Consolidate processing.

CHANGELOG.md

@@ -3,20 +3,22 @@
 ## 8.0.0 - 2024-mm-dd
 
 ### Added
-- Add TypedLiteral, UntypedLiteral encoders and decoders to more generally
-  handle type-specific behavior such as cryptosuite string encoding as well
-  as pass through uncompressible untyped values.
-
-### CHANGED
-- Restructure term registry system to be more general with four tables which
-  can all be passed by the user during encoding and decoding:
-    - keywordsTable: for JSON-LD keywords (e.g. '@context', '@type')
-    - stringTable: for JSON-LD string values. This subsumes 'appContextMap',
-      as context strings can be included here, but includes arbitrary string
-      values as well.
-    - urlSchemeTable: for URL schemes (i.e. 'http://')
-    - typedLiteralTable: for values associated with JSON-LD types. This
-      subsumes type-specific codecs like the cryptosuite codec.
+- Added support for passing through (without semantic compression) terms not
+  defined in contexts and other plain values.
+
+### Changed
+- Restructure term registry system to be more general with a type table. The
+  type table expresses type-namespaced tables of values that can be used when encoding and decoding. This type table (of tables) can be passed by the user
+  when encoding and decoding and will be given preference over any processing-
+  mode-specific codecs (such as multibase codecs).
+  The supported types include:
+    - context: for JSON-LD context URLs, this subsumes the old 'appContextMap'
+      and any contexts expressed in the old arbitrary string table as well.
+    - url: for any JSON-LD URL value, such as values associated with `@id`
+      or `@type` (or their aliases) JSON keys.
+    - none: for any untyped or plain literal values.
+    - <any JSON-LD type expressed as URL>: for any values associated with
+      any custom JSON-LD types.
 - Restructure CBOR-LD tag system to use a range of tags where
   the tag value informs what values for the tables above should be used
   via a registry. Legacy tags (0x0501) are still supported, and new tags are

lib/Compressor.js

@@ -1,32 +1,16 @@
 /*!
- * Copyright (c) 2021-2023 Digital Bazaar, Inc. All rights reserved.
+ * Copyright (c) 2021-2024 Digital Bazaar, Inc. All rights reserved.
  */
 import * as cborg from 'cborg';
 import {
   FIRST_CUSTOM_TERM_ID,
-  KEYWORDS_TABLE,
-  URL_SCHEME_TABLE
+  KEYWORDS_TABLE
 } from './tables.js';
 import {CborldEncoder} from './codecs/CborldEncoder.js';
 import {ContextEncoder} from './codecs/ContextEncoder.js';
 import {inspect} from './util.js';
-
-import {MultibaseEncoder} from './codecs/MultibaseEncoder.js';
 import {Transformer} from './Transformer.js';
-import {TypedLiteralEncoder} from './codecs/TypedLiteralEncoder.js';
-import {UntypedLiteralEncoder} from './codecs/UntypedLiteralEncoder.js';
-import {UriEncoder} from './codecs/UriEncoder.js';
-import {VocabTermEncoder} from './codecs/VocabTermEncoder.js';
-import {XsdDateEncoder} from './codecs/XsdDateEncoder.js';
-import {XsdDateTimeEncoder} from './codecs/XsdDateTimeEncoder.js';
-
-export const TYPE_ENCODERS = new Map([
-  ['@id', UriEncoder],
-  ['@vocab', VocabTermEncoder],
-  ['https://w3id.org/security#multibase', MultibaseEncoder],
-  ['http://www.w3.org/2001/XMLSchema#date', XsdDateEncoder],
-  ['http://www.w3.org/2001/XMLSchema#dateTime', XsdDateTimeEncoder]
-]);
+import {ValueEncoder} from './codecs/ValueEncoder.js';
 
 const CONTEXT_TERM_ID = KEYWORDS_TABLE.get('@context');
 const CONTEXT_TERM_ID_PLURAL = CONTEXT_TERM_ID + 1;
@@ -49,24 +33,13 @@ export class Compressor extends Transformer {
    * @param {object} options - The options to use.
    * @param {documentLoaderFunction} options.documentLoader -The document
    *   loader to use when resolving JSON-LD Context URLs.
-   * @param {Map} options.stringTable - A map of string values and
-   *  their associated CBOR-LD integer values.
-   * @param {Map} options.typedLiteralTable - A map of JSON-LD types
-   *  to maps that map values of that type to their associated CBOR-LD
-   *  integer values.
+   * @param {Map} options.typeTable - A map of possible value types, including
+   *  `context`, `url`, `none`, and any JSON-LD type, each of which maps to
+   *   another map of values of that type to their associated CBOR-LD integer
+   *   values.
    */
-  constructor({
-    documentLoader,
-    stringTable,
-    typedLiteralTable
-  } = {}) {
-    super({
-      documentLoader,
-      stringTable,
-      typedLiteralTable
-    });
-    this.keywordsTable = new Map(KEYWORDS_TABLE);
-    this.urlSchemeTable = new Map(URL_SCHEME_TABLE);
+  constructor({documentLoader, typeTable} = {}) {
+    super({documentLoader, typeTable});
   }
 
