basedpl 0.1.79__tar.gz → 0.1.80__tar.gz
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Potentially problematic release.
This version of basedpl might be problematic. Click here for more details.
- {basedpl-0.1.79 → basedpl-0.1.80}/.github/workflows/test.yml +2 -1
- {basedpl-0.1.79 → basedpl-0.1.80}/Cargo.lock +9 -9
- {basedpl-0.1.79 → basedpl-0.1.80}/Cargo.toml +3 -3
- {basedpl-0.1.79 → basedpl-0.1.80}/DEV.md +8 -8
- {basedpl-0.1.79 → basedpl-0.1.80}/PKG-INFO +1 -1
- {basedpl-0.1.79 → basedpl-0.1.80}/build.rs +8 -8
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/cli.qmd +1 -1
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/composition.qmd +1 -1
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/left-tail-arrow.qmd +1 -1
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/right-tail-arrow.qmd +1 -1
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/llms.txt +1 -1
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/web.ipynb +1 -1
- {basedpl-0.1.79 → basedpl-0.1.80}/py/src/lib.rs +59 -31
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/bpltests.py +39 -4
- {basedpl-0.1.79 → basedpl-0.1.80}/src/agreement.rs +5 -6
- {basedpl-0.1.79 → basedpl-0.1.80}/src/array/storage.rs +88 -64
- {basedpl-0.1.79 → basedpl-0.1.80}/src/array/text.rs +23 -23
- {basedpl-0.1.79 → basedpl-0.1.80}/src/array.rs +145 -95
- {basedpl-0.1.79 → basedpl-0.1.80}/src/batch.rs +66 -75
- {basedpl-0.1.79 → basedpl-0.1.80}/src/canvas.rs +5 -3
- {basedpl-0.1.79 → basedpl-0.1.80}/src/csv.rs +20 -19
- {basedpl-0.1.79 → basedpl-0.1.80}/src/data.rs +52 -54
- {basedpl-0.1.79 → basedpl-0.1.80}/src/date.rs +2 -2
- {basedpl-0.1.79 → basedpl-0.1.80}/src/derivative.rs +89 -108
- {basedpl-0.1.79 → basedpl-0.1.80}/src/display.rs +29 -29
- {basedpl-0.1.79 → basedpl-0.1.80}/src/distribution.rs +16 -17
- {basedpl-0.1.79 → basedpl-0.1.80}/src/element.rs +10 -10
- {basedpl-0.1.79 → basedpl-0.1.80}/src/error.rs +32 -23
- {basedpl-0.1.79 → basedpl-0.1.80}/src/eval/assign.rs +42 -28
- {basedpl-0.1.79 → basedpl-0.1.80}/src/eval/binder.rs +131 -60
- {basedpl-0.1.79 → basedpl-0.1.80}/src/eval/bindings.rs +36 -11
- {basedpl-0.1.79 → basedpl-0.1.80}/src/eval/idioms.rs +10 -18
- {basedpl-0.1.79 → basedpl-0.1.80}/src/eval/operators.rs +192 -139
- {basedpl-0.1.79 → basedpl-0.1.80}/src/eval/session.rs +83 -63
- {basedpl-0.1.79 → basedpl-0.1.80}/src/eval/tree.rs +152 -73
- {basedpl-0.1.79 → basedpl-0.1.80}/src/eval.rs +132 -108
- {basedpl-0.1.79 → basedpl-0.1.80}/src/execution.rs +29 -16
- {basedpl-0.1.79 → basedpl-0.1.80}/src/explain.rs +58 -40
- {basedpl-0.1.79 → basedpl-0.1.80}/src/files.rs +5 -10
- {basedpl-0.1.79 → basedpl-0.1.80}/src/fusion.rs +80 -52
- {basedpl-0.1.79 → basedpl-0.1.80}/src/host.rs +1 -1
- {basedpl-0.1.79 → basedpl-0.1.80}/src/host_web.rs +1 -1
- {basedpl-0.1.79 → basedpl-0.1.80}/src/image/codecs.rs +1 -1
- {basedpl-0.1.79 → basedpl-0.1.80}/src/image/imageio.rs +5 -6
- {basedpl-0.1.79 → basedpl-0.1.80}/src/image.rs +5 -7
- {basedpl-0.1.79 → basedpl-0.1.80}/src/inspection.rs +1 -1
- {basedpl-0.1.79 → basedpl-0.1.80}/src/jit.rs +181 -127
- {basedpl-0.1.79 → basedpl-0.1.80}/src/js.rs +8 -8
- {basedpl-0.1.79 → basedpl-0.1.80}/src/json.rs +10 -5
- {basedpl-0.1.79 → basedpl-0.1.80}/src/keyed.rs +5 -3
- {basedpl-0.1.79 → basedpl-0.1.80}/src/lib.rs +4 -4
- {basedpl-0.1.79 → basedpl-0.1.80}/src/linear.rs +88 -75
- {basedpl-0.1.79 → basedpl-0.1.80}/src/name.rs +5 -1
- {basedpl-0.1.79 → basedpl-0.1.80}/src/number.rs +120 -98
- {basedpl-0.1.79 → basedpl-0.1.80}/src/number_theory.rs +7 -8
- {basedpl-0.1.79 → basedpl-0.1.80}/src/optimize.rs +11 -14
- {basedpl-0.1.79 → basedpl-0.1.80}/src/pervasive.rs +289 -230
- {basedpl-0.1.79 → basedpl-0.1.80}/src/plan.rs +58 -34
- {basedpl-0.1.79 → basedpl-0.1.80}/src/plot.rs +6 -10
- {basedpl-0.1.79 → basedpl-0.1.80}/src/polynomial.rs +9 -11
- {basedpl-0.1.79 → basedpl-0.1.80}/src/primitive/indexing.rs +135 -100
- {basedpl-0.1.79 → basedpl-0.1.80}/src/primitive/numeric.rs +74 -61
- {basedpl-0.1.79 → basedpl-0.1.80}/src/primitive/search.rs +33 -28
- {basedpl-0.1.79 → basedpl-0.1.80}/src/primitive/structural.rs +26 -15
- {basedpl-0.1.79 → basedpl-0.1.80}/src/primitive.rs +405 -152
- {basedpl-0.1.79 → basedpl-0.1.80}/src/program/blocks.rs +267 -142
- {basedpl-0.1.79 → basedpl-0.1.80}/src/program/contraction.rs +18 -16
- {basedpl-0.1.79 → basedpl-0.1.80}/src/program/layout.rs +28 -22
- {basedpl-0.1.79 → basedpl-0.1.80}/src/program/run.rs +28 -15
- {basedpl-0.1.79 → basedpl-0.1.80}/src/program/trace.rs +72 -41
- {basedpl-0.1.79 → basedpl-0.1.80}/src/program/verify.rs +4 -4
- {basedpl-0.1.79 → basedpl-0.1.80}/src/program.rs +111 -72
- {basedpl-0.1.79 → basedpl-0.1.80}/src/protocol.rs +7 -6
- {basedpl-0.1.79 → basedpl-0.1.80}/src/reference.rs +74 -51
- {basedpl-0.1.79 → basedpl-0.1.80}/src/regex.rs +1 -3
- {basedpl-0.1.79 → basedpl-0.1.80}/src/search.rs +23 -23
- {basedpl-0.1.79 → basedpl-0.1.80}/src/selection.rs +50 -31
- {basedpl-0.1.79 → basedpl-0.1.80}/src/settings.rs +3 -1
- {basedpl-0.1.79 → basedpl-0.1.80}/src/symbols.rs +9 -8
- {basedpl-0.1.79 → basedpl-0.1.80}/src/syntax.rs +121 -100
- {basedpl-0.1.79 → basedpl-0.1.80}/src/system.rs +45 -39
- {basedpl-0.1.79 → basedpl-0.1.80}/src/timing.rs +61 -27
- {basedpl-0.1.79 → basedpl-0.1.80}/src/transform.rs +148 -83
- {basedpl-0.1.79 → basedpl-0.1.80}/src/xml.rs +1 -1
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/core.rs +7 -3
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/derivatives.rs +10 -3
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/explain.rs +22 -20
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/jit.rs +146 -61
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/reference.rs +20 -17
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/scaling.rs +3 -2
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/settings.rs +17 -10
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/speed/README.md +1 -1
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/speed/arrays.bpl +14 -11
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/speed/maps.bpl +1 -1
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/speed/models.bpl +7 -7
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/speed/structure.bpl +5 -5
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/speed.rs +5 -5
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/units.rs +14 -5
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/wasm-pkg.mjs +2 -2
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/wasm-speed.mjs +6 -6
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/wasm.mjs +24 -19
- {basedpl-0.1.79 → basedpl-0.1.80}/.cargo/config.toml +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/.gitattributes +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/.github/workflows/ci.yml +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/.github/workflows/docs.yml +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/.github/workflows/speed.yml +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/.gitignore +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/.node-version +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/CONTRIBUTING.md +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/LICENSE +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/README.md +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/_config.yml +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/_layouts/default.html +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/editors/jetbrains/BPL.tmbundle/Syntaxes/bpl.tmLanguage.json +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/editors/jetbrains/BPL.tmbundle/Syntaxes/hpl.tmLanguage.json +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/editors/jetbrains/BPL.tmbundle/info.plist +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/editors/jetbrains/BPL.xml +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/editors/jetbrains/README.md +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/editors/macos/BasedPL-de.txt +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/editors/macos/BasedPL-es.txt +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/editors/macos/BasedPL-fr.txt +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/editors/macos/BasedPL-uk.txt +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/editors/macos/BasedPL-us.txt +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/editors/vim/ftdetect/bpl.vim +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/editors/vim/syntax/bpl.vim +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/editors/windows/README.md +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/editors/windows/build.ps1 +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/editors/windows/install.ps1 +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/editors/windows/keyboard.c +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/editors/windows/keyboard.def +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/editors/windows/keyboard.rc +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/editors/windows/layout.h +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/index_files/figure-commonmark/cell-17-output-1.svg +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/index_files/figure-commonmark/cell-18-output-1.svg +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/index_files/figure-html/cell-17-output-1.svg +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/index_files/figure-html/cell-18-output-1.svg +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/lib/LICENSE-april +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/lib/README.md +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/lib/array.bpl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/lib/dyalog.bpl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/lib/graph.bpl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/lib/numeric.bpl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/lib/power.bpl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/lib/regex.bpl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/lib/string.bpl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/lib/tree.bpl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/.gitignore +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/_quarto.yml +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/arrays.ipynb +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/assignment.ipynb +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/autodiff.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/bpl.xml +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/bqn.xml +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/bracket-types.ipynb +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/comparison.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/data.ipynb +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/dfns.ipynb +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/distributions.ipynb +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/driver.js +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/dyalog.ipynb +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/dyalog.xml +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/evaluation.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/fonts/BQN386B.woff2 +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/fonts/README.md +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/fonts/SAX2B.woff2 +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/fonts.css +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/gallery/attractor.bpl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/gallery/boids.bpl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/gallery/dragon.bpl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/gallery/flow.bpl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/gallery/mandelbrot.bpl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/gallery/mandelbrot32.bpl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/gallery/orbits.bpl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/gallery/phyllotaxis.bpl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/gallery/prefs.bpl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/gallery/reaction.bpl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/gallery.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/getting-started.ipynb +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/_metadata.yml +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/alpha-underbar.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/alpha.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/at.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/backslash-bar.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/backslash.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/braces.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/brackets.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/bullet.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/circle-backslash.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/circle-bar.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/circle-diaeresis.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/circle-divide.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/circle-slash.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/circle-star.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/circle-stile.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/circle-times.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/circle.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/colon.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/comma-bar.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/comma.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/del-diaeresis.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/del-stile.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/del.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/delta-stile.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/diaeresis.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/diamond-underbar.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/diamond.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/domino.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/dot.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/down-arrow.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/down-shoe.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/down-tack.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/downstile.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/epsilon-underbar.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/epsilon.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/equal-underbar-slash.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/equal-underbar.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/equals.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/error-guard.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/exclamation.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/greater-than-or-equal.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/greater-than.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/hash.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/hydrant.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/infinity.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/inverted-question.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/iota-underbar.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/iota.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/jot-diaeresis.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/jot.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/lamp.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/left-arrow.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/left-shoe-underbar.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/left-shoe.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/left-tack.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/less-than-or-equal.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/less-than.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/minus.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/not-equals.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/obelus.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/omega-underbar.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/omega.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/overbar.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/parentheses.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/partial.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/pi.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/plus.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/quad-colon.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/quad-diamond.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/quad-equal.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/quad-jot.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/quad.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/question.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/quote.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/radical.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/rho.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/right-arrow-from-bar.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/right-left-arrows.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/right-shoe.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/right-tack.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/slash-bar.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/slash.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/squad.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/star-diaeresis.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/star.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/stile-tilde.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/stile.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/superscript-b.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/thorn.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/tilde-diaeresis.