From 0c7fb2ee56d7fb9ef4aea3839c6d60307be84987 Mon Sep 17 00:00:00 2001
From: Dmitriy Pleshevskiy <dmitriy@ideascup.me>
Date: Mon, 13 Jun 2022 17:55:47 +0300
Subject: [PATCH] par(md): add simple list
---
par/md.test.ts | 67 ++++++++++++++++++++++++++++++++
par/md.ts | 102 +++++++++++++++++++++++++++++++++++++++++++++----
2 files changed, 161 insertions(+), 8 deletions(-)
diff --git a/par/md.test.ts b/par/md.test.ts
index 08e18e8..46dbf84 100644
--- a/par/md.test.ts
+++ b/par/md.test.ts
@@ -191,3 +191,70 @@ Deno.test({
);
},
});
+
+// List
+
+Deno.test({
+ name: "should parse list with empty items",
+ fn: () => {
+ const par = new MarkdownParser();
+
+ assertEquals(
+ ren.render(par.parse("-")),
+ "<ul><li></li></ul>",
+ );
+ assertEquals(
+ ren.render(par.parse("- ")),
+ "<ul><li></li></ul>",
+ );
+ },
+});
+
+Deno.test({
+ name: "should parse list if line contains additional spaces",
+ fn: () => {
+ const expected = "<ul><li>hello</li></ul>";
+ const par = new MarkdownParser();
+ assertEquals(ren.render(par.parse(" - hello")), expected);
+ assertEquals(ren.render(par.parse(" - hello")), expected);
+ assertEquals(ren.render(par.parse(" - hello")), expected);
+ },
+});
+
+Deno.test({
+ name: "should not display a single paragraph in the list",
+ fn: () => {
+ const par = new MarkdownParser();
+
+ assertEquals(
+ ren.render(par.parse("- hello")),
+ "<ul><li>hello</li></ul>",
+ );
+ assertEquals(
+ ren.render(par.parse(`\
+- hello
+world`)),
+ "<ul><li>hello world</li></ul>",
+ );
+ assertEquals(
+ ren.render(par.parse(`\
+- hello
+ world`)),
+ "<ul><li>hello world</li></ul>",
+ );
+ },
+});
+
+Deno.test({
+ name: "should parse many items in the list",
+ fn: () => {
+ const par = new MarkdownParser();
+
+ assertEquals(
+ ren.render(par.parse(`\
+- hello
+- world`)),
+ "<ul><li>hello</li><li>world</li></ul>",
+ );
+ },
+});
diff --git a/par/md.ts b/par/md.ts
index 4de23f4..78987fc 100644
--- a/par/md.ts
+++ b/par/md.ts
@@ -7,6 +7,8 @@ const RE_EMPTY_LINE = /^\s*$/;
const RE_OPEN_ATX_HEADING = /^\s{0,3}(#{1,6})(\s|$)/;
const RE_CLOSE_ATX_HEADING = /(^|\s+)#*\s*$/;
+const RE_LIST_ITEM = /^\s{0,3}([-+*])(\s|$)/;
+
// TODO: make better regex for destination
const RE_LINK = /\[([\s\S]*?)]\((?:([^\s]*)|<(.+?)>)(?: ('|")(.+?)\4)?\)/;
@@ -18,6 +20,7 @@ export class MarkdownParser implements Parser {
while (readStr.length) {
const newReadStr = skipEmptyLine(readStr) ??
parseAtxHeading(astDoc, readStr) ??
+ parseList(astDoc, readStr) ??
