hydra-jvm (empty) → 0.17.0

raw patch · 5 files changed

+1664/−0 lines, 5 filesdep +basedep +hydra-kernel

Dependencies added: base, hydra-kernel

Files

• CHANGELOG.md +1372/−0
• LICENSE +201/−0
• NOTICE +3/−0
• hydra-jvm.cabal +38/−0
• src/main/haskell/Hydra/Jvm/Serde.hs +50/−0

@@ -0,0 +1,1372 @@+# Hydra changelog++All notable changes to the Hydra project are documented in this file.++This changelog tracks changes across all Hydra implementations+(Haskell, Java, Python, Scala, TypeScript, Clojure, Common Lisp, Emacs Lisp, Scheme)+and supporting infrastructure.++The format is inspired by [Keep a Changelog](https://keepachangelog.com/en/1.0.0/)+and [Apache TinkerPop](https://github.com/apache/tinkerpop/blob/master/CHANGELOG.asciidoc).++Pre-1.0 versions are listed below.+Untagged versions correspond to in-repository version bumps that were not pushed as Git tags or Hackage releases;+they are documented here for completeness.++---++## [0.17.0] - 2026-07-01++Major themes: first-class support for effects and I/O in the kernel, TypeScript and+Scala reaching publishable parity with the other hosts, and continued maturation of+the published-host build model.++### Highlights++- **Effects and I/O** ([#286](https://github.com/CategoricalData/hydra/issues/286),+  [#288](https://github.com/CategoricalData/hydra/issues/288),+  [#494](https://github.com/CategoricalData/hydra/issues/494)): a new `effect` variant+  on `hydra.core.Type` for representing effectful computations; new kernel modules+  `hydra.file` (`FilePath`, `FileType`, `FileStatus`, `FileExtension`), `hydra.time`+  (`Timespec`), and `hydra.error.file`; new primitive libraries `hydra.lib.effects`,+  `hydra.lib.files`, and `hydra.lib.text` with byte-level I/O; and a new+  `EffectfulTestCase` mechanism so the common test suite can exercise effectful+  primitives across hosts.+- **TypeScript is a complete implementation**+  ([#462](https://github.com/CategoricalData/hydra/issues/462)): self-hosts and passes+  the common test suite, with a per-package npm publish pipeline+  ([#492](https://github.com/CategoricalData/hydra/issues/492)).+- **Scala publishing**: per-package Maven Central publish infrastructure+  ([#491](https://github.com/CategoricalData/hydra/issues/491)).+- **Published-host build model**: published mode now compiles the kernel from the+  co-generated `dist/haskell` rather than Hackage, removing the need for a+  `haskell=local` shim ([#500](https://github.com/CategoricalData/hydra/issues/500));+  the oil-and-water principle and cold-build workflow for `dist/haskell` are documented+  ([#376](https://github.com/CategoricalData/hydra/issues/376)).+- **Kernel hygiene**: `uniqueLabel` (apostrophe suffixes) replaced by `chooseUniqueLabel`+  (numeric suffixes) in `hydra.names`+  ([#436](https://github.com/CategoricalData/hydra/issues/436)).++### Implementations++Eight hosts self-host and pass the common test suite across five families: Haskell,+Java, Python, Scala, and the four Lisp dialects (Clojure, Common Lisp, Scheme, Emacs+Lisp), with TypeScript now joining as a complete host/target. Go remains a "head bud"+(the coder still has emission bugs and the head's runtime is largely placeholder; it+does not yet host the test suite).++### New Features++- Translingual cross-reference convention for `doc` strings: doc-escape tags render to+  each host's native doc syntax (Javadoc, Haddock, Sphinx, TSDoc)+  ([#433](https://github.com/CategoricalData/hydra/issues/433)).+- Neo4j JSON-artifact validation demo, exercised across Java, Python, Haskell, and+  TypeScript ([#510](https://github.com/CategoricalData/hydra/issues/510)).++### Improvements++- The Java/Maven publish set now covers every official Hydra target language, one coder+  package per implementation family+  ([#468](https://github.com/CategoricalData/hydra/issues/468)).+- JVM Maven artifact groups are bifurcated to `net.fortytwo.hydra.java` and+  `net.fortytwo.hydra.scala`, so the Java and Scala (and future Clojure) publish sets+  never collide on coordinates+  ([#519](https://github.com/CategoricalData/hydra/issues/519)).+- Host-specific third-party integrations (the former `bindings/` tree) are folded into+  `overlay/`, unifying hand-written host-native source under one tree+  ([#511](https://github.com/CategoricalData/hydra/issues/511)).+- The coder de-parameterizes context and error types, simplifying generated coder+  signatures ([#518](https://github.com/CategoricalData/hydra/issues/518)).+- TypeScript-to-Java generation now emits the `hydra.lib.*` aggregate namespace+  interfaces (`Maps`, `Strings`, etc.)+  ([#499](https://github.com/CategoricalData/hydra/issues/499)).++---++## [0.16.1] - 2026-06-16++Major themes: the first release built by bootstrapping from previously published+packages, the first full 9 × 9 cross-implementation bootstrapping sweep, and+published-package self-containment fixes for the artifacts that bootstrap rests on.++### Highlights++- **Bootstrap from published packages**: the build links the published 0.16.0 hosts+  (Hackage / Maven Central / PyPI) and uses them to build the next version of Hydra,+  resting on 0.16.0's forward-compatibility guarantees — the published-host consume+  model with a `--local-host` shim, used for the first time in a release+  ([#370](https://github.com/CategoricalData/hydra/issues/370)). 0.16.1 still builds+  against **local** Haskell and Python hosts (`hostOverrides`), because the published+  0.16.0 artifacts it would otherwise consume are the very ones being repaired below.+- **First full 9 × 9 bootstrapping sweep** — every implementation generating every+  other; dozens of cross-host issues surfaced and fixed+  ([#409](https://github.com/CategoricalData/hydra/issues/409), ongoing).+- **Published-package self-containment**: relocate each host's hand-written runtime+  into `overlay/<lang>/hydra-kernel/` so it ships inside the published `hydra-kernel`+  artifact — Python `hydra.python.util` into the wheel+  ([#461](https://github.com/CategoricalData/hydra/issues/461),+  [#472](https://github.com/CategoricalData/hydra/issues/472)) and the Haskell+  authoring DSL + `Libraries`/`Terms` into the sdist+  ([#473](https://github.com/CategoricalData/hydra/issues/473)). Generalized to a+  uniform `overlay/ = transform(packages) + copy(overlay)` model across all hosts+  ([#434](https://github.com/CategoricalData/hydra/issues/434)).+- **Kernel self-containment via generated `hydra.lib.*` registries**: primitive+  names are derived from generated `PrimitiveDefinition` def-modules rather than+  hand-written name strings, in every host+  ([#473](https://github.com/CategoricalData/hydra/issues/473)).+- **In-memory encode/decode synthesis**: derived `hydra.{encode,decode}.*` modules+  are synthesized at build time and excluded from `dist/`, with the synthesizer (not+  inference) as the authority for their annotations+  ([#448](https://github.com/CategoricalData/hydra/issues/448),+  [#476](https://github.com/CategoricalData/hydra/issues/476)).+- **Release hardening**: Apache source-release compliance (signed source archive,+  LICENSE/NOTICE bundling, KEYS) ([#441](https://github.com/CategoricalData/hydra/issues/441)),+  per-package publish orchestrators ([#449](https://github.com/CategoricalData/hydra/issues/449)),+  and packaging-boundary verification gated in CI and at publish time+  ([#472](https://github.com/CategoricalData/hydra/issues/472)).++### New features++- Published-host consume model wired through `bin/sync.sh` for all three hosts+  (`--published-host` default, `--local-host` shim), with the version basis keyed+  off `hydra.json` and the `VERSION` file retired+  ([#370](https://github.com/CategoricalData/hydra/issues/370),+  [#347](https://github.com/CategoricalData/hydra/issues/347)).+- Java and Python coder DSL sources made the sole source of truth; the legacy+  Haskell DSL copies were deleted and cross-package references resolve by name+  ([#346](https://github.com/CategoricalData/hydra/issues/346)).+- Per-host `verify-distribution.sh` self-containment gates (Haskell stages the+  per-package trio and `stack build`s it; Python installs the wheels into an+  isolated venv and imports the kernel modules; Java resolves the published jars+  from an offline consumer), run in CI and as a hard gate in `bin/prepare-release.sh`+  ([#472](https://github.com/CategoricalData/hydra/issues/472)).+- Apache-compliant canonical source archive: signed + checksummed `git archive`+  tarball as the release of record, root `NOTICE`, `KEYS`, and a release-verify CI+  workflow ([#441](https://github.com/CategoricalData/hydra/issues/441)).+- Per-package publish orchestrators `publish-maven.sh` / `publish-pypi.sh` with+  dependency-closure and leaves-first ordering checks+  ([#449](https://github.com/CategoricalData/hydra/issues/449)).+- `hydra-typescript` published as a target Java/Maven + PyPI artifact+  ([#468](https://github.com/CategoricalData/hydra/issues/468)).+- Java coder emits a fluent builder for every generated record type+  ([#465](https://github.com/CategoricalData/hydra/issues/465)); Python coder emits+  per-field `with_<field>` copy-update methods and a fluent builder+  ([#466](https://github.com/CategoricalData/hydra/issues/466)).++### Improvements++- Drop the vestigial `InferenceContext` parameter from the `Primitive.implementation`+  carrier across all hosts; add a graph-free `funT` coder and six `lists` default+  implementations ([#446](https://github.com/CategoricalData/hydra/issues/446)).+- Derive package routing from declared modules; split derived generation into broad+  `mainDslModules` and scoped `mainEncodingModules`+  ([#474](https://github.com/CategoricalData/hydra/issues/474),+  [#475](https://github.com/CategoricalData/hydra/issues/475)).+- Hash JSON content into the per-package input digest; add `digest-check+  refresh-input` and a CI guard that force-regenerates and diffs+  `dist/haskell/hydra-{java,python}` ([#469](https://github.com/CategoricalData/hydra/issues/469)).+- Per-package inference seeds downstream packages from each package's full universe+  while inferring only write targets, fixing `hydra-scala` inference within the CI+  heap budget; declare the `hydra-java` dependency in `hydra-scala` for cold+  ordering ([#470](https://github.com/CategoricalData/hydra/issues/470)).+- Remove the vestigial `Lib/Defaults` registry and empty `defaultLibModules`+  manifest field from all hosts ([#437](https://github.com/CategoricalData/hydra/issues/437)).+- Bundle LICENSE + NOTICE into Java jars (`META-INF`) and Python wheels+  (`dist-info`); declare Apache-2.0 in the Scala build+  ([#441](https://github.com/CategoricalData/hydra/issues/441)).+- Mark lazy parameters on the Scala and Common Lisp primitive registries so eager+  coders thunk correctly ([#453](https://github.com/CategoricalData/hydra/issues/453),+  [#477](https://github.com/CategoricalData/hydra/issues/477),+  [#425](https://github.com/CategoricalData/hydra/issues/425)).+- Staging-loop WIP-commit gate to keep `WIP:` commits off `origin/main`+  ([#466](https://github.com/CategoricalData/hydra/issues/466)).++### Bug fixes++- **Python wheel self-containment**: the `hydra-kernel` wheel now ships+  `hydra.python.util`; gated by a packaging-boundary smoke test in CI and+  `publish-pypi.sh` ([#472](https://github.com/CategoricalData/hydra/issues/472),+  [#461](https://github.com/CategoricalData/hydra/issues/461)).+- **Haskell sdist self-containment**: track the generated `Hydra/Dsl/Terms.hs`+  (was gitignored) so the committed dist exports `ToPrimName`, fixing the cold+  build of the published `hydra-kernel`+  ([#473](https://github.com/CategoricalData/hydra/issues/473)).+- Decode synthesizer threads the fully-applied result type into wrap/record/union+  decoder bodies, so derived decoders emit `DataRow<v>` rather than raw `DataRow`,+  fixing Java and Scala generic decode compilation+  ([#476](https://github.com/CategoricalData/hydra/issues/476)).+- Scala coder emits explicit type arguments for polymorphic non-primitive+  references under a type application (e.g. `requireField`), fixing `Any`-inference+  of curried decoder lambda parameters in `decode/typed`+  ([#434](https://github.com/CategoricalData/hydra/issues/434)).+- Common Lisp and Emacs Lisp: move the hand-written runtime to a dialect-tier path+  so it no longer collides with the generated `hydra.lib.*` def-modules+  ([#434](https://github.com/CategoricalData/hydra/issues/434)).+- Python coder uses the binding `TypeScheme` as the authoritative source for+  parameter and return types ([#488](https://github.com/CategoricalData/hydra/issues/488));+  fixes a duplicate `self` parameter in builders+  ([#478](https://github.com/CategoricalData/hydra/issues/478)).+- Sort term definitions by name in source-file generation so non-Haskell hosts emit+  deterministic alphabetical Haskell output (host-independent of `Maps.elems`+  iteration order) ([#489](https://github.com/CategoricalData/hydra/issues/489)).