@wdprlib/parser 3.2.0 → 4.0.0

package/src/parser/rules/block/module/listpages/compiler.ts

@@ -20,20 +20,9 @@
  * @module
  */
 
-import type { CompiledTemplate, VariableContext, PageData, UserInfo } from "./types";
-
-/** Default character count for `%%preview%%` when no length is specified (Wikidot default). */
-const DEFAULT_PREVIEW_LENGTH = 200;
-
-/**
- * Regex pattern for matching ListPages template variables with all parameter variants.
- *
- * Captures: [1] variable name, [2] brace parameter, [3] paren parameter, [4] format string.
- * The format portion allows single `%` characters (for strftime tokens like `%Y`)
- * but stops at `%%` (which terminates the variable).
- */
-const VARIABLE_REGEX =
-  /%%([a-z_]+)(?:\{([^}]+)\})?(?:\((\d+)\))?(?:\|([^%]*(?:%(?!%)[^%]*)*))?%%/gi;
+import type { CompiledTemplate, VariableContext } from "./types";
+import { createVariableGetter } from "./template/getters";
+import { scanTemplateVariables } from "./template/syntax";
 
 /**
  * Compile a ListPages template string into an executable function.
@@ -50,19 +39,22 @@ export function compileTemplate(template: string): CompiledTemplate {
   let lastIndex = 0;
 
   // Split template into static and dynamic parts
-  for (const match of template.matchAll(VARIABLE_REGEX)) {
+  for (const match of scanTemplateVariables(template)) {
     // Add static part before this match
-    if (match.index !== undefined && match.index > lastIndex) {
+    if (match.index > lastIndex) {
       parts.push(template.slice(lastIndex, match.index));
     }
 
     // Convert variable to getter function
-    const [, varName, braceParam, parenParam, format] = match;
-    if (!varName) continue;
-    const getter = createVariableGetter(varName.toLowerCase(), braceParam, parenParam, format);
+    const getter = createVariableGetter(
+      match.name.toLowerCase(),
+      match.braceParam,
+      match.parenParam,
+      match.format,
+    );
     parts.push(getter);
 
-    lastIndex = match.index !== undefined ? match.index + match[0].length : lastIndex;
+    lastIndex = match.index + match.raw.length;
   }
 