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/tilde.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/times.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/tricolon.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/up-arrow.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/up-down-arrow.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/up-shoe.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/up-tack.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/upstile.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/vee-bar.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/vee.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/wedge-bar.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/wedge.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs/zilde.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs-study.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/glyphs.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/hanzi-notation-zh.ipynb +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/hanzi-notation.ipynb +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/hpl.xml +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/index-notation.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/index.ipynb +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/indexing.ipynb +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/introspection.ipynb +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/keyboard.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/keyed.ipynb +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/magics.ipynb +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/mathish.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/modules.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/nbdev.yml +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/notebooks.ipynb +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/numbers.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/overview.ipynb +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/papers/aplwiki-arrays.md +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/papers/brickman-depth-1-scalars.md +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/papers/mercer-based-system.md +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/performance.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/playground/bpl.js +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/playground/gallery.js +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/playground/monaco-worker.js +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/playground/page.js +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/playground/worker.js +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/playground.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/plot.ipynb +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/principles.ipynb +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/processes.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/puppy.jpg +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/python.ipynb +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/regex.ipynb +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/repl.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/scripts.ipynb +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/sidebar.yml +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/strings.ipynb +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/study/01_basics.ipynb +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/study/02_functions.ipynb +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/study/03_operators.ipynb +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/study/04_nested-arrays.ipynb +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/study/05_searching-indexing.ipynb +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/study/06_rearranging-arrays.ipynb +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/study/07_numerical-tools.ipynb +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/style.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/styles.css +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/syntax.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/_metadata.yml +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/c.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/canvas.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/constants.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/copy.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/csv.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/date.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/decompose.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/deflate.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/delay.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/detach.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/distribution.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/edge.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/element.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/ex.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/explain.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/fetch.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/float.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/hash.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/host.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/image.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/index.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/js.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/json.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/literal.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/load.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/memory.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/metadata.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/mime.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/mkdir.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/nc.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/nget.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/nl.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/normalize.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/nput.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/path.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/plot.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/prefs.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/r.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/rand.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/readdir.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/remove.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/rename.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/signal.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/src.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/storage.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/svg.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/time.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/ucs.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/uuid.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/vfi.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system/xml.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/system-functions.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/terms.qmd +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/nbs/xml.ipynb +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/package-lock.json +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/package.json +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/provenance/README.md +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/provenance/aplcart.jsonl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/provenance/april.jsonl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/provenance/bqn.jsonl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/provenance/core.jsonl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/provenance/dyalog.jsonl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/provenance/ngn.jsonl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/py/Cargo.toml +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/py/build.rs +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/pyproject.toml +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/__init__.py +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/_modidx.py +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/assets/__init__.py +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/assets/fonts/BQN386B.ttf +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/assets/fonts/LICENSE-BQN386 +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/assets/fonts/LICENSE-SAX2 +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/assets/fonts/SAX2B.ttf +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/assets/macos/BasedPL-de.bundle/Contents/Info.plist +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/assets/macos/BasedPL-de.bundle/Contents/Resources/BasedPL-de.icns +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/assets/macos/BasedPL-de.bundle/Contents/Resources/BasedPL-de.keylayout +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/assets/macos/BasedPL-de.bundle/Contents/Resources/en.lproj/InfoPlist.strings +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/assets/macos/BasedPL-de.bundle/Contents/version.plist +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/assets/macos/BasedPL-es.bundle/Contents/Info.plist +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/assets/macos/BasedPL-es.bundle/Contents/Resources/BasedPL-es.icns +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/assets/macos/BasedPL-es.bundle/Contents/Resources/BasedPL-es.keylayout +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/assets/macos/BasedPL-es.bundle/Contents/Resources/en.lproj/InfoPlist.strings +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/assets/macos/BasedPL-es.bundle/Contents/version.plist +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/assets/macos/BasedPL-fr.bundle/Contents/Info.plist +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/assets/macos/BasedPL-fr.bundle/Contents/Resources/BasedPL-fr.icns +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/assets/macos/BasedPL-fr.bundle/Contents/Resources/BasedPL-fr.keylayout +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/assets/macos/BasedPL-fr.bundle/Contents/Resources/en.lproj/InfoPlist.strings +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/assets/macos/BasedPL-fr.bundle/Contents/version.plist +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/assets/macos/BasedPL-uk.bundle/Contents/Info.plist +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/assets/macos/BasedPL-uk.bundle/Contents/Resources/BasedPL-uk.icns +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/assets/macos/BasedPL-uk.bundle/Contents/Resources/BasedPL-uk.keylayout +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/assets/macos/BasedPL-uk.bundle/Contents/Resources/en.lproj/InfoPlist.strings +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/assets/macos/BasedPL-uk.bundle/Contents/version.plist +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/assets/macos/BasedPL-us.bundle/Contents/Info.plist +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/assets/macos/BasedPL-us.bundle/Contents/Resources/BasedPL-us.icns +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/assets/macos/BasedPL-us.bundle/Contents/Resources/BasedPL-us.keylayout +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/assets/macos/BasedPL-us.bundle/Contents/Resources/en.lproj/InfoPlist.strings +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/assets/macos/BasedPL-us.bundle/Contents/version.plist +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/bpl.tmLanguage.json +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/dyalog.py +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/editors.py +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/font.py +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/fonts.css +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/functions.py +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/hpl.tmLanguage.json +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/input.js +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/install.py +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/ipython.py +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/keyboards.json +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/keyboards.py +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/layout.json +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/lb.js +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/notebook_runner.py +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/notebooks.py +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/printing.py +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/reference.py +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/syntax.py +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/windows_keyboard.py +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/python/basedpl/worker.py +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/rustfmt.toml +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/src/canvas.js +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/src/eval/dissect.rs +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/common/mod.rs +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/conftest.py +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/highlighting.mjs +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/input.mjs +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/instructions.txt +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/reference/LICENSE-aplcart +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/reference/LICENSE-april +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/reference/LICENSE-bqn +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/reference/LICENSE-bqncrate +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/reference/LICENSE-dyalog +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/reference/LICENSE-ngn +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/reference/README.md +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/reference/aplcart.bpl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/reference/april.bpl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/reference/bqn.bpl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/reference/core.bpl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/reference/distributions.bpl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/reference/dyalog.bpl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/reference/graphics.bpl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/reference/lib.bpl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/reference/ngn.bpl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/reference/regex.bpl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/reference/static-int32.bpl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/reference/static.bpl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/reference/support/april.bpl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/reference/support/avl.bpl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/reference/support/redblack.bpl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/reference/support/sbst.bpl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/reference/support/splay.bpl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/speed/calls.bpl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/speed/images.bpl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/speed/tables.bpl +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/test_api.py +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/test_basic.py +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/test_bpltests.py +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/test_glyphs.py +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/test_graphics.py +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/test_install.py +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/test_instructions.py +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/test_ipython.py +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/test_kernel.py +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/test_keylayout.py +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/test_notebooks.py +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/test_reference.py +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/test_repl.py +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/test_windows_keyboard.py +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/test_worker.py +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/tests/windows_keyboard.c +0 -0
- {basedpl-0.1.79 → basedpl-0.1.80}/wheel/data/share/jupyter/kernels/bpl/kernel.json +0 -0
|
@@ -9,12 +9,13 @@ permissions:
|
|
|
9
9
|
contents: read
|
|
10
10
|
|
|
11
11
|
# In each job, every test step after the first runs even when an earlier step fails, to report every test's result in one run.
|
|
12
|
-
# A build step that runs past 10 minutes has stalled. These builds normally take about 3.
|
|
13
12
|
jobs:
|
|
14
13
|
test:
|
|
15
14
|
runs-on: ubicloud-standard-4-ubuntu-2604
|
|
16
15
|
steps:
|
|
17
16
|
- uses: actions/checkout@v7
|
|
17
|
+
# The speed tests' costs for large new arrays depend on whether the kernel gives them transparent huge pages.
|
|
18
|
+
- run: cat /sys/kernel/mm/transparent_hugepage/enabled
|
|
18
19
|
# `tests/highlighting.mjs` reads the TextMate grammar through `shiki`, which `package.json` pins.
|
|
19
20
|
- run: npm ci
|
|
20
21
|
- uses: dtolnay/rust-toolchain@stable
|
|
@@ -99,7 +99,7 @@ checksum = "ac07cdecf99051d9a5238b80f35af32cdeba5b336e55d957b318b50137e18da5"
|
|
|
99
99
|
|
|
100
100
|
[[package]]
|
|
101
101
|
name = "basedpl"
|
|
102
|
-
version = "0.1.
