Compare commits

...

2 Commits

Author SHA1 Message Date
Dmitriy Pleshevskiy 53c4f4039c
par(md): add softbreak 2022-06-12 01:28:56 +03:00
Dmitriy Pleshevskiy 9551f3ae1e
par(md): skip empty line...
... instead of clearing new line character
2022-06-12 00:58:24 +03:00
2 changed files with 56 additions and 30 deletions

View File

@ -7,12 +7,13 @@ const ren = new HtmlStrRenderer();
// Misc
Deno.test({
name: "should skip new line character",
name: "should skip empty line",
fn: () => {
const par = new MarkdownParser();
assertEquals(ren.render(par.parse("\n")), "");
assertEquals(ren.render(par.parse("\r\n")), "");
assertEquals(ren.render(par.parse("\n\r\n")), "");
assertEquals(ren.render(par.parse("\n \n")), "");
},
});
@ -56,8 +57,6 @@ Deno.test({
assertEquals(ren.render(par.parse(" # hello")), "<h1>hello</h1>");
assertEquals(ren.render(par.parse(" # hello")), "<h1>hello</h1>");
assertEquals(ren.render(par.parse(" # hello")), "<h1>hello</h1>");
assertEquals(ren.render(par.parse("\n # hello")), "<h1>hello</h1>");
assertEquals(ren.render(par.parse("\r\n # hello")), "<h1>hello</h1>");
},
});
@ -101,3 +100,16 @@ Deno.test({
assertEquals(ren.render(par.parse("hello")), "<p>hello</p>");
},
});
Deno.test({
name: "should parse paragraph with softbreak",
fn: () => {
const par = new MarkdownParser();
const input = `\
hello
world`;
assertEquals(ren.render(par.parse(input)), "<p>hello world</p>");
},
});

View File

@ -2,7 +2,7 @@ import { AnyNode, Elem, Fragment, TextNode } from "../core/node.ts";
import { isNil } from "../core/utils.ts";
import { Parser } from "./types.ts";
const RE_NEW_LINE = /^\r?\n/;
const RE_EMPTY_LINE = /^\s*$/;
const RE_OPEN_ATX_HEADING = /^\s{0,3}(#{1,6})(\s|$)/;
const RE_CLOSE_ATX_HEADING = /(^|\s+)#*\s*$/;
@ -12,17 +12,9 @@ export class MarkdownParser implements Parser {
const astDoc: AstDocument = { kind: AstKind.Document, content: [] };
let readStr = input;
while (readStr.length) {
// 1. clear new line character
const match = RE_NEW_LINE.exec(readStr);
if (!isNil(match)) {
readStr = readStr.slice(match[0].length);
continue;
}
// 2. try to find atx heading sequence
const newReadStr = parseAtxHeading(astDoc, readStr) ??
const newReadStr = skipEmptyLine(readStr) ??
parseAtxHeading(astDoc, readStr) ??
parseParagraph(astDoc, readStr);
if (isNil(newReadStr)) break;
readStr = newReadStr;
@ -42,11 +34,15 @@ function DocChild(content: AstDocumentChild): Elem {
}
function Heading(ast: AstAtxHeading): Elem {
return new Elem(`h${ast.level}`, {}, ast.content.map(Text));
return new Elem(`h${ast.level}`, {}, ast.content.map(InlineContent));
}
function Paragraph(ast: AstParagraph): Elem {
return new Elem("p", {}, ast.content.map(Text));
return new Elem("p", {}, ast.content.map(InlineContent));
}
function InlineContent(ast: AstInlineContent): TextNode {
return Text(ast);
}
function Text(ast: AstText): TextNode {
@ -55,6 +51,12 @@ function Text(ast: AstText): TextNode {
// parse utils
function skipEmptyLine(readStr: string): string | null {
const match = RE_EMPTY_LINE.exec(readStr);
if (isNil(match)) return null;
return readStr.slice(match[0].length);
}
function parseAtxHeading(ast: AstDocument, readStr: string): string | null {
const match = RE_OPEN_ATX_HEADING.exec(readStr);
if (isNil(match)) return null;
@ -94,13 +96,22 @@ function parseParagraph(ast: AstDocument, readStr: string): string | null {
};
ast.content.push(paragraph);
const paragraphInlineContent = readStr.includes("\n")
? readStr.slice(0, readStr.indexOf("\n") + 1)
: readStr;
let paragraphInlineContent = "";
while (!RE_EMPTY_LINE.test(readStr)) {
console.log({ readStr });
paragraphInlineContent += readStr.includes("\n")
? readStr.slice(0, readStr.indexOf("\n") + 1)
: readStr;
readStr = readStr.slice(paragraphInlineContent.length);
}
parseInlineContent(paragraph, paragraphInlineContent);
console.log({ paragraphInlineContent, readStr });
return readStr.slice(paragraphInlineContent.length);
if (paragraphInlineContent.length) {
parseInlineContent(paragraph, paragraphInlineContent);
}
return readStr;
}
function parseInlineContent(
@ -109,11 +120,11 @@ function parseInlineContent(
): string | null {
if (!readStr.length) return null;
const text: AstText = {
kind: AstKind.Text,
content: readStr.trim(),
};
ast.content.push(text);
const parts = readStr.split("\n").filter(Boolean).map(
(textPart): AstText => ({ kind: AstKind.Text, content: textPart }),
);
ast.content = parts;
return readStr;
}
@ -123,16 +134,19 @@ function parseInlineContent(
type AstDocument = BaseAstItem<AstKind.Document, AstDocumentChild[]>;
type AstDocumentChild = AstAtxHeading | AstParagraph;
interface AstAtxHeading extends BaseAstItem<AstKind.AtxHeading, AstText[]> {
interface AstAtxHeading
extends BaseAstItem<AstKind.AtxHeading, AstInlineContent[]> {
level: HeadingLevel;
}
type AstParagraph = BaseAstItem<AstKind.Paragraph, AstText[]>;
type HeadingLevel = 1 | 2 | 3 | 4 | 5 | 6;
type AstParagraph = BaseAstItem<AstKind.Paragraph, AstInlineContent[]>;
type AstInlineContent = AstText;
type AstText = BaseAstItem<AstKind.Text, string>;
type HeadingLevel = 1 | 2 | 3 | 4 | 5 | 6;
interface BaseAstItem<K extends AstKind, Cont> {
kind: K;
content: Cont;