scheme-langserver

Release History

2.1.7

Bug-fix release improving log replay robustness, custom file-filter support, identifier analysis for script-level define-record-type, and various internal fixes.

What’s new in 2.1.7

• Log replay:
  • bin/log-debug.sps and bin/parallel-log-debug.sps now support multi-line JSON messages in production logs by buffering body lines until the next read-message/send-message marker.
• Workspace / file-filter:
  • New file-filter API supports presets ('akku, 'txt, 'scheme), custom extension lists, and predicate procedures. CLI split into --package-manager/-p (preset) and --file-filter/-f (extension list).
  • .sld is included in the standard Scheme extension set.
• Identifier analysis:
  • Fixed a crash when a define-record-type with single-argument immutable fields (e.g. (immutable expression)) appears at the top level of a script file. The rule now correctly falls back to the document’s ordered reference list when the parent index-node is ().
  • dereference-index-node is inlined and repeated dereferences are avoided for better performance.
• Type inference:
  • Trivial type rule for include/resolve libraries now resolves pointer references to real definitions.
• Internal:
  • Replaced custom I/O helpers with Chez built-ins.
  • Multi-thread server initialization now uses Chez’s built-in threaded? check.
  • ufo-match wildcard pattern uses :_ instead of _ to comply with R6RS restrictions on syntax-rules auxiliary literals.

2.1.6

Bug-fix release completing shared/cyclic literal handling in binding forms and enabling workspace cache saves from compiled binaries.

What’s new in 2.1.6

• Identifier analysis:
  • All remaining binding-form rules (let, let*, letrec, let-values, do, fluid-let, let-syntax, letrec-syntax, with-syntax, syntax-case, syntax-rules, define-syntax, define-record-type, s7/lambda*, s7/define*) now dereference shared index-nodes before accessing children, preventing crashes or infinite loops on cyclic parameter/binding lists.
  • New regression test tests/analysis/identifier/rules/test-shared-reference-binding-forms.sps covers these edge cases.
• Build:
  • Release and development builds now use compile-chez-program --full-chez, so the compiled ./run binary retains the $write-fasl-bytevectors primitive and can save the workspace FASL cache.
  • Dockerfile, Dockerfile.musl, and CI workflows updated to install the static libraries (libuuid, libncurses, libtinfo) required for linking against full-chez.a.

2.1.5

Bug-fix release fixing infinite recursion and memory growth caused by cyclic/shared Scheme literals.

What’s new in 2.1.5

• AST:
  • init-index-node now detects cyclic and shared reader graph notation (#n= / #n#) via an internal compound->node hashtable.
  • Reference occurrences become leaf index-nodes with a new shared-reference field pointing to the defining node, keeping the AST acyclic.
• Bug fixes:
  • Fixes the memory leak/hang when analyzing files like swish src/swish/db.ss, which contains #0=(101 . #0#).
• Type inference:
  • Reference nodes generate a type constraint equating their type with the referenced definition node.

2.1.4

Feature and bug-fix release improving rest/dotted parameter handling and type inference.

What’s new in 2.1.4

• Type inference:
  • Rest/dotted parameters in define, lambda, and case-lambda now produce function signatures containing (inner:list? something? ...).
• Identifier analysis:
  • Rest/dotted parameter bindings in define, case-lambda, and do loop variables now point to the correct per-symbol index-node.
• AST:
  • Dotted-pair structure is now preserved in the index-node tree via synthetic annotations, preventing dot-position symbols from being dropped.

2.1.3

Feature release adding a Chez FASL workspace cache with incremental refresh.

What’s new in 2.1.3

• Workspace cache:
  • New --cache-path <dir> CLI option persists the analyzed workspace state to a Chez FASL file (<cache-path>/workspace.fasl).
  • On restart, unchanged files skip the expensive init-references phase; changed/added/deleted files are refreshed incrementally.
  • Cache manifest includes langserver version, Chez version, machine type, and record-layout fingerprint; any mismatch falls back to a cold start.
  • Non-serializable procedure fields (index-node-expansion-generator, identifier-reference-syntax-expander) are cleared before save and regenerated on demand.
  • Typical speedups: ~20x on small fixtures, ~30–50x on larger projects.
• Docs: Updated README.md, AGENTS.md, doc/analysis/workspace.md, and doc/build-and-startup.md with cache usage and implementation details.

2.1.2

Bug-fix release restoring bracket-mismatch diagnostics in the fault-tolerant tokenizer.

What’s new in 2.1.2

• Tokenizer:
  • Restore clear diagnostics for unmatched parentheses and brackets (e.g. unclosed parenthesis, unexpected close bracket) during fault-tolerant parsing, while keeping the R7RS/S7 compatibility fixes from 2.1.1.