|
|
102
|
+
version = "0.1.80"
|
|
103
103
|
dependencies = [
|
|
104
104
|
"base64 0.23.1",
|
|
105
105
|
"bytemuck",
|
|
@@ -148,7 +148,7 @@ dependencies = [
|
|
|
148
148
|
|
|
149
149
|
[[package]]
|
|
150
150
|
name = "basedpl-cli"
|
|
151
|
-
version = "0.1.
|
|
151
|
+
version = "0.1.80"
|
|
152
152
|
dependencies = [
|
|
153
153
|
"async-trait",
|
|
154
154
|
"basedpl",
|
|
@@ -162,7 +162,7 @@ dependencies = [
|
|
|
162
162
|
|
|
163
163
|
[[package]]
|
|
164
164
|
name = "basedpl-py"
|
|
165
|
-
version = "0.1.
|
|
165
|
+
version = "0.1.80"
|
|
166
166
|
dependencies = [
|
|
167
167
|
"basedpl",
|
|
168
168
|
"ctrlc",
|
|
@@ -180,7 +180,7 @@ dependencies = [
|
|
|
180
180
|
|
|
181
181
|
[[package]]
|
|
182
182
|
name = "basedpl-wasm"
|
|
183
|
-
version = "0.1.
|
|
183
|
+
version = "0.1.80"
|
|
184
184
|
dependencies = [
|
|
185
185
|
"basedpl",
|
|
186
186
|
"js-sys",
|
|
@@ -1697,9 +1697,9 @@ dependencies = [
|
|
|
1697
1697
|
|
|
1698
1698
|
[[package]]
|
|
1699
1699
|
name = "rustymath"
|
|
1700
|
-
version = "0.1.
|
|
1700
|
+
version = "0.1.53"
|
|
1701
1701
|
source = "registry+https://github.com/rust-lang/crates.io-index"
|
|
1702
|
-
checksum = "
|
|
1702
|
+
checksum = "f23d74fe738a2a8ed77303288e1cfb1eccf87005fb55c2876814e41374d1f54a"
|
|
1703
1703
|
dependencies = [
|
|
1704
1704
|
"foldhash",
|
|
1705
1705
|
"half",
|
|
@@ -1718,9 +1718,9 @@ dependencies = [
|
|
|
1718
1718
|
|
|
1719
1719
|
[[package]]
|
|
1720
1720
|
name = "rustymath-buffer"
|
|
1721
|
-
version = "0.1.
|
|
1721
|
+
version = "0.1.53"
|
|
1722
1722
|
source = "registry+https://github.com/rust-lang/crates.io-index"
|
|
1723
|
-
checksum = "
|
|
1723
|
+
checksum = "acf6a5013fe3dbbc1b1883fd10b26848e464fdc185837164352a1cca162bc246"
|
|
1724
1724
|
dependencies = [
|
|
1725
1725
|
"pyo3",
|
|
1726
1726
|
"rustymath",
|
|
@@ -2464,7 +2464,7 @@ checksum = "3ad82d2a33cdc9674dc7465672f271e096168fcdbe0f799d9e6db8c5892679dc"
|
|
|
2464
2464
|
|
|
2465
2465
|
[[package]]
|
|
2466
2466
|
name = "xtask"
|
|
2467
|
-
version = "0.1.
|
|
2467
|
+
version = "0.1.80"
|
|
2468
2468
|
dependencies = [
|
|
2469
2469
|
"anyhow",
|
|
2470
2470
|
"rustymath",
|
|
@@ -5,14 +5,14 @@ members = ["py"]
|
|
|
5
5
|
join_comments = true
|
|
6
6
|
|
|
7
7
|
[workspace.package]
|
|
8
|
-
version = "0.1.
|
|
8
|
+
version = "0.1.80"
|
|
9
9
|
edition = "2024"
|
|
10
10
|
rust-version = "1.98"
|
|
11
11
|
license = "Apache-2.0"
|
|
12
12
|
|
|
13
13
|
[workspace.dependencies]
|
|
14
|
-
rustymath = { version = ">=0.1.
|
|
15
|
-
rustymath-buffer = ">=0.1.
|
|
14
|
+
rustymath = { version = ">=0.1.53, <1", default-features = false }
|
|
15
|
+
rustymath-buffer = ">=0.1.53, <1"
|
|
16
16
|
pyo3 = ">=0.29.2, <1"
|
|
17
17
|
pyo3-build-config = ">=0.29.2, <1"
|
|
18
18
|
ctrlc = ">=3.5.2, <4"
|
|
@@ -36,7 +36,7 @@ These design notes in `meta/` guide work still ahead:
|
|
|
36
36
|
- `linear.md`: the analysis that carries over to deferred Jacobians. Jeremy replaced its notation on 17 October 2026, as "Language changes under study" in `programs.md` describes.
|
|
37
37
|
- `stage3-build.md`'s "Step 4: shared values as outputs": the plan for bringing outputs back.
|
|
38
38
|
- `stage4-build.md`'s "Step 5 in outline", its choice 8 and "Not in stage 4": the stage's last step and what it leaves out.
|
|
39
|
-
- `design-audit-summary-2026-10-26.md`: the combined view of four audits of stages 1 to 4, whose order of work the release plan
|
|
39
|
+
- `design-audit-summary-2026-10-26.md`: the combined view of four audits of stages 1 to 4, whose order of work the indexing release's plan replaced. The four `design-audit-*` notes beside it give the evidence.
|
|
40
40
|
- `learned-2026-10-26.md`: what 26 October 2026 established, where each record is, and what's planned.
|
|
41
41
|
- `open.md`: every open item.
|
|
42
42
|
|
|
@@ -76,11 +76,11 @@ Costs in copies can't show a per-call change of a few percent, because code layo
|
|
|
76
76
|
|
|
77
77
|
`call_profile(case, function)` counts every instruction that one call of `function` retires, with the calls it makes, by function, on macOS on Apple silicon. `lldb` runs the case once per step in a fresh process, lets `skip=` calls of `function` run, then single-steps the next one to its return. `function` is a regex for a demangled name, such as `binder::derived`, and needs code of its own. `base=` names a saved copy whose counts appear beside the build's. The allocator's functions, with the functions that they call, share one row, because what they retire depends on the heap's state. A second total leaves that row out. These counts are exact, where `profile`'s come from time. `profile_diff` once put about 50 instructions of a value mark's rise on a conversion loop whose machine code hadn't changed, and single-stepping found them in `derivative::converted`. Use `profile_diff` to find the function whose call changed, then `call_profile` on that function to settle the change. It counts calls of up to 100,000 instructions, at about 450 a second.
|
|
78
78
|
|
|
79
|
-
`bpl-tests NAME ARGUMENT...` runs one of these functions from the shell and prints its result: `check_file`, `check_page`, `time_cases`, `instructions`, `instruction_budgets`, `profile`, `profile_diff`, `call_profile`, `disasm`, `jit_disasm`, `speed_costs`, `thread_scaling`, `cargo_costs`, `pytest_costs`, `compare_costs` or `
|
|
79
|
+
`bpl-tests NAME ARGUMENT...` runs one of these functions from the shell and prints its result: `check_file`, `check_page`, `time_cases`, `instructions`, `instruction_budgets`, `profile`, `profile_diff`, `call_profile`, `disasm`, `jit_disasm`, `speed_costs`, `thread_scaling`, `cargo_costs`, `pytest_costs`, `compare_costs`, `wasm_test` or `wasm_stack`. For example, `bpl-tests time_cases '+/v' --threads 4` times `+/v` on four threads. `time_cases`, `instructions` and `speed_costs` take cases as positional arguments. When a case starts with `-`, put the cases last, after `--`, as in `bpl-tests speed_costs -- '-.×m' '+/v'`. `bpl-tests NAME -h` lists a function's arguments. Called in a kernel, the same functions return Python objects, such as the rows that `compare_costs` reads.
|
|
80
80
|
|
|
81
|
-
On each push, the `Test` workflow logs every native cost in its `test` job and every wasm cost in its `wasm` job. The speed test that writes them times the test-profile build that the budgets limit. `bpl --costs PATH` times a `.bpl` file or a directory of them as the speed test does, and prints each cost with this machine's budget, without failing on a budget. The manual `Speed` workflow runs it on a `dist` build, the profile releases use, on CI's runner, beside its wasm timing. The `Release` workflow runs it on each binary it builds, on GitHub's runners. Their logs show what releases cost, for comparison with the `Test` workflow's. `logged_costs` reads all three workflows' logs. When you review a run, look for costs over their budgets and for costs that changed by more than 20% since the run before. `compare_costs(before, after)` lists both, from two runs' logs or from the rows that `speed_costs` and the other timing functions give. It measures each change on times per item where both runs give them, so a change to `COPIES` doesn't show as one. Every log names its CPU in its copy lines. Study a slowdown until you know its cause only when it appears in more than one CI run, or when two or more similar cases slow by similar amounts. Treat any other slowdown as normal variation.
|
|
81
|
+
On each push, the `Test` workflow logs every native cost in its `test` job and every wasm cost in its `wasm` job. The speed test that writes them times the test-profile build that the budgets limit. `bpl --costs PATH` times a `.bpl` file or a directory of them as the speed test does, and prints each cost with this machine's budget, then each case still over its budget after the retries, marked `over`, without failing on a budget. The manual `Speed` workflow runs it on a `dist` build, the profile releases use, on CI's runner, beside its wasm timing. The `Release` workflow runs it on each binary it builds, on GitHub's runners. Their logs show what releases cost, for comparison with the `Test` workflow's. `logged_costs` reads all three workflows' logs. When you review a run, look for costs over their budgets and for costs that changed by more than 20% since the run before. `compare_costs(before, after)` lists both, from two runs' logs or from the rows that `speed_costs` and the other timing functions give. It measures each change on times per item where both runs give them, so a change to `COPIES` doesn't show as one. Every log names its CPU in its copy lines. Study a slowdown until you know its cause only when it appears in more than one CI run, or when two or more similar cases slow by similar amounts. Treat any other slowdown as normal variation.
