From 0d67fcada30b7d52014eb8dab3d5e3c320ca4b23 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?wifi=E6=AD=AAf?= <1402772884@qq.com> Date: Tue, 4 Feb 2025 14:09:29 +0800 Subject: [PATCH] =?UTF-8?q?feat(mini-markdown-ast-parser):=20ast=E5=A2=9E?= =?UTF-8?q?=E9=87=8F=E6=9E=84=E5=BB=BA?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- .../src/core/parse/index.ts | 35 ++++-- .../src/core/parse/parseIncrementally.ts | 118 ++++++++++++++++++ 2 files changed, 141 insertions(+), 12 deletions(-) create mode 100644 packages/mini-markdown-ast-parser/src/core/parse/parseIncrementally.ts diff --git a/packages/mini-markdown-ast-parser/src/core/parse/index.ts b/packages/mini-markdown-ast-parser/src/core/parse/index.ts index 71141f5..a092a5f 100644 --- a/packages/mini-markdown-ast-parser/src/core/parse/index.ts +++ b/packages/mini-markdown-ast-parser/src/core/parse/index.ts @@ -1,19 +1,30 @@ -import { tokenizer } from './tokenizer' -import type { RootTokens } from '../../types/tokens' +import { tokenizer } from "./tokenizer"; +import type { RootTokens } from "../../types/tokens"; +import { parseIncrementally } from "./parseIncrementally"; + +export interface IncrementalParseOptions { + prevMarkdown?: string; + prevRoot?: RootTokens; +} // 语法分析 -export const parseMarkdown = (markdown: string): RootTokens => { - const lines = markdown.split('\n') - // 根节点 +export const parseMarkdown = (markdown: string, options?: IncrementalParseOptions): RootTokens => { + const lines = markdown.split("\n"); + // 如果有增量解析选项且存在前一个解析结果 + if (options?.prevMarkdown && options?.prevRoot) { + return parseIncrementally(markdown, lines, options); + } + + // 如果没有增量解析选项，执行完整解析 const root: RootTokens = { - type: 'root', + type: "root", children: [], position: { start: { line: 1, column: 1, offset: 0 }, - end: { line: lines.length, column: 1, offset: markdown.length } - } - } + end: { line: lines.length, column: 1, offset: markdown.length }, + }, + }; // 语法解析（词法分析器） - tokenizer(lines, root) - return root -} + tokenizer(lines, root); + return root; +}; diff --git a/packages/mini-markdown-ast-parser/src/core/parse/parseIncrementally.ts b/packages/mini-markdown-ast-parser/src/core/parse/parseIncrementally.ts new file mode 100644 index 0000000..8789422 --- /dev/null +++ b/packages/mini-markdown-ast-parser/src/core/parse/parseIncrementally.ts @@ -0,0 +1,118 @@ +import { RootTokens } from "@/types/tokens"; +import { IncrementalParseOptions } from "."; +import { tokenizer } from "./tokenizer"; + +// 增量解析方法 +export const parseIncrementally = ( + markdown: string, + lines: string[], + options: IncrementalParseOptions, +): RootTokens => { + const prevLines = options.prevMarkdown?.split("\n") || []; + // 找到所有变化的区间 + const changeRanges: { start: number; end: number }[] = []; + let currentRange: { start: number; end: number } | null = null; + + // 遍历所有行，找出所有变化的区间 + for (let i = 0; i < Math.max(lines.length, prevLines.length); i++) { + const isLineDifferent = lines[i] !== prevLines[i]; + + if (isLineDifferent && !currentRange) { + currentRange = { start: i, end: i + 1 }; + } else if (isLineDifferent && currentRange) { + currentRange.end = i + 1; + } else if (!isLineDifferent && currentRange) { + changeRanges.push(currentRange); + currentRange = null; + } + } + + // 处理最后一个区间 + if (currentRange) { + changeRanges.push(currentRange); + } + + // 如果没有变化，直接返回之前的解析结果 + if (changeRanges.length === 0 && options.prevRoot) { + return options.prevRoot; + } + + // 如果没有变化但也没有之前的解析结果，创建新的根节点 + if (changeRanges.length === 0) { + return { + type: "root", + children: [], + position: { + start: { line: 1, column: 1, offset: 0 }, + end: { line: lines.length, column: 1, offset: markdown.length }, + }, + }; + } + + // 创建新的根节点 + const root: RootTokens = { + type: "root", + children: [], + position: { + start: { line: 1, column: 1, offset: 0 }, + end: { line: lines.length, column: 1, offset: markdown.length }, + }, + }; + + // 复制第一个变化区间之前的节点 + root.children = (options.prevRoot?.children || []).filter( + (node) => node.position.end.line <= changeRanges[0].start, + ); + + // 处理每个变化区间 + let lastEndLine = changeRanges[0].start; + for (const range of changeRanges) { + // 复制区间之间的未变化节点 + if (range.start > lastEndLine && options.prevRoot?.children) { + const unchangedNodes = options.prevRoot.children.filter( + (node) => node.position.start.line > lastEndLine && node.position.end.line <= range.start, + ); + root.children.push(...unchangedNodes); + } + + // 解析变化的部分 + const changedText = lines.slice(range.start, range.end).join("\n"); + const changedOffset = lines.slice(0, range.start).join("\n").length + (range.start > 0 ? 1 : 0); + const tempRoot: RootTokens = { + type: "root", + children: [], + position: { + start: { line: range.start + 1, column: 1, offset: changedOffset }, + end: { line: range.end, column: 1, offset: changedOffset + changedText.length }, + }, + }; + + tokenizer(lines.slice(range.start, range.end), tempRoot); + root.children.push(...tempRoot.children); + + lastEndLine = range.end; + } + + // 复制最后一个变化区间之后的节点，并调整位置信息 + const lineDiff = lines.length - prevLines.length; + const remainingNodes = options.prevRoot?.children + .filter((node) => node.position.start.line > changeRanges[changeRanges.length - 1].end) + .map((node) => ({ + ...node, + position: { + start: { + ...node.position.start, + line: node.position.start.line + lineDiff, + }, + end: { + ...node.position.end, + line: node.position.end.line + lineDiff, + }, + }, + })); + + if (remainingNodes) { + root.children.push(...remainingNodes); + } + return root; +}; -- Gitee