   // Add remaining static part
@@ -79,375 +71,3 @@ export function compileTemplate(template: string): CompiledTemplate {
     return result;
   };
 }
-
-/**
- * Create a getter function for a specific template variable.
- *
- * Handles all variable variants: parameterized (`{param}`), parenthesized (`(param)`),
- * formatted (`|format`), and simple. Returns a function that extracts the appropriate
- * value from a `VariableContext`.
- *
- * Unknown variable names return a function that always returns an empty string,
- * matching Wikidot's behavior of silently ignoring unknown variables.
- *
- * @param name - Lowercase variable name
- * @param braceParam - Content inside `{...}`, if present
- * @param parenParam - Content inside `(...)`, if present
- * @param format - Content after `|`, if present
- * @returns A function that extracts the variable's value from a VariableContext
- */
-function createVariableGetter(
-  name: string,
-  braceParam?: string,
-  parenParam?: string,
-  format?: string,
-): (ctx: VariableContext) => string {
-  // Parameterized variables with {param}
-  if (braceParam !== undefined) {
-    switch (name) {
-      case "content": {
-        // %%content{n}%% - n is 1-indexed, split content by ==== on demand
-        const idx = Number(braceParam) - 1;
-        return (ctx) => {
-          if (!ctx.page.content) return "";
-          const sections = splitContentSections(ctx.page.content);
-          return sections[idx] ?? "";
-        };
-      }
-      case "form_data":
-        return (ctx) => ctx.page.formData?.[braceParam] ?? "";
-      case "form_raw":
-        return (ctx) => ctx.page.formRaw?.[braceParam] ?? "";
-      case "form_label":
-        return (ctx) => ctx.page.formLabel?.[braceParam] ?? "";
-      case "form_hint":
-        return (ctx) => ctx.page.formHint?.[braceParam] ?? "";
-    }
-  }
-
-  // Parameterized variables with (param)
-  if (parenParam !== undefined && name === "preview") {
-    const len = Number(parenParam);
-    return (ctx) => (ctx.page.content ?? "").slice(0, len);
-  }
-
-  // Date variables with format
-  if (format !== undefined) {
-    switch (name) {
-      case "created_at":
-        return (ctx) => formatDate(ctx.page.createdAt, format);
-      case "updated_at":
-        return (ctx) => formatDate(ctx.page.updatedAt, format);
-      case "commented_at":
-        return (ctx) => (ctx.page.commentedAt ? formatDate(ctx.page.commentedAt, format) : "");
-    }
-  }
-
-  // tags_linked with prefix
-  if (name === "tags_linked") {
-    const prefix = format ?? "/system:page-tags/tag/";
-    return (ctx) => formatTagsLinked(ctx.page.tags, prefix);
-  }
-  if (name === "_tags_linked") {
-    const prefix = format ?? "/system:page-tags/tag/";
-    return (ctx) => formatTagsLinked(ctx.page.hiddenTags, prefix);
-  }
-
-  // Simple variables
-  const getter = SIMPLE_GETTERS[name];
-  if (getter) return getter;
-
-  // Unknown variable - return empty string
-  return () => "";
-}
-
-// =============================================================================
-// Simple Variable Getters
-// =============================================================================
-
-const SIMPLE_GETTERS: Record<string, (ctx: VariableContext) => string> = {
-  // Lifecycle - created
-  created_at: (ctx) => formatDate(ctx.page.createdAt),
-  created_by: (ctx) => ctx.page.createdBy?.name ?? "Anonymous",
-  created_by_unix: (ctx) => ctx.page.createdBy?.unixName ?? "",
-  created_by_id: (ctx) => String(ctx.page.createdBy?.id ?? 0),
-  created_by_linked: (ctx) => formatUserLinked(ctx.page.createdBy),
-
-  // Lifecycle - updated
-  updated_at: (ctx) => formatDate(ctx.page.updatedAt),
-  updated_by: (ctx) => ctx.page.updatedBy?.name ?? "Anonymous",
-  updated_by_unix: (ctx) => ctx.page.updatedBy?.unixName ?? "",
-  updated_by_id: (ctx) => String(ctx.page.updatedBy?.id ?? 0),
-  updated_by_linked: (ctx) => formatUserLinked(ctx.page.updatedBy),
-
-  // Lifecycle - commented
-  commented_at: (ctx) => (ctx.page.commentedAt ? formatDate(ctx.page.commentedAt) : ""),
-  commented_by: (ctx) => ctx.page.commentedBy?.name ?? "",
-  commented_by_unix: (ctx) => ctx.page.commentedBy?.unixName ?? "",
-  commented_by_id: (ctx) => String(ctx.page.commentedBy?.id ?? 0),
