Repository navigation
Add superscript Rank notation and improve native array execution - #142
Merged
Merged
Conversation
Use function superscripts for Rank and bare ⁻ for inverses, with matching parser/editor support and migrated documentation, notebooks and library examples. Validate captured layouts before replaying differentiated programs. Reuse block scratch and shared Mix assembly, and floor dynamic reals directly into the selected integer width. Preserve test inventories while reducing setup and diagnostic overhead. See Scripts, Composition, Autodiff and Development.
Keep the full fixture catalogue available while selecting execution by code or exact file/section/code identity. Width comparisons can read the exact section even when that section has no selected row. Reuse exact selection for retries instead of measuring extra sections and filtering afterward. Preserve the three-row uncompiled guard and all existing budgets.
Update displayed inverse spelling, unseeded examples and native image outputs in the distributions, basics, web and XML notebooks. Notebook sources, assertions, IDs and metadata remain unchanged.
Share native map input/output handling with rustymath and write directly into stage destinations. Validate changed capture descriptions on replay and preserve batched reads. Move reference and timing support into testkit. Run correctness, timing, instruction and allocation checks through nextest. Keep prerelease on the normal profile without duplicating native timing suites, and exclude performance measurement from CI. DEV.md documents the shared kernel boundary, emitted-code checks and operator-level performance measurement.
Construct prerelease steps without parsing new fixture syntax through the previously installed extension. Discover browser timing batches in the fresh wasm test process after development installation, retaining exact selectors, per-batch execution limits and complete failure reporting.
Keep native instruction-only cases out of browser timing batches. Stream each completed batch report through the existing CLI before advancing, preserving the output if a later batch times out or is cancelled. Retain exact selectors and per-batch limits.
Keep the outputs from the completed Mac prerelease run. Only the three random draw and deal result lines change.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Use function superscripts for Rank and bare ⁻ for inverse across the parser, Python, editor and documentation. Validate input and captured layouts before AD replay. Share native map input/output handling with rustymath. Separate image, JIT and testkit build layers, and include correctness, timing, instruction and allocation checks in default tests.