Publish as versioned Gitea module; move dictionary pipeline out

- Rename module to gitea.iliadenisov.ru/developer/scrabble-solver so it can be
  consumed as a versioned dependency (no go.work replace / CI clone).
- De-internalize wordlist and dictdawg as public packages.
- Remove cmd/builddict, dictprep/, the dictionaries submodule and the dawg
  Makefile: the word-list parsing and DAWG build now live in the separate
  scrabble-dictionary repository, which publishes the DAWG set as a release artifact.
- internal/dict loads the committed dawg/en_sowpods.dawg fixture for cmd/stress.
- Update README/CLAUDE docs accordingly.
This commit is contained in:
Ilia Denisov
2026-06-04 19:11:46 +02:00
parent 63a7c663bf
commit 256999b42c
41 changed files with 93 additions and 402477 deletions
+13 -43
View File
@@ -1,24 +1,18 @@
// Package dict loads the English test dictionary as a DAWG, preferring the serialized
// cache under testdata and falling back to building from the dictionaries submodule.
// Paths are resolved relative to the repository root so it works both from the repo root
// (commands) and from a package directory (tests).
// Package dict loads the English test dictionary as a DAWG from the committed
// dawg/en_sowpods.dawg fixture, for the cmd/stress benchmark. The dictionary build
// pipeline (word-list parsing and DAWG construction from sources) now lives in the
// separate scrabble-dictionary repository; this package only loads the committed
// artifact. Paths are resolved relative to the repository root so it works both from
// the repo root (commands) and from a package directory (tests).
package dict
import (
"os"
"path/filepath"
"github.com/iliadenisov/alphabet"
dawg "github.com/iliadenisov/dafsa"
"scrabble-solver/internal/dictdawg"
"scrabble-solver/internal/wordlist"
)
// MinLen and MaxLen bound playable word lengths (a 15x15 board holds at most 15).
const (
MinLen = 2
MaxLen = 15
"gitea.iliadenisov.ru/developer/scrabble-solver/dictdawg"
)
func exists(p string) bool { _, err := os.Stat(p); return err == nil }
@@ -42,35 +36,11 @@ func Root() string {
}
}
// DAWGCache and WordlistPath locate the English cache file and source word list,
// relative to the repository root.
func DAWGCache() string { return filepath.Join(Root(), "testdata", "sowpods.dawg") }
func WordlistPath() string { return filepath.Join(Root(), "dictionaries", "english", "sowpods.txt") }
// DAWGCache locates the committed English DAWG, relative to the repository root.
func DAWGCache() string { return filepath.Join(Root(), "dawg", "en_sowpods.dawg") }
// EnglishAvailable reports whether the English dictionary can be loaded (cache or source).
func EnglishAvailable() bool {
return exists(DAWGCache()) || exists(WordlistPath())
}
// EnglishAvailable reports whether the committed English DAWG is present.
func EnglishAvailable() bool { return exists(DAWGCache()) }
// EnglishWords returns the encoded English word list (from the submodule source).
func EnglishWords() ([][]byte, error) {
return wordlist.Read(WordlistPath(), alphabet.Latin(), MinLen, MaxLen)
}
// EnglishDAWG returns the English DAWG, loading the cache if present, otherwise building
// it from the word list and caching it (best effort).
func EnglishDAWG() (dawg.Finder, error) {
if exists(DAWGCache()) {
return dictdawg.Load(DAWGCache())
}
words, err := EnglishWords()
if err != nil {
return nil, err
}
f, err := dictdawg.Build(alphabet.Latin(), words)
if err != nil {
return nil, err
}
_ = dictdawg.Save(f, DAWGCache())
return f, nil
}
// EnglishDAWG loads the committed English DAWG.
func EnglishDAWG() (dawg.Finder, error) { return dictdawg.Load(DAWGCache()) }