https://nativegate.dev/blog/ Notes on turning legacy C++ and Fortran into live APIs: generated bindings, isolated builds, and numeric drift detection. en https://nativegate.dev/blog/first-contact-netlib-specfun.html https://nativegate.dev/blog/first-contact-netlib-specfun.html Wed, 23 Sep 2026 00:00:00 GMT nativegate run on a third-party codebase: netlib specfun regenerated in CI on two OSes, 14 golden values pinned, three generator defects found and fixed. https://nativegate.dev/blog/exposing-petroleum-models-through-python.html https://nativegate.dev/blog/exposing-petroleum-models-through-python.html Wed, 28 Oct 2026 00:00:00 GMT How PVT correlations and well-property models become callable Python and HTTP without rewriting the validated Fortran. https://nativegate.dev/blog/containerizing-legacy-scientific-software.html https://nativegate.dev/blog/containerizing-legacy-scientific-software.html Sat, 24 Oct 2026 00:00:00 GMT What the image needs, what must match the orchestrator, and what correctness rules the container cannot solve. https://nativegate.dev/blog/fortran-90-95-to-python.html https://nativegate.dev/blog/fortran-90-95-to-python.html Tue, 20 Oct 2026 00:00:00 GMT Easier to bind than F77 — but each modern feature carries a practical consideration when exposed to Python. https://nativegate.dev/blog/fortran-77-to-python.html https://nativegate.dev/blog/fortran-77-to-python.html Sun, 18 Oct 2026 00:00:00 GMT What actually changes when F77 becomes reachable from Python, and what deliberately doesn't. https://nativegate.dev/blog/how-to-wrap-legacy-fortran-with-f2py.html https://nativegate.dev/blog/how-to-wrap-legacy-fortran-with-f2py.html Wed, 14 Oct 2026 00:00:00 GMT A practical walkthrough: the happy path, the legacy-deck path, and what remains to build after the extension module compiles. https://nativegate.dev/blog/f2py-when-it-works-where-it-stops.html https://nativegate.dev/blog/f2py-when-it-works-where-it-stops.html Sun, 11 Oct 2026 00:00:00 GMT Where NumPy's F2PY is genuinely the right answer, and the honest list of what it does not address. https://nativegate.dev/blog/legacy-scientific-software-modernization.html https://nativegate.dev/blog/legacy-scientific-software-modernization.html Wed, 07 Oct 2026 00:00:00 GMT How organizations actually modernize legacy scientific software without rewriting the math. https://nativegate.dev/blog/deploying-legacy-engineering-models-as-apis.html https://nativegate.dev/blog/deploying-legacy-engineering-models-as-apis.html Sun, 04 Oct 2026 00:00:00 GMT Stateful models, one-call-at-a-time semantics, and what the generated service gives you — and deliberately doesn't. https://nativegate.dev/blog/modernizing-legacy-engineering-software.html https://nativegate.dev/blog/modernizing-legacy-engineering-software.html Wed, 30 Sep 2026 00:00:00 GMT Rewrites transfer decades of invisible discipline decisions by hand, with no oracle. Interface-first modernization with a numerical baseline works. https://nativegate.dev/blog/fortran-reservoir-model-to-python-api.html https://nativegate.dev/blog/fortran-reservoir-model-to-python-api.html Sat, 26 Sep 2026 00:00:00 GMT A walkthrough of the committed example: 4,800 lines of unmodified Fortran, a verified service, and the two details that decide correctness. https://nativegate.dev/blog/modernizing-legacy-petroleum-engineering-software.html https://nativegate.dev/blog/modernizing-legacy-petroleum-engineering-software.html Tue, 22 Sep 2026 00:00:00 GMT PVT decks, COMMON blocks, and validation as the asset: why petroleum software is the hard version of the modernization problem. https://nativegate.dev/blog/nativegate-vs-f90wrap.html https://nativegate.dev/blog/nativegate-vs-f90wrap.html Sat, 19 Sep 2026 00:00:00 GMT f90wrap handles modern, object-shaped Fortran with real derived-type support. An honest comparison of where interface generation ends and a full modernization workflow begins. https://nativegate.dev/blog/nativegate-vs-f2py.html https://nativegate.dev/blog/nativegate-vs-f2py.html Mon, 14 Sep 2026 00:00:00 GMT F2PY is NumPy's own Fortran-to-Python tool and remains the right choice for many jobs. An honest comparison of scope: bindings vs the whole modernization workflow. https://nativegate.dev/blog/fortran-to-python.html https://nativegate.dev/blog/fortran-to-python.html Tue, 08 Sep 2026 00:00:00 GMT The Fortran-to-Python boundary layer is mostly solved. Where the work actually piles up: intent, COMMON state, builds, packaging, and evidence the numbers didn't move. https://nativegate.dev/blog/modernize-legacy-fortran.html https://nativegate.dev/blog/modernize-legacy-fortran.html Tue, 01 Sep 2026 00:00:00 GMT Keep the numerics, replace the interface. The path from a validated deck to a Python package, an API, and a verified numerical baseline. https://nativegate.dev/blog/isolated-native-builds.html https://nativegate.dev/blog/isolated-native-builds.html Thu, 20 Aug 2026 00:00:00 GMT The build container a generated service runs in: non-root, network-isolated, and why a compiler you don't control still doesn't get to touch your host. https://nativegate.dev/blog/generated-bindings-vs-handwritten.html https://nativegate.dev/blog/generated-bindings-vs-handwritten.html Thu, 13 Aug 2026 00:00:00 GMT Why nativegate generates pybind11 and f2py bindings from what the source actually declares, instead of asking you to maintain wrappers by hand.