-  commented_by_linked: (ctx) => formatUserLinked(ctx.page.commentedBy),
-
-  // Structure - page
-  name: (ctx) => ctx.page.name,
-  category: (ctx) => ctx.page.category,
-  fullname: (ctx) => ctx.page.fullname,
-  title: (ctx) => ctx.page.title,
-  title_linked: (ctx) => `[[[${ctx.page.fullname} | ${ctx.page.title}]]]`,
-  link: (ctx) => `https://${ctx.site.domain}/${ctx.page.fullname}`,
-
-  // Structure - parent
-  parent_name: (ctx) => ctx.page.parentName ?? "",
-  parent_category: (ctx) => ctx.page.parentCategory ?? "",
-  parent_fullname: (ctx) => ctx.page.parentFullname ?? "",
-  parent_title: (ctx) => ctx.page.parentTitle ?? "",
-  parent_title_linked: (ctx) =>
-    ctx.page.parentFullname ? `[[[${ctx.page.parentFullname} | ${ctx.page.parentTitle}]]]` : "",
-
-  // Content
-  content: (ctx) => ctx.page.content ?? "",
-  preview: (ctx) => (ctx.page.content ?? "").slice(0, DEFAULT_PREVIEW_LENGTH),
-  summary: (ctx) => getSummary(ctx.page),
-  first_paragraph: (ctx) => getFirstParagraph(ctx.page.content),
-
-  // Tags
-  tags: (ctx) => ctx.page.tags.join(" "),
-  _tags: (ctx) => ctx.page.hiddenTags.join(" "),
-
-  // Metrics
-  children: (ctx) => String(ctx.page.children),
-  comments: (ctx) => String(ctx.page.comments),
-  size: (ctx) => String(ctx.page.size),
-  rating: (ctx) => String(ctx.page.rating),
-  rating_votes: (ctx) => String(ctx.page.ratingVotes),
-  rating_percent: (ctx) => String(ctx.page.ratingPercent ?? 0),
-  revisions: (ctx) => String(ctx.page.revisions),
-
-  // Pagination
-  index: (ctx) => String(ctx.index),
-  total: (ctx) => String(ctx.total),
-  limit: (ctx) => (ctx.limit !== undefined ? String(ctx.limit) : ""),
-  total_or_limit: (ctx) =>
-    String(ctx.limit !== undefined ? Math.min(ctx.total, ctx.limit) : ctx.total),
-
-  // Site context
-  site_title: (ctx) => ctx.site.title,
-  site_name: (ctx) => ctx.site.name,
-  site_domain: (ctx) => ctx.site.domain,
-};
-
-// =============================================================================
-// Helper Functions
-// =============================================================================
-
-/**
- * Format a Date object using an optional strftime-like format string.
- *
- * When no format is provided, returns the ISO 8601 string representation.
- *
- * @param date - The date to format
- * @param format - Optional strftime format string (e.g., `"%Y-%m-%d"`)
- * @returns Formatted date string
- */
-function formatDate(date: Date, format?: string): string {
-  if (!format) {
-    // Default ISO format
-    return date.toISOString();
-  }
-  return strftime(date, format);
-}
-
-/**
- * Format a user as Wikidot's `[[*user name]]` inline syntax for linked display.
- *
- * @param user - User info, or undefined for anonymous display
- * @returns Wikidot user link syntax, or "Anonymous" if no user
- */
-function formatUserLinked(user?: UserInfo): string {
-  if (!user) return "Anonymous";
-  return `[[*user ${user.name}]]`;
-}
-
-/**
- * Format an array of tags as space-separated Wikidot link syntax.
- *
- * Each tag becomes `[prefix/tag tag]` where prefix defaults to
- * `/system:page-tags/tag/` unless overridden by the template's pipe format.
- *
- * @param tags - Array of tag names
- * @param prefix - URL prefix for tag links
- * @returns Space-separated string of Wikidot link syntax, or empty string if no tags
- */
-function formatTagsLinked(tags: string[], prefix: string): string {
-  if (tags.length === 0) return "";
-  return tags.map((tag) => `[${prefix}${tag} ${tag}]`).join(" ");
-}
-
-/**
- * Split content by ==== separators
- *
- * wikidot uses preg_split('/^([=]{4,})$/m', $source) to split content
- * %%content{n}%% references sections[n-1] (1-indexed)
- */
-function splitContentSections(content: string): string[] {
-  return content.split(/^={4,}$/m).map((section) => section.trim());
-}
-
-/**
- * Extract a summary from page content.
- *
- * If the content contains `====` section separators, returns the first section
- * (equivalent to `%%content{1}%%`). Otherwise, returns the first paragraph.
- *
- * @param page - Page data containing content
- * @returns Summary text extracted from the page content
- */
-function getSummary(page: PageData): string {
-  if (page.content) {
-    const sections = splitContentSections(page.content);
-    if (sections.length > 1) {
