@samva/markup 0.8.0 → 0.10.0

package/dist/lexer.js

@@ -1,281 +0,0 @@
-//#region src/lexer.ts
-const CHAR_TAB = 9;
-const CHAR_LINE_FEED = 10;
-const CHAR_CARRIAGE_RETURN = 13;
-const CHAR_SPACE = 32;
-const CHAR_BANG = 33;
-const CHAR_DOUBLE_QUOTE = 34;
-const CHAR_HASH = 35;
-const CHAR_DOLLAR = 36;
-const CHAR_AMPERSAND = 38;
-const CHAR_SINGLE_QUOTE = 39;
-const CHAR_SLASH = 47;
-const CHAR_SEMICOLON = 59;
-const CHAR_LESS_THAN = 60;
-const CHAR_EQUALS = 61;
-const CHAR_GREATER_THAN = 62;
-const CHAR_OPEN_BRACE = 123;
-const CHAR_CLOSE_BRACE = 125;
-const CHAR_DOT = 46;
-const CHAR_UNDERSCORE = 95;
-const CHAR_0 = 48;
-const CHAR_9 = 57;
-const CHAR_UPPER_A = 65;
-const CHAR_UPPER_X = 88;
-const CHAR_UPPER_Z = 90;
-const CHAR_LOWER_A = 97;
-const CHAR_LOWER_X = 120;
-const CHAR_LOWER_Z = 122;
-const TokenKind = {
-	EndOfFile: 0,
-	/** `<` opening a start tag (text mode). */
-	LessThan: 1,
-	/** `</` opening a close tag (text mode). */
-	LessThanSlash: 2,
-	GreaterThan: 3,
-	SlashGreaterThan: 4,
-	Identifier: 5,
-	Equals: 6,
-	/** Quoted attribute value; getStringValue decodes entities. */
-	String: 7,
-	OpenBrace: 8,
-	CloseBrace: 9,
-	/** Unsigned integer inside a brace attribute value. */
-	Number: 10,
-	/** Raw text run; getTextValue decodes entities. */
-	Text: 11,
-	/** `{{name}}` or `${name}`; name/syntax carried on the token. */
-	Interpolation: 12,
-	/** Lexically invalid input; error carried on the token. */
-	Unknown: 13
-};
-const isWhitespace = (code) => code === CHAR_SPACE || code === CHAR_TAB || code === CHAR_LINE_FEED || code === CHAR_CARRIAGE_RETURN;
-const isDigit = (code) => code >= CHAR_0 && code <= CHAR_9;
-const isLetter = (code) => code >= CHAR_UPPER_A && code <= CHAR_UPPER_Z || code >= CHAR_LOWER_A && code <= CHAR_LOWER_Z;
-const isNameChar = (code) => isLetter(code) || isDigit(code) || code === CHAR_UNDERSCORE || code === CHAR_DOT;
-const NAMED_ENTITIES = {
-	amp: "&",
-	lt: "<",
-	gt: ">",
-	quot: "\"",
-	apos: "'",
-	nbsp: "\xA0"
-};
-const numericEntityCodePoint = (digits, radix) => {
-	const codePoint = Number.parseInt(digits, radix);
-	return Number.isFinite(codePoint) && codePoint <= 1114111 ? codePoint : void 0;
-};
-/**
-* Validates the entity starting at `offset` (which points at `&`).
-* Returns the end offset (past `;`) or -1 when malformed.
