ceph/src/arrow/js/src/builder/dictionary.ts

   1 // Licensed to the Apache Software Foundation (ASF) under one
   2 // or more contributor license agreements.  See the NOTICE file
   3 // distributed with this work for additional information
   4 // regarding copyright ownership.  The ASF licenses this file
   5 // to you under the Apache License, Version 2.0 (the
   6 // "License"); you may not use this file except in compliance
   7 // with the License.  You may obtain a copy of the License at
   8 //
   9 //   http://www.apache.org/licenses/LICENSE-2.0
  10 //
  11 // Unless required by applicable law or agreed to in writing,
  12 // software distributed under the License is distributed on an
  13 // "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
  14 // KIND, either express or implied.  See the License for the
  15 // specific language governing permissions and limitations
  16 // under the License.
  17
  18 import { Vector } from '../vector';
  19 import { IntBuilder } from './int';
  20 import { Dictionary, DataType } from '../type';
  21 import { Builder, BuilderOptions } from '../builder';
  22
  23 type DictionaryHashFunction = (x: any) => string | number;
  24
  25 export interface DictionaryBuilderOptions<T extends DataType = any, TNull = any> extends BuilderOptions<T, TNull> {
  26     dictionaryHashFunction?: DictionaryHashFunction;
  27 }
  28
  29 /** @ignore */
  30 export class DictionaryBuilder<T extends Dictionary, TNull = any> extends Builder<T, TNull> {
  31
  32     protected _dictionaryOffset: number;
  33     protected _dictionary?: Vector<T['dictionary']>;
  34     protected _keysToIndices: { [key: string]: number };
  35     public readonly indices: IntBuilder<T['indices']>;
  36     public readonly dictionary: Builder<T['dictionary']>;
  37
  38     constructor({ 'type': type, 'nullValues': nulls, 'dictionaryHashFunction': hashFn }: DictionaryBuilderOptions<T, TNull>) {
  39         super({ type: new Dictionary(type.dictionary, type.indices, type.id, type.isOrdered) as T });
  40         this._nulls = <any> null;
  41         this._dictionaryOffset = 0;
  42         this._keysToIndices = Object.create(null);
  43         this.indices = Builder.new({ 'type': this.type.indices, 'nullValues': nulls }) as IntBuilder<T['indices']>;
  44         this.dictionary = Builder.new({ 'type': this.type.dictionary, 'nullValues': null }) as Builder<T['dictionary']>;
  45         if (typeof hashFn === 'function') {
  46             this.valueToKey = hashFn;
  47         }
  48     }
  49
  50     public get values() { return this.indices.values; }
  51     public get nullCount() { return this.indices.nullCount; }
  52     public get nullBitmap() { return this.indices.nullBitmap; }
  53     public get byteLength() { return this.indices.byteLength + this.dictionary.byteLength; }
  54     public get reservedLength() { return this.indices.reservedLength + this.dictionary.reservedLength; }
  55     public get reservedByteLength() { return this.indices.reservedByteLength + this.dictionary.reservedByteLength; }
  56     public isValid(value: T['TValue'] | TNull) { return this.indices.isValid(value); }
  57     public setValid(index: number, valid: boolean) {
  58         const indices = this.indices;
  59         valid = indices.setValid(index, valid);
  60         this.length = indices.length;
  61         return valid;
  62     }
  63     public setValue(index: number, value: T['TValue']) {
  64         const keysToIndices = this._keysToIndices;
  65         const key = this.valueToKey(value);
  66         let idx = keysToIndices[key];
  67         if (idx === undefined) {
  68             keysToIndices[key] = idx = this._dictionaryOffset + this.dictionary.append(value).length - 1;
  69         }
  70         return this.indices.setValue(index, idx);
  71     }
  72     public flush() {
  73         const type = this.type;
  74         const prev = this._dictionary;
  75         const curr = this.dictionary.toVector();
  76         const data = this.indices.flush().clone(type);
  77         data.dictionary = prev ? prev.concat(curr) : curr;
  78         this.finished || (this._dictionaryOffset += curr.length);
  79         this._dictionary = data.dictionary as Vector<T['dictionary']>;
  80         this.clear();
  81         return data;
  82     }
  83     public finish() {
  84         this.indices.finish();
  85         this.dictionary.finish();
  86         this._dictionaryOffset = 0;
  87         this._keysToIndices = Object.create(null);
  88         return super.finish();
  89     }
  90     public clear() {
  91         this.indices.clear();
  92         this.dictionary.clear();
  93         return super.clear();
  94     }
  95     public valueToKey(val: any): string | number {
  96         return typeof val === 'string' ? val : `${val}`;
  97     }
  98 }