+- Java coder serializes the `static` method modifier (previously a non-exhaustive+  case) ([#465](https://github.com/CategoricalData/hydra/issues/465)); JavaDoc+  `@link`/`@param` sanitization for strict doclint+  ([#449](https://github.com/CategoricalData/hydra/issues/449)).+- Scala host writer appends a trailing newline when missing, fixing a Lisp-target+  byte mismatch ([#463](https://github.com/CategoricalData/hydra/issues/463)).+- Clojure host: silent wrapping `bigint_to_int{8,16,32,64}` casts per the kernel+  contract ([#483](https://github.com/CategoricalData/hydra/issues/483)); load-order+  and `--kernel-only` filter fixes after kernel renames+  ([#425](https://github.com/CategoricalData/hydra/issues/425)); load the+  `hydra-haskell` baseline for Clojure→Haskell rebuilds+  ([#487](https://github.com/CategoricalData/hydra/issues/487),+  [#460](https://github.com/CategoricalData/hydra/issues/460)).+- Lisp-dialect bootstrap fixes: emit JSON object entries as 2-element `Pair` lists+  rather than dotted cons ([#452](https://github.com/CategoricalData/hydra/issues/452),+  also fixing [#16](https://github.com/CategoricalData/hydra/issues/16)); read+  module names via the correct struct accessor+  ([#471](https://github.com/CategoricalData/hydra/issues/471)); drop stale+  surplus flag arguments from `generate-source-files` calls+  ([#457](https://github.com/CategoricalData/hydra/issues/457),+  [#458](https://github.com/CategoricalData/hydra/issues/458),+  [#464](https://github.com/CategoricalData/hydra/issues/464)); rename eta-wrapped+  projection/unwrap parameters to avoid kernel shadowing+  ([#428](https://github.com/CategoricalData/hydra/issues/428)); propagate a+  non-zero exit from the Scheme test runner on failures and fix Scheme float+  scientific-notation formatting for `|x| >= 1e7`+  ([#447](https://github.com/CategoricalData/hydra/issues/447)).+- Cross-host target dispatch: Common Lisp and Clojure hosts dispatch Scala and+  TypeScript targets ([#482](https://github.com/CategoricalData/hydra/issues/482),+  [#483](https://github.com/CategoricalData/hydra/issues/483)); Scala/Java host+  target-dispatch completions ([#5](https://github.com/CategoricalData/hydra/issues/5),+  [#7](https://github.com/CategoricalData/hydra/issues/7),+  [#17](https://github.com/CategoricalData/hydra/issues/17)).+- TypeScript test runtime reads the kernel JSON field names `body`/`signature`+  (not legacy `term`/`typeScheme`); `lists.nub` dedups by canonical JSON+  ([#485](https://github.com/CategoricalData/hydra/issues/485)).+- Sync robustness: always reassemble `dist/python` under `HYDRA_IN_SYNC` to heal+  partial-dist state ([#480](https://github.com/CategoricalData/hydra/issues/480));+  persist the run dir on bootstrap prep-phase failures+  ([#481](https://github.com/CategoricalData/hydra/issues/481)); complete per-host+  static deps so the bootstrap demo finds fresh dist+  ([#445](https://github.com/CategoricalData/hydra/issues/445)).+- Pin the Python target test interpreter to Python ≥ 3.12+  ([#486](https://github.com/CategoricalData/hydra/issues/486)); Emacs Lisp / Scheme+  target test-runner re-copy robustness+  ([#456](https://github.com/CategoricalData/hydra/issues/456)).+- JavaDoc Pages publish rewritten to build a combined site from the CI-generated+  `dist/java/` artifact, fixing the failure mode where `pages.yml` ran against a+  bare checkout with no `dist/java/` present+  ([#450](https://github.com/CategoricalData/hydra/issues/450)).++### Documentation++- Document the `overlay/` self-containment model, the published-host consume model+  for all three hosts, the `hydra.json` version / host-version basis, and a+  migration-shims recipe ([#434](https://github.com/CategoricalData/hydra/issues/434),+  [#370](https://github.com/CategoricalData/hydra/issues/370),+  [#347](https://github.com/CategoricalData/hydra/issues/347)).+- Document the kernel self-containment / `hydra.lib.*` def-module model and the+  derived-module no-inference invariant+  ([#473](https://github.com/CategoricalData/hydra/issues/473),+  [#466](https://github.com/CategoricalData/hydra/issues/466)); generated Java and+  Python record builders documented in the DSL guides+  ([#465](https://github.com/CategoricalData/hydra/issues/465),+  [#466](https://github.com/CategoricalData/hydra/issues/466)).+- Use absolute GitHub URLs in package READMEs so links resolve when published+  standalone ([#454](https://github.com/CategoricalData/hydra/issues/454)); bundle a+  README into each Python package dir for sdist metadata+  ([#418](https://github.com/CategoricalData/hydra/issues/418)).++---++## [0.16.0] - 2026-06-09++Major themes: primitive-metadata reification, self-hosting coder DSLs with all+eight hosts passing the common test suite, finalized `hydra.packaging`, and a+deep kernel-vocabulary cleanup.++### Highlights++- **`PrimitiveDefinition` and per-namespace registry modules** with host-independent+  specs for every standard-library primitive ([#156](https://github.com/CategoricalData/hydra/issues/156), [#319](https://github.com/CategoricalData/hydra/issues/319)).+- **Self-hosting coder DSLs; all eight hosts pass the common test suite**, including+  the completed TypeScript head ([#344](https://github.com/CategoricalData/hydra/issues/344), [#126](https://github.com/CategoricalData/hydra/issues/126)).+- **Finalized `hydra.packaging`** and a unified `Module.dependencies` field ([#369](https://github.com/CategoricalData/hydra/issues/369), [#354](https://github.com/CategoricalData/hydra/issues/354)).+- **Kernel-vocabulary cleanup**: `maybe`→`optional` across kernel and hosts, plus+  order-preserving JSON objects ([#401](https://github.com/CategoricalData/hydra/issues/401), [#398](https://github.com/CategoricalData/hydra/issues/398)).++### New features++- Reify `PrimitiveDefinition` in `hydra.packaging` with per-namespace `hydra.lib.<sub>` registries ([#156](https://github.com/CategoricalData/hydra/issues/156)).+- Host-independent `comments` specifications for all standard-library primitives ([#319](https://github.com/CategoricalData/hydra/issues/319)).+- Kernel-oriented style checks in `hydra.validate.*` ([#321](https://github.com/CategoricalData/hydra/issues/321), [#351](https://github.com/CategoricalData/hydra/issues/351), [#352](https://github.com/CategoricalData/hydra/issues/352)), on a reshaped validation API ([#320](https://github.com/CategoricalData/hydra/issues/320)).+- Promote the JSON Schema coder to a full DSL coder ([#350](https://github.com/CategoricalData/hydra/issues/350)).+- Transpose coder sources into host-native DSLs for self-hosting ([#344](https://github.com/CategoricalData/hydra/issues/344), [#358](https://github.com/CategoricalData/hydra/issues/358)).+- `Term`-typed annotations replacing `Map<Name, Term>` ([#386](https://github.com/CategoricalData/hydra/issues/386)).+- Language-specific EDSL features on the kernel `Language` ([#233](https://github.com/CategoricalData/hydra/issues/233)).+- Primitive-metadata-driven laziness in eager coders ([#391](https://github.com/CategoricalData/hydra/issues/391)).+- `/test` skill and `bin/test.sh` for target-language test validation ([#387](https://github.com/CategoricalData/hydra/issues/387)).+- Automatic differentiation: a `grad` transform ([#324](https://github.com/CategoricalData/hydra/issues/324)).+- `hydra.show.error.pg` for `InvalidGraphError` ([#374](https://github.com/CategoricalData/hydra/issues/374)).++### Improvements++- Finalize `hydra.packaging`: `Binding`→`TypeDefinition`, `Module`/`Package` `comments`, `moduleName` cleanup ([#369](https://github.com/CategoricalData/hydra/issues/369), [#396](https://github.com/CategoricalData/hydra/issues/396), [#402](https://github.com/CategoricalData/hydra/issues/402), [#404](https://github.com/CategoricalData/hydra/issues/404)).+- Merge `Module.termDependencies`/`typeDependencies` into one `dependencies` field ([#354](https://github.com/CategoricalData/hydra/issues/354)).+- Resolve the `maybe`/`optional` ambiguity across kernel and all hosts ([#401](https://github.com/CategoricalData/hydra/issues/401)).+- Order-preserving JSON objects; order-sensitive `Value` equality ([#398](https://github.com/CategoricalData/hydra/issues/398)).+- Rename `Namespace`→`ModuleName` ([#316](https://github.com/CategoricalData/hydra/issues/316)) and the phantom DSL types to `Typed*` ([#397](https://github.com/CategoricalData/hydra/issues/397)).+- Move from a type-class enum to type-class names ([#275](https://github.com/CategoricalData/hydra/issues/275)).+- Repackage "eval lib" modules as kernel lib modules; promote remaining sources ([#260](https://github.com/CategoricalData/hydra/issues/260), [#355](https://github.com/CategoricalData/hydra/issues/355), [#356](https://github.com/CategoricalData/hydra/issues/356)).+- Reshape `hydra.context` into `InferenceContext` ([#368](https://github.com/CategoricalData/hydra/issues/368)).+- Streamline the Haskell, Java, Python, and Scala syntax models, re-pinning Java to SE 21 and Python to the 3.14 PEG grammar ([#297](https://github.com/CategoricalData/hydra/issues/297)).+- Adopt persistent immutable collections across hosts ([#359](https://github.com/CategoricalData/hydra/issues/359), [#360](https://github.com/CategoricalData/hydra/issues/360), [#361](https://github.com/CategoricalData/hydra/issues/361), [#362](https://github.com/CategoricalData/hydra/issues/362)).+- Migrate underscored kernel definition names to camelCase ([#348](https://github.com/CategoricalData/hydra/issues/348)).+- Tidy qualified/unqualified imports in the Haskell DSL sources ([#308](https://github.com/CategoricalData/hydra/issues/308)).+- Eliminate the `Hydra.Module.Compat` shim ([#315](https://github.com/CategoricalData/hydra/issues/315)).+- Homogenize writer conventions across target languages ([#339](https://github.com/CategoricalData/hydra/issues/339)).+- Disambiguate format-version field naming across JSON metadata ([#412](https://github.com/CategoricalData/hydra/issues/412)).+- Per-package DSL→JSON build with per-package incremental inference ([#381](https://github.com/CategoricalData/hydra/issues/381), [#395](https://github.com/CategoricalData/hydra/issues/395)).+- Prune stale `dist/` outputs; self-heal digests; exclude digests from VCS; reconcile drift; fail loud ([#357](https://github.com/CategoricalData/hydra/issues/357), [#393](https://github.com/CategoricalData/hydra/issues/393), [#405](https://github.com/CategoricalData/hydra/issues/405), [#378](https://github.com/CategoricalData/hydra/issues/378), [#379](https://github.com/CategoricalData/hydra/issues/379), [#392](https://github.com/CategoricalData/hydra/issues/392), [#414](https://github.com/CategoricalData/hydra/issues/414)).+- Accept all valid (host, target) pairs in the generation drivers ([#421](https://github.com/CategoricalData/hydra/issues/421)).+- Move Gradle build infrastructure into `heads/java/`; remove top-level `pixi.toml` ([#384](https://github.com/CategoricalData/hydra/issues/384), [#385](https://github.com/CategoricalData/hydra/issues/385)).+- Eliminate post-generation test patches; fix loader/runner kludges; re-enable test timing ([#307](https://github.com/CategoricalData/hydra/issues/307), [#309](https://github.com/CategoricalData/hydra/issues/309), [#311](https://github.com/CategoricalData/hydra/issues/311)).+- Performance/benchmarks and usability assessments ([#277](https://github.com/CategoricalData/hydra/issues/277), [#283](https://github.com/CategoricalData/hydra/issues/283)).+- Split the Hackage distribution into per-package distributions ([#418](https://github.com/CategoricalData/hydra/issues/418)).++### Bug fixes++- Encode `uint32` JSON literals as numbers, not strings ([#431](https://github.com/CategoricalData/hydra/issues/431)).+- Satisfy or suppress warnings in generated Java ([#349](https://github.com/CategoricalData/hydra/issues/349)).+- Restore JSON parser test cases (polymorphic case handling, ordering fallback) ([#336](https://github.com/CategoricalData/hydra/issues/336)).+- Keep annotations through `termAlternatives`/`typeAlternatives` adaptation ([#353](https://github.com/CategoricalData/hydra/issues/353)).+- Scala host decodes module JSON containing term definitions ([#342](https://github.com/CategoricalData/hydra/issues/342)).+- Java coder: fix exponential `let`-bound `cases` cost ([#372](https://github.com/CategoricalData/hydra/issues/372)) and explicit type args on `PersistentMap.ofEntries` ([#394](https://github.com/CategoricalData/hydra/issues/394)).+- Python-host codegen: drop extra `pyStringToPyStrings` wrapping ([#367](https://github.com/CategoricalData/hydra/issues/367)).+- `escape_literal_char` no longer mangles RDF triples ([#363](https://github.com/CategoricalData/hydra/issues/363)).+- Sync robustness: stale coder JSON, omitted native modules, dead helper, bad source dirs ([#406](https://github.com/CategoricalData/hydra/issues/406), [#400](https://github.com/CategoricalData/hydra/issues/400), [#411](https://github.com/CategoricalData/hydra/issues/411), [#408](https://github.com/CategoricalData/hydra/issues/408)).