|
|
82
82
|
|
|
83
|
-
A budget is a case's expected cost times a multiplier: 2 under 50 copies, 1.5 from 50 to 500, and 1.3 over 500. The multiplier is largest for cheap cases, which vary most between runs. The expected cost starts from a first-principles analysis of what the case should cost. A later expected cost comes from a fuller understanding of the case than that analysis gives, such as a platform difference or a cost the analysis missed, and the case's comment records it. For a case with a known gap, the multiplier applies to today's measured cost, held until the gap is fixed, and the case's comment names the gap. A platform's cost that no run has measured there, such as an estimate for Linux or wasm, keeps twice the estimate until CI measures it. Budgets set before 27 October 2026 are still twice their expected cost, until a survey of logged costs applies the steps to them. A case over its budget needs study, which finds either a speed bug to fix or something the analysis missed. Avoid raising a budget to match a run without understanding why the case costs what it does. Each `⍝⍝:` footer in a BPL speed file holds a native budget for macOS on Apple silicon, then one for Linux on CI's x86 runner, then, in basedpl's file, a wasm budget for Node on CI's x86 runner. One expected cost serves every platform at first, until a specific difference on one platform is found. macOS running matrix products and decompositions on Accelerate, where CI runs rustymath's own kernels, is such a difference. A comment above each case, or above its group, gives the expected cost with its reason, and any platform's own cost with the reason for it. `set_budgets(expected)` applies the multiplier and writes the budgets for the cases that `expected` names, a dict from each case's code to its expected cost. `builds=` limits the change to one platform's column, for a cost that differs there. `estimate=True` keeps twice the cost, for a platform's estimates. A case can appear in more than one section, such as `copies` and `widths`. `sections=` limits the change to the sections it names. `path=` selects a BPL speed file or directory, or one of Rustymath's table-format files, `tests/speed.txt` or `tests/speed_python.txt`.
|
|
83
|
+
A budget is a case's expected cost times a multiplier: 2 under 50 copies, 1.5 from 50 to 500, and 1.3 over 500. The multiplier is largest for cheap cases, which vary most between runs. The expected cost starts from a first-principles analysis of what the case should cost. A later expected cost comes from a fuller understanding of the case than that analysis gives, such as a platform difference or a cost the analysis missed, and the case's comment records it. For a case with a known gap, the multiplier applies to today's measured cost, held until the gap is fixed, and the case's comment names the gap. A platform's cost that no run has measured there, such as an estimate for Linux or wasm, keeps twice the estimate until CI measures it. Budgets set before 27 October 2026 are still twice their expected cost, until a survey of logged costs applies the steps to them. A case over its budget needs study, which finds either a speed bug to fix or something the analysis missed. Avoid raising a budget to match a run without understanding why the case costs what it does. Each `⍝⍝:` footer in a BPL speed file holds a native budget for macOS on Apple silicon, then one for Linux on CI's x86 runner, then, in basedpl's file, a wasm budget for Node on CI's x86 runner. A speed test fails on a budget only on the CPU that the budgets were set for, `rustymath::timing::BUDGET_CPU`: the development Mac's for macOS, and CI's runner's for Linux and wasm. On any other CPU, it prints every cost, then each case still over its budget after the retries, marked `over`, and fails on none. A cost in copies depends on a CPU's caches as well as its speed, as rustymath's DEV.md shows for the 9950X that joined CI's runner pool in October 2026. One expected cost serves every platform at first, until a specific difference on one platform is found. macOS running matrix products and decompositions on Accelerate, where CI runs rustymath's own kernels, is such a difference. A comment above each case, or above its group, gives the expected cost with its reason, and any platform's own cost with the reason for it. `set_budgets(expected)` applies the multiplier and writes the budgets for the cases that `expected` names, a dict from each case's code to its expected cost. `builds=` limits the change to one platform's column, for a cost that differs there. `estimate=True` keeps twice the cost, for a platform's estimates. A case can appear in more than one section, such as `copies` and `widths`. `sections=` limits the change to the sections it names. `path=` selects a BPL speed file or directory, or one of Rustymath's table-format files, `tests/speed.txt` or `tests/speed_python.txt`.
|
|
84
84
|
|
|
85
85
|
`tests/explain.rs`'s `speed_cases_take_their_routes` captures every speed case after warm-up, with its normal setup and size, forcing the result outside timing (`timing::routes`). It also captures each reference case that has route directives, once under ordinary preferences. Within `cargo test`, only this test checks route directives. `check_file` checks a file's directives too, through `reference::check`. Put `⍝⍝⍝ route NAME` above a case to require that fast route to execute. A permitted slow event never excuses a missing required route.
|
|
86
86
|
|
|
@@ -98,7 +98,7 @@ In a Claude Code session, every exhash edit is recorded in `/tmp/exhash-journal/
|
|
|
98
98
|
|
|
99
99
|
Cooperating tools share the machine's timing lock, `/tmp/aai-ws-timing.lock`. Timings take it exclusively. `cargo develop`, `cargo wait`, `cargo wasm`, the reference test, pytest, `check_file`, `check_page`, `bpl-nb` and `wasm_test` without `speed` take it shared. Rust code takes it through `rustymath::timing`. When a kernel calls `time_cases` or native `speed_costs`, the `bpl` worker that the call starts takes the lock, not the kernel. Python code takes it with `fastws.cargo.timing_lock`. Node can't lock a file. `wasm_test` and `speed_costs(wasm=DIR)` hold the lock in Python while Node runs. Other Rust tests and bare Cargo commands do not all take it. After five minutes a timing fails. A build or test run then goes ahead without the lock. After 30 seconds, an exclusive wait prints the processes with the lock file open. The list can include processes that are only waiting for the lock. Never take the lock yourself in a kernel process. Never coordinate timings by hand. A hold that outlasts its call blocks the timings that later calls launch. Native timings run on one thread and request user-interactive macOS QoS, which favours the performance cores. QoS is not CPU affinity.
|
|
100
100
|
|
|
101
|
-
Run `cargo wasm` during development only for browser work. Agents use `cargo wasm --out DIR`, then run the browser tests against that build with `basedpl.bpltests.wasm_test(DIR)` for `tests/wasm.mjs` and `wasm_test(DIR, speed=True)` for `tests/wasm-speed.mjs`.
|
|
101
|
+
Run `cargo wasm` during development only for browser work. Agents use `cargo wasm --out DIR`, then run the browser tests against that build with `basedpl.bpltests.wasm_test(DIR)` for `tests/wasm.mjs` and `wasm_test(DIR, speed=True)` for `tests/wasm-speed.mjs`. On this Mac, that speed test prints every cost and marks each case still over its wasm budget without failing, because those budgets were set for CI's runner. This avoids replacing a module another agent is using. For other edits under `cfg(web)` or `host_web.rs`, check compilation with `cargo cs -p basedpl-wasm --target wasm32-unknown-unknown`, which checks the browser wrapper crate together with basedpl's browser code. `cargo cs` alone compiles neither.
|
|
102
102
|
|
|
103
103
|
### Machine code
|
|
104
104
|
|
|
@@ -128,7 +128,7 @@ Basedpl's crates pass on rustymath's default `native` feature, which rustymath's
|
|
|
128
128
|
|
|
129
129
|
Use published dependency ranges with the lower bounds required by the APIs in use. `Cargo.toml` names a published rustymath range; the `~/aai-ws` Cargo patch selects the sibling checkout locally. CI resolves published crates. Release rustymath changes before relying on them in basedpl CI. Don't edit fastws-generated Cargo patches, `.git/fastws-cargo-key` or workspace configuration, and don't commit the workspace-generated `Cargo.lock`. `meta/` is ignored planning material, not contributor documentation. `[tool.uv] cache-keys` tracks packaging manifests, not Rust sources.
|
|
130
130
|
|
|
131
|
-
`Test` runs native and wasm jobs on pushes to main and pull requests. Native CI runs `cargo test`, editable installation, `cargo develop`, pytest, `bpl-nb` and `nbdev-test`. Python tests use the same test-profile build. The wasm job builds with `cargo wasm --strip` and runs both Node suites. Both jobs run on `ubicloud-standard-4-ubuntu-2604`, which is Ubuntu 26.04 on the `ubicloud-standard-4` VM size that speed budgets are set for.
|
|
131
|
+
`Test` runs native and wasm jobs on pushes to main and pull requests. Native CI runs `cargo test`, editable installation, `cargo develop`, pytest, `bpl-nb` and `nbdev-test`. Python tests use the same test-profile build. The wasm job builds with `cargo wasm --strip` and runs both Node suites. Both jobs run on `ubicloud-standard-4-ubuntu-2604`, which is Ubuntu 26.04 on the `ubicloud-standard-4` VM size that speed budgets are set for. They generate a lockfile before restoring `Swatinem/rust-cache`, making dependency changes part of the cache key. `Speed` is manually triggered. Its `wasm` job restores the `Test` workflow's wasm cache. Its native job builds the `dist` profile without one.
|
|
132
132
|
|
|
133
133
|
`Release` runs separately on `v*` tags on GitHub's runners. It builds dist wheels, sdist, standalone CLI artifacts and the Windows keyboard archive, then publishes Python artifacts and the GitHub release. Its npm job publishes the stripped browser package through trusted publishing and purges jsDelivr's package files to avoid stale unversioned URLs. Release neither runs tests nor waits for the Test workflow.
|
|
134
134
|
|
|
@@ -195,7 +195,7 @@ An assignment compiles to one instruction, which runs `Session::assigned`, the s
|
|
|
195
195
|
|
|
196
196
|
Each item of a bracket literal is a level whose value goes into the literal at once, as `Session::item_result` converts it, and `Session::bracket` builds the literal for `resolve` and for the compiled `Array` instruction alike. A literal with a computed key that isn't constant stays with `resolve`. Each statement keeps the recording of its first evaluation (`Statement::recording`). A later call whose `⍺` is absent where that evaluation's held an argument, or the reverse, resumes the binder where the statement reads `⍺`.
|
|
197
197
|
|
|
198
|
-
Tail calls reuse frames while preserving lexical dependencies. Handlers disable tail reuse. Evaluation depth is capped at 20,000 natively and at
|
|
198
|
+
Tail calls reuse frames while preserving lexical dependencies. Handlers disable tail reuse. Evaluation depth is capped at 20,000 natively and at 244 in the browser, whose stack runs out first. Each call of a function other than a primitive counts toward it, and so do each statement of a dfn other than its last, each group, each run that ends in a dyadic operator and each bracket item, compiled or not. A compiled statement or level counts with `Session::descend` alone, without the interrupt check and stack growth of `Session::nested`. Calls of primitives other than `⍎` don't count, because a primitive never calls back into BPL. They skip the depth, capture and stack bookkeeping, and on numbers a form that applies a kernel to each pair of items runs that kernel at once (`Primitive::scalar_call`). A separate 20,000-frame limit bounds retained lexical frames. Late-bound name resolution stops at 128 levels. These are execution limits, not a sandbox against arbitrarily deep source, arrays or function graphs.
