utilitas 1995.2.21 → 1995.2.23

package/lib/email.mjs

@@ -3,11 +3,13 @@ import {
     which
 } from './utilitas.mjs';
 
+import { convert } from './storage.mjs';
 import { isPrimary } from './callosum.mjs';
 
-const _NEED = ['form-data', 'mailgun.js', 'node-mailjet'];
+const _NEED = ['form-data', 'mailgun.js', 'mailparser', 'node-mailjet'];
 const getSenderName = () => senderName;
 const log = (content) => _log(content, import.meta.url);
+const TEXT = 'TEXT';
 
 let domain, senderName, senderEmail, provider, client;
 
@@ -90,13 +92,22 @@ const send = async (email, subject, text, html, args, options) => {
         senderName: options.senderName, senderEmail: options.senderEmail,
         to: uniqueArray(email), subject, text
     }, html ? { html } : {}));
-}
+};
+
+const parse = async (input, options) => {
+    const { simpleParser } = await need('mailparser');
+    const email = await convert(input, {
+        input: TEXT, expected: TEXT, ...options || {},
+    });
+    return await simpleParser(email, options);
+};
 
 export default init;
 export {
     _NEED,
     getSenderName,
     init,
+    parse,
     rawSend,
     send,
 };

package/lib/manifest.mjs

@@ -1,7 +1,7 @@
 const manifest = {
     "name": "utilitas",
     "description": "Just another common utility for JavaScript.",
-    "version": "1995.2.21",
+    "version": "1995.2.23",
     "private": false,
     "homepage": "https://github.com/Leask/utilitas",
     "main": "index.mjs",
@@ -44,6 +44,7 @@ const manifest = {
         "jsdom": "^23.0.1",
         "lorem-ipsum": "^2.0.8",
         "mailgun.js": "^9.3.0",
+        "mailparser": "^3.6.5",
         "mime-types": "^2.1.35",
         "mysql2": "^3.6.5",
         "node-mailjet": "^6.0.5",

package/lib/uoid.mjs

@@ -1,6 +1,6 @@
 import { assertUrl, basename, ensureString, rotate } from './utilitas.mjs';
 import { hexToBigInt, md5, randomString, uniqueString } from './encryption.mjs';
-import { v1 as uuidv1, v5 as uuidv5 } from 'uuid';
+import { v4 as uuidv4, v5 as uuidv5 } from 'uuid';
 import manifest from './manifest.mjs';
 
 // https://stackoverflow.com/questions/7905929/how-to-test-valid-uuid-guid
@@ -19,7 +19,7 @@ const getTimestampFromUuid = (uuid) => uuid ? Number((BigInt(
 ) - 122192928000000000n) / 10000n) : 0;
 
 const create = (options) => {
-    options = Object.assign({ file: import.meta.url, id: uuidv1() }, options || {});
+    options = Object.assign({ file: import.meta.url, id: uuidv4() }, options || {});
     options.type = options.type || basename(options.file);
     if ((options.security = ~~options.security) === 1) {
         options.security = 128;

package/lib/vision.mjs

@@ -7,8 +7,10 @@ import {
     trim,
 } from './utilitas.mjs';
 
+import { v4 as uuidv4 } from 'uuid';
 import { getApiKeyCredentials } from './encryption.mjs';
 import fs from 'node:fs';
+import path from 'node:path';
 
 const _NEED = [
     '@google-cloud/vision', 'office-text-extractor', 'pdfjs-dist',
@@ -18,14 +20,14 @@ const _NEED = [
 const [BASE64, BUFFER, FILE, DEFAULT_LANG] = ['BASE64', 'BUFFER', 'FILE', 'eng'];
 const ceil = num => num.toFixed(4);
 const errorMessage = 'Invalid input data.';
-const getTextFromBatch = bt => bt.responses.map(p => p.fullTextAnnotation.text);
+const getTextFromBatch = b => b.responses.map(p => p?.fullTextAnnotation?.text || '');
 const DOCUMENT_TEXT_DETECTION = 'DOCUMENT_TEXT_DETECTION';
 const features = [{ type: DOCUMENT_TEXT_DETECTION }];
 const mimeType = 'application/pdf';
 const pages = [1, 2, 3, 4, 5]; // max 5 pages limit for batchAnnotateFiles API
 const log = content => _log(content, import.meta.url);
 