-      return sections[0]?.trim() ?? "";
-    }
-  }
-  return getFirstParagraph(page.content);
-}
-
-/**
- * Extract the first paragraph from wikitext content.
- *
- * Strips headings, TOC directives, div blocks, and module blocks before
- * splitting on double newlines to find the first content paragraph.
- *
- * @param content - Raw wikitext content, or undefined
- * @returns The first paragraph text, or empty string if no content
- */
-function getFirstParagraph(content?: string): string {
-  if (!content) return "";
-
-  // Remove headings, TOC, div, module blocks
-  const s = content
-    .replace(/^(\+{1,6}) (.*)/gm, "")
-    .replace(/^\[\[toc(\s[^\]]+)?\]\]/gim, "")
-    .replace(/^\[\[\/?div(\s[^\]]+)?\]\]/gim, "")
-    .replace(/^\[\[\/?module(\s[^\]]+)?\]\]/gim, "")
-    .trim();
-
-  // Split by double newlines and take first
-  const paragraphs = s.split(/\n{2,}/);
-  return paragraphs[0]?.trim() ?? "";
-}
-
-/** Full month names for strftime `%B` token. Static to avoid per-call allocation. */
-const MONTHS = [
-  "January",
-  "February",
-  "March",
-  "April",
-  "May",
-  "June",
-  "July",
-  "August",
-  "September",
-  "October",
-  "November",
-  "December",
-];
-/** Abbreviated month names for strftime `%b` token. */
-const MONTHS_SHORT = [
-  "Jan",
-  "Feb",
-  "Mar",
-  "Apr",
-  "May",
-  "Jun",
-  "Jul",
-  "Aug",
-  "Sep",
-  "Oct",
-  "Nov",
-  "Dec",
-];
-/** Full day names for strftime `%A` token. */
-const DAYS = ["Sunday", "Monday", "Tuesday", "Wednesday", "Thursday", "Friday", "Saturday"];
-/** Abbreviated day names for strftime `%a` token. */
-const DAYS_SHORT = ["Sun", "Mon", "Tue", "Wed", "Thu", "Fri", "Sat"];
-
-/** Pre-compiled regex matching strftime tokens (`%X` where X is a supported letter or `%`). */
-const STRFTIME_REGEX = /%([YymdHeHIMSpbBaAwjZz%])/g;
-
-/**
- * Minimal strftime implementation supporting common tokens.
- *
- * Uses a single-pass regex replace for O(n) performance. All dates are
- * treated as UTC to match Wikidot's server-side rendering behavior.
- *
- * Supported tokens: `%Y` (4-digit year), `%y` (2-digit year), `%m` (month),
- * `%d` (zero-padded day), `%e` (day), `%H` (24h hour), `%I` (12h hour),
- * `%M` (minute), `%S` (second), `%p` (AM/PM), `%b`/`%B` (month name),
- * `%a`/`%A` (day name), `%w` (weekday number), `%j` (day of year),
- * `%Z`/`%z` (timezone), `%%` (literal percent).
- *
- * @param date - The date to format
- * @param format - strftime format string
- * @returns Formatted date string
- */
-function strftime(date: Date, format: string): string {
-  const pad = (n: number, len = 2) => String(n).padStart(len, "0");
-
-  return format.replace(STRFTIME_REGEX, (_, token: string) => {
-    switch (token) {
-      case "Y":
-        return String(date.getFullYear());
-      case "y":
-        return String(date.getFullYear()).slice(-2);
-      case "m":
-        return pad(date.getMonth() + 1);
-      case "d":
-        return pad(date.getDate());
-      case "e":
-        return String(date.getDate());
-      case "H":
-        return pad(date.getHours());
-      case "I":
-        return pad(date.getHours() % 12 || 12);
-      case "M":
-        return pad(date.getMinutes());
-      case "S":
-        return pad(date.getSeconds());
-      case "p":
-        return date.getHours() < 12 ? "AM" : "PM";
-      case "b":
-        return MONTHS_SHORT[date.getMonth()] ?? "";
-      case "B":
-        return MONTHS[date.getMonth()] ?? "";
-      case "a":
-        return DAYS_SHORT[date.getDay()] ?? "";
-      case "A":
-        return DAYS[date.getDay()] ?? "";
-      case "w":
-        return String(date.getDay());
-      case "j":
-        return pad(getDayOfYear(date), 3);
-      case "Z":
-        return "UTC";
-      case "z":
-        return "+0000";
-      case "%":
-        return "%";
-      default:
-        return `%${token}`;
-    }
-  });
-}
-
-/**
- * Calculate the day of year (1-366) for a given date.
- *
- * @param date - The date to calculate for
- * @returns Day of year as an integer (1 = January 1st)
- */
-function getDayOfYear(date: Date): number {
-  const start = new Date(date.getFullYear(), 0, 0);
-  const diff = date.getTime() - start.getTime();
-  const oneDay = 1000 * 60 * 60 * 24;
-  return Math.floor(diff / oneDay);
-}