rubika 1.2.3 → 1.2.4

package/utils/parser.ts

@@ -1,173 +1,137 @@
-import { MarkdownType, MetaDataPart, MetadataResult } from "../bot/types/utils";
-
-const MENTION_PREFIX_TYPES: Record<string, string> = {
-  u: "User",
-  g: "Group",
-  c: "Channel",
-  b: "Bot",
+type MarkdownEntity = {
+  type: string;
+  from_index: number;
+  length: number;
+  language?: string;
+  mention_text_user_id?: string;
+  mention_text_object_guid?: string;
+  mention_text_object_type?: string;
+  link?: { type: string; hyperlink_data: { url: string } };
 };
 
-function buildUtf16PrefixLengths(text: string): number[] {
-  const prefixLengths: number[] = [0];
-  let total = 0;
-  for (const char of text) {
-    total += char.codePointAt(0)! > 0xffff ? 2 : 1;
-    prefixLengths.push(total);
-  }
-  return prefixLengths;
-}
-
-const MARKDOWN_RE =
-  /(?:^(?:> ?[^\n]*\n?)+)|```([\s\S]*?)```|\*\*([^\n*]+?)\*\*|`([^\n`]+?)`|__([^\n_]+?)__|--([^\n-]+?)--|~~([^\n~]+?)~~|\|\|([^\n|]+?)\|\||\[([^\]]+?)\]\((\S+)\)/gms;
-
-const MARKDOWN_TYPE_SEQUENCE: [string, [MarkdownType, number | null]][] = [
-  [">", ["Quote", null]],
-  ["```", ["Pre", 1]],
-  ["**", ["Bold", 2]],
-  ["`", ["Mono", 3]],
-  ["__", ["Italic", 4]],
-  ["--", ["Underline", 5]],
-  ["~~", ["Strike", 6]],
-  ["||", ["Spoiler", 7]],
-  ["[", ["Link", 8]],
-];
-
-export default class Markdown {
-  static toMetadata(text: string): MetadataResult {
-    const metaDataParts: MetaDataPart[] = [];
-    let currentText = text;
+class Markdown {
+  static markdownRegExp =
+    /(^|\s|\n)(````?)([\s\S]+?)(````?)([\s\n.,:?!;]|$)|(^|\s)(`|\*\*|__|~~|--|\|\||\^\^)([^\n]+?)\7([\s.,:?!;]|$)|@([a-zA-Z0-9]+)\s*\((.+?)\)|\[(.+?)\]\((.+?)\)/m;
+
+  static markdownEntities: Record<string, string> = {
+    "`": "Mono",
+    "**": "Bold",
+    __: "Italic",
+    "||": "Spoiler",
+    "~~": "Strike",
+    "--": "Underline",
+    "^^": "Quote",
+  };
+
+  static toMetadata(text: string): {
+    text: string;
+    metadata?: { meta_data_parts: MarkdownEntity[] };
+  } {
+    const entities: MarkdownEntity[] = [];
+    let match: RegExpMatchArray | null;
+    let remainingText = text;
+    const parsedTextParts: string[] = [];
     let offset = 0;
-    let charOffset = 0;
-    const utf16Prefix = buildUtf16PrefixLengths(text);
-
-    let match: RegExpExecArray | null;
-    const globalRegex = new RegExp(MARKDOWN_RE, "gms");
-
-    while ((match = globalRegex.exec(text)) !== null) {
-      const group = match[0];
-      const start = match.index;
-      const end = start + group.length;
-      const adjustedStart = utf16Prefix[start] - offset;
-      const adjustedCharStart = start - charOffset;
-
-      for (const [prefix, [mdType, groupIdx]] of MARKDOWN_TYPE_SEQUENCE) {
-        if (group.startsWith(prefix)) {
-          let content = "";
-          let contentLength = 0;
-          let charContentLength = 0;
-
-          if (mdType === "Quote") {
-            const quoteLines = group.split("\n");
-            const contentLines = quoteLines.map((line) => {
-              if (line.startsWith("> ")) return line.slice(2);
-              if (line.startsWith(">")) return line.slice(1);
-              return line;
-            });
-            content = contentLines.join("\n");
-            charContentLength = content.length;
-            contentLength = Markdown.getUtf16Length(content);
-
-            const innerMeta = this.toMetadata(content);
-            content = innerMeta.text;
-            contentLength = Markdown.getUtf16Length(content);
-            charContentLength = content.length;
-
-            if (innerMeta.metadata) {
-              for (const part of innerMeta.metadata.meta_data_parts) {
-                part.from_index += adjustedStart;
-                metaDataParts.push(part);
-              }
-            }
-          } else {
-            if (groupIdx !== null && match[groupIdx] !== undefined) {
-              content = match[groupIdx] || "";
-              const groupStart = match.index + group.indexOf(content);
-              const groupEnd = groupStart + content.length;
-              contentLength = utf16Prefix[groupEnd] - utf16Prefix[groupStart];
-              charContentLength = content.length;
-
-              if (mdType !== "Pre" && mdType !== "Link") {
-                const innerMeta = this.toMetadata(content);
-                content = innerMeta.text;
-                contentLength = Markdown.getUtf16Length(content);
-                charContentLength = content.length;
-
-                if (innerMeta.metadata) {
-                  for (const part of innerMeta.metadata.meta_data_parts) {
-                    part.from_index += adjustedStart;
-                    metaDataParts.push(part);
-                  }
-                }
-              }
-            } else {
-              content = "";
-              contentLength = 0;
-              charContentLength = 0;
-            }
+
+    while ((match = remainingText.match(this.markdownRegExp))) {
+      const matchIndex = offset + (match.index || 0);
+      parsedTextParts.push(remainingText.substring(0, match.index!));
+      let matchedContent = match[3] || match[8] || match[11] || match[13];
+
+      if (matchedContent) {
+        offset -= matchedContent.length;
+        matchedContent = matchedContent.trim();
+        offset += matchedContent.length;
+
+        if (/^`*$/.test(matchedContent)) {
+          parsedTextParts.push(match[0]);
+        } else if (match[3]) {
+          if (match[5] === "\n") {
+            match[5] = "";
+            offset -= 1;
+          }
+          parsedTextParts.push(match[1] + matchedContent + match[5]);
+          entities.push({
+            type: "Pre",
+            language: "",
+            from_index: matchIndex + match[1].length,
+            length: matchedContent.length,
+          });
+          offset -= match[2].length + match[4].length;
+        } else if (match[7]) {
+          parsedTextParts.push(match[6] + matchedContent + match[9]);
+          entities.push({
+            type: this.markdownEntities[match[7]],
+            from_index: matchIndex + match[6].length,
+            length: matchedContent.length,
+          });
+          offset -= 2 * match[7].length;
+        } else if (match[11]) {
+          parsedTextParts.push(matchedContent);
+          entities.push({
+            type: "MentionText",
+            mention_text_object_guid: match[10],
+            from_index: matchIndex,
+            length: matchedContent.length,
+            mention_text_object_type: "User",
+          });
+          offset -= match[0].length - matchedContent.length;
+        } else if (match[12]) {
+          const [label, url] = [match[12], match[13]];
+
+          let mentionType: string | undefined;
+          if (url.length === 32) {
+            if (url.startsWith("u")) mentionType = "User";
+            else if (url.startsWith("g")) mentionType = "Group";
+            else if (url.startsWith("c")) mentionType = "Channel";
           }
 