-let client, officeParser, xlsx;
+let client;
 
 const init = async (options) => {
     if (options) {
@@ -49,16 +51,22 @@ const parseOfficeFile = async (source, options) => {
     }];
     switch (ensureString(options?.lib || OTE, { case: 'LOW' })) {
         case OTE:
-            officeParser || (officeParser = (
-                await need(OTE)
-            ).getTextExtractor());
+            const officeParser = (await need(OTE)).getTextExtractor();
             const input = await convert(source, { ...opts, expected: BUFFER });
-            return await officeParser.extractText({
-                input, type: BUFFER.toLowerCase(),
-            });
+            try {
+                return await officeParser.extractText({
+                    input, type: BUFFER.toLowerCase(),
+                });
+            } catch (err) {
+                if (/application\/x\-cfb$/.test(err.message)) {
+                    err.message += ` (https://github.com/gamemaker1/office-text-extractor/issues/10)`;
+                }
+                throwError(err.message);
+            }
         case XLSX:
             // https://docs.sheetjs.com/docs/getting-started/installation/nodejs
-            xlsx || ((xlsx = await need(XLSX)) && xlsx.set_fs(fs));
+            const xlsx = await need(XLSX)
+            xlsx.set_fs(fs);
             const { content, cleanup } = await convert(source, {
                 ...opts, expected: FILE, withCleanupFunc: true
             });
@@ -181,7 +189,7 @@ const read = async (image, options) => {
     const content = await convert(image, {
         input: options?.input, expected: BASE64, errorMessage,
     });
-    let result = await client.batchAnnotateFiles({
+    const result = await client.batchAnnotateFiles({
         requests: [{ inputConfig: { mimeType, content }, features, pages }],
     });
     return options?.raw ? result : getTextFromBatch(result[0].responses[0]);
@@ -191,7 +199,8 @@ const readAll = async (image, options) => {
     assert(client, 'Vision API has not been initialized.', 500);
     const result = {};
     result.upload = await uploadToCloud(image, {
-        gzip: false, prefix: options?.prefix || '_vision', ...options || {},
+        destination: path.join(options?.prefix || '_vision', `${uuidv4()}.pdf`),
+        ...options || {},
     });
     const uri = result.upload?.gs;
     const destination = `${uri}_result/`;
@@ -240,7 +249,7 @@ const getPdfInfo = async (file, options) => {
     const result = {
         numPages: doc.numPages,
         info: data.info,
-        metadata: { ...data.metadata.getAll() },
+        metadata: { ...data.metadata?.getAll() },
         pages: options?.withPages ? await getPdfPages(doc) : null,
     };
     return result;

package/lib/web.mjs

@@ -1,5 +1,6 @@
 import { assembleUrl, ignoreErrFunc, need, throwError } from './utilitas.mjs';
 import { getJson, getParsedHtml } from './shot.mjs';
+import { convert } from './storage.mjs';
 
 const _NEED = [
     'jsdom', 'youtube-transcript', '@mozilla/readability', '@ngrok/ngrok'
@@ -9,8 +10,12 @@ const _NEED = [
 const YT_REGEXP = /^((?:https?:)?\/\/)?((?:www|m)\.)?((?:youtube(-nocookie)?\.com|youtu.be))(\/(?:[\w\-]+\?v=|embed\/|v\/)?)([\w\-]+)(\S+)?$/i;
 const isYoutubeUrl = url => (url || '').match(YT_REGEXP)?.[6];
 const distillPage = async (url, op) => (await getParsedHtml(url, op))?.content;
+const TEXT = 'TEXT';
 
-const distillHtml = async (html, options) => {
+const distillHtml = async (input, options) => {
+    const html = await convert(input, {
+        input: TEXT, expected: TEXT, ...options || {},
+    });
     const lib = await Promise.all([need('jsdom'), need('@mozilla/readability')]);
     const [JSDOM, Readability] = [lib[0].JSDOM, lib[1].Readability];
     const _doc = new JSDOM(html)?.window?.document;

package/package.json

@@ -1,7 +1,7 @@
 {
     "name": "utilitas",
     "description": "Just another common utility for JavaScript.",
-    "version": "1995.2.21",
+    "version": "1995.2.23",
     "private": false,
     "homepage": "https://github.com/Leask/utilitas",
     "main": "index.mjs",
@@ -55,6 +55,7 @@
         "jsdom": "^23.0.1",
         "lorem-ipsum": "^2.0.8",
         "mailgun.js": "^9.3.0",
+        "mailparser": "^3.6.5",
         "mime-types": "^2.1.35",
         "mysql2": "^3.6.5",
         "node-mailjet": "^6.0.5",