+- Cross-host self-hosting fixes (Scala, TypeScript, and the four Lisp dialects) ([#422](https://github.com/CategoricalData/hydra/issues/422), [#423](https://github.com/CategoricalData/hydra/issues/423), [#424](https://github.com/CategoricalData/hydra/issues/424), [#429](https://github.com/CategoricalData/hydra/issues/429), [#444](https://github.com/CategoricalData/hydra/issues/444), [#366](https://github.com/CategoricalData/hydra/issues/366), [#389](https://github.com/CategoricalData/hydra/issues/389), [#407](https://github.com/CategoricalData/hydra/issues/407), [#425](https://github.com/CategoricalData/hydra/issues/425), [#426](https://github.com/CategoricalData/hydra/issues/426), [#427](https://github.com/CategoricalData/hydra/issues/427), [#432](https://github.com/CategoricalData/hydra/issues/432), [#438](https://github.com/CategoricalData/hydra/issues/438), [#439](https://github.com/CategoricalData/hydra/issues/439), [#443](https://github.com/CategoricalData/hydra/issues/443)).++### Removed++- Removed the `bigfloat` literal type from `FloatType` and `FloatValue` ([#330](https://github.com/CategoricalData/hydra/issues/330)).+  Arbitrary-precision floating-point values were inconsistently supported+  across host languages (Java/Scala in particular lacked a true IEEE 754+  arbitrary-precision implementation), and nothing in the codebase relied on+  them. `FloatType` and `FloatValue` now have only `float32` and `float64`+  variants. Removed library primitives: `bigfloatToBigint`, `bigfloatToFloat32`,+  `bigfloatToFloat64`, `bigintToBigfloat`, `float32ToBigfloat`,+  `float64ToBigfloat`, `readBigfloat`, `showBigfloat`, `roundBigfloat`. New+  primitives `float32ToFloat64` and `float64ToFloat32` (the latter lossy)+  replace the removed bigfloat-routed conversions.+- Removed `heads/haskell/bin/assemble-hackage-sdist.sh` ([#418](https://github.com/CategoricalData/hydra/issues/418)), the 0.15-era+  bridge that flattened the multi-source-dir Haskell head into one self-contained+  `hydra` sdist. Superseded by the per-package assembler (see Improvements).++### Documentation++- Comprehensive documentation refresh for 0.16 ([#383](https://github.com/CategoricalData/hydra/issues/383))+  spanning nine tracks across READMEs, `docs/`, the wiki, CLAUDE.md, and+  in-source kernel comments.+- New `docs/json-format.md` reference describing the canonical JSON wire format+  (tagged-union duality, optional-field rules, IEEE 754 sentinels, integer+  threshold).+- New wiki pages: `Design` (rationale behind each major choice) and `Inference`+  (HM type inference with class constraints).+- Cross-checked every prose doc against current code; corrected drift in+  primitive-checking sections of `docs/recipes/maintenance.md`, the `Primitive`+  shape in `docs/implementation.md` and `docs/recipes/adding-primitives.md`,+  and the build-system narrative.+- Wove motivation ("the why") through the existing doc surface rather than+  collecting it into separately-titled sidebars; landed an inference-overview+  centerpiece on the wiki ([#377](https://github.com/CategoricalData/hydra/issues/377)).+- Aligned doc vocabulary with the LambdaGraph paper and the literature review+  for the design space Hydra inhabits.+- Net-new READMEs for `packages/hydra-kernel`, `packages/hydra-ext`,+  `packages/hydra-coq`, and `packages/hydra-wasm`; top-level `README.md`+  now carries a canonical packages table.+- Audited and refreshed the Claude surface: CLAUDE.md "Where to look up X"+  table extended with rows for `docs/getting-started.md`, `docs/json-format.md`,+  and the `Design` / `Inference` wiki pages; hard-rules block reordered by+  violation frequency; `claude/pitfalls.md` superseded entries folded into+  `docs/troubleshooting.md` / recipes.+- Project-descriptor strings reviewed and aligned across 24 build / packaging+  files (Hackage synopsis, Maven Central description, PyPI / conda-forge+  classifiers, Hydra-format `package.json` files).+- `docs/index.html` (the GitHub Pages landing page) refreshed with audience+  hops into wiki Concepts / Design, `docs/getting-started.md`, and the GitHub+  repo, plus the full published-artifact list across Hackage, Maven Central,+  PyPI, and conda-forge.+- In-source kernel-code rationale comments added at key Sources sites covering+  injection / pair / either / unit-void duality, error taxonomy, variant+  records, inference-context threading, validation profiles, reduction state,+  lexical filters, name mangling, scoping, variable handling, defaults+  integration, and Tarjan SCC choice rationale. JSON encode/decode modules+  cross-reference `docs/json-format.md`.+- Filed `Primitive.implementation` carrier-simplification follow-up+  ([#446](https://github.com/CategoricalData/hydra/issues/446))+  to drop the unused `InferenceContext` and `Graph` parameters once the+  `Lib/Defaults/` integration ([#437](https://github.com/CategoricalData/hydra/issues/437))+  lands.++---++## [0.15.0] - 2026-04-29++A structural-cleanup release.+The repository is reorganized into a three-tree+`packages/` (DSL sources) / `heads/` (per-host runtime) / `dist/` (generated output)+layout ([#290](https://github.com/CategoricalData/hydra/issues/290)), the kernel is simplified ([#251](https://github.com/CategoricalData/hydra/issues/251), [#292](https://github.com/CategoricalData/hydra/issues/292), [#332](https://github.com/CategoricalData/hydra/issues/332)), and Java + Python now+ship as per-package publishable artifacts on Maven Central / PyPI / conda-forge+([#305](https://github.com/CategoricalData/hydra/issues/305)).+Incremental, content-hash-based caches accelerate the sync pipeline ([#247](https://github.com/CategoricalData/hydra/issues/247)).+Three new generation-only targets — Coq ([#326](https://github.com/CategoricalData/hydra/issues/326)), WebAssembly ([#325](https://github.com/CategoricalData/hydra/issues/325)), and an+automatic-differentiation demo ([#324](https://github.com/CategoricalData/hydra/issues/324)) — are in progress as Claude collaborations.++### Highlights++- **Three-tree repository layout**: `packages/` (DSL), `heads/` (host runtime),+  `dist/` (generated) ([#290](https://github.com/CategoricalData/hydra/issues/290)).+- **Per-package distributions** for Java (4 Maven artifacts) and Python (5 wheels):+  `hydra-kernel`, `hydra-pg`, `hydra-rdf`, `hydra-{java,python}`, plus `hydra-ext`+  on PyPI ([#305](https://github.com/CategoricalData/hydra/issues/305)).+- **Kernel simplifications**: `Function`/`Elimination` removed ([#332](https://github.com/CategoricalData/hydra/issues/332)),+  `FunctionPrimitive` removed ([#251](https://github.com/CategoricalData/hydra/issues/251)), `Hydra.Ext.*` prefix retired ([#331](https://github.com/CategoricalData/hydra/issues/331)),+  `Context` error wrapper gone ([#292](https://github.com/CategoricalData/hydra/issues/292)).+- **Incremental caches** across the sync pipeline; ~70x speedup on no-op resync ([#247](https://github.com/CategoricalData/hydra/issues/247)).+- **JSON kernel format v1** with formal wire-format spec and four-field rename ([#343](https://github.com/CategoricalData/hydra/issues/343)).+- **Three new generation-only targets** (in progress): Coq ([#326](https://github.com/CategoricalData/hydra/issues/326)), WebAssembly ([#325](https://github.com/CategoricalData/hydra/issues/325)),+  automatic-differentiation demo ([#324](https://github.com/CategoricalData/hydra/issues/324)).++### New features++- Per-package distributions ([#305](https://github.com/CategoricalData/hydra/issues/305)): standalone Maven / PyPI / conda-forge builds,+  generated `build.gradle` / `pyproject.toml` per package, transitive dep resolution.+- JSON kernel format v1 ([#343](https://github.com/CategoricalData/hydra/issues/343)): formal spec at `docs/json-format.md`,+  four-field rename to `typeScheme`, `Module` field reorder, `formatVersion` stamp.+- `decimal` literal type in the kernel ([#338](https://github.com/CategoricalData/hydra/issues/338)) across all hosts.+- Type-directed JSON encoder with idiomatic optional encoding ([#314](https://github.com/CategoricalData/hydra/issues/314), [#318](https://github.com/CategoricalData/hydra/issues/318)).+- Incremental type inference plus content-hash caches ([#247](https://github.com/CategoricalData/hydra/issues/247)).+- Coq generation target ([#326](https://github.com/CategoricalData/hydra/issues/326), in progress): `hydra-coq` package, 131/131 `.v` files pass `coqc`.+- WebAssembly target ([#325](https://github.com/CategoricalData/hydra/issues/325), in progress): `hydra-wasm` package, M4 closure mechanism.+- Automatic-differentiation demo ([#324](https://github.com/CategoricalData/hydra/issues/324), in progress): symbolic source-to-source.+- `hydra.show.error` module ([#265](https://github.com/CategoricalData/hydra/issues/265)): consolidated error-message builders.+- `hydra.lib.maybes.toList` primitive ([#257](https://github.com/CategoricalData/hydra/issues/257)).+- NaN / Inf round-tripping through JSON and per-language serdes ([#312](https://github.com/CategoricalData/hydra/issues/312), [#318](https://github.com/CategoricalData/hydra/issues/318), [#330](https://github.com/CategoricalData/hydra/issues/330)).+- Typeclass inference test group ([#274](https://github.com/CategoricalData/hydra/issues/274)).+- Term and type linters ([#232](https://github.com/CategoricalData/hydra/issues/232)).+- Validation for modules and graphs ([#155](https://github.com/CategoricalData/hydra/issues/155)).+- User-defined-function (UDF) support and DSL-vs-primitive overrides+  ([#62](https://github.com/CategoricalData/hydra/issues/62), [#63](https://github.com/CategoricalData/hydra/issues/63), [#102](https://github.com/CategoricalData/hydra/issues/102), [#158](https://github.com/CategoricalData/hydra/issues/158)).+- Stale-source detection across sync entrypoints ([#228](https://github.com/CategoricalData/hydra/issues/228)).++### Improvements++- Sync infrastructure: matrix tool, per-package orchestrator, per-language wrappers ([#290](https://github.com/CategoricalData/hydra/issues/290)).+- TestGraph post-generation patches eliminated; DSL emits `TestEnv` refs directly.+- Minimized standard imports in generated Haskell ([#161](https://github.com/CategoricalData/hydra/issues/161)); improved auto-aliases ([#322](https://github.com/CategoricalData/hydra/issues/322)).+- Removed kernel term dependencies from the interpreter ([#257](https://github.com/CategoricalData/hydra/issues/257)).+- Removed `Graph` and `Context` arguments from primitive implementations ([#266](https://github.com/CategoricalData/hydra/issues/266)).+- Refactored kernel term modules: split rewriting/schemas, renamed show.meta → codegen,+  moved `Coder` to `coders`, merged `extract.helpers` ([#221](https://github.com/CategoricalData/hydra/issues/221)).+- Removed Aeson and HsYAML dependencies; YAML coder moved to `hydra-ext` ([#261](https://github.com/CategoricalData/hydra/issues/261)).+- Lisp recursive let bindings via SCC ([#341](https://github.com/CategoricalData/hydra/issues/341)); parenthesis cleanup ([#84](https://github.com/CategoricalData/hydra/issues/84)).+- Everything-to-everything bootstrapping demo ([#254](https://github.com/CategoricalData/hydra/issues/254)).+- Re-enabled skipped Python tests; `disabledForPython` retired ([#263](https://github.com/CategoricalData/hydra/issues/263)).+- Promoted host-specific code in `packages/` into per-head locations ([#337](https://github.com/CategoricalData/hydra/issues/337)).+- Migrated `hydra.common` Java module out ([#10](https://github.com/CategoricalData/hydra/issues/10)).+- Test infrastructure unification under `UniversalTestCase` ([#246](https://github.com/CategoricalData/hydra/issues/246)).+- Default property graph → RDF mapping ([#296](https://github.com/CategoricalData/hydra/issues/296)).+- 13 unsafe partial primitives replaced with `Maybe`-returning alternatives ([#201](https://github.com/CategoricalData/hydra/issues/201)).+- Java generated code uses standard `java.util` collection interfaces ([#313](https://github.com/CategoricalData/hydra/issues/313)).+- `Term.union` renamed to `Term.inject` ([#334](https://github.com/CategoricalData/hydra/issues/334)).+- Inconsistencies in domain DSLs cleaned up ([#219](https://github.com/CategoricalData/hydra/issues/219)).+- Coder packages moved out of hydra-haskell ([#182](https://github.com/CategoricalData/hydra/issues/182)).+- Java package-naming scheme revised ([#223](https://github.com/CategoricalData/hydra/issues/223)).+- Decimal used (instead of bigfloat) for JSON numbers ([#340](https://github.com/CategoricalData/hydra/issues/340)).+- Hydra kernel moved out of hydra-haskell ([#282](https://github.com/CategoricalData/hydra/issues/282)).++### Bug fixes++- `substInTypeScheme` capture-unsafe substitution under nested forall binders.+- Java `Float64` arithmetic primitives corrected for current `PrimitiveFunction` API.+- Clojure decimal handling: value-based equality, `BigInteger` return from `decimal_to_bigint`.+- Scheme runtime R7RS imports for `sets.scm` / `maps.scm` / `eithers.map_set`.+- Python coder defaults corrected from `Py.Name` to `Py.Expression`.+- Scala host JSON-decode failure on modules containing term definitions ([#342](https://github.com/CategoricalData/hydra/issues/342)).