alice-words

This fixture builds a HAMT cataloging the locations of all words appearing in the first chapter of Lewis Carroll's Alice in Wonderland.

• The reference text is available in words.txt.
• Words are matched using a simple /\w+/ regular expression.
• The HAMT uses a bitWidth of 5 (maximum of 32 entries or buckets per node) and a bucketSize of 3.
• Blocks are encoded using DAG-CBOR and CIDs use a SHA2-256 multihash.
• Each entry is keyed by the word (as Bytes as per the HAMT spec) and the values are stored inline.
• Values are an array of the line and column location of each occurrence of the word, in order of appearance.
• Values take the following form:

type Value [Datum]
type Datum struct {
  line Int
  column Int
}

Once compiled:

• The root CID of such a HAMT should be bafyreic672jz6huur4c2yekd3uycswe2xfqhjlmtmm5dorb6yoytgflova. This root is listed as the root of the CAR containing the blocks.
• There will be 36 blocks: 1 root and 25 non-roots, matching the schema in the HAMT spec.

Files:

• hamt.car - CAR containing 36 blocks
• hamt.json - A sorted JSON representation of the data contained in the HAMT, with a mapping of word to locations in the text.
• words.txt - The source text used to generate the data.