-          const metaDataPart: MetaDataPart = {
-            type: mdType,
-            from_index: adjustedStart,
-            length: contentLength,
+          parsedTextParts.push(label);
+          const metaDataPart: MarkdownEntity = {
+            type: mentionType ? "MentionText" : "Link",
+            from_index: matchIndex,
+            length: label.length,
           };
 
-          // تنظیمات بر اساس نوع
-          if (mdType === "Pre") {
-            const lines = content.split("\n", 2);
-            const language =
-              lines.length > 1 && lines[0].trim() ? lines[0].trim() : "";
-            metaDataPart.language = language;
-            if (language) {
-              content = lines.slice(1).join("\n");
-              contentLength = Markdown.getUtf16Length(content);
-              charContentLength = content.length;
-            }
-          } else if (mdType === "Link") {
-            const url = match[9];
-            const mentionType = MENTION_PREFIX_TYPES[url?.[0]] || "hyperlink";
-
-            if (mentionType === "hyperlink") {
-              metaDataPart.link_url = url;
-              metaDataPart.link = {
-                type: mentionType,
-                hyperlink_data: { url },
-              };
-            } else {
-              metaDataPart.type = "MentionText";
-              metaDataPart.mention_text_object_guid = url;
-              metaDataPart.mention_text_user_id = url;
-              metaDataPart.mention_text_object_type = mentionType;
-            }
+          if (mentionType) {
+            metaDataPart.mention_text_object_guid = url;
+            metaDataPart.mention_text_user_id = url;
+            metaDataPart.mention_text_object_type = mentionType;
+          } else {
+            metaDataPart.link = { type: "hyperlink", hyperlink_data: { url } };
           }
 
-          metaDataParts.push(metaDataPart);
-          currentText =
-            currentText.slice(0, adjustedCharStart) +
-            content +
-            currentText.slice(end - charOffset);
-
-          const markupLength = utf16Prefix[end] - utf16Prefix[start];
-          const charMarkupLength = end - start;
-          offset += markupLength - contentLength;
-          charOffset += charMarkupLength - charContentLength;
-
-          break;
+          entities.push(metaDataPart);
+          offset -= match[0].length - label.length;
         }
+
+        remainingText = remainingText.substring(match.index! + match[0].length);
+        offset += match.index! + match[0].length;
       }
     }
 
-    return {
-      text: currentText.trim(),
-      ...(metaDataParts.length > 0 && {
-        metadata: { meta_data_parts: metaDataParts },
-      }),
+    parsedTextParts.push(remainingText);
+    let resultText = parsedTextParts.join("");
+    if (!resultText.replace(/\s+/g, "").length) {
+      resultText = text;
+      entities.splice(0, entities.length);
+    }
+    if (!entities.length) {
+      resultText = resultText.trim();
+    }
+
+    const returnData: {
+      text: string;
+      metadata?: { meta_data_parts: MarkdownEntity[] };
+    } = {
+      text: resultText,
     };
-  }
 
-  static getUtf16Length(str: string): number {
-    return str.split("").reduce((len, char) => {
-      return len + (char.codePointAt(0)! > 0xffff ? 2 : 1);
-    }, 0);
+    if (entities.length > 0)
+      returnData.metadata = { meta_data_parts: entities };
+
+    return returnData;
   }
 }
+
+export default Markdown;

package/utils/errors.ts

@@ -1,26 +0,0 @@
-import checkFilters from "./checkFilter";
-
-class Logger<T> {
-  constructor(
-    private errors: any,
-    private bot: T,
-  ) {}
-
-  async error(text: string, type: "error" | "warn") {
-    const time = new Date();
-    for (let { filters, handler } of this.errors) {
-      const error = {
-        message: `${time.toISOString()} | ⟮ ${type} ⟯ ----> ${text}\n`,
-        bot: this.bot,
-      };
-
-      const passed = await checkFilters(text, filters);
-
-      if (passed) {
-        await handler(error);
-      }
-    }
-  }
-}
-
-export default Logger;