+- Lisp coder mutually-recursive let bindings for Clojure ([#341](https://github.com/CategoricalData/hydra/issues/341)).++### Documentation++- Documentation refresh for the 0.15 packaging restructure ([#290](https://github.com/CategoricalData/hydra/issues/290), [#331](https://github.com/CategoricalData/hydra/issues/331)); CLAUDE.md,+  READMEs, recipes, demos, and seven wiki pages updated.+- Code-generation recipe rewritten for the per-package dist layout ([#282](https://github.com/CategoricalData/hydra/issues/282)).+- New `maintenance.md` recipe consolidates non-source-file scans, stale generated+  detection, design-violation checks, and freshness checks.+- Documentation style guide added at `docs/documentation-style-guide.md`.+- Cross-worktree messaging protocol documented in CLAUDE.md.+- Decimal type documented in the lexicon and per-language READMEs ([#338](https://github.com/CategoricalData/hydra/issues/338)).+- JSON encoding format docs updated for idiomatic optional encoding ([#314](https://github.com/CategoricalData/hydra/issues/314)).++---++## [0.14.1] - 2026-03-30++Patch release.+Regenerates all implementations after a fix to the group eval primitive ([#281](https://github.com/CategoricalData/hydra/issues/281)),+and relocates the property-graph and RDF/SHACL "ext" modules into hydra-java and hydra-python.++### Improvements++- Moved PG, RDF, and SHACL ext modules from hydra-ext into hydra-java and hydra-python.++### Bug fixes++- Group eval primitive fix; all targets regenerated ([#281](https://github.com/CategoricalData/hydra/issues/281)).+- JavaDoc HTML-entity escaping in generated `DestructuringPattern.java`.++---++## [0.14.0] - 2026-03-29++Major release adding four new complete Hydra implementations — Scala ([#273](https://github.com/CategoricalData/hydra/issues/273)),+Clojure ([#278](https://github.com/CategoricalData/hydra/issues/278)), Common Lisp, and Scheme — bringing the total to seven bootstrapping hosts+(plus Emacs Lisp, which passes the test suite but does not yet participate in bootstrapping).+Also includes kernel simplifications, an Avro bidirectional coder ([#301](https://github.com/CategoricalData/hydra/issues/301)),+eval primitives ([#281](https://github.com/CategoricalData/hydra/issues/281)), new demos, and significant performance work on the Scheme host.++### Highlights++- **Four new complete implementations**: Hydra-Scala ([#273](https://github.com/CategoricalData/hydra/issues/273)), Hydra-Clojure ([#278](https://github.com/CategoricalData/hydra/issues/278)),+  Hydra-Common Lisp, and Hydra-Scheme pass the common test suite and can serve as+  bootstrapping hosts.+  The four Lisp dialects share a single coder and serializer.+- **Kernel simplifications**: removed `RowType`/`WrappedType` ([#82](https://github.com/CategoricalData/hydra/issues/82)),+  unified `Graph`/`TypeContext`/`InferenceContext` ([#192](https://github.com/CategoricalData/hydra/issues/192)),+  replaced `Module.elements` with `Module.definitions` ([#214](https://github.com/CategoricalData/hydra/issues/214)),+  consolidated error types ([#268](https://github.com/CategoricalData/hydra/issues/268)), and promoted all staging modules to DSL sources ([#267](https://github.com/CategoricalData/hydra/issues/267)).+- **Avro bidirectional coder** ([#301](https://github.com/CategoricalData/hydra/issues/301)) with property graph pipeline demo.+- **Scheme bootstrap performance**: 58 minutes → 5.7 minutes via O(1) vhash data structures+  and IEEE 754 float precision fixes.++### Breaking changes++- Removed legacy adapter framework; replaced by the staging-promoted variant ([#236](https://github.com/CategoricalData/hydra/issues/236), [#295](https://github.com/CategoricalData/hydra/issues/295)).+- Replaced legacy Flow monad with Either ([#245](https://github.com/CategoricalData/hydra/issues/245)).+- Removed `Module.elements`; use `Module.definitions` ([#214](https://github.com/CategoricalData/hydra/issues/214)).+- Removed `RowType` and `WrappedType` ([#82](https://github.com/CategoricalData/hydra/issues/82)).+- Renamed `hydra.accessors` to `hydra.paths` ([#271](https://github.com/CategoricalData/hydra/issues/271)).+- Renamed `hydra.tarjan` and merged into `hydra.sorting` ([#220](https://github.com/CategoricalData/hydra/issues/220)).+- Merged `hydra.constraints` into `hydra.query` ([#272](https://github.com/CategoricalData/hydra/issues/272)).+- Moved `hydra.workflow` to hydra-ext ([#270](https://github.com/CategoricalData/hydra/issues/270)).+- Consolidated `hydra.compute` into `hydra.util` ([#269](https://github.com/CategoricalData/hydra/issues/269)).+- Removed `hydra.describe` ([#216](https://github.com/CategoricalData/hydra/issues/216)) and types from `hydra.mantle` ([#217](https://github.com/CategoricalData/hydra/issues/217)).+- Replaced `OtherError` with task-specific error types ([#268](https://github.com/CategoricalData/hydra/issues/268)).+- Java: lowered minimum Java version to 11 ([#249](https://github.com/CategoricalData/hydra/issues/249)).+- Java: `Pair` replaces `Tuple2` ([#252](https://github.com/CategoricalData/hydra/issues/252)); `Comparable` implemented for generated classes ([#131](https://github.com/CategoricalData/hydra/issues/131)).++### New features++- Hydra-Scala ([#273](https://github.com/CategoricalData/hydra/issues/273)): complete implementation, sbt/Scala 3, 3043 tests.+- Hydra-Lisp: Clojure ([#278](https://github.com/CategoricalData/hydra/issues/278)), Common Lisp, Emacs Lisp, Scheme — four dialects,+  one coder. CI workflows for Clojure, Common Lisp, and Scheme.+- Avro bidirectional coder ([#301](https://github.com/CategoricalData/hydra/issues/301)) with demo.+- Eval primitives ([#281](https://github.com/CategoricalData/hydra/issues/281)): Logic, Equality, Math, collections, and Groups.+- Generated DSL modules ([#180](https://github.com/CategoricalData/hydra/issues/180)): `hydra.dsls` generates DSL helpers from type modules.+- New demos: GraphQL+JSON ([#279](https://github.com/CategoricalData/hydra/issues/279)), PG validation ([#284](https://github.com/CategoricalData/hydra/issues/284)), SHACL ([#294](https://github.com/CategoricalData/hydra/issues/294)), Avro+PG.+- Packaging module ([#290](https://github.com/CategoricalData/hydra/issues/290)): `hydra.packaging` with `Package` type and validation.+- New primitives: `lists.foldr` ([#280](https://github.com/CategoricalData/hydra/issues/280)), `maybes.toList`, float rounding ([#264](https://github.com/CategoricalData/hydra/issues/264), [#285](https://github.com/CategoricalData/hydra/issues/285)).+- Cross-language benchmarking suite ([#234](https://github.com/CategoricalData/hydra/issues/234)).+- Limited typeclass inference ([#164](https://github.com/CategoricalData/hydra/issues/164)).+- Standardized validation patterns across modules ([#291](https://github.com/CategoricalData/hydra/issues/291)).++### Improvements++- Promoted all staging modules to DSL sources ([#267](https://github.com/CategoricalData/hydra/issues/267)).+- Removed legacy adapters, JSON/YAML coders, grammar framework ([#236](https://github.com/CategoricalData/hydra/issues/236), [#295](https://github.com/CategoricalData/hydra/issues/295)).+- Java: efficient immutable data structures ([#193](https://github.com/CategoricalData/hydra/issues/193)); flattened term naming.+- Java: hidden specialized collection classes in type-level files ([#313](https://github.com/CategoricalData/hydra/issues/313)).+- Inference error messages improved ([#231](https://github.com/CategoricalData/hydra/issues/231)).+- Combined nested let bindings in the Haskell coder ([#248](https://github.com/CategoricalData/hydra/issues/248)).+- Simplified `hydra.adapt.simple` ([#255](https://github.com/CategoricalData/hydra/issues/255)).+- Removed Haskell operators dependency from tests ([#288](https://github.com/CategoricalData/hydra/issues/288)).+- Regression tests for non-Haskell/Java coders ([#85](https://github.com/CategoricalData/hydra/issues/85)).+- Regex support added ([#293](https://github.com/CategoricalData/hydra/issues/293)).+- `hydra.annotations.isNativeType` revisited ([#214](https://github.com/CategoricalData/hydra/issues/214)).+- Migrated grammar-generated models ([#295](https://github.com/CategoricalData/hydra/issues/295)).+- `<<= -shouldBe-> <<~` operator change ([#276](https://github.com/CategoricalData/hydra/issues/276)).+- Minimize Generation.hs ([#225](https://github.com/CategoricalData/hydra/issues/225)).+- Implemented missing Flows primitives ([#183](https://github.com/CategoricalData/hydra/issues/183)).+- Removed JavaDocs from the repository ([#262](https://github.com/CategoricalData/hydra/issues/262)).+- Floating-point test portability fixes for Linux CI ([#264](https://github.com/CategoricalData/hydra/issues/264), [#285](https://github.com/CategoricalData/hydra/issues/285)).+- Moved YAML coder to hydra-ext ([#224](https://github.com/CategoricalData/hydra/issues/224)).+- Minimized native Python and Java primitives in favor of DSL implementations ([#244](https://github.com/CategoricalData/hydra/issues/244)).+- Removed the `disabledForPython` test tag ([#250](https://github.com/CategoricalData/hydra/issues/250)).++### Release tooling++- 0.14 release tracking ([#304](https://github.com/CategoricalData/hydra/issues/304)).++### Documentation++- Thorough cleanup of user documentation ([#302](https://github.com/CategoricalData/hydra/issues/302), [#303](https://github.com/CategoricalData/hydra/issues/303)).+- Coding style guide added to wiki.+- Updated 'new implementation' recipe with learnings from Clojure head ([#278](https://github.com/CategoricalData/hydra/issues/278)).++---++## [0.13.0] - 2026-02-27++Major release completing Hydra-Python ([#66](https://github.com/CategoricalData/hydra/issues/66)) and Hydra-Java ([#166](https://github.com/CategoricalData/hydra/issues/166)) as self-hosting Hydra implementations+and demonstrating mutual self-hosting across all three languages.+Significant improvements to type inference, type checking, rewriting, and adaptation.+New language features include first-class Either types, typeclass inference, binary data support,+and a native JSON parser/writer.+Comprehensive tooling for cross-implementation bootstrapping+and provisional support for Go, Rust, and JavaScript targets.++### Highlights++- **Mutual self-hosting across three implementations**: Hydra is now fully self-hosting+  in Haskell ([#179](https://github.com/CategoricalData/hydra/issues/179)), Java, and Python, and can cross-generate between every combination+  of host and target language (9 paths total).+- **Hydra-Java is complete** ([#166](https://github.com/CategoricalData/hydra/issues/166)): all kernel and generation tests pass.+- **Hydra-Python is complete** ([#66](https://github.com/CategoricalData/hydra/issues/66)): 100% test parity with Haskell.+- **Language coders promoted into the Hydra kernel** ([#176](https://github.com/CategoricalData/hydra/issues/176)): Haskell, Java, and Python+  coders are now defined as Hydra DSL modules, making them self-hosting.+- **Either type support** ([#210](https://github.com/CategoricalData/hydra/issues/210)) with full inference, checking, and library support.+- **Native JSON parser and writer** ([#188](https://github.com/CategoricalData/hydra/issues/188), [#243](https://github.com/CategoricalData/hydra/issues/243), [#253](https://github.com/CategoricalData/hydra/issues/253)): Hydra's own JSON parser and writer+  replace the Aeson dependency; bootstrap-from-JSON architecture for cross-implementation+  code generation.++### Breaking changes++- Renamed `Optional` to `Maybe` throughout the codebase ([#204](https://github.com/CategoricalData/hydra/issues/204)):+  `hydra.lib.optionals` → `hydra.lib.maybes`; all `optional` variants renamed to `maybe`.+- Removed unlabeled product and sum types; replaced with pair types ([#212](https://github.com/CategoricalData/hydra/issues/212)).+- Renamed `hydra.core.TypedTerm` to `hydra.core.TypeApplicationTerm`.+- Renamed `hydra.json` to `hydra.json.model`.+- Renamed `hydra.lib.maps.remove` to `hydra.lib.maps.delete`.+- Removed deprecated `hydra.decoding` module.+- Removed old `hydra.mantle.Either` type (replaced by core `Either`).+- Removed `hydra.lib.tuples` library.+- Removed C# coder stub (C# syntax model remains).+- DSL syntax migration: removed `OverloadedStrings` for term expressions ([#238](https://github.com/CategoricalData/hydra/issues/238)).+- Removed `Hydra.Dsl.ShorthandTypes` ([#222](https://github.com/CategoricalData/hydra/issues/222)).++### New features++- Self-hosting in Haskell ([#179](https://github.com/CategoricalData/hydra/issues/179)) and full mutual cross-generation (Haskell ↔ Java ↔ Python).+- New bootstrapping demo validating all 9 paths across 249 modules.+- Promoted Haskell, Java, and Python coders from staging code into Hydra DSL sources ([#176](https://github.com/CategoricalData/hydra/issues/176)).+- JSON-as-source-of-truth bootstrapping architecture ([#243](https://github.com/CategoricalData/hydra/issues/243), [#253](https://github.com/CategoricalData/hydra/issues/253)):+  new executables and scripts enabling `bootstrap-from-json` for cross-implementation+  code generation.