-*/
-const scanEntity = (source, offset) => {
-	let index = offset + 1;
-	if (source.charCodeAt(index) === CHAR_HASH) {
-		index++;
-		const marker = source.charCodeAt(index);
-		const hex = marker === CHAR_LOWER_X || marker === CHAR_UPPER_X;
-		if (hex) index++;
-		const digitsStart = index;
-		while (index < source.length) {
-			const code = source.charCodeAt(index);
-			if (!(isDigit(code) || hex && (code >= 97 && code <= 102 || code >= 65 && code <= 70))) break;
-			index++;
-		}
-		if (index === digitsStart) return -1;
-		if (numericEntityCodePoint(source.slice(digitsStart, index), hex ? 16 : 10) === void 0) return -1;
-		return source.charCodeAt(index) === CHAR_SEMICOLON ? index + 1 : -1;
-	}
-	const nameStart = index;
-	while (index < source.length && isLetter(source.charCodeAt(index))) index++;
-	const name = source.slice(nameStart, index);
-	if (!Object.hasOwn(NAMED_ENTITIES, name)) return -1;
-	return source.charCodeAt(index) === CHAR_SEMICOLON ? index + 1 : -1;
-};
-/** Decodes entities in text that already passed `scanEntity` validation. */
-const decodeEntities = (raw) => {
-	if (!raw.includes("&")) return raw;
-	let result = "";
-	let index = 0;
-	while (index < raw.length) {
-		const ampersand = raw.indexOf("&", index);
-		if (ampersand === -1) {
-			result += raw.slice(index);
-			break;
-		}
-		result += raw.slice(index, ampersand);
-		const semicolon = raw.indexOf(";", ampersand);
-		if (semicolon === -1) {
-			result += raw.slice(ampersand);
-			break;
-		}
-		const body = raw.slice(ampersand + 1, semicolon);
-		if (body.startsWith("#x") || body.startsWith("#X")) {
-			const codePoint = numericEntityCodePoint(body.slice(2), 16);
-			result += codePoint === void 0 ? raw.slice(ampersand, semicolon + 1) : String.fromCodePoint(codePoint);
-		} else if (body.startsWith("#")) {
-			const codePoint = numericEntityCodePoint(body.slice(1), 10);
-			result += codePoint === void 0 ? raw.slice(ampersand, semicolon + 1) : String.fromCodePoint(codePoint);
-		} else result += Object.hasOwn(NAMED_ENTITIES, body) ? NAMED_ENTITIES[body] : raw.slice(ampersand, semicolon + 1);
-		index = semicolon + 1;
-	}
-	return result;
-};
-const createLexer = (source) => {
-	let position = 0;
-	const length = source.length;
-	const token = (kind, start, extra) => {
-		return {
-			kind,
-			start,
-			end: position,
-			...extra
-		};
-	};
-	const unknown = (start, code, message) => {
-		if (position === start) position = Math.min(start + 1, length);
-		return token(TokenKind.Unknown, start, { error: {
-			code,
-			message
-		} });
-	};
-	const scanInterpolation = (start, syntax) => {
-		position = start + 2;
-		while (position < length && isWhitespace(source.charCodeAt(position))) position++;
-		if (syntax === "dollar" && source.charCodeAt(position) === CHAR_SLASH) position++;
-		const nameStart = position;
-		while (position < length && isNameChar(source.charCodeAt(position))) position++;
-		const name = source.slice(nameStart, position);
-		while (position < length && isWhitespace(source.charCodeAt(position))) position++;
-		const closed = syntax === "curly" ? source.charCodeAt(position) === CHAR_CLOSE_BRACE && source.charCodeAt(position + 1) === CHAR_CLOSE_BRACE : source.charCodeAt(position) === CHAR_CLOSE_BRACE;
-		if (name.length === 0 || !closed) {
-			position = start;
-			return unknown(start, "invalid-interpolation", `malformed interpolation — write ${syntax === "curly" ? "{{variableName}}" : "${variableName}"} using letters, digits, underscore, or dot`);
-		}
-		position += syntax === "curly" ? 2 : 1;
-		return token(TokenKind.Interpolation, start, {
-			name,
-			syntax
-		});
-	};
-	const scanTextToken = () => {
-		const start = position;
-		if (position >= length) return token(TokenKind.EndOfFile, start);
-		const first = source.charCodeAt(position);
-		if (first === CHAR_LESS_THAN) {
-			const next = source.charCodeAt(position + 1);
