples.size());
+  for (auto &mol : samples) {
+    std::string pickled;
+    MolPickler::pickleMol(mol, pickled);
+    pickles.push_back(std::move(pickled));
+  }
+  BENCHMARK("MolPickler::molFromPickle") {
+    auto total_atoms = 0;
+    for (auto &pickled : pickles) {
+      ROMol res(pickled);
+      total_atoms += res.getNumAtoms();
+    }
+    REQUIRE(total_atoms > 0);
+    return total_atoms;
+  };
+}

rdkit/source/Code/Bench/smiles.cpp

@@ -0,0 +1,46 @@
+#include <catch2/catch_all.hpp>
+#include <string>
+
+#include "bench_common.hpp"
+
+#include <GraphMol/ROMol.h>
+#include <GraphMol/SmilesParse/SmilesParse.h>
+#include <GraphMol/SmilesParse/SmilesWrite.h>
+
+using namespace RDKit;
+
+TEST_CASE("SmilesToMol", "[smiles]") {
+  BENCHMARK("SmilesToMol") {
+    auto total_atoms = 0;
+    for (auto smiles : bench_common::SAMPLES) {
+      auto mol = v2::SmilesParse::MolFromSmiles(smiles);
+      REQUIRE(mol);
+      total_atoms += mol->getNumAtoms();
+    }
+    return total_atoms;
+  };
+}
+
+TEST_CASE("MolToSmiles", "[smiles]") {
+  auto samples = bench_common::load_samples();
+  BENCHMARK("MolToSmiles") {
+    auto total_length = 0;
+    for (auto &mol : samples) {
+      auto smiles = MolToSmiles(mol);
+      total_length += smiles.size();
+    }
+    return total_length;
+  };
+}
+
+TEST_CASE("MolToCXSmiles", "[smiles]") {
+  auto samples = bench_common::load_samples();
+  BENCHMARK("MolToCXSmiles") {
+    auto total_length = 0;
+    for (auto &mol : samples) {
+      auto smiles = MolToCXSmiles(mol);
+      total_length += smiles.size();
+    }
+    return total_length;
+  };
+}

rdkit/source/Code/Bench/stereo.cpp

@@ -0,0 +1,72 @@
+#include <catch2/catch_all.hpp>
+#include <string>
+
+#include "bench_common.hpp"
+
+#include <GraphMol/CIPLabeler/CIPLabeler.h>
+#include <GraphMol/Chirality.h>
+#include <GraphMol/ROMol.h>
+#include <GraphMol/SmilesParse/SmilesParse.h>
+#include <GraphMol/MolOps.h>
+#include <GraphMol/test_fixtures.h>
+
+using namespace RDKit;
+
+TEST_CASE("Chirality::findPotentialStereo", "[stereo]") {
+  auto samples = bench_common::load_samples();
+
+  BENCHMARK("Chirality::findPotentialStereo") {
+    auto total = 0;
+
+    for (auto &mol : samples) {
+      auto stereo_infos = Chirality::findPotentialStereo(mol);
+
+      // workaround for https://github.com/rdkit/rdkit/issues/8880
+      mol.clearComputedProps();
+
+      for (auto &info : stereo_infos) {
+        total += info.controllingAtoms.size();
+      }
+    }
+
+    return total;
+  };
+}
+
+TEST_CASE("CIPLabeler::assignCIPLabels", "[stereo]") {
+  auto samples = bench_common::load_samples();
+  BENCHMARK("CIPLabeler::assignCIPLabels") {
+    for (auto &mol : samples) {
+      CIPLabeler::assignCIPLabels(mol);
+    }
+  };
+}
+
+TEST_CASE("MolOps::assignStereochemistry", "[stereo]") {
+  const auto cleanIt = true;
+  const auto force = true;
+  const auto flagPossibleStereoCenters = true;
+
+  auto samples = bench_common::load_samples();
+
+  const auto legacy = GENERATE(true, false);
+  UseLegacyStereoPerceptionFixture fx(legacy);
+  auto str_legacy = std::string(legacy ? "true" : "false");
+
+  BENCHMARK("MolOps::assignStereochemistry legacy=" + str_legacy) {
+    auto total = 0;
+
+    for (auto &mol : samples) {
+      MolOps::assignStereochemistry(mol, cleanIt, force,
+                                    flagPossibleStereoCenters);
+      for (auto &atom : mol.atoms()) {
+        total += atom->getChiralTag();
+      }
+
+      // workaround for https://github.com/rdkit/rdkit/issues/8880
+      mol.clearComputedProps();
+    }
+
+    return total;
+  };
+}