   /**
@@ -82,6 +55,8 @@ export class Compressor extends Transformer {
    * @returns {Promise<Uint8Array>} - The compressed CBOR-LD bytes.
    */
   async compress({jsonldDocument, diagnose} = {}) {
+    this._reset();
+
     const transformMaps = await this._createTransformMaps({jsonldDocument});
     if(diagnose) {
       diagnose('Diagnostic CBOR-LD compression transform map(s):');
@@ -91,9 +66,6 @@ export class Compressor extends Transformer {
   }
 
   async _createTransformMaps({jsonldDocument}) {
-    // initialize state
-    this._reset();
-
     // handle single or multiple JSON-LD docs
     const transformMaps = [];
     const isArray = Array.isArray(jsonldDocument);
@@ -118,8 +90,7 @@ export class Compressor extends Transformer {
     const isArray = Array.isArray(context);
     const contexts = isArray ? context : [context];
     for(const value of contexts) {
-      const encoder = ContextEncoder.createEncoder(
-        {value, transformer: this});
+      const encoder = ContextEncoder.createEncoder({value, transformer: this});
       entries.push(encoder || value);
     }
     const id = isArray ? CONTEXT_TERM_ID_PLURAL : CONTEXT_TERM_ID;
@@ -158,51 +129,53 @@ export class Compressor extends Transformer {
     return entries;
   }
 
+  _getObjectTypes({obj, typeTerms}) {
+    const objectTypes = new Set();
+    for(const term of typeTerms) {
+      const types = obj[term];
+      if(types !== undefined) {
+        if(Array.isArray(types)) {
+          types.forEach(objectTypes.add, objectTypes);
+        } else {
+          objectTypes.add(types);
+        }
+      }
+    }
+    return objectTypes;
+  }
+
   _reset() {
     this.contextMap = new Map();
     this.nextTermId = FIRST_CUSTOM_TERM_ID;
-    this.termToId = this.keywordsTable;
-  }
-
-  _transformObjectId({transformMap, termInfo, value}) {
-    const {termId} = termInfo;
-    const encoder = UriEncoder.createEncoder(
-      {value, transformer: this, termInfo});
-    transformMap.set(termId, encoder || value);
-  }
-
-  _transformObjectType({transformMap, termInfo, value}) {
-    const {termId, plural} = termInfo;
-    const values = plural ? value : [value];
-    const entries = [];
-    for(const value of values) {
-      const encoder = VocabTermEncoder.createEncoder(
-        {value, transformer: this, termInfo});
-      entries.push(encoder || value);
-    }
-    transformMap.set(termId, plural ? entries : entries[0]);
+    this.termToId = new Map(KEYWORDS_TABLE);
   }
 
-  _transformTypedValue({entries, termType, value, termInfo}) {
-    const encoder = TypedLiteralEncoder.createEncoder({
-      value,
-      transformer: this,
-      termType
-    }) || TYPE_ENCODERS.get(termType)?.createEncoder({
-      value,
-      transformer: this,
-      termInfo
-    });
-    if(encoder) {
-      entries.push(encoder);
-      return true;
+  _transformValue({entries, termType, value, termInfo}) {
+    if(typeof value === 'object') {
+      return;
     }
+    const encoder = ValueEncoder.createEncoder(
+      {value, transformer: this, termInfo, termType});
+    entries.push(encoder ?? value);
+    return true;
   }
 
   async _transformArray({entries, contextStack, value}) {
     // recurse into array
     const children = [];
     for(const obj of value) {
+      if(Array.isArray(obj)) {
+        // recurse into array of arrays
+        await this._transformArray(
+          {entries: children, contextStack, value: obj});
+        continue;
+      }
+      // handle non-object
+      if(!(obj && typeof obj === 'object')) {
+        children.push(obj);
+        continue;
+      }
+      // handle object
       const childMap = new Map();
       children.push(childMap);
       await this._transform({obj, transformMap: childMap, contextStack});
@@ -217,14 +190,6 @@ export class Compressor extends Transformer {
     await this._transform({obj: value, transformMap, contextStack});
   }
 
-  async _transformUntypedValue({entries, value}) {
-    const encoder = UntypedLiteralEncoder.createEncoder({
-      value,
-      transformer: this
-    });
-    entries.push(encoder);
-  }
-
   _assignEntries({entries, transformMap, termInfo}) {
     const {termId, plural} = termInfo;
     transformMap.set(termId, plural ? entries : entries[0]);