-			if (next === CHAR_SLASH) {
-				position += 2;
-				return token(TokenKind.LessThanSlash, start);
-			}
-			if (isLetter(next)) {
-				position += 1;
-				return token(TokenKind.LessThan, start);
-			}
-			if (next === CHAR_BANG) return unknown(start, "comment-not-supported", source.startsWith("<!--", position) ? "comments are not part of SML — remove the <!-- ... --> block" : "unexpected <! — SML has no doctype or declaration syntax; remove it");
-			return unknown(start, "stray-angle-bracket", "stray < in text — escape it as &lt;");
-		}
-		if (first === CHAR_OPEN_BRACE && source.charCodeAt(position + 1) === CHAR_OPEN_BRACE) return scanInterpolation(start, "curly");
-		if (first === CHAR_DOLLAR && source.charCodeAt(position + 1) === CHAR_OPEN_BRACE) return scanInterpolation(start, "dollar");
-		while (position < length) {
-			const code = source.charCodeAt(position);
-			if (code === CHAR_LESS_THAN) break;
-			if (code === CHAR_OPEN_BRACE && source.charCodeAt(position + 1) === CHAR_OPEN_BRACE) break;
-			if (code === CHAR_DOLLAR && source.charCodeAt(position + 1) === CHAR_OPEN_BRACE) break;
-			if (code === CHAR_AMPERSAND) {
-				const entityEnd = scanEntity(source, position);
-				if (entityEnd === -1) {
-					if (position > start) return token(TokenKind.Text, start);
-					return unknown(start, "bare-ampersand", "bare & in text — escape it as &amp; (or use a supported entity like &lt;)");
-				}
-				position = entityEnd;
-				continue;
-			}
-			position++;
-		}
-		return token(TokenKind.Text, start);
-	};
-	const scanTagToken = () => {
-		while (position < length && isWhitespace(source.charCodeAt(position))) position++;
-		const start = position;
-		if (position >= length) return token(TokenKind.EndOfFile, start);
-		const code = source.charCodeAt(position);
-		switch (code) {
-			case CHAR_GREATER_THAN:
-				position++;
-				return token(TokenKind.GreaterThan, start);
-			case CHAR_SLASH:
-				if (source.charCodeAt(position + 1) === CHAR_GREATER_THAN) {
-					position += 2;
-					return token(TokenKind.SlashGreaterThan, start);
-				}
-				return unknown(start, "unexpected-character", "unexpected / inside tag — write /> to self-close");
-			case CHAR_EQUALS:
-				position++;
-				return token(TokenKind.Equals, start);
-			case CHAR_OPEN_BRACE:
-				position++;
-				return token(TokenKind.OpenBrace, start);
-			case CHAR_CLOSE_BRACE:
-				position++;
-				return token(TokenKind.CloseBrace, start);
-			case CHAR_DOUBLE_QUOTE:
-			case CHAR_SINGLE_QUOTE:
-				position++;
-				while (position < length) {
-					const current = source.charCodeAt(position);
-					if (current === code) break;
-					if (current === CHAR_AMPERSAND) {
-						const entityEnd = scanEntity(source, position);
-						if (entityEnd === -1) return unknown(position, "bare-ampersand", "bare & in attribute value — escape it as &amp;");
-						position = entityEnd;
-						continue;
-					}
-					if (current === CHAR_LESS_THAN || current === CHAR_LINE_FEED) break;
-					position++;
-				}
-				if (source.charCodeAt(position) !== code) {
-					const quote = String.fromCharCode(code);
-					return {
-						kind: TokenKind.Unknown,
-						start,
-						end: position,
-						error: {
-							code: "unterminated-string",
-							message: `unterminated attribute value — add a closing ${quote}`
-						}
-					};
-				}
-				position++;
-				return token(TokenKind.String, start);
-			default:
-				if (isLetter(code)) {
-					position++;
-					while (position < length) {
-						const current = source.charCodeAt(position);
-						if (!isLetter(current) && !isDigit(current)) break;
-						position++;
-					}
-					return token(TokenKind.Identifier, start);
-				}
-				if (isDigit(code)) {
-					position++;
-					while (position < length && isDigit(source.charCodeAt(position))) position++;
-					return token(TokenKind.Number, start);
-				}