2.1.1

Bug-fix release with R7RS/S7 tokenizer compatibility, identifier analysis fixes, and improved diagnostics.

What’s new in 2.1.1

• Bug fixes:
  • Fix typed-lambda / lambda dotted formals crash when encountering typed pair parameters like (ht hash-table?).
  • Fix identifier-compare? to guard symbol->string with symbol? checks.
  • Fix rename / alias import unused-import false positives.
  • Skip reference initialization for files with empty index-node-list.
• Tokenizer:
  • R7RS compatibility: #u8(...) → #vu8(...), #\null → #\nul, #\escape → #\esc, #; datum comments.
  • S7 compatibility: #<eof>, #<undefined>, #<fails:...>, #<predicate?>, #_id, #"..." raw strings.
• Diagnostics: Analysis errors now use display-condition for clearer compound-condition output.
• Docs: Added doc/top-environment.md documenting the top-environment mechanism.

2.1.0

Major release with expanded diagnostics, macro auto-resolution, performance optimizations, and Docker CI upgraded to Chez 10.4.1.

What’s new in 2.1.0

• LSP protocol: workspace/symbol search is now supported.
• Diagnostics:
  • Diagnostics now include standard LSP source and code fields.
  • Duplicate identifier detection in binding forms (lambda, case-lambda, let, letrec, let-values, do, define, with-syntax).
  • Unused import detection for only, except, rename, and alias modifiers.
  • Tokenizer syntax errors are now surfaced as document diagnostics.
  • Silent type-inference and type-rule failures are diagnosed.
  • Empty diagnostic arrays are sent to clear stale client-side errors.
• Macro auto-resolution: extended from syntax-rules to syntax-case, let-syntax, and letrec-syntax. Multi-layer macro cascade reference propagation is fixed.
• Type inference: define-record-type now infers record types; car/cdr family macros (caar, cadr, caddr, cadddr, caaar, cadar, etc.) have dedicated type rules.
• Performance: OPT-1~5 optimizations (expander-doc caching, hashtable reverse maps, tail-recursive accumulators, incremental all-pairs maintenance); MEM-1/3/6 memory optimizations for auto macro expansion; dedupe and reference hot-paths switched to eq? hashtables; matrix operations rewritten with cons+reverse.
• Robustness: hardened LSP message parsing against EOF, invalid Content-Length, and malformed JSON; shutdown/exit lifecycle fully compliant with the LSP spec; didChange auto-cancellation removed to comply with the spec; request-queue concurrency hardened with cancel barriers and log-mutex.
• Infrastructure: Docker build chain upgraded from Chez 9.6.4 to 10.4.1; chez-exe switched to the ufo5260987423/chez-exe fork for 10.x compatibility; test suite refactored to use AST search instead of hard-coded positions.

2.0.3

Fixed pretty-print bugs that were mixed with standard I/O.

2.0.2

Publish diagnoses, though now only can figure out “fail to find library”.

2.0.1

Fix many bugs.

2.0.0

Fix many bugs and switch between different top environments.

1.2.9

Now, enjoy type inference!

1.2.8

Now hover and auto completion is ready for use. I also have done many things about parsing fault tolerance.

1.2.7

Fix bugs on uri parsing, do you know LSP request uri may wrongly process escape characters?

1.2.6

Fault tolerant parser.

1.2.5

Fix: Some protocol api bugs. And now it’s basically smooth with Magic Scheme and Vscode.

1.2.4

Fix: hover api. It failed when processing meta.

1.2.3

Why completion api doesn’t work well? I don’t know and just fix.

1.2.2

I just fixed some bugs processing my own other projects.

1.2.1

I just fixed some bugs processing SS/SCM codes.

1.2.0

Re-construct the identifier catching mechanism with abstract interpreter.

1.1.1

Scheme-langserver now releases type information used in corresponding libraries! Its soundness is still not guaranteed!

1.1.0

Type inference has been embedded into autocompletion! And it uses a homemade DSL.

1.0.13

Fix bug: sometimes can’t shutdown server. Optimization: re-construct document-sync mechanism.

1.0.12

Add ss/scm-import-rnrs option.

1.0.11

Gradual Typing system, all basic rules have been passed.

1.0.10

Fix bugs in 1.0.9.

1.0.9

Abandoned: add parallel and synchronize mechanism.

1.0.8

Build index as document synchronizing instead of workspace initializing.

1.0.7

Catch syntax-* identifier bindings.

This site is open source. Improve this page.