|
|
199
199
|
|
|
200
200
|
## Values, layouts and assignment
|
|
201
201
|
|
|
@@ -327,7 +327,7 @@ Browser `Session.run` streams MIME events and returns a protocol error or null.
|
|
|
327
327
|
|
|
328
328
|
The playground and gallery load released assets from jsDelivr, or local `nbs/playground/` assets on localhost. Shared `bpl.js` owns editors, worker startup and output. Restarting a trapped/stopped worker loses its state. `worker.js`, `page.js` and `gallery.js` own their respective protocols/UI; `nbs/driver.js` documents canvas drawing data. Gallery demos are `.bpl` sources with shared preferences, checked natively by `language_examples`. Browser recursion tests cover both V8 compilers and fail with LIMIT before stack exhaustion.
|
|
329
329
|
|
|
330
|
-
CI's two `wasm` jobs, in the `Test` and `Speed` workflows, take Node's version from `.node-version`. TurboFan's frame sizes vary between V8 versions by enough to decide whether a recursion test overflows. On 28 October 2026, one build's `each` shape needed 0.68 MB of stack to reach 171 levels under Node 24's TurboFan, against 0.62 MB under Node 26's. `tests/wasm.mjs` runs its depth checks with Node's 192 KB reserve plus 80% of the roughly 470 KB that a Chrome worker gives, 0.5575 MB in all. A build whose frames grow into the remaining 20% fails the test. In wasm, `Binder::reduce` and `Bindings::enter` stay out of line, to keep their locals out of the frames that stay on the stack at every level of recursion. To find a frame that grew, print each function's TurboFan frame size, in "stack slots", with `node --no-liftoff --print-wasm-code-function-index=N`.
|
|
330
|
+
CI's two `wasm` jobs, in the `Test` and `Speed` workflows, take Node's version from `.node-version`. TurboFan's frame sizes vary between V8 versions by enough to decide whether a recursion test overflows. On 28 October 2026, one build's `each` shape needed 0.68 MB of stack to reach 171 levels under Node 24's TurboFan, against 0.62 MB under Node 26's. `tests/wasm.mjs` runs its depth checks with Node's 192 KB reserve plus 80% of the roughly 470 KB that a Chrome worker gives, 0.5575 MB in all. A build whose frames grow into the remaining 20% fails the test. CI's x86-64 V8 takes more stack per level than Apple silicon's. The browser's depth limit, `MAX_CALL_DEPTH` in `src/eval.rs`, allows for levels up to 1.48 times their size on Apple silicon. `basedpl.bpltests.wasm_stack(DIR)` measures the stack that each recursion shape takes per level and below its first level. The limit is set from those figures. In wasm, `Binder::reduce` and `Bindings::enter` stay out of line, to keep their locals out of the frames that stay on the stack at every level of recursion. To find a frame that grew, print each function's TurboFan frame size, in "stack slots", with `node --no-liftoff --print-wasm-code-function-index=N`. `wasm_stack` runs each trial in a fresh `node` process whose worker compiles the module, as the test's worker does. Code compiled once and shared across workers gets different frames.
|
|
331
331
|
|
|
332
332
|
## System functions and rendering
|
|
333
333
|
|
|
@@ -8,13 +8,10 @@ fn page_text(path: &Path) -> String {
|
|
|
8
8
|
let mut text = String::new();
|
|
9
9
|
for cell in notebook["cells"].as_array().unwrap() {
|
|
10
10
|
let source = match &cell["source"] {
|
|
11
|
-
serde_json::Value::String(s) => s.clone(),
|
|
12
|
-
lines => lines.as_array().unwrap().iter().map(|s| s.as_str().unwrap()).collect(),
|
|
11
|
+
serde_json::Value::String(s) => s.clone(), lines => lines.as_array().unwrap().iter().map(|s| s.as_str().unwrap()).collect()
|
|
13
12
|
};
|
|
14
13
|
match cell["cell_type"].as_str().unwrap() {
|
|
15
|
-
"markdown" => text.push_str(&source),
|
|
16
|
-
"code" => text.push_str(&format!("```{language}\n{source}\n```")),
|
|
17
|
-
_ => continue,
|
|
14
|
+
"markdown" => text.push_str(&source), "code" => text.push_str(&format!("```{language}\n{source}\n```")), _ => continue
|
|
18
15
|
}
|
|
19
16
|
text.push_str("\n\n");
|
|
20
17
|
}
|
|
@@ -46,10 +43,13 @@ fn main() {
|
|
|
46
43
|
help = help.replace("](../", "](");
|
|
47
44
|
}
|
|
48
45
|
let title = help.lines().find_map(|l| l.strip_prefix("# ")).unwrap_or(stem);
|
|
49
|
-
let aliases: Vec<&str> = if dir == "glyphs" { vec![stem] }
|
|
46
|
+
let aliases: Vec<&str> = if dir == "glyphs" { vec![stem] }
|
|
47
|
+
else if dir == "system" {
|
|
50
48
|
let mut fenced = false;
|
|
51
|
-
help.lines().filter(|line| {
|
|
52
|
-
|
|
49
|
+
help.lines().filter(|line| {
|
|
50
|
+
fenced ^= line.starts_with("```");
|
|
51
|
+
!fenced && line.starts_with('#')
|
|
52
|
+
}).flat_map(|line| line.split('`').skip(1).step_by(2)).filter(|name| name.starts_with(['•', '$'])).collect()
|
|
53
53
|
} else { Vec::new() };
|
|
54
54
|
entries.push(format!("Page {{ name: {name:?}, title: {title:?}, aliases: &{aliases:?}, text: {help:?} }}"));
|
|
55
55
|
}
|
|
@@ -52,6 +52,6 @@ Use actual glyphs in files, pipes and `-e`; backtick names are a [REPL input met
|
|
|
52
52
|
| `bpl --help` | Usage |
|
|
53
53
|
| `bpl --version` | Version |
|
|
54
54
|
| `bpl --worker` | JSON-lines requests with deadlines and interruption |
|
|
55
|
-
| `bpl --costs
|
|
55
|
+
| `bpl --costs PATH` | Times the cases in a speed file in the format of basedpl's `tests/speed/*.bpl`, or in a directory of them, as the speed test does. Prints each cost with this machine's budget, then each case still over its budget after the retries, marked `over`. A case over its budget leaves the exit status at 0 |
|
|
56
56
|
|
|
57
57
|
See [process interfaces](processes.qmd) for machine-readable output. The exit status is 0 on success, 1 after an evaluation or I/O error, and 2 for invalid command arguments. An error doesn't discard the output written before it.
|
|
@@ -66,7 +66,7 @@ n←⊽⁻¹+÷ ⋄ n 1.5 ⍝ 1.416666667
|
|
|
66
66
|
-↢10⁝3 ⍝ ¯7
|
|
67
67
|
```
|
|
68
68
|
|
|
69
|
-
|
|
69
|
+
With two function operands, `↣` and `↢` form monadic hooks. `f↣g Y` is `(f Y)g Y` and `f↢g Y` is `Y f(g Y)`. Two-function trains in BPL are atops. `(3<)↣# Y` is `(3<Y)#Y`. It keeps the items of `Y` that are greater than 3:
|
|
70
70
|
|
|
71
71
|
```bpl
|
|
72
72
|
(3<)↣# 2 7 1 8 ⍝ 7 8
|
|
@@ -2,7 +2,7 @@
|
|
|
2
2
|
|
|
3
3
|
Dyadic operator. Keys: `Alt-,`. Hanzi: `后` (`後`)
|
|
4
4
|
|
|
5
|
-
`f↢g Y` is `Y f(g Y)`. The left argument of `f` is `Y` itself. For `f(g Y)`, use [Atop](jot.qmd), `f∘g`.
|
|
5
|
+
The monadic hook `f↢g Y` is `Y f(g Y)`. The left argument of `f` is `Y` itself. For `f(g Y)`, use [Atop](jot.qmd), `f∘g`.
|
|
6
6
|
|
|
7
7
|
```bpl
|
|
8
8
|
×↢- 5 ⍝ ¯25
|
|
@@ -2,7 +2,7 @@
|
|
|
2
2
|
|
|
3
3
|
Dyadic operator. Keys: `Alt-.`. Hanzi: `先`
|
|
4
4
|
|
|
5
|
-
`f↣g Y` is `(f Y)g Y`, where `f` prepares the left argument of `g`. Operators bind from the left. For example, `3↣<↣#` is `(3↣<)↣#`, which keeps the items greater than 3.
|
|
5
|
+
The monadic hook `f↣g Y` is `(f Y)g Y`, where `f` prepares the left argument of `g`. Operators bind from the left. For example, `3↣<↣#` is `(3↣<)↣#`, which keeps the items greater than 3.
|
|
6
6
|
|
|
7
7
|
```bpl
|
|
8
8
|
-↣+ 5 ⍝ 0
|
|
@@ -14,7 +14,7 @@ Entry: Dyalog diff, example, defining page. Not the reference: read the page bef
|
|
|
14
14
|
|
|
15
15
|
## Language
|
|
16
16
|
|
|
17
|
-
- [Syntax](syntax.html.md): BQN stranding: mark strands with `[]` (per BQN's 5-year retrospective), except spaced literals (step 1) and subjects side by side (step 3); like BQN, a space is semantic. `⁝`
|
|
17
|
+
- [Syntax](syntax.html.md): BQN stranding: mark strands with `[]` (per BQN's 5-year retrospective), except spaced literals (step 1) and subjects side by side (step 3); like BQN, a space is semantic. `⁝` separates tokens and stops literal stranding without ending the run: `h⁝m`≡`h(m)`, `f⍣3⁝0`≡`(f⍣3)0`. 3 steps: 1. spaced literals join, 1 vector: `1 2`, `"ab" "cd"`; even operand+arg: `+/⍠2 3 3 3↕m` has operand `2 3 3 3`, write `+/⍠2 3⁝3 3↕m`. 2. each other run→`(run)`: `a+b × c+d`≡`(a+b)(×)(c+d)`; fn between two runs: space both sides: `a×b + c`≡`(a×b)+c`; `a×b +c`≡`[a×b ; +c]`; `a×b+ c`≡`(a×b+)c`, train `a×(b+c)`; `1+k (⌈/⊢)⌸ a`≡`(1+k)((⌈/⊢)⌸)(a)`. 3. read as APL, right to left; subjects side by side strand: `a b`≡`(a)(b)`≡`[a b]` only if neither touches; `1+a b`≡`(1+a)(b)`→`11 20`, 1 plus both: `1+[a b]`→`11 21`; `n 2⍴0`≡`(n)(2⍴0)`→`[3 [0 0]]`, names never join literal strands: write `[n 2]⍴0`. no needless spaces; join by deleting spaces, not adding parens: `1+k(⌈/⊢)⌸a`. name fn, name arg: `h⁝m` `h[m n]`. name operand, glyph arg: `+/⍠k⍵`. list: `[+/x ; +/x ÷ ≢x]`. name digits: `x_1`; `x1`≡`x(1)`.