-				return unknown(start, "unexpected-character", `unexpected character ${JSON.stringify(String.fromCharCode(code))} inside tag`);
-		}
-	};
-	return {
-		position: () => position,
-		scanTextToken,
-		scanTagToken,
-		getText: (t) => source.slice(t.start, t.end),
-		getTextValue: (t) => decodeEntities(source.slice(t.start, t.end)),
-		getStringValue: (t) => decodeEntities(source.slice(t.start + 1, t.end - 1))
-	};
-};
-//#endregion
-export { TokenKind, createLexer, decodeEntities };
-
-//# sourceMappingURL=lexer.js.map

package/dist/model.d.ts

@@ -1,24 +0,0 @@
-import { S as SmlTag, b as SmlDocument } from "./ast-BrDtApwS.js";
-import { Result } from "better-result";
-//#region src/model.d.ts
-/** The subset of an element's attributes the navigator and agent tools read. */
-type SmlNodeKeyAttrs = Readonly<Record<string, string | number>>;
-interface SmlNodeIndexEntry {
-  readonly id: string;
-  readonly tag: SmlTag;
-  readonly order: number;
-  readonly parentId: string | null;
-  readonly childIndex: number;
-  readonly depth: number;
-  readonly path: string;
-  readonly textPreview: string;
-  readonly attrs: SmlNodeKeyAttrs;
-}
-interface SmlNodeIndex {
-  readonly nodes: readonly SmlNodeIndexEntry[];
-}
-declare const deriveNodeIndexFromDocument: (document: SmlDocument) => SmlNodeIndex;
-declare const deriveNodeIndex: (source: string) => Result<SmlNodeIndex, string>;
-//#endregion
-export { SmlNodeIndex, SmlNodeIndexEntry, SmlNodeKeyAttrs, deriveNodeIndex, deriveNodeIndexFromDocument };
-//# sourceMappingURL=model.d.ts.map

package/dist/model.js

@@ -1,77 +0,0 @@
-import { parseSml } from "./parser.js";
-import { Result } from "better-result";
-//#region src/model.ts
-const KEY_ATTRS = /* @__PURE__ */ new Set([
-	"alt",
-	"class",
-	"dir",
-	"height",
-	"hidden",
-	"href",
-	"label",
-	"lang",
-	"level",
-	"name",
-	"scheme",
-	"src",
-	"test",
-	"valign",
-	"width"
-]);
-const attrValue = (value) => value.value;
-const truncate = (value, length) => value.length <= length ? value : `${value.slice(0, length - 3)}...`;
-const normalizePreview = (value) => truncate(value.replace(/\s+/g, " ").trim(), 96);
-const textPreview = (node) => {
-	const parts = [];
-	const visit = (child) => {
-		if (child.kind === "text") {
-			parts.push(child.value);
-			return;
-		}
-		if (child.kind === "interpolation") {
-			parts.push(`{{${child.name}}}`);
-			return;
-		}
-		for (const nested of child.children) visit(nested);
-	};
-	for (const child of node.children) visit(child);
-	return normalizePreview(parts.join(" "));
-};
-const keyAttrs = (node) => Object.fromEntries(node.attributes.filter((attribute) => KEY_ATTRS.has(attribute.name)).map((attribute) => [attribute.name, attrValue(attribute.value)]));
-const deriveNodeIndexFromDocument = (document) => {
-	const nodes = [];
-	let order = 0;
-	const visit = (node, parentId, childIndex, depth, parentPath) => {
-		if (node.kind !== "element") return;
-		if (node.tag === "Source") return;
-		const pathSegment = `${node.tag}[${childIndex}]`;
-		const path = parentPath ? `${parentPath}/${pathSegment}` : pathSegment;
-		nodes.push({
-			id: node.id,
-			tag: node.tag,
-			order: order++,
-			parentId,
-			childIndex,
-			depth,
-			path,
-			textPreview: textPreview(node),
-			attrs: keyAttrs(node)
-		});
-		let elementChildIndex = 0;
-		for (const child of node.children) if (child.kind === "element") {
-			visit(child, node.id, elementChildIndex, depth + 1, path);
-			elementChildIndex++;
-		}
-	};
-	visit(document.root, null, 0, 0, "");
-	return { nodes };
-};
-const deriveNodeIndex = (source) => {
-	const parsed = parseSml(source);
-	if (Result.isError(parsed)) return Result.err(parsed.error.diagnostics.map((diagnostic) => diagnostic.message).join("\n"));
-	return Result.ok(deriveNodeIndexFromDocument(parsed.value));
-};
-//#endregion
-export { deriveNodeIndex, deriveNodeIndexFromDocument };
-
-//# sourceMappingURL=model.js.map