rdkit/source/Code/Bench/substruct_match.cpp

@@ -0,0 +1,109 @@
+#include <catch2/catch_all.hpp>
+#include <string>
+#include <vector>
+#include <istream>
+#include <filesystem>
+
+#include "bench_common.hpp"
+
+#include <GraphMol/ROMol.h>
+#include <GraphMol/SmilesParse/SmilesParse.h>
+#include <GraphMol/Substruct/SubstructMatch.h>
+
+using namespace RDKit;
+
+namespace bench_substruct_match {
+
+std::filesystem::path relative_to_rdbase(
+    const std::filesystem::path &relative) {
+  char *rdbase = std::getenv("RDBASE");
+  if (!rdbase) {
+    throw std::runtime_error("RDBASE environment variable not set");
+  }
+  std::filesystem::path path(rdbase);
+  path /= relative;
+  return path;
+}
+
+std::vector<std::string> parse_smarts_examples(std::istream &in) {
+  std::vector<std::string> examples;
+  std::string line;
+  while (std::getline(in, line)) {
+    if (line.starts_with("#") || line.starts_with(" ") || line.empty()) {
+      continue;
+    }
+    std::istringstream ss(line);
+    std::string smarts;
+    std::getline(ss, smarts, ' ');
+    examples.push_back(line);
+  }
+  return examples;
+}
+
+std::vector<std::string> load_smarts_examples(std::filesystem::path &path) {
+  std::ifstream file(path);
+  return parse_smarts_examples(file);
+}
+
+std::vector<ROMol> load_smarts_queries(const std::filesystem::path &relative) {
+  auto path = relative_to_rdbase(relative);
+  auto examples = load_smarts_examples(path);
+  std::vector<ROMol> queries;
+  for (const auto &example : examples) {
+    auto query = v2::SmilesParse::MolFromSmarts(example);
+    REQUIRE(query);
+    queries.push_back(std::move(*query));
+  }
+  return queries;
+}
+
+}  // namespace bench_substruct_match
+
+TEST_CASE("ROMol::GetSubstructMatch", "[substruct_match]") {
+  auto samples = bench_common::load_samples();
+
+  auto query = v2::SmilesParse::MolFromSmarts(
+      "[#6,#7]1:[#6,#7]:[#6](:[#6,#7]:[#6,#7]:[#6]:1-[#17,#35,#53,#9])-[#5](-[#8])-[#8]");
+  BENCHMARK("ROMol::GetSubstructMatch") {
+    auto total = 0;
+    for (auto &mol : samples) {
+      auto matches = SubstructMatch(mol, *query);
+      total += matches.size();
+    }
+    return total;
+  };
+}
+
+TEST_CASE("ROMol::GetSubstructMatch RLewis", "[substruct_match]") {
+  auto mols = bench_common::load_samples();
+  auto queries = bench_substruct_match::load_smarts_queries(
+      "Data/SmartsLib/RLewis_smarts.txt");
+
+  BENCHMARK("ROMol::GetSubstructMatch RLewis") {
+    auto total = 0;
+    for (const auto &mol : mols) {
+      for (const auto &query : queries) {
+        auto matches = SubstructMatch(mol, query);
+        total += matches.size();
+      }
+    }
+    return total;
+  };
+}
+
+TEST_CASE("ROMol::GetSubstructMatch patty", "[substruct_match]") {
+  auto mols = bench_common::load_samples();
+  auto queries = bench_substruct_match::load_smarts_queries(
+      "Data/SmartsLib/patty_rules.txt");
+
+  BENCHMARK("ROMol::GetSubstructMatch patty") {
+    auto total = 0;
+    for (const auto &mol : mols) {
+      for (const auto &query : queries) {
+        auto matches = SubstructMatch(mol, query);
+        total += matches.size();
+      }
+    }
+    return total;
+  };
+}

rdkit/source/Code/CMakeLists.txt

@@ -0,0 +1,44 @@
+createExportTestHeaders()
+add_subdirectory(RDGeneral)
+if(RDK_BUILD_PYTHON_WRAPPERS)
+  add_subdirectory(RDBoost)
+endif(RDK_BUILD_PYTHON_WRAPPERS)
+add_subdirectory(RDStreams)
+add_subdirectory(DataStructs)
+add_subdirectory(Geometry)
+
+add_subdirectory(Numerics)
+add_subdirectory(ForceField)
+add_subdirectory(DistGeom)
+add_subdirectory(Catalogs)
+
+add_subdirectory(GraphMol)
+add_subdirectory(Query)
+add_subdirectory(Features)
+
+add_subdirectory(DataManip)
+if(RDK_BIG_ENDIAN)
+  message("Skipping build of SimDivPickers on big endian system")
+else(RDK_BIG_ENDIAN)
+  add_subdirectory(SimDivPickers)
+endif(RDK_BIG_ENDIAN)
+
+add_subdirectory(ML)
+add_subdirectory(ChemicalFeatures)
+
+if(RDK_BUILD_SWIG_WRAPPERS)
+  add_subdirectory(JavaWrappers)
+endif()
+
+if(RDK_BUILD_PGSQL)
+  add_subdirectory(PgSQL/rdkit)
+endif(RDK_BUILD_PGSQL)
+
+add_subdirectory(MinimalLib)
+
+
+if(RDK_BUILD_FUZZ_TARGETS)
+  add_subdirectory(Fuzz)
+endif(RDK_BUILD_FUZZ_TARGETS)
+
+add_subdirectory(Bench)

rdkit/source/Code/Catalogs/CMakeLists.txt

@@ -0,0 +1,11 @@
+
+rdkit_library(Catalogs 
+              Catalog.cpp CatalogParams.cpp CatalogEntry.cpp
+              LINK_LIBRARIES RDGeneral)
+target_compile_definitions(Catalogs PRIVATE RDKIT_CATALOGS_BUILD)
+
+rdkit_headers(CatalogEntry.h
+              Catalog.h
+              CatalogParams.h DEST Catalogs)
+
+

rdkit/source/Code/Catalogs/Catalog.cpp

@@ -0,0 +1,15 @@
+// $Id$
+//
+//  Copyright (C) 2003-2006 Rational Discovery LLC
+//
+//  @@ All Rights Reserved @@
+//  This file is part of the RDKit.
+//  The contents are covered by the terms of the BSD license
+//  which is included in the file license.txt, found at the root
+//  of the RDKit source tree.
+//
+//
+
+#include "Catalog.h"
+
+namespace RDCatalog {}