|
|
18
18
|
- [Evaluation](evaluation.html.md): run ending in fn: train. `+/÷≢ x` mean; `+/÷≢x`≡`+/÷(≢x)`. `⁝` doesn't group a train: `+/÷≢⁝x` still means the latter. A space before an arg makes the run on its left a train: `10÷√1++/⍠i x` calls dyadic `√`, write `10÷√1++/⍠i⁝x`; `+/,m+⍤2 P` applies the train `+/,m+⍤2` to `P`, write `+/,m(+⍤2)P`.
|
|
19
19
|
- [Composition](composition.html.md): read before writing tests or examples. trains: `f⁝g Y`≡`f(g Y)`, `f⁝g⁝h Y`≡`(f Y)g(h Y)`; `∘` atop, `⍥` over, `↣` `↢` bind left, right; `⍨` self, commute; `⊢` `⊣` stand for args in trains; `⁻¹` inverse; `@` under. worse→better, same results: `0.5×⊢+2÷⊢`→`⊽⁻¹+÷`; `{⍵,1}`→`,↢1`; `×⍨x`→`x²`; `≥⍣3⊢5`→`≥⍣3⁝5`; `+/⍠i⊢m`→`+/⍠i⁝m`. Prefer spaces for top-level groups/subtrains: `+/÷≢ ⊼v`, `mean_square←+/÷≢ ⊼`, `f⁝g⁝h x`. A space groups the whole left run, including any prefix. Nested `()`: try the inner pair first. With only a right arg, prefer `⁝` where a boundary is needed: `f⍣3⁝0`, `(+/÷≢)⍤1⁝2 3⍴⍳6`. With both args, group the derived fn: `10 20(+⍤1)m`. Keep compound operands grouped: `⌊@(10×)x`. Add nothing where tokens already stay distinct: `+/⍤1m`.
|
|
20
20
|
- [Agreement](evaluation.html.md#agreement-and-pervasion): pervasive fns, Each, Rank align leading axes: `[1 2 3⋄4 5 6]+10 20`→`[11 12 13⋄24 25 26]`. length-1 axes expand.
|
|
@@ -774,7 +774,7 @@
|
|
|
774
774
|
"- File reads and `•fetch` are synchronous `XMLHttpRequest`s, relative to the URL you pass to `configure`.\n",
|
|
775
775
|
"- `•r` uses JavaScript's `RegExp`, with its syntax and replacement templates.\n",
|
|
776
776
|
"- `•date⁻¹` has no `locale` option.\n",
|
|
777
|
-
"-
|
|
777
|
+
"- Evaluation nests at most 244 levels deep. A call of a function other than a primitive, a group and each statement of a dfn before its last each count a level. A recursive dfn such as `{⍵=0?0;1+∇ ⍵-1}` takes two levels a call, for 120 calls. A deeper evaluation is a `LIMIT` error.\n",
|
|
778
778
|
"- Fused computations run on kernels built into the module. The browser build compiles no machine code at run time, where native builds compile large fused computations, as [Performance](performance.qmd) describes.\n",
|
|
779
779
|
"- Every call runs on one thread, whatever the `threads` setting."
|
|
780
780
|
]
|
|
@@ -1,15 +1,16 @@
|
|
|
1
|
-
use basedpl::{
|
|
1
|
+
use basedpl::{
|
|
2
|
+
EvalOptions, Evaluation, Function, InterruptHandle, Number, Operand, Session, Value, buffer::{Buffer, DType, FloatWidth, IntBuf, View}, keyed::{self, Texts}
|
|
3
|
+
};
|
|
2
4
|
use num_bigint::BigInt;
|
|
3
5
|
use num_rational::BigRational;
|
|
4
6
|
use pyo3::{
|
|
5
|
-
exceptions::{PyBufferError, PyKeyError, PyTypeError, PyValueError}, ffi, prelude::*, sync::PyOnceLock,
|
|
6
|
-
types::{PyBool, PyBoolMethods, PyComplex, PyComplexMethods, PyDict, PyFloat, PyInt, PyList, PyRange, PySlice, PyString, PyTuple, PyType}
|
|
7
|
+
IntoPyObjectExt, exceptions::{PyBufferError, PyKeyError, PyTypeError, PyValueError}, ffi, prelude::*, sync::PyOnceLock,
|
|
8
|
+
types::{PyBool, PyBoolMethods, PyComplex, PyComplexMethods, PyDict, PyFloat, PyInt, PyList, PyRange, PySlice, PyString, PyTuple, PyType}
|
|
7
9
|
};
|
|
8
10
|
use rustymath::{layout::Keys, structure::IndexPart};
|
|
9
11
|
use rustymath_buffer::walk::{Collect, walked, within};
|
|
10
|
-
use std::{ffi::c_int, sync::{atomic::{AtomicPtr, Ordering}
|
|
12
|
+
use std::{ffi::c_int, sync::{Arc, Mutex, atomic::{AtomicPtr, Ordering}}};
|
|
11
13
|
|
|
12
|
-
/// The memory guard stops a runaway computation at half of physical memory.
|
|
13
14
|
#[global_allocator]
|
|
14
15
|
static GUARD: rustymath::memory::Guard = rustymath::memory::Guard::new();
|
|
15
16
|
|
|
@@ -72,7 +73,10 @@ fn numpy_types(py: Python<'_>) -> PyResult<Option<&'static (Py<PyType>, Py<PyTyp
|
|
|
72
73
|
/// The value of a NumPy scalar or array, or `None` for any other object. A scalar of a numeric dtype keeps its dtype. Arrays of `str` or objects convert item by item, as do unsigned 64-bit values beyond `int64`, which stay exact.
|
|
73
74
|
fn numpy(obj: &Bound<'_, PyAny>, items: &mut Items) -> PyResult<Option<Value>> {
|
|
74
75
|
let Some((generic, ndarray)) = numpy_types(obj.py())? else { return Ok(None) };
|
|
75
|
-
let dtype = |o: &Bound<'_, PyAny>| -> PyResult<(char, usize)> {
|
|
76
|
+
let dtype = |o: &Bound<'_, PyAny>| -> PyResult<(char, usize)> {
|
|
77
|
+
let d = o.getattr("dtype")?;
|
|
78
|
+
Ok((d.getattr("kind")?.extract()?, d.getattr("itemsize")?.extract()?))
|
|
79
|
+
};
|
|
76
80
|
if obj.is_instance(generic.bind(obj.py()))? {
|
|
77
81
|
if !"biufc".contains(dtype(obj)?.0) { return element(&obj.call_method0("item")?, items).map(Some); }
|
|
78
82
|
let value = numeric(&obj.call_method0("__array__")?)?.ok_or_else(|| PyTypeError::new_err("unsupported NumPy dtype"))?;
|
|
@@ -82,7 +86,9 @@ fn numpy(obj: &Bound<'_, PyAny>, items: &mut Items) -> PyResult<Option<Value>> {
|
|
|
82
86
|
let (mut kind, size) = dtype(obj)?;
|
|
83
87
|
if !"biufcUO".contains(kind) || kind == 'f' && size > 8 || kind == 'c' && size > 16 { return Err(PyTypeError::new_err("unsupported NumPy dtype")); }
|
|
84
88
|
let mut obj = obj.clone();
|
|
85
|
-
if kind == 'u' && size == 8 && obj.getattr("size")?.extract::<usize>()? > 0 && obj.call_method0("max")?.ge(1u64 << 63)? {
|
|
89
|
+
if kind == 'u' && size == 8 && obj.getattr("size")?.extract::<usize>()? > 0 && obj.call_method0("max")?.ge(1u64 << 63)? {
|
|
90
|
+
(obj, kind) = (obj.call_method1("astype", ("O",))?, 'O');
|
|
91
|
+
}
|
|
86
92
|
if !"UO".contains(kind) { return numeric(&obj); }
|
|
87
93
|
let data = obj.call_method0("ravel")?.call_method0("tolist")?.try_iter()?.map(|o| element(&o?, items)).collect::<PyResult<Vec<_>>>()?;
|
|
88
94
|
let prototype = if kind == 'U' { Value::Character(' ') } else { zero(items.width) };
|
|
@@ -170,9 +176,7 @@ fn array(py: Python<'_>, a: &Value) -> PyResult<Py<PyAny>> {
|
|
|
170
176
|
#[pyclass(name = "_Array")]
|
|
171
177
|
struct PyArray { inner: Option<Value> }
|
|
172
178
|
|
|
173
|
-
impl From<Value> for PyArray {
|
|
174
|
-
fn from(inner: Value) -> Self { Self { inner: Some(inner) } }
|
|
175
|
-
}
|
|
179
|
+
impl From<Value> for PyArray { fn from(inner: Value) -> Self { Self { inner: Some(inner) } } }
|
|
176
180
|
|
|
177
181
|
impl PyArray {
|
|
178
182
|
/// The value, or an error once `own` has moved its items.