parseParagraph(astDoc, readStr);
if (isNil(newReadStr)) break;
readStr = newReadStr;
@@ -27,12 +30,33 @@ export class MarkdownParser implements Parser {
}
}
-function DocChild(content: AstDocumentChild): Elem {
- switch (content.kind) {
+function List(ast: AstList): Elem {
+ // switch (ast.kind)
+ return BulletList(ast);
+}
+
+function BulletList(ast: AstBulletList): Elem {
+ return new Elem("ul", {}, ast.content.map(ListItem));
+}
+
+function ListItem(ast: AstListItem): Elem {
+ return new Elem(
+ "li",
+ {},
+ ast.content.length === 1 && ast.content[0].kind === AstKind.Paragraph
+ ? ast.content[0].content.map(InlineContent)
+ : ast.content.map(DocChild),
+ );
+}
+
+function DocChild(ast: AstDocumentChild): Elem {
+ switch (ast.kind) {
case AstKind.AtxHeading:
- return Heading(content);
+ return Heading(ast);
case AstKind.Paragraph:
- return Paragraph(content);
+ return Paragraph(ast);
+ case AstKind.List:
+ return List(ast);
}
}
@@ -97,7 +121,45 @@ function parseAtxHeading(ast: AstDocument, readStr: string): string | null {
);
}
-function parseParagraph(ast: AstDocument, readStr: string): string | null {
+function parseList(ast: AstDocument, readStr: string): string | null {
+ if (!readStr.length) return null;
+
+ let listMatch = RE_LIST_ITEM.exec(readStr);
+ if (isNil(listMatch)) return null;
+
+ const astList: AstBulletList = {
+ kind: AstKind.List,
+ type: AstListType.Bullet,
+ bulletChar: listMatch[1] as ListBulletChar,
+ content: [],
+ };
+ ast.content.push(astList);
+
+ do {
+ const astListItem: AstListItem = {
+ kind: AstKind.ListItem,
+ content: [],
+ };
+ astList.content.push(astListItem);
+
+ readStr = readStr.slice(listMatch[0].length);
+
+ const newReadStr = // parseAtxHeading(astList, readStr) ??
+ // parseList(astList, readStr) ??
+ parseParagraph(astListItem, readStr);
+ if (isNil(newReadStr)) break;
+ readStr = newReadStr;
+
+ listMatch = RE_LIST_ITEM.exec(readStr);
+ } while (!isNil(listMatch));
+
+ return readStr;
+}
+
+function parseParagraph(
+ ast: AstDocument | AstListItem,
+ readStr: string,
+): string | null {
if (!readStr.length) return null;
const paragraph: AstParagraph = {
@@ -108,7 +170,8 @@ function parseParagraph(ast: AstDocument, readStr: string): string | null {
let paragraphInlineContent = "";
while (!RE_EMPTY_LINE.test(readStr)) {
- console.log({ readStr });
+ const listMatch = RE_LIST_ITEM.exec(readStr);
+ if (!isNil(listMatch)) break;
paragraphInlineContent += readStr.includes("\n")
? readStr.slice(0, readStr.indexOf("\n") + 1)
: readStr;
@@ -161,7 +224,10 @@ function parseText(
if (!readStr.length) return null;
const parts = readStr.split("\n").filter(Boolean).map(
- (textPart): AstText => ({ kind: AstKind.Text, content: textPart }),
+ (textPart): AstText => ({
+ kind: AstKind.Text,
+ content: textPart.trimStart(),
+ }),
);
ast.content.push(...parts);
@@ -172,7 +238,20 @@ function parseText(
// AST
type AstDocument = BaseAstItem<AstKind.Document, AstDocumentChild[]>;
-type AstDocumentChild = AstAtxHeading | AstParagraph;
+type AstDocumentChild = AstAtxHeading | AstBulletList | AstParagraph | AstList;
+
+type AstList = AstBulletList; // | AstOrderedList
+
+enum AstListType {
+ Bullet,
+ // Ordered,
+}
+
+type ListBulletChar = "-" | "+" | "*";
+
+type AstListItem = BaseAstItem<AstKind.ListItem, AstListItemChild[]>;
+
+type AstListItemChild = AstDocumentChild;
interface AstAtxHeading
extends BaseAstItem<AstKind.AtxHeading, AstInlineContent[]> {
@@ -181,6 +260,11 @@ interface AstAtxHeading
type HeadingLevel = 1 | 2 | 3 | 4 | 5 | 6;
+interface AstBulletList extends BaseAstItem<AstKind.List, AstListItem[]> {
+ type: AstListType.Bullet;
+ bulletChar: ListBulletChar;
+}
+
type AstParagraph = BaseAstItem<AstKind.Paragraph, AstInlineContent[]>;
type AstInlineContent = AstText | AstLink;
@@ -201,6 +285,8 @@ enum AstKind {
Document,
AtxHeading,
Paragraph,
+ List,
+ ListItem,
Link,
Text,
}