+- Generated term encoders and decoders replace legacy hand-coded modules ([#47](https://github.com/CategoricalData/hydra/issues/47)).+- Hydra-Python completion ([#66](https://github.com/CategoricalData/hydra/issues/66)): restructured project layout with `src/main` and+  `src/gen-main` ([#191](https://github.com/CategoricalData/hydra/issues/191)); full library implementations including `hydra.lib.flows`;+  `@lru_cache` optimization for nullary bindings; PyPy compatibility.+- Hydra-Java completion ([#166](https://github.com/CategoricalData/hydra/issues/166)): all kernel and generation tests pass; deep changes+  to Java coder; `Lazy` utility class for delayed evaluation; array-type support+  in Java serde; tail-call optimization in the Java coder; `Tuple` implements `Comparable`.+- Either type ([#210](https://github.com/CategoricalData/hydra/issues/210)): `EitherType`, `either` term constructor, `hydra.lib.eithers` library+  with `either`, `map`, `mapList`, `mapMaybe`, `bimap`, `bind`, `fromLeft`, `fromRight`,+  `isLeft`, `isRight`, `lefts`, `rights`, `partitionEithers`.+- Pair type as a core construct ([#211](https://github.com/CategoricalData/hydra/issues/211)).+- Void type ([#237](https://github.com/CategoricalData/hydra/issues/237)): uninhabited `void` type as the dual of `unit`; type-level only.+- Native JSON parser and writer ([#188](https://github.com/CategoricalData/hydra/issues/188), [#242](https://github.com/CategoricalData/hydra/issues/242)):+  bidirectional JSON encoder/decoder; replaces the Aeson dependency;+  cross-checked via a special test runner.+- Typeclass inference ([#164](https://github.com/CategoricalData/hydra/issues/164)):+  type schemes with typeclass constraints via `TypeVariableMetadata`;+  typeclass metadata on primitive definitions; Haskell coder uses typeclass info.+- Binary/`ByteString` support ([#172](https://github.com/CategoricalData/hydra/issues/172)): binary data via `ByteString` instead of `String`;+  Haskell coder uses `Data.ByteString`; new `binaryToBytes` primitive.+- Parser combinators: new `hydra.parsing` module with combinator types and DSL.+- Higher-order primitive interpreter ([#198](https://github.com/CategoricalData/hydra/issues/198)): all primitives are fully interpretable+  across all three implementations; legacy `requiresInterp` tag removed.+- Let hoisting and flattening: hoisting polymorphic let terms;+  refined hoisting for monomorphic bindings inside polymorphic ones;+  more efficient `liftLambdaAboveLet` ([#202](https://github.com/CategoricalData/hydra/issues/202));+  case-statement hoisting ([#241](https://github.com/CategoricalData/hydra/issues/241)); subterm hoisting helpers.+- Provisional language targets: Go ([#65](https://github.com/CategoricalData/hydra/issues/65)), Rust (with experimental type-level coder),+  JavaScript (syntax model, serde, DSL, generated sources), and Java/Python syntax DSLs.+- New adapter framework ([#236](https://github.com/CategoricalData/hydra/issues/236)): all external coders refactored+  (Haskell, YAML, Scala, JSON Schema, C++, PDL, Protobuf, GraphQL).+- Compact labeled records and variants ([#122](https://github.com/CategoricalData/hydra/issues/122)).+- New primitives:+  `hydra.lib.lists.find`, `hydra.lib.lists.partition`,+  `hydra.lib.flows.withDefault`, `hydra.lib.flows.foldl`,+  `hydra.lib.math.{max,min,abs,pred,signum,succ,even,odd}`,+  `hydra.lib.pairs.bimap`, `hydra.lib.eithers.bind`,+  `hydra.lib.literals.binaryToBytes`, complete `readXxx`/`showXxx` families,+  `hydra.lexical.chooseUniqueName`, `neg` renamed to `negate`,+  full floating-point math primitives in Python ([#208](https://github.com/CategoricalData/hydra/issues/208)).+- DeepCore DSL: a layer one level deeper than the Meta DSLs,+  for programs that construct programs that construct terms.+- `typeOf` implementation ([#168](https://github.com/CategoricalData/hydra/issues/168)).+- Implement `Comparable` for generated Java classes ([#131](https://github.com/CategoricalData/hydra/issues/131)); update of Hydra-Java ([#166](https://github.com/CategoricalData/hydra/issues/166)).+- Compact labeled records exploration ([#122](https://github.com/CategoricalData/hydra/issues/122)).+- Comparison primitives without typeclasses ([#187](https://github.com/CategoricalData/hydra/issues/187)).+- Investigated core support for Either and Flow monads ([#200](https://github.com/CategoricalData/hydra/issues/200)).++### Improvements++- Type system:+  - Refactored `hydra.inference` to create new module `hydra.checking`.+  - Inference checks after unification to prevent schema names from being unified+    with inferred type variables.+  - Type application terms preserved when preparing application terms for Python.+  - Sanity-check tuple projection index against arity during inference.+  - Bind lost type variables in term annotations.+  - Made `extendTypeContextForLet` tolerant of untyped bindings.+- Code generation:+  - Python: support for deeply-nested match statements; more complete `let` support;+    Pythonic syntax for polymorphic function definitions; refinements for complex+    case-hoisting scenarios; performance improvements ([#209](https://github.com/CategoricalData/hydra/issues/209), [#239](https://github.com/CategoricalData/hydra/issues/239), [#240](https://github.com/CategoricalData/hydra/issues/240)); inline+    let/case rewriting for Python targets ([#203](https://github.com/CategoricalData/hydra/issues/203)).+  - Java: method declaration style consistent with calling style.+  - Haskell: standardized typeclass names.+  - Increased maximum trace depth to support more complex sources.+- Common test suite promotion: all tests promoted into a shared, implementation-independent+  test kernel ([#213](https://github.com/CategoricalData/hydra/issues/213)).+- Statically-compiled test suite in Haskell ([#207](https://github.com/CategoricalData/hydra/issues/207)).+- Included all Hydra kernel types in test schema ([#205](https://github.com/CategoricalData/hydra/issues/205)).+- Reorganized Haskell DSL definitions; standardized imports.+- Added TODO comments to all unsafe primitives ([#201](https://github.com/CategoricalData/hydra/issues/201)).+- New `rewriteAndFoldTerm` utility for simultaneous rewriting and folding.+- Lexical helper for dereferencing schema types through aliases.+- Made JSON parser lazy with new `lazy` parser combinator.+- More efficient substitution helpers for empty substitutions.+- Precomputed type/inference context in kernel test runner for performance.+- Property graph encoding and decoding modules promoted into DSL.+- `hydra.tabular` and `hydra.pg.graphson` modules promoted into DSL.+- `hydra.show.*` modules generated into Java and Python.+- Aligned epsilon encoding with the LambdaGraph spec ([#89](https://github.com/CategoricalData/hydra/issues/89)).+- Fixed asymmetry of introductions and eliminations ([#86](https://github.com/CategoricalData/hydra/issues/86), [#134](https://github.com/CategoricalData/hydra/issues/134)).+- Minimized inferred type annotations ([#114](https://github.com/CategoricalData/hydra/issues/114)).+- Investigated combining products with records, sums with unions ([#196](https://github.com/CategoricalData/hydra/issues/196), [#212](https://github.com/CategoricalData/hydra/issues/212)).+- Refined type reduction in adapters ([#144](https://github.com/CategoricalData/hydra/issues/144)).+- Investigation of unification issues ([#163](https://github.com/CategoricalData/hydra/issues/163)).+- Refactored RowType ([#82](https://github.com/CategoricalData/hydra/issues/82)).+- LLM-assisted PG schemas and mappings ([#171](https://github.com/CategoricalData/hydra/issues/171)).+- Simplified unit-valued variants in Python ([#206](https://github.com/CategoricalData/hydra/issues/206)).+- Checked for inconsistencies between Term/TTerm and Type/TType DSLs ([#218](https://github.com/CategoricalData/hydra/issues/218)).+- GenPG demo: refactored to support Haskell, Python, and Java; Python runner;+  Java generation; finalized output format.+- Testing: extended test runners for let hoisting (Haskell, Python, Java);+  floating-point precision tests; ordering tests for `hydra.lib.maps`/`hydra.lib.sets`;+  inference tests for inferred System F terms; case-hoisting test cases; new JSON coder tests;+  thorough test cases for all primitives ([#199](https://github.com/CategoricalData/hydra/issues/199)).++### Bug fixes++- Fixed interpreter bugs causing test failures ([#235](https://github.com/CategoricalData/hydra/issues/235)).+- Fixed bug in type checking in connection with dead code.+- Fixed bug in `removeTypesFromTerm`.+- Fixed bug in `typeOfMap`.+- Fixed 32-bit max int value in `hydra.constants`.+- Fixed numeric precision in Python.+- Added `BigDecimal` support and fixed `uint8` (short) support in Java.+- Fixed Haskell coder with respect to imports for binary literals.+- Excluded `Prelude.encodeFloat` and `Prelude.decodeFloat` from generated Haskell+  due to name collisions.+- Fixed consistency issues in Python primitives:+  `hydra.lib.sets.toList`, `hydra.lib.maps.union` (precedence),+  `hydra.lib.strings.lines`, `hydra.lib.strings.readString` (dequoting),+  `hydra.lib.lists.apply`, `hydra.lib.maps.toList`.+- Fixed Haskell type signature of `hydra.lists.span`.+- Fixed issue with eta expansion of typed terms.+- Fixed shadowing issue in generated encoding modules.+- Added missing alternatives in `hydra.rewriting.rewriteTermM`.++### Documentation++- Updated main README with complete implementation status.+- Added Documentation sections to all implementation READMEs.+- Fixed outdated Code-organization.md (Python now uses `src/gen-main`).+- Fixed incorrect code paths in Implementation.md.+- Added Java section to Testing wiki.+- Comprehensive developer recipes in `docs/recipes/`.+- Documentation for tail-call optimization implementation.+- Added Haddock comments to Haskell primitives.+- Created wiki page on the release process ([#194](https://github.com/CategoricalData/hydra/issues/194)).++### Community++- Accepted babeloff's `isTrivialTerm` changes.++---++## [0.12.0] - 2025-08-28++A consolidation release: significant Python progress, kernel-level cleanup,+and a wave of bookkeeping closures for Hydra-Java module ports.++### Breaking changes++- Removed `hydra.decoding` module ([#190](https://github.com/CategoricalData/hydra/issues/190)).+- Flattened dependency tiers; eliminated tier-1, tier-2, tier-3 organization ([#184](https://github.com/CategoricalData/hydra/issues/184)).+- Removed `typed` term variant ([#173](https://github.com/CategoricalData/hydra/issues/173), [#162](https://github.com/CategoricalData/hydra/issues/162)).+- Module naming changes: moved tabular and relational modules into kernel ([#152](https://github.com/CategoricalData/hydra/issues/152)).+- Standardized on uncurried helper functions in term-level DSLs ([#174](https://github.com/CategoricalData/hydra/issues/174)).+- Renamed `hydra.lib.flows.traverseOptional` to `hydra.lib.flows.mapOptional`.+- Eliminated camel-cased namespace parts from the kernel ([#152](https://github.com/CategoricalData/hydra/issues/152)).+- Replaced list and optional eliminations with primitive functions ([#150](https://github.com/CategoricalData/hydra/issues/150)).+- Moved `Hydra.Ext` to hydra-ext ([#178](https://github.com/CategoricalData/hydra/issues/178)); moved 'ext' sources into hydra-ext ([#189](https://github.com/CategoricalData/hydra/issues/189)).++### New features++- Significant progress toward Hydra-Python completion:+  generated all kernel types and terms into Python;+  added `Decimal` support for bigfloat values;+  environment tracking for variable scoping;+  generated `hydra.languages` into Python;+  `__hash__` and `__eq__` on serializable generated classes ([#160](https://github.com/CategoricalData/hydra/issues/160)).+- Designated `unit` term and type variants ([#186](https://github.com/CategoricalData/hydra/issues/186)).+- `hydra.lib.flows.mapElems` and `hydra.lib.flows.mapKeys` utilities.+- Promoted JSON coder into the DSL ([#181](https://github.com/CategoricalData/hydra/issues/181)).+- Promoted JSON utilities into the DSL ([#181](https://github.com/CategoricalData/hydra/issues/181)).+- C# coder ([#139](https://github.com/CategoricalData/hydra/issues/139)).+- GQL model and parser ([#140](https://github.com/CategoricalData/hydra/issues/140)).+- Tabular adapters ([#142](https://github.com/CategoricalData/hydra/issues/142)).+- Type-level C++ coder ([#170](https://github.com/CategoricalData/hydra/issues/170)).+- Variable-types-to-Haskell-typedefs transformation ([#70](https://github.com/CategoricalData/hydra/issues/70)).+- A newline at the end of each generated file ([#154](https://github.com/CategoricalData/hydra/issues/154)).+- Refactor kernel modules and primitives by connectivity ([#177](https://github.com/CategoricalData/hydra/issues/177)).++### Improvements++- Condensed repeated elements in error traces ([#165](https://github.com/CategoricalData/hydra/issues/165)).+- Fixed transitive schema dependencies ([#185](https://github.com/CategoricalData/hydra/issues/185)).