|
|
@@ -193,22 +197,27 @@ fn index_error(error: basedpl::Error) -> PyErr {
|
|
|
193
197
|
fn selection(value: &Value, index: &Bound<'_, PyAny>, convert: &Bound<'_, PyAny>) -> PyResult<IndexSelection> {
|
|
194
198
|
let converted = |obj: &Bound<'_, PyAny>| -> PyResult<Value> { Ok(convert.call1((obj,))?.extract::<PyRef<'_, PyArray>>()?.value()?.clone()) };
|
|
195
199
|
let parts = rustymath_buffer::selection::parts(index)?;
|
|
196
|
-
let values = parts.iter().map(|p| p.as_ref().try_map(|p| if rustymath_buffer::selection::native(p) { Ok(None) } else { converted(p).map(Some) }))
|
|
197
|
-
|
|
198
|
-
|
|
199
|
-
|
|
200
|
-
|
|
201
|
-
|
|
202
|
-
|
|
203
|
-
|
|
204
|
-
|
|
200
|
+
let values = parts.iter().map(|p| p.as_ref().try_map(|p| if rustymath_buffer::selection::native(p) { Ok(None) } else { converted(p).map(Some) }))
|
|
201
|
+
.collect::<PyResult<Vec<_>>>()?;
|
|
202
|
+
if let Some(IndexPart::Axis(Some(first))) = values.iter().find(|p| matches!(p, IndexPart::Axis(_) | IndexPart::All))
|
|
203
|
+
&& value.picks_paths(first).map_err(index_error)? {
|
|
204
|
+
let paths = parts.iter().zip(values).map(|(p, v)| {
|
|
205
|
+
v.try_map(|v| match (p.axis(), v) {
|
|
206
|
+
(_, Some(v)) if !matches!(v.view(), Some(View::Booleans(_))) => Ok(v), (Some(p), None) if !p.is_instance_of::<PySlice>() => converted(p),
|
|
207
|
+
_ => Err(PyTypeError::new_err("an index with pick paths takes positions, keys, full slices, None and ..., not partial slices or masks"))
|
|
208
|
+
})
|
|
209
|
+
}).collect::<PyResult<Vec<_>>>()?;
|
|
210
|
+
return Ok(IndexSelection::Rich(paths));
|
|
211
|
+
}
|
|
205
212
|
let layout = value.core_layout();
|
|
206
213
|
value.planned(&values, |i, axis| {
|
|
207
214
|
let (IndexPart::Axis(p), IndexPart::Axis(v)) = (&parts[i], &values[i]) else { unreachable!("planned decodes axis parts") };
|
|
208
215
|
match v {
|
|
209
216
|
None => Some(rustymath_buffer::selection::selector(layout, axis, p, || unreachable!("a native selector decodes without converting"))),
|
|
210
217
|
Some(v) => match v.view() {
|
|
211
|
-
Some(View::Booleans(_)) =>
|
|
218
|
+
Some(View::Booleans(_)) => {
|
|
219
|
+
Some(rustymath_buffer::selection::array_selector(layout, axis, v.core_layout().clone(), std::sync::Arc::new(v.clone())))
|
|
220
|
+
}
|
|
212
221
|
_ => None,
|
|
213
222
|
},
|
|
214
223
|
}
|
|
@@ -231,8 +240,7 @@ impl Owned {
|
|
|
231
240
|
// SAFETY: `bool` and `u8`, and `char` and `i32`, have the same size and alignment, and every `bool` is a valid `u8` and every `char` a valid `i32`.
|
|
232
241
|
let mut items = match items {
|
|
233
242
|
Buffer::Booleans(v) => Buffer::Integers(IntBuf::U8(unsafe { retyped(v) })),
|
|
234
|
-
Buffer::Characters(v) => Buffer::Integers(IntBuf::I32(unsafe { retyped(v) })),
|
|
235
|
-
items => items,
|
|
243
|
+
Buffer::Characters(v) => Buffer::Integers(IntBuf::I32(unsafe { retyped(v) })), items => items
|
|
236
244
|
};
|
|
237
245
|
let data = AtomicPtr::new(items.as_mut_ptr());
|
|
238
246
|
let (shape, strides) = rustymath_buffer::geometry(dtype, shape);
|
|
@@ -256,7 +264,10 @@ impl Owned {
|
|
|
256
264
|
unsafe fn __getbuffer__(slf: Bound<'_, Self>, view: *mut ffi::Py_buffer, flags: c_int) -> PyResult<()> {
|
|
257
265
|
let this = slf.get();
|
|
258
266
|
// SAFETY: `data` addresses `items`, which the object owns with the shape and strides and never changes, and the export holds the object.
|
|
259
|
-
unsafe {
|
|
267
|
+
unsafe {
|
|
268
|
+
rustymath_buffer::export(
|
|
269
|
+
view, flags, slf.as_any(), this.dtype, this.data.load(Ordering::Relaxed), this.items.len(), &this.shape, &this.strides, false)
|
|
270
|
+
}
|
|
260
271
|
}
|
|
261
272
|
}
|
|
262
273
|
|
|
@@ -272,7 +283,8 @@ impl PyArray {
|
|
|
272
283
|
}
|
|
273
284
|
fn with_names(&self, names: Option<Vec<Option<String>>>) -> PyResult<Self> {
|
|
274
285
|
let value = self.value()?;
|
|
275
|
-
let layout = value.core_layout().clone().with_names(names.unwrap_or_default().into_iter().map(|n| n.map(Arc::from)).collect())
|
|
286
|
+
let layout = value.core_layout().clone().with_names(names.unwrap_or_default().into_iter().map(|n| n.map(Arc::from)).collect())
|
|
287
|
+
.map_err(rustymath_buffer::py_err)?;
|
|
276
288
|
Ok(value.clone().with_core_layout(layout).map_err(|k| PyValueError::new_err(k.to_string()))?.into())
|
|
277
289
|
}
|
|
278
290
|
#[getter]
|
|
@@ -284,10 +296,12 @@ impl PyArray {
|
|
|
284
296
|
Ok(value.clone().with_core_layout(layout).map_err(|k| PyValueError::new_err(k.to_string()))?.into())
|
|
285
297
|
}
|
|
286
298
|
fn __rustymath_array__(&self, py: Python<'_>) -> PyResult<(Py<PyAny>, Py<PyAny>, Vec<Option<String>>)> {
|
|
287
|
-
if !self.exports()? {
|
|
299
|
+
if !self.exports()? {
|
|
300
|
+
return Err(PyTypeError::new_err("only ordinary compact values can cross to Rustymath; convert exact or mixed values explicitly"));
|
|
301
|
+
}
|
|
288
302
|
let value = self.value()?;
|
|
289
303
|
let owner = match value.widened_for_export() {
|
|
290
|
-
Some(items) => Owned::new(items, value.shape()).into_py_any(py)?, None => self.copy()?.into_py_any(py)
|
|
304
|
+
Some(items) => Owned::new(items, value.shape()).into_py_any(py)?, None => self.copy()?.into_py_any(py)?
|
|
291
305
|
};
|
|
292
306
|
Ok((owner, self.keys()?.into_py_any(py)?, self.names()?))
|
|
293
307
|
}
|
|
@@ -324,7 +338,11 @@ impl PyArray {
|
|
|
324
338
|
let (items, offset, strides) = lendable(&value).ok_or_else(|| PyBufferError::new_err("an array of mixed or exact items has no buffer"))?;
|
|
325
339
|
let dtype = items.view().expect("lendable items are compact").dtype();
|
|
326
340
|
// SAFETY: the offset and strides reach only items of `items`, as the copy's map does. An atom holds its items inline, so the pointer comes from the copy that the export holds.
|
|
327
|
-
unsafe {
|
|
341
|
+
unsafe {
|
|
342
|
+
rustymath_buffer::lend(
|
|
343
|
+
view, flags, slf.py(), items, |v| v.view().expect("the items are compact").as_ptr().add(offset * dtype.bytes()), dtype, value.shape(),
|
|
344
|
+
strides.as_deref())
|
|
345
|
+
}
|
|
328
346
|
}
|
|
329
347
|
/// The items for NumPy to view writably: moved with no copy where nothing else holds them, and copied otherwise. Integers that static kinds store narrower than their kind come widened to it (`Value::widened_for_export`). For items it can't move, such as exact numbers, it still empties the array and returns `None`, so the caller copies them first.
|
|
330
348
|
fn own(&mut self) -> PyResult<Option<Owned>> {
|
|
@@ -364,7 +382,9 @@ impl PyArray {
|
|
|
364
382
|
let (value, values, selection) = (self.value()?.clone(), values.value()?.clone(), selection(self.value()?, index, convert)?);
|
|
365
383
|
py.detach(move || {
|
|
366
384
|
let _turn = Session::current();
|
|
367
|
-
match selection {
|
|
385
|
+
match selection {
|
|
386
|
+
IndexSelection::Planned(plan, atom) => value.assigned(plan, atom, &values), IndexSelection::Rich(parts) => value.assigned_parts(&parts, &values)
|
|
387
|
+
}
|
|
368
388
|
}).map(Self::from).map_err(index_error)
|
|
369
389
|
}
|
|
370
390
|
/// Whether this array holds `query`, a native array or a `str`. A one-character string is an atom. A longer one is one item where the major cells are items, as a list of strings holds them, and otherwise a character vector, which matches a row of characters.
|
|
@@ -372,12 +392,17 @@ impl PyArray {
|
|
|
372
392
|
let value = self.value()?.clone();
|
|
373
393
|
let query = match query.cast::<PyString>() {
|
|
374
394
|
Ok(s) => match text(s, &mut Texts::default())? {
|
|
375
|
-
chars @ Value::Array(_) if value.shape().len() == 1 =>
|
|
395
|
+
chars @ Value::Array(_) if value.shape().len() == 1 => {
|
|
396
|
+
Value::from_parts(Vec::<usize>::new(), vec![chars], Value::Character(' ')).map_err(value_error)?
|
|
397
|
+
}
|
|
376
398
|
item => item,
|
|
377
399
|
},
|
|
378
400
|
Err(_) => query.extract::<PyRef<'_, PyArray>>()?.value()?.clone(),
|
|
379
401
|
};
|
|
380
|
-
py.detach(move || {
|
|
402
|
+
py.detach(move || {
|
|
403
|
+
let _turn = Session::current();
|
|
404
|
+
value.contains(&query)
|
|
405
|
+
}).map_err(|e| match &e.kind {
|
|
381
406
|
basedpl::ErrorKind::Array(e) if e.kind == rustymath::error::Kind::Rank => PyTypeError::new_err(e.message.to_string()),
|
|
382
407
|
_ => PyValueError::new_err(e.to_string()),
|
|
383
408
|
})
|
|
@@ -505,7 +530,10 @@ impl PySession {
|
|
|
505
530
|
fn make_interactive(&self) { Session::make_interactive() }
|
|
506
531
|
#[pyo3(signature = (prefix="", classes=vec![2, 3, 4], complete=false))]
|
|
507
532
|
fn names(&self, py: Python<'_>, prefix: &str, classes: Vec<i64>, complete: bool) -> Vec<String> {
|
|
508
|
-
py.detach(|| {
|
|
533
|
+
py.detach(|| {
|
|
534
|
+
let session = Session::current();
|
|
535
|
+
if complete { session.complete(prefix) } else { session.name_list(&classes, prefix) }
|
|
536
|
+
})
|
|
509
537
|
}
|
|
510
538
|
fn interrupt(&self) { if let Some(active) = &*self.active.lock().unwrap() { active.interrupt(); } }
|
|
511
539
|
}
|
|
@@ -242,7 +242,7 @@ _SPEED = _ROOT/'tests'/'speed'
|
|
|
242
242
|
_BUDGETS = ('macos', 'linux', 'wasm')
|
|
243
243
|
|
|
244
244
|
def _platform():
|
|
245
|
-
"This machine's platform, whose native budget column
|
|
245
|
+
"This machine's platform, whose native budget column its speed tests compare costs with, as `rustymath::timing::NATIVE_COLUMN` picks it, or `None` on a machine with no budgets, such as Linux on Arm."
