feat: support characters function (#1022)

* feat: support characters function

* fix: support negative numbers

packages/helix-shared-indexer/src/index-resource.js

@@ -19,6 +19,37 @@ import { toText } from 'hast-util-to-text';
 import { toHtml } from 'hast-util-to-html';
 import { unified } from 'unified';
 
+/**
+ * Returns the substring from `start` to `end` of the given text. If `start` or `end` are
+ * negative, they address the position counted from the end of the text. `end` is optional,
+ * and defaults to the length of the text.
+ *
+ * Examples:
+ * - characters('hello, world.', 0, 5) = 'hello'
+ * - characters('hello, world.', 7) = 'world.'
+ * - characters('hello, world.', -6, -1) = 'world'
+ *
+ * @param {string|Array<string>} text
+ * @param {number} start The start position.
+ * @param {number} [end] The end position.
+ * @returns {Array<string>}
+ */
+const characters = (text, start, end) => {
+  if (Array.isArray(text)) {
+    // eslint-disable-next-line no-param-reassign
+    text = text.join(' ');
+  }
+  if (start < 0) {
+    // eslint-disable-next-line no-param-reassign
+    start += text.length;
+  }
+  if (end <= 0) {
+    // eslint-disable-next-line no-param-reassign
+    end += text.length;
+  }
+  return [text.substring(start, end)];
+};
+
 const helpers = {
   dateValue: (elements, format) => {
     const result = helpers.parseTimestamp(elements, format);
@@ -93,6 +124,7 @@ const helpers = {
     }
     return [text.split(/\s+/g).slice(start, end).join(' ')];
   },
+  characters,
   replace: (s, searchValue, replaceValue) => [s.replace(searchValue, replaceValue)],
   replaceAll: (s, searchValue, replaceValue) => [s.replaceAll(searchValue, replaceValue)],
 };
@@ -129,6 +161,14 @@ function evaluate(expression, context) {
       case Jsep.LITERAL: {
         return node.value;
       }
+      case Jsep.UNARY_EXP: {
+        const value = evalNode(node.argument);
+        if (node.operator === '-') {
+          return -value;
+        }
+        log.warn('operator not supported: ', node.operator);
+        return undefined;
+      }
       default: {
         log.warn('evaluate type not supported: ', node.type);
       }

packages/helix-shared-indexer/test/index-resource.test.js

@@ -60,22 +60,38 @@ indices:
         select: main > div:nth-of-type(5)
         value: |
           innerHTML(el)
-      teaser:
+      words:
         select: main > div:nth-child(n+4) p
         value: |
           words(textContent(el), 0, 20)
-      last-modified:
+      non-array-words:
         select: none
         value: |
-          parseTimestamp(headers['last-modified'], 'ddd, DD MMM YYYY hh:mm:ss GMT')
-      last-modified-raw:
+          words(headers['last-modified'])
+      characters:
+        select: main > div:nth-child(n+4) p
+        value: |
+          characters(textContent(el), 0, 20)
+      characters-negative-start:
+        select: main > div:nth-child(n+4) p
+        value: |
+          characters(textContent(el), -5)
+      characters-negative-end:
+        select: main > div:nth-child(n+4) p
+        value: |
+          characters(textContent(el), -5, -1)
+      non-array-characters:
         select: none
         value: |
-          headers['last-modified']
-      non-array-words:
+          characters(headers['last-modified'], 0, 5)
+      parse-timestamp:
         select: none
         value: |
-          words(headers['last-modified'])
+          parseTimestamp(headers['last-modified'], 'ddd, DD MMM YYYY hh:mm:ss GMT')
+      headers-last-modified:
+        select: none
+        value: |
+          headers['last-modified']
       match-simple:
         select: meta[name="x-source-hash"]
         value: |
@@ -124,6 +140,9 @@ indices:
         selectFirst: foo
         value: |
           attribute(el, 'href')
+      unknown-operator:
+        select: none
+        value: '+5'
   `;
 
 const BODY = `
@@ -193,31 +212,36 @@ describe('Index Resource Tests', () => {
       author: 'Max',
       'bad-selector': '',
       'call-unknown-function': '',
+      characters: 'Lorem ipsum dolor si',
+      'characters-negative-start': 'after',
+      'characters-negative-end': 'afte',
       'condition-unsupported': '',
       date: 44313,
       'first-alternate': 'before',
       'first-href-in-a': 'https://my.domain.com/assets/asset-link',
       'first-href-in-foo': '',
-      'second-alternate': 'before<br>after',
-      'last-modified': 1614007680,
-      'last-modified-raw': 'Mon, 22 Feb 2021 15:28:00 GMT',
+      'first-segment': 'abc',
+      'headers-last-modified': 'Mon, 22 Feb 2021 15:28:00 GMT',
       'match-simple': '',
       'member-unknown-var': '',
       'member-without-get': '',
       'missing-header': '',
+      'non-array-characters': 'Mon, ',
       'non-array-words': 'Mon, 22 Feb 2021 15:28:00 GMT',
       paragraph: '\n    <p>consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua</p>\n  ',
-      'first-segment': 'abc',
+      'parse-timestamp': 1614007680,
       'replace-path': '/zc/de/ab/fg/abcd',
       'replaceAll-path': '/zc/de/z/fg/zcd',
+      'second-alternate': 'before<br>after',
       sourceHash: 'JJYxCM1NDG4ahJm9f',
-      teaser: 'Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua Ut',
       title: 'I feel good',
       topics: [
         'A',
         'B',
         'C',
       ],
+      'unknown-operator': '',
+      words: 'Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua Ut',
     });
   });
 });