+- Improved adapter solution with respect to literals.+- Better handling of type variables in Python coder.+- Enhanced term-level Python generation with proper ordering to minimize forward references.+- Topo-sort helper function for dependency management.+- Updated all Python libraries with latest changes.+- Distinguished between deannotation and detyping of terms.+- Adapted primitives when applying language constraints to a graph.+- Use unqualified names where possible in generated code ([#153](https://github.com/CategoricalData/hydra/issues/153)).+- Eliminated `Data.Graph` (containers) dependency ([#167](https://github.com/CategoricalData/hydra/issues/167)).+- "Close the loop" demonstration ([#175](https://github.com/CategoricalData/hydra/issues/175)).+- Implemented `typeOf` ([#168](https://github.com/CategoricalData/hydra/issues/168)).++### Bug fixes++- Fixed System F terms for records and case statements ([#168](https://github.com/CategoricalData/hydra/issues/168)).+- Fixed issue with transitive type-level module dependencies ([#185](https://github.com/CategoricalData/hydra/issues/185)).+- Corrected handling of variable references in Python application terms.+- Fixed nullary functions in Python (added empty parens).+- Added missing alternatives in `hydra.rewriting.rewriteTermM`.++### Bookkeeping++- Closed long-standing Hydra-Java module-port issues+  ([#9](https://github.com/CategoricalData/hydra/issues/9), [#11](https://github.com/CategoricalData/hydra/issues/11), [#15](https://github.com/CategoricalData/hydra/issues/15), [#16](https://github.com/CategoricalData/hydra/issues/16), [#18](https://github.com/CategoricalData/hydra/issues/18), [#19](https://github.com/CategoricalData/hydra/issues/19), [#20](https://github.com/CategoricalData/hydra/issues/20), [#21](https://github.com/CategoricalData/hydra/issues/21), [#22](https://github.com/CategoricalData/hydra/issues/22), [#23](https://github.com/CategoricalData/hydra/issues/23), [#25](https://github.com/CategoricalData/hydra/issues/25), [#44](https://github.com/CategoricalData/hydra/issues/44), [#45](https://github.com/CategoricalData/hydra/issues/45));+  most of the underlying work shipped in earlier releases.+- Closed Hydra-Java DSL-port issues ([#24](https://github.com/CategoricalData/hydra/issues/24), [#26](https://github.com/CategoricalData/hydra/issues/26), [#27](https://github.com/CategoricalData/hydra/issues/27), [#28](https://github.com/CategoricalData/hydra/issues/28), [#29](https://github.com/CategoricalData/hydra/issues/29)) and+  the Scala/Avro tracking issues ([#77](https://github.com/CategoricalData/hydra/issues/77), [#81](https://github.com/CategoricalData/hydra/issues/81)); these had been resolved+  by earlier releases (Scala in 0.14.0, Avro bidi in 0.14.0).++### Documentation++- Updated Hydra-Haskell README.+- Updated Hydra-Java README with corrected links.+- Updated JavaDocs.++---++## [0.11.0] - 2025-03-16++Untagged in-repository version.+A type-inference rebase release: major restructuring of inference internals+to align with Algorithm W, plus a sweeping removal of explicit type annotations+across the kernel.++### Highlights++- **Inference rebased on Algorithm W** ([#118](https://github.com/CategoricalData/hydra/issues/118)): unify-early strategy ([#146](https://github.com/CategoricalData/hydra/issues/146)),+  infinite-type checks restored ([#116](https://github.com/CategoricalData/hydra/issues/116)), recursive-element inference ([#90](https://github.com/CategoricalData/hydra/issues/90)),+  unified handling of elements and let bindings ([#112](https://github.com/CategoricalData/hydra/issues/112)), minimized manual+  type annotations ([#119](https://github.com/CategoricalData/hydra/issues/119)).+- **`hydra.inference` module** introduced for inference type definitions;+  substitution and unification moved up a level.++### Breaking changes++- Argument order of `Logic.ifElse` changed ([#147](https://github.com/CategoricalData/hydra/issues/147)).++### New features++- Inference test cases promoted into the common test suite ([#148](https://github.com/CategoricalData/hydra/issues/148));+  added "kernel examples" inference tests; more recursion / mutual recursion cases.++### Improvements++- Removed type annotations across all kernel sources ([#119](https://github.com/CategoricalData/hydra/issues/119)).+- Migrated to fully-applied primitive library DSLs ([#157](https://github.com/CategoricalData/hydra/issues/157)).+- Aligned `Graph` and `Element` with `LetBinding` ([#159](https://github.com/CategoricalData/hydra/issues/159)).+- `hydra.core.Unit` recognized by the Haskell coder.+- Refactored inference tests.++### Bug fixes++- Fix for underdetermined type in `hydra.decode`.++---++## [0.10.0] - 2025-02-19++Untagged in-repository version.+Adds Hydra-Python (initial completion), a GraphSON coder,+provisional Hydra-Go support, and dot-separated namespace conventions.++### Highlights++- **Hydra-Python initial completion** ([#80](https://github.com/CategoricalData/hydra/issues/80), [#66](https://github.com/CategoricalData/hydra/issues/66)): Python coder, generated kernel+  modules in Python, and full library implementations across kernel namespaces.+- **GraphSON coder** ([#79](https://github.com/CategoricalData/hydra/issues/79)).+- **Provisional Hydra-Go** ([#65](https://github.com/CategoricalData/hydra/issues/65)).+- **Namespace formatting** standardized to dot-separated form ([#151](https://github.com/CategoricalData/hydra/issues/151)).++### Breaking changes++- All namespaces use dot-separated form ([#151](https://github.com/CategoricalData/hydra/issues/151)).++### New features++- Hydra-Python: Python coder; generated `hydra.coreEncoding`, `hydra.strip`,+  `hydra.literals` and other kernel modules in Python; tuple-expression support;+  updated `hydra.constants` in Python.+- GraphSON coder ([#79](https://github.com/CategoricalData/hydra/issues/79)).+- Provisional Hydra-Go ([#65](https://github.com/CategoricalData/hydra/issues/65)).+- `string-to-char-list` and `char-list-to-string` primitives ([#149](https://github.com/CategoricalData/hydra/issues/149)).+- Restored normalization for generated term variables ([#145](https://github.com/CategoricalData/hydra/issues/145)).++### Improvements++- Re-leveled dependency tiers ([#135](https://github.com/CategoricalData/hydra/issues/135)).+- Updated all Java, Haskell, hydra-ext, and Python primitives to current sources.++---++## [0.9.0] - 2025-01-06++Untagged in-repository version.+Bootstraps Hydra-Python and the JSON Schema coder; introduces a fluent Java DSL for flows.++### Highlights++- **JSON Schema coder** ([#141](https://github.com/CategoricalData/hydra/issues/141)).+- **Python coder scaffold**: serde, language constraints, and generated Haskell+  sources for Python.+- **Fluent-style Java DSL for flows** ([#143](https://github.com/CategoricalData/hydra/issues/143)).++### New features++- JSON Schema coder ([#141](https://github.com/CategoricalData/hydra/issues/141)).+- Python coder scaffold and serde.+- Provisional Python language constraints module.+- Fluent-style Java DSL for flows ([#143](https://github.com/CategoricalData/hydra/issues/143)); moved Flows DSL into `hydra.dsl`.+- New `hydra.coders` DSL.+- New DSL functions for core variants.++### Improvements++- Generalized `Namespaces` so it can be used for languages other than Haskell.+- Local-name-only Python type aliases (the directory tree provides the namespace).+- Coders DSL used to simplify Java language-constraint sources.+- Removed redundant productions in the Python grammar.+- Added unit tests for fluent flows.++### Documentation++- Updated JavaDocs for 0.9.0.++---++## [0.8.1] - 2024-09-24++Patch release.++### Improvements++- Generalized the property graph merging utility to accept type systems other than Hydra Core.++---++## [0.8.0] - 2024-09-09++C# support ([#139](https://github.com/CategoricalData/hydra/issues/139)), Graphviz DOT coder, term accessors, and a small breaking change in Java+to support Spark.++### Breaking changes++- Made `FlowException` serializable for Spark support (may affect existing error handling).++### New features++- C# syntax module based on the Microsoft ANTLR grammar ([#139](https://github.com/CategoricalData/hydra/issues/139)).+- Graphviz DOT coder ([#136](https://github.com/CategoricalData/hydra/issues/136)):+  full support for lambdas and recursive `let` statements;+  compact 'accessor graph' visualization option;+  customizable label styles;+  highlighted let-bound terms; edge labels.+- Term accessor type to facilitate lenses and path-aware transformations.+- OpenGQL grammar and generated Haskell for the OpenGQL model ([#140](https://github.com/CategoricalData/hydra/issues/140)).++### Improvements++- Added `</>` as alternative to `@@` application operator in Haskell DSLs.+- Utilities for working with term accessors.+- More specific exception class for JSON decoding.+- Additional convenience methods on `JsonDecoding.java`.+- Used transitive dependencies in hydra-java and hydra-ext.+- Upgraded ANTLR to address a vulnerability.+- Moved TinkerPop utilities from `hydra/ext/org/apache/tinkerpop` into `hydra/pg`.++---++## [0.7.0] - 2024-08-21++Major refactoring of module organization and namespace management.+Establishes the DNS-style namespace convention in hydra-ext+and lifts property-graph models into a top-level `hydra/pg` namespace.++### Breaking changes++- Namespace reorganization: migrated to DNS-based module naming in hydra-ext ([#138](https://github.com/CategoricalData/hydra/issues/138)):+  `hydra/ext/avro` → `hydra/ext/org/apache/avro`,+  `hydra/ext/graphql` → `hydra/ext/org/graphql`,+  `hydra/ext/yaml` → `hydra/ext/org/yaml`,+  `hydra/ext/json/decoding` → `hydra/ext/org/json/decoding`,+  `hydra/ext/rdf` → `hydra/ext/org/w3/rdf`,+  `hydra/ext/shacl` → `hydra/ext/org/w3/shacl`,+  `hydra/ext/tinkerpop` → `hydra/ext/org/apache/tinkerpop`.+- Property graphs: promoted property graph modules from+  `hydra/ext/org/apache/tinkerpop` to `hydra/pg`.+- Project structure: renamed `hydra-extensions` to `hydra-ext`;+  renamed `hydra/langs` to `hydra/ext` ([#138](https://github.com/CategoricalData/hydra/issues/138));+  made hydra-ext into a Gradle subproject alongside hydra-java;+  added top-level Gradle build.++### Improvements++- Moved TinkerPop modules from hydra-haskell into hydra-ext for better separation.+- Moved miscellaneous models to `Hydra.Ext.Other` ([#138](https://github.com/CategoricalData/hydra/issues/138)).+- Moved XML Schema, SQL, ShEx, and OWL models to hydra-ext.+- Moved KQL, Parquet, and Python modules to hydra-ext.+- Separated JavaDocs for hydra-java and hydra-ext.+- Updated publishing configuration in `build.gradle`.++---++## [0.6.0] - 2024-08-19++Type-system improvements (System F type abstraction/application; eliminated polymorphic+types from Hydra Core), expanded language support (Graphviz DOT model, generated Java+field-name constants), and revamped annotation handling.++### Breaking changes++- Eliminated `InferenceContext` helper type ([#103](https://github.com/CategoricalData/hydra/issues/103)).+- Changed typing environment from `Map Name Type` to `Map Name TypeScheme` in graphs ([#76](https://github.com/CategoricalData/hydra/issues/76)).+- Updated annotations and flows to use `map<Name, Term>` instead of `map<string, Term>` ([#133](https://github.com/CategoricalData/hydra/issues/133)).+- Unified `Name` and `FieldName`; aliased to `string` ([#121](https://github.com/CategoricalData/hydra/issues/121)).+- Renamed `hydra/core.UnitType` to `hydra/core.Unit` ([#123](https://github.com/CategoricalData/hydra/issues/123)).+- Eliminated annotation classes; Hydra has a single built-in notion of annotations ([#113](https://github.com/CategoricalData/hydra/issues/113)).+- Unified the `Nominal` type with `RowType` ([#115](https://github.com/CategoricalData/hydra/issues/115)).+- Removed the `extends` parameter for row types ([#132](https://github.com/CategoricalData/hydra/issues/132)).+- Replaced `hydra/Kv` with `hydra/Annotations` (rename revisited).++### New features++- System F support: `typeAbstraction` and `typeApplication` term constructors.+- Python3 syntax model based on the official Python BNF ([#80](https://github.com/CategoricalData/hydra/issues/80)).+- Graphviz DOT model and initial support ([#136](https://github.com/CategoricalData/hydra/issues/136)).+- Field-name constants generated in Java code generation ([#137](https://github.com/CategoricalData/hydra/issues/137)).+- Provisional Algorithm W implementation in Haskell (originally from @wisnesky), kept for reference.++### Improvements++- Primitive organization: Hydra primitives organized into libraries for namespace management.+- Code generation:+  - Escaped field name constants and `with_` methods in Java coder.+  - Generated field name constants in Java ([#137](https://github.com/CategoricalData/hydra/issues/137)).+  - Avoided duplicated comments in wrapper classes in generated Java.+  - Removed superfluous newtype comments in Haskell coder.+- Updated Delta Parquet model to follow the Java API more closely.