|
|
246
246
|
import rustymath
|
|
247
247
|
column = rustymath.timing().column
|
|
248
248
|
return None if column is None else _BUDGETS[column]
|
|
@@ -657,6 +657,10 @@ def _node():
|
|
|
657
657
|
if not ran.returncode and want in (None, ran.stdout.lstrip('v').split('.')[0]): return node
|
|
658
658
|
raise ValueError(f'`.node-version` names Node {want}, which this machine lacks: `brew install node@{want}` installs it' if want else "`node` on `PATH` doesn't run")
|
|
659
659
|
|
|
660
|
+
def _wasm_env(wasm):
|
|
661
|
+
"The environment in which `tests/wasm-pkg.mjs` loads the browser build that `cargo wasm --out` wrote to `wasm`."
|
|
662
|
+
return os.environ | dict(BASEDPL_WASM=str(Path(wasm).expanduser()))
|
|
663
|
+
|
|
660
664
|
def speed_costs(
|
|
661
665
|
wasm:Path=None, # A directory that `cargo wasm --out` wrote; None: a fresh native worker
|
|
662
666
|
native=False, # Time Rustymath's own linear algebra instead of Accelerate
|
|
@@ -678,7 +682,7 @@ def speed_costs(
|
|
|
678
682
|
f'process.stdout.write(JSON.stringify(costs(JSON.parse(readFileSync(0, "utf8")), {json.dumps(cases)}, {json.dumps(prefs)})))')
|
|
679
683
|
with timing_lock(exclusive=True):
|
|
680
684
|
p = subprocess.run([_node(), '--no-liftoff', '--input-type=module', '-e', script], input=json.dumps(files), capture_output=True,
|
|
681
|
-
text=True, timeout=timeout, env=
|
|
685
|
+
text=True, timeout=timeout, env=_wasm_env(wasm))
|
|
682
686
|
if p.returncode: raise ValueError(p.stderr)
|
|
683
687
|
if p.stderr: sys.stderr.write(p.stderr)
|
|
684
688
|
rows = json.loads(p.stdout)
|
|
@@ -789,11 +793,42 @@ def wasm_test(
|
|
|
789
793
|
test = _SPEED.with_name('wasm-speed.mjs' if speed else 'wasm.mjs')
|
|
790
794
|
with timing_lock(exclusive=speed):
|
|
791
795
|
p = subprocess.run([_node(), *(['--no-liftoff'] if speed else []), '--test', str(test)], cwd=_ROOT, capture_output=True,
|
|
792
|
-
text=True, timeout=timeout, env=
|
|
796
|
+
text=True, timeout=timeout, env=_wasm_env(wasm) | (dict(BASEDPL_COSTS='1') if speed else {}))
|
|
793
797
|
report = Report(p.stdout + p.stderr)
|
|
794
798
|
if p.returncode: raise ValueError(f'\n{report}')
|
|
795
799
|
return report
|
|
796
800
|
|
|
801
|
+
class StackUse(list):
|
|
802
|
+
"Each recursion shape's worker stack per level and below its first level, in KB, as `(shape, per_level, base)` rows."
|
|
803
|
+
def __repr__(self): return '\n'.join(f'{per:6.2f} KB a level {base:7.1f} KB base {shape}' for shape, per, base in self)
|
|
804
|
+
|
|
805
|
+
def wasm_stack(
|
|
806
|
+
wasm, # A directory that `cargo wasm --out` wrote
|
|
807
|
+
):
|
|
808
|
+
"The worker stack, in KB, that each recursion shape of `tests/wasm.mjs` takes per level and below its first level, under V8's TurboFan alone (`--no-liftoff`). The browser's depth limit, `MAX_CALL_DEPTH` in `src/eval.rs`, is set from these. For each shape, it bisects to 1 KB the least stack at which the shape's recursion reaches the test's depth, and half that depth. Each trial runs `tests/wasm.mjs depths` in a fresh `node` process. Its worker compiles the module, as the test's worker does. The bisections run at once, one for each core, with the machine's timing lock held shared."
|
|
809
|
+
node, test, env, most = _node(), _SPEED.with_name('wasm.mjs'), _wasm_env(wasm), 2048
|
|
810
|
+
def depths(kb, *shape):
|
|
811
|
+
"What `tests/wasm.mjs depths` gives with `kb` KB of worker stack, for every shape or for `shape`, a name and a depth: each shape's depth, the depth it reached, and what deeper recursion gave. `None` where the stack is too small for the worker to load the module."
|
|
812
|
+
p = subprocess.run([node, '--no-liftoff', test, 'depths', str(kb), *map(str, shape)], cwd=_ROOT, capture_output=True, text=True, env=env)
|
|
813
|
+
if p.returncode and 'RangeError: Maximum call stack size exceeded' in p.stderr: return None
|
|
814
|
+
if p.returncode: raise ValueError(p.stderr)
|
|
815
|
+
return json.loads(p.stdout)
|
|
816
|
+
def least(job):
|
|
817
|
+
"The least worker stack, in KB, at which shape `name` reaches depth `n`, searched from the 192 KB that Node keeps in reserve."
|
|
818
|
+
(name, n), lo, hi = job, 192, most
|
|
819
|
+
while hi - lo > 1:
|
|
820
|
+
mid = (lo+hi)//2
|
|
821
|
+
if (got := depths(mid, name, n)) and got[name][1] == n: hi = mid
|
|
822
|
+
else: lo = mid
|
|
823
|
+
return hi
|
|
824
|
+
with timing_lock():
|
|
825
|
+
full = depths(most)
|
|
826
|
+
if short := [name for name, (n, reached, *_) in full.items() if reached != n]: raise ValueError(f'With {most} KB of stack, {", ".join(short)} fall short of their depths')
|
|
827
|
+
jobs = [(name, d) for name, (n, *_) in full.items() for d in (n//2, n)]
|
|
828
|
+
with ThreadPoolExecutor(os.cpu_count()) as pool: kb = dict(zip(jobs, pool.map(least, jobs)))
|
|
829
|
+
per = {name: (kb[name, n] - kb[name, n//2])/(n - n//2) for name, (n, *_) in full.items()}
|
|
830
|
+
return StackUse((name, per[name], kb[name, n] - per[name]*n) for name, (n, *_) in full.items())
|
|
831
|
+
|
|
797
832
|
def _timed_build(name, cmd, env):
|
|
798
833
|
"The seconds that the build `cmd` takes in basedpl's root with `env` added, holding the timing lock shared. A failed build raises `ValueError` with its output."
|
|
799
834
|
start = time.monotonic()
|
|
@@ -984,7 +1019,7 @@ def prerelease(
|
|
|
984
1019
|
return result
|
|
985
1020
|
|
|
986
1021
|
|
|
987
|
-
_COMMANDS = 'check_file check_page time_cases instructions instruction_budgets profile profile_diff call_profile disasm jit_disasm speed_costs thread_scaling cargo_costs pytest_costs run_pytest compare_costs wasm_test target_builds prerelease'.split()
|
|
1022
|
+
_COMMANDS = 'check_file check_page time_cases instructions instruction_budgets profile profile_diff call_profile disasm jit_disasm speed_costs thread_scaling cargo_costs pytest_costs run_pytest compare_costs wasm_test wasm_stack target_builds prerelease'.split()
|
|
988
1023
|
|
|
989
1024
|
def main():
|
|
990
1025
|
"Run a function in `_COMMANDS` from the shell, as `bpl-tests NAME ARGUMENT...`, and print its result. `bpl-tests NAME -h` lists NAME's arguments."
|
|
@@ -1,4 +1,4 @@
|
|
|
1
|
-
use crate::{array::{
|
|
1
|
+
use crate::{ErrorKind, Value, array::{Layout, generated_len}};
|
|
2
2
|
use rustymath::{agreement::Mapping, error::Error, gather::Steps};
|
|
3
3
|
use std::borrow::Cow;
|
|
4
4
|
|
|
@@ -47,7 +47,9 @@ impl Agree for Agreement {
|
|
|
47
47
|
}
|
|
48
48
|
|
|
49
49
|
/// `core`, the pairing of `left` and `right`, with the renderer their layouts agree on.
|
|
50
|
-
fn with_renderer(core: Result<Core, Error>, left: &Layout, right: &Layout) -> Result<Agreement, ErrorKind> {
|
|
50
|
+
fn with_renderer(core: Result<Core, Error>, left: &Layout, right: &Layout) -> Result<Agreement, ErrorKind> {
|
|
51
|
+
Ok(core?.map_layout(|core| Layout::paired(core, left, right)))
|
|
52
|
+
}
|
|
51
53
|
|
|
52
54
|
/// The tangent `t` laid out as `layout`, its primal's: repeated where it was smaller, as broadcasting repeats it. A linear operation, whose transpose sums back. A dual keeps the smaller tangent that a pervasive rule gives, which later calls broadcast as they go, and only code that copies items expands it (`Dual::full_tangent`).
|
|
53
55
|
pub(crate) fn expand(t: &Value, layout: &crate::array::Layout) -> Value {
|
|
@@ -66,10 +68,7 @@ pub(crate) fn broadcast_tables(mapping: &Mapping, frame: &[usize], cell: &[usize
|
|
|
66
68
|
pub(crate) fn laid_cells(value: &Value, rank: usize, mapping: &Mapping, frame: &Layout) -> Result<Value, ErrorKind> {
|
|
67
69
|
let n = value.shape().len();
|
|
68
70
|
let layout = frame.concat(&value.layout().axes(n - rank..n));
|
|
69
|
-
if value.is_atom() {
|
|
70
|
-
generated_len(layout.shape())?;
|
|
71
|
-
return Ok(layout.filled(value));
|
|
72
|
-
}
|
|
71
|
+
if value.is_atom() { generated_len(layout.shape())?; return Ok(layout.filled(value)); }
|
|
73
72
|
// A value whose frame is `frame` already holds each cell where the mapping reads it.
|
|
74
73
|
if *value.layout().axes(0..n - rank) == **frame { return value.relaid(layout); }
|
|
75
74
|
crate::primitive::copied(value, layout, &mapping.with_steps(0, broadcast_tables(mapping, frame.shape(), &value.shape()[n - rank..])))
|