+- Enriched OpenCypher features module with the complete list of standard Cypher functions.+- Added convenience methods to `JsonEncoding.java`.+- Restored `Hydra.Inference` subdirectory.+- Allowed arbitrary case for element names ([#7](https://github.com/CategoricalData/hydra/issues/7)).+- Corrected `uint8` representation in Java to `Short` ([#120](https://github.com/CategoricalData/hydra/issues/120)).++### Bug fixes++- Fixed handling of type annotations in term adapters.+- Fixed decoding of encoded terms annotated with a type.+- Corrected encoding of universal types in `showTerm` ([#117](https://github.com/CategoricalData/hydra/issues/117)).++---++## [0.5.3] - 2024-08-06++Patch release.+Preserves order of map key/value pairs and set elements during JSON serialization.++### Bug fixes++- Preserved ordering of key/value pairs and set elements in JSON output.++### New features++- Polymorphic `equal` primitive in Haskell and Java.+- Generated Hydra type definitions together with native instantiations of Hydra types+  in Haskell (a stepping stone toward generated coders, [#47](https://github.com/CategoricalData/hydra/issues/47)).++---++## [0.5.1] - 2024-08-01++Patch release.+Dependency updates and minor improvements following the 0.5.0 Hydra Core overhaul.++### Improvements++- Updated dependencies.++---++## [0.5.0] - 2024-07-26++Untagged in-repository version.+Major Hydra Core overhaul: eliminates polymorphic types in favor of explicit+TypeScheme; consolidates the annotation system; revamps `Let` term encoding.++### Breaking changes++- Eliminated polymorphic types from Hydra Core in Haskell ([#125](https://github.com/CategoricalData/hydra/issues/125)):+  removed type parameters from `hydra/core` types in Hydra-Java; replaced type schemes+  with forall types ([#76](https://github.com/CategoricalData/hydra/issues/76)).+- Revamped core encoding and `Let`: bindings are now a list of+  `(name, term, optional type scheme)` triplets instead of a map.+- Unified typed-term types and moved `TypeScheme` into Hydra Core.+- Eliminated `AnnotationClass`; Hydra has a single built-in notion of annotations now ([#113](https://github.com/CategoricalData/hydra/issues/113)).+- Renamed and refactored phantom-type wrappers ([#128](https://github.com/CategoricalData/hydra/issues/128)); removed deprecated `Reference`+  type from `hydra/phantoms`.+- Removed incomplete support for stream types and stream terms ([#88](https://github.com/CategoricalData/hydra/issues/88)).++### New features++- `LiteralTypes` DSL module so literals can be built independently of terms.+- `Hydra.Minimal` module to facilitate collaboration on Hydra application prototypes.+- `TypeConstraint` type added to the kernel.+- `foldl` primitive in Haskell and Java; DSL uses `foldl` instead of the built-in+  list elimination term by default.++### Improvements++- Improvements to Java and JSON coders; improved untyped JSON coder.+- Pre-order and post-order term-traversal test cases added.+- Refactored type inference tests.+- Restored generated Java (Core through Tier 3) with only a few manual tweaks.+- Element names allow arbitrary case ([#7](https://github.com/CategoricalData/hydra/issues/7)).++---++## [0.4.0] - 2024-07-10++Untagged in-repository version.+Adds null checks to generated Java (breaking, but justified — Hydra-Java disallows nulls),+the Gremlin model, and Protobuf annotation refinements.++### Breaking changes++- Null checks added to constructors and `withXXX()` methods in generated Java;+  null values are no longer permitted in Hydra-Java.+- Removed Java 8 restriction (raised minimum Java version).++### New features++- Gremlin model ([#127](https://github.com/CategoricalData/hydra/issues/127)); generated Haskell for Gremlin.+- `hydra/lib/lists.safeHead` primitive in Haskell (necessary in Java due to eager evaluation).+- Support for Protobuf `deprecated=true` annotation; non-string Protobuf options.+- Java options in generated Protobuf.++### Improvements++- Used `java.util.Objects.requireNonNull` instead of a custom null check in generated Java.+- Added 'automatically generated file' disclaimer to generated Java files.+- Added autogen comments and null checks to (most) generated Java.+- Added necessary schema-level dependencies for built-in modules.+- Added null checks to the Flows DSL in Java.++---++## [0.3.0] - 2024-06-05++Untagged in-repository version.+Adds an OpenCypher parser based on an ANTLR grammar ([#124](https://github.com/CategoricalData/hydra/issues/124)),+a KQL (Kusto Query Language) model and serializer,+a Delta Parquet model,+and an early property graph queries model in the style of Cypher.++### New features++- OpenCypher parser based on the OpenCypher M23 ANTLR grammar ([#124](https://github.com/CategoricalData/hydra/issues/124));+  ANTLR added as a build dependency;+  large Cypher test suite drawn from the Cypher Manual.+- Cypher-to-PG-query transformer (Java; partially implemented).+- Property graph queries DSL and generated property graph query classes in Java.+- Property graph queries model (Cypher / future GQL style).+- KQL (Kusto Query Language) initial model and basic serializer.+- Delta Parquet model with generated sources.++### Improvements++- Java options and `deprecated=true` support in generated Protobuf.+- Convenience methods in Java.++---++## [0.2.0] - 2024-01-10++Hydra-Java 0.2.0 release.+Significant Java tooling and DSL build-out, Protobuf bidirectional coder,+Python3 syntax model ([#80](https://github.com/CategoricalData/hydra/issues/80)), property graph validation ([#100](https://github.com/CategoricalData/hydra/issues/100)),+JSON bidirectional support in Java ([#104](https://github.com/CategoricalData/hydra/issues/104)),+OpenCypher initial model ([#108](https://github.com/CategoricalData/hydra/issues/108)),+Cypher unification, and the start of the typed-DSL story in Haskell.++### New features++- Property graph validation: validator type, optionality on property types,+  customizable id/property handling in the merging adapter, fine-tuned validation,+  comprehensive unit tests in Java ([#100](https://github.com/CategoricalData/hydra/issues/100)).+- Bidirectional JSON serialization/deserialization in Java ([#104](https://github.com/CategoricalData/hydra/issues/104)).+- Checkstyle configuration based on Google style guide ([#111](https://github.com/CategoricalData/hydra/issues/111)).+- Python3 syntax model based on the official Python BNF ([#80](https://github.com/CategoricalData/hydra/issues/80)).+- Initial OpenCypher model and parser scaffold ([#108](https://github.com/CategoricalData/hydra/issues/108)).+- Protobuf coder ([#99](https://github.com/CategoricalData/hydra/issues/99)): type encoder, field numbering, formatting, deprecated annotation,+  Java options, non-string options.+- Adapter framework tolerance of recursive types ([#58](https://github.com/CategoricalData/hydra/issues/58)); default branch for case+  statements ([#60](https://github.com/CategoricalData/hydra/issues/60)).+- Topological sort in type inference ([#67](https://github.com/CategoricalData/hydra/issues/67)).+- `hydra/query` model ([#73](https://github.com/CategoricalData/hydra/issues/73)).+- Records-to-TinkerPop-elements coder ([#64](https://github.com/CategoricalData/hydra/issues/64)).+- Equality primitives for all literal types ([#92](https://github.com/CategoricalData/hydra/issues/92)); polymorphic logic primitives ([#95](https://github.com/CategoricalData/hydra/issues/95)).+- `mapKeys` primitive ([#93](https://github.com/CategoricalData/hydra/issues/93)).+- Tuple support in Java ([#94](https://github.com/CategoricalData/hydra/issues/94)).+- Java 8 as an optional target ([#96](https://github.com/CategoricalData/hydra/issues/96)); optionally generate Java classes as `Serializable` ([#97](https://github.com/CategoricalData/hydra/issues/97)).+- Bidirectional "merged vertex" and "merged edge" coders ([#106](https://github.com/CategoricalData/hydra/issues/106)).+- Nullability and parameterization in the tabular model ([#107](https://github.com/CategoricalData/hydra/issues/107)).+- Term-level / type-level dependency distinction for modules ([#109](https://github.com/CategoricalData/hydra/issues/109)).+- Flatten nested `let` terms for transformation into Java ([#110](https://github.com/CategoricalData/hydra/issues/110)).+- Various `Flow` utilities in Java: `bind3`; convenience methods for consuming results;+  `fromFlow` variants including one that throws.++### Improvements++- Hydra-Java module ports ([#10](https://github.com/CategoricalData/hydra/issues/10), [#12](https://github.com/CategoricalData/hydra/issues/12), [#13](https://github.com/CategoricalData/hydra/issues/13), [#14](https://github.com/CategoricalData/hydra/issues/14), [#17](https://github.com/CategoricalData/hydra/issues/17), [#30](https://github.com/CategoricalData/hydra/issues/30), [#31](https://github.com/CategoricalData/hydra/issues/31), [#32](https://github.com/CategoricalData/hydra/issues/32), [#33](https://github.com/CategoricalData/hydra/issues/33), [#34](https://github.com/CategoricalData/hydra/issues/34), [#35](https://github.com/CategoricalData/hydra/issues/35),+  [#38](https://github.com/CategoricalData/hydra/issues/38), [#39](https://github.com/CategoricalData/hydra/issues/39), [#41](https://github.com/CategoricalData/hydra/issues/41), [#43](https://github.com/CategoricalData/hydra/issues/43), [#48](https://github.com/CategoricalData/hydra/issues/48), [#71](https://github.com/CategoricalData/hydra/issues/71), [#72](https://github.com/CategoricalData/hydra/issues/72), [#75](https://github.com/CategoricalData/hydra/issues/75), [#83](https://github.com/CategoricalData/hydra/issues/83)):+  `Common`, `CoreEncoding`, `CoreLanguage`, `Kernel`, `Flows`, `Lists`, `Literals`,+  `Math`, `Sets`, `Strings`, `Optionals`, `Maps`, `Flows` primitives, `Sets` primitives.+- Common DSLs in Haskell: Flow support ([#68](https://github.com/CategoricalData/hydra/issues/68)), Let support ([#69](https://github.com/CategoricalData/hydra/issues/69)).+- Renamed annotation type parameter `m` to `a` ([#41](https://github.com/CategoricalData/hydra/issues/41)).+- Updated wrapper, element, and variable types ([#75](https://github.com/CategoricalData/hydra/issues/75)).+- Removed `schema` field from `Element` ([#83](https://github.com/CategoricalData/hydra/issues/83)).+- GraphQL coder ([#78](https://github.com/CategoricalData/hydra/issues/78)).+- Code quality:+  - Updated main and test sources to conform to Checkstyle configuration ([#111](https://github.com/CategoricalData/hydra/issues/111)).+  - Extended line-length limit from 100 to 120 characters.+  - Disabled overly restrictive indentation and import-order rules.+  - Disabled `RightCurlyAlone` for compact inline map definitions.+- Streamlined property-graph element merging to unify properties with identical+  keys and types ([#106](https://github.com/CategoricalData/hydra/issues/106)).+- Convenience functions for printing literals and literal types in the Java DSL.+- Minor addition to JSON decoding.+- DSL convenience: alphabetic-case flexibility for element names ([#7](https://github.com/CategoricalData/hydra/issues/7)).++### Bug fixes++- Fixed type annotations for the Sets Java port ([#48](https://github.com/CategoricalData/hydra/issues/48)).+- Various minor fixes during the Java module ports.++---++## [0.1.1] - 2022-12-04++Patch release the same day as 0.1.0.+README polish only.++---++## [0.1.0] - 2022-12-04++First packaged release of Hydra (Hackage).+This release contains the complete foundation of Hydra.++### Core language++- Hydra's core type and data languages (`hydra.core`).+- Core models for graphs and modules.+- Computation model with the `Flow` monad.+- BNF grammars support.+- Phantom types.+- Basic operations on types and terms.++### Type system++- Hindley–Milner-style type inference.+- Type schemes with polymorphism.+- Type/term validation.++### Transformations++- Adapter system: type-to-type rewriting and transformation.+- Coders:+  Haskell coder (types and terms),+  Java coder (types and terms),+  partial Scala coder (terms only),+  Avro coder,+  JSON coder,+  PDL (Pegasus Data Language) coder,+  RDF + SHACL coder,+  YAML coder.++### Language support++- Models for GraphQL, OWL (Web Ontology Language), ShEx (Shape Expressions),+  and TinkerPop-style property graphs.++### Developer tools++- Type-construction DSL and term-construction DSL.+- QuickCheck property-based tests.++### Implementations++- Hydra-Haskell: bootstrapping implementation with full kernel.+- Hydra-Java: Java implementation with mature tooling.++### Community++- Set up the LambdaGraph Discord server ([#2](https://github.com/CategoricalData/hydra/issues/2)).++---++## Contributing++We welcome contributions! Please see:++- [Developer recipes](https://github.com/CategoricalData/hydra/tree/main/docs/recipes)+- [LambdaGraph Discord](https://bit.ly/lg-discord)++## Release process++See the [Release policy](https://github.com/CategoricalData/hydra/wiki/Release-policy)+wiki page and [docs/release-workflow.md](docs/release-workflow.md)+for information on how Hydra releases are versioned, built, and published.

@@ -0,0 +1,201 @@+                                 Apache License+                           Version 2.0, January 2004+                        http://www.apache.org/licenses/++   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION++   1. Definitions.++      "License" shall mean the terms and conditions for use, reproduction,+      and distribution as defined by Sections 1 through 9 of this document.++      "Licensor" shall mean the copyright owner or entity authorized by+      the copyright owner that is granting the License.++      "Legal Entity" shall mean the union of the acting entity and all+      other entities that control, are controlled by, or are under common+      control with that entity. For the purposes of this definition,+      "control" means (i) the power, direct or indirect, to cause the+      direction or management of such entity, whether by contract or+      otherwise, or (ii) ownership of fifty percent (50%) or more of the+      outstanding shares, or (iii) beneficial ownership of such entity.++      "You" (or "Your") shall mean an individual or Legal Entity+      exercising permissions granted by this License.++      "Source" form shall mean the preferred form for making modifications,+      including but not limited to software source code, documentation+      source, and configuration files.++      "Object" form shall mean any form resulting from mechanical+      transformation or translation of a Source form, including but+      not limited to compiled object code, generated documentation,+      and conversions to other media types.++      "Work" shall mean the work of authorship, whether in Source or+      Object form, made available under the License, as indicated by a+      copyright notice that is included in or attached to the work+      (an example is provided in the Appendix below).++      "Derivative Works" shall mean any work, whether in Source or Object+      form, that is based on (or derived from) the Work and for which the+      editorial revisions, annotations, elaborations, or other modifications+      represent, as a whole, an original work of authorship. For the purposes+      of this License, Derivative Works shall not include works that remain+      separable from, or merely link (or bind by name) to the interfaces of,+      the Work and Derivative Works thereof.++      "Contribution" shall mean any work of authorship, including+      the original version of the Work and any modifications or additions+      to that Work or Derivative Works thereof, that is intentionally+      submitted to Licensor for inclusion in the Work by the copyright owner+      or by an individual or Legal Entity authorized to submit on behalf of+      the copyright owner. For the purposes of this definition, "submitted"+      means any form of electronic, verbal, or written communication sent+      to the Licensor or its representatives, including but not limited to+      communication on electronic mailing lists, source code control systems,+      and issue tracking systems that are managed by, or on behalf of, the+      Licensor for the purpose of discussing and improving the Work, but+      excluding communication that is conspicuously marked or otherwise+      designated in writing by the copyright owner as "Not a Contribution."++      "Contributor" shall mean Licensor and any individual or Legal Entity+      on behalf of whom a Contribution has been received by Licensor and+      subsequently incorporated within the Work.++   2. Grant of Copyright License. Subject to the terms and conditions of+      this License, each Contributor hereby grants to You a perpetual,+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable+      copyright license to reproduce, prepare Derivative Works of,+      publicly display, publicly perform, sublicense, and distribute the+      Work and such Derivative Works in Source or Object form.++   3. Grant of Patent License. Subject to the terms and conditions of+      this License, each Contributor hereby grants to You a perpetual,+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable+      (except as stated in this section) patent license to make, have made,+      use, offer to sell, sell, import, and otherwise transfer the Work,+      where such license applies only to those patent claims licensable+      by such Contributor that are necessarily infringed by their+      Contribution(s) alone or by combination of their Contribution(s)+      with the Work to which such Contribution(s) was submitted. If You+      institute patent litigation against any entity (including a+      cross-claim or counterclaim in a lawsuit) alleging that the Work+      or a Contribution incorporated within the Work constitutes direct+      or contributory patent infringement, then any patent licenses+      granted to You under this License for that Work shall terminate+      as of the date such litigation is filed.++   4. Redistribution. You may reproduce and distribute copies of the+      Work or Derivative Works thereof in any medium, with or without+      modifications, and in Source or Object form, provided that You+      meet the following conditions:++      (a) You must give any other recipients of the Work or+          Derivative Works a copy of this License; and++      (b) You must cause any modified files to carry prominent notices+          stating that You changed the files; and++      (c) You must retain, in the Source form of any Derivative Works+          that You distribute, all copyright, patent, trademark, and+          attribution notices from the Source form of the Work,+          excluding those notices that do not pertain to any part of+          the Derivative Works; and++      (d) If the Work includes a "NOTICE" text file as part of its+          distribution, then any Derivative Works that You distribute must+          include a readable copy of the attribution notices contained+          within such NOTICE file, excluding those notices that do not+          pertain to any part of the Derivative Works, in at least one+          of the following places: within a NOTICE text file distributed+          as part of the Derivative Works; within the Source form or+          documentation, if provided along with the Derivative Works; or,+          within a display generated by the Derivative Works, if and+          wherever such third-party notices normally appear. The contents+          of the NOTICE file are for informational purposes only and+          do not modify the License. You may add Your own attribution+          notices within Derivative Works that You distribute, alongside+          or as an addendum to the NOTICE text from the Work, provided+          that such additional attribution notices cannot be construed+          as modifying the License.++      You may add Your own copyright statement to Your modifications and+      may provide additional or different license terms and conditions+      for use, reproduction, or distribution of Your modifications, or+      for any such Derivative Works as a whole, provided Your use,+      reproduction, and distribution of the Work otherwise complies with+      the conditions stated in this License.++   5. Submission of Contributions. Unless You explicitly state otherwise,+      any Contribution intentionally submitted for inclusion in the Work+      by You to the Licensor shall be under the terms and conditions of+      this License, without any additional terms or conditions.+      Notwithstanding the above, nothing herein shall supersede or modify+      the terms of any separate license agreement you may have executed+      with Licensor regarding such Contributions.++   6. Trademarks. This License does not grant permission to use the trade+      names, trademarks, service marks, or product names of the Licensor,+      except as required for reasonable and customary use in describing the+      origin of the Work and reproducing the content of the NOTICE file.++   7. Disclaimer of Warranty. Unless required by applicable law or+      agreed to in writing, Licensor provides the Work (and each+      Contributor provides its Contributions) on an "AS IS" BASIS,+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or+      implied, including, without limitation, any warranties or conditions+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A+      PARTICULAR PURPOSE. You are solely responsible for determining the+      appropriateness of using or redistributing the Work and assume any+      risks associated with Your exercise of permissions under this License.++   8. Limitation of Liability. In no event and under no legal theory,+      whether in tort (including negligence), contract, or otherwise,+      unless required by applicable law (such as deliberate and grossly+      negligent acts) or agreed to in writing, shall any Contributor be+      liable to You for damages, including any direct, indirect, special,+      incidental, or consequential damages of any character arising as a+      result of this License or out of the use or inability to use the+      Work (including but not limited to damages for loss of goodwill,+      work stoppage, computer failure or malfunction, or any and all+      other commercial damages or losses), even if such Contributor+      has been advised of the possibility of such damages.++   9. Accepting Warranty or Additional Liability. While redistributing+      the Work or Derivative Works thereof, You may choose to offer,+      and charge a fee for, acceptance of support, warranty, indemnity,+      or other liability obligations and/or rights consistent with this+      License. However, in accepting such obligations, You may act only+      on Your own behalf and on Your sole responsibility, not on behalf+      of any other Contributor, and only if You agree to indemnify,+      defend, and hold each Contributor harmless for any liability+      incurred by, or claims asserted against, such Contributor by reason+      of your accepting any such warranty or additional liability.++   END OF TERMS AND CONDITIONS++   APPENDIX: How to apply the Apache License to your work.++      To apply the Apache License to your work, attach the following+      boilerplate notice, with the fields enclosed by brackets "[]"+      replaced with your own identifying information. (Don't include+      the brackets!)  The text should be enclosed in the appropriate+      comment syntax for the file format. We also recommend that a+      file or class name and description of purpose be included on the+      same "printed page" as the copyright notice for easier+      identification within third-party archives.++   Copyright [yyyy] [name of copyright owner]++   Licensed under the Apache License, Version 2.0 (the "License");+   you may not use this file except in compliance with the License.+   You may obtain a copy of the License at++       http://www.apache.org/licenses/LICENSE-2.0++   Unless required by applicable law or agreed to in writing, software+   distributed under the License is distributed on an "AS IS" BASIS,+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.+   See the License for the specific language governing permissions and+   limitations under the License.

@@ -0,0 +1,3 @@+Hydra+Copyright 2021-2026 The Hydra authors+https://github.com/CategoricalData/hydra

@@ -0,0 +1,38 @@+cabal-version: 1.12++-- This file has been generated from package.yaml by hpack version 0.39.1.+--+-- see: https://github.com/sol/hpack++name:           hydra-jvm+version:        0.17.0+synopsis:       Shared JVM support for Hydra's Java, Scala, and Clojure coders+description:    Hydra is an implementation of the LambdaGraph data model, which takes advantage of an isomorphism between labeled hypergraphs and typed lambda calculus: in Hydra, "graphs are programs, and programs are graphs". Common JVM support for Hydra (shared by hydra-java, hydra-scala, hydra-clojure)+category:       Data+homepage:       https://github.com/CategoricalData/hydra#readme+bug-reports:    https://github.com/CategoricalData/hydra/issues+author:         Joshua Shinavier <josh@fortytwo.net>+maintainer:     Joshua Shinavier <josh@fortytwo.net>+license:        Apache-2.0+license-file:   LICENSE+build-type:     Simple+extra-source-files:+    LICENSE+    NOTICE+    CHANGELOG.md++source-repository head+  type: git+  location: https://github.com/CategoricalData/hydra++library+  exposed-modules:+      Hydra.Jvm.Serde+  other-modules:+      Paths_hydra_jvm+  hs-source-dirs:+      src/main/haskell+  build-depends:+      base >=4.19.0 && <4.22+    , hydra-kernel ==0.17.0+  default-language: Haskell2010

@@ -0,0 +1,50 @@+-- Note: this is an automatically generated file. Do not edit.+-- | Common JVM serialization helpers: Java/Scala string and character escaping++module Hydra.Jvm.Serde where+import qualified Hydra.Overlay.Haskell.Lib.Lists as Lists+import qualified Hydra.Overlay.Haskell.Lib.Logic as Logic+import qualified Hydra.Overlay.Haskell.Lib.Strings as Strings+import Prelude hiding  (Enum, Ordering, decodeFloat, encodeFloat, fail, map, pure, sum)++-- | Escape a single character for use in a Java/Scala string or character literal+escapeJavaChar :: Int -> String+escapeJavaChar c =+    Logic.ifElse (c == 34) "\\\""+    (Logic.ifElse (c == 92) "\\\\"+    (Logic.ifElse (c == 10) "\\n"+    (Logic.ifElse (c == 13) "\\r"+    (Logic.ifElse (c == 9)  "\\t"+    (Logic.ifElse (c == 8)  "\\b"+    (Logic.ifElse (c == 12) "\\f"+    (Logic.ifElse (c >= 32 && c < 127) (Strings.fromList [c])+    (javaUnicodeEscape c))))))))++-- | Escape a string for use in a Java/Scala string literal+escapeJavaString :: String -> String+escapeJavaString s = Strings.cat (Lists.map escapeJavaChar (Strings.toList s))++-- | Convert an integer in [0,15] to the ASCII code of the corresponding hex digit+hexDigit :: Int -> Int+hexDigit n = Logic.ifElse (n < 10) (n + 48) (n - 10 + 65)++-- | Produce a \\uXXXX Unicode escape for the given code point+javaUnicodeEscape :: Int -> String+javaUnicodeEscape n =+    Logic.ifElse (n > 65535)+      (let n' = n - 65536+           hi = 55296 + (n' `div` 1024)+           lo = 56320 + (n' `mod` 1024)+       in "\\u" ++ padHex4 hi ++ "\\u" ++ padHex4 lo)+      ("\\u" ++ padHex4 n)++-- | Pad an integer to 4 hex digits+padHex4 :: Int -> String+padHex4 n =+    let d3 = n `div` 4096+        r3 = n `mod` 4096+        d2 = r3 `div` 256+        r2 = r3 `mod` 256+        d1 = r2 `div` 16+        d0 = r2 `mod` 16+    in Strings.fromList [hexDigit d3, hexDigit d2, hexDigit d1, hexDigit d0]