feat: truncate content exceeding the byte limit (#129)

Bumps [@typescript-eslint/parser](https://github.com/typescript-eslint/typescript-eslint/tree/HEAD/packages/parser) from 5.53.0 to 5.54.0.
- [Release notes](https://github.com/typescript-eslint/typescript-eslint/releases)
- [Changelog](https://github.com/typescript-eslint/typescript-eslint/blob/main/packages/parser/CHANGELOG.md)
- [Commits](https://github.com/typescript-eslint/typescript-eslint/commits/v5.54.0/packages/parser)

---
updated-dependencies:
- dependency-name: "@typescript-eslint/parser"
  dependency-type: direct:development
  update-type: version-update:semver-minor
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>

.github/workflows/update-major-version.yml

@@ -0,0 +1,31 @@
+name: Update Major Version
+run-name: Update ${{ github.event.inputs.main_version }} to ${{ github.event.inputs.target }}
+
+on:
+  workflow_dispatch:
+    inputs:
+      target:
+        description: The target tag or reference
+        required: true
+      main_version:
+        type: choice
+        description: The major version tag to update
+        options:
+          - v3
+
+jobs:
+  tag:
+    runs-on: ubuntu-latest
+    steps:
+    - uses: actions/checkout@v3
+      with:
+        token: ${{ secrets.ACTIONS_BOT_TOKEN }}
+        fetch-depth: 0
+    - name: Git config
+      run: |
+        git config user.name actions-bot
+        git config user.email actions-bot@users.noreply.github.com
+    - name: Tag new target
+      run: git tag -f ${{ github.event.inputs.main_version }} ${{ github.event.inputs.target }}
+    - name: Push new tag
+      run: git push origin ${{ github.event.inputs.main_version }} --force

README.md

@@ -30,8 +30,10 @@ This is useful if you `docker push` your images to Docker Hub. It provides an ea
 
 #### Content limits
 
-DockerHub has content limits, which if exceeded will result in the content being automatically truncated.
+DockerHub has content limits.
 The readme content is limited to 25,000 bytes, and `short-description` is limited to 100 characters.
+This action truncates content to prevent the request being rejected.
+If the content has been truncated a warning will be issued in the run log.
 
 #### Specifying the file path
 

__test__/readme-helper.unit.test.ts

@@ -1,4 +1,4 @@
-import {completeRelativeUrls} from '../src/readme-helper'
+import {completeRelativeUrls, truncateToBytes} from '../src/readme-helper'
 
 describe('complete relative urls tests', () => {
   const GITHUB_SERVER_URL = process.env['GITHUB_SERVER_URL']
@@ -333,3 +333,12 @@ describe('complete relative urls tests', () => {
     )
   })
 })
+
+describe('truncate to bytes tests', () => {
+  test('unicode aware truncation to a number of bytes', async () => {
+    expect(truncateToBytes('test string to be truncated', 10)).toEqual(
+      'test strin'
+    )
+    expect(truncateToBytes('😀😁😂🤣😃😄😅', 10)).toEqual('😀😁')
+  })
+})

dist/index.js

@@ -311,9 +311,10 @@ var __awaiter = (this && this.__awaiter) || function (thisArg, _arguments, P, ge
     });
 };
 Object.defineProperty(exports, "__esModule", ({ value: true }));
-exports.completeRelativeUrls = exports.getReadmeContent = exports.ENABLE_URL_COMPLETION_DEFAULT = exports.IMAGE_EXTENSIONS_DEFAULT = exports.README_FILEPATH_DEFAULT = void 0;
+exports.completeRelativeUrls = exports.getReadmeContent = exports.truncateToBytes = exports.ENABLE_URL_COMPLETION_DEFAULT = exports.IMAGE_EXTENSIONS_DEFAULT = exports.README_FILEPATH_DEFAULT = void 0;
 const core = __importStar(__nccwpck_require__(2186));
 const fs = __importStar(__nccwpck_require__(7147));
+const unicodeSubstring = __nccwpck_require__(6986);
 exports.README_FILEPATH_DEFAULT = './README.md';
 exports.IMAGE_EXTENSIONS_DEFAULT = 'bmp,gif,jpg,jpeg,png,svg,webp';
 exports.ENABLE_URL_COMPLETION_DEFAULT = false;
@@ -321,6 +322,15 @@ const TITLE_REGEX = `(?: +"[^"]+")?`;
 const REPOSITORY_URL = `${process.env['GITHUB_SERVER_URL']}/${process.env['GITHUB_REPOSITORY']}`;
 const BLOB_PREFIX = `${REPOSITORY_URL}/blob/${process.env['GITHUB_REF_NAME']}/`;
 const RAW_PREFIX = `${REPOSITORY_URL}/raw/${process.env['GITHUB_REF_NAME']}/`;
+const MAX_BYTES = 25000;
+function truncateToBytes(s, n) {
+    let len = n;
+    while (Buffer.byteLength(s) > n) {
+        s = unicodeSubstring(s, 0, len--);
+    }
+    return s;
+}
+exports.truncateToBytes = truncateToBytes;
 function getReadmeContent(readmeFilepath, enableUrlCompletion, imageExtensions) {
     return __awaiter(this, void 0, void 0, function* () {
         // Fetch the readme content
@@ -328,7 +338,11 @@ function getReadmeContent(readmeFilepath, enableUrlCompletion, imageExtensions)
             encoding: 'utf8'
         });
         readmeContent = completeRelativeUrls(readmeContent, readmeFilepath, enableUrlCompletion, imageExtensions);
-        return readmeContent;
+        const truncatedReadmeContent = truncateToBytes(readmeContent, MAX_BYTES);
+        if (truncatedReadmeContent.length !== readmeContent.length) {
+            core.warning(`The README content exceeds DockerHub's limit and has been truncated to ${MAX_BYTES} bytes.`);
+        }
+        return truncatedReadmeContent;
     });
 }
 exports.getReadmeContent = getReadmeContent;
@@ -6430,6 +6444,64 @@ if (process.env.NODE_DEBUG && /\btunnel\b/.test(process.env.NODE_DEBUG)) {
 exports.debug = debug; // for test
 
 
+/***/ }),
+
+/***/ 6986:
+/***/ ((module) => {
+
+"use strict";
+
+
+function charAt(string, index) {
+  var first = string.charCodeAt(index);
+  var second;
+  if (first >= 55296 && first <= 56319 && string.length > index + 1) {
+    second = string.charCodeAt(index + 1);
+    if (second >= 56320 && second <= 57343) {
+      return string.substring(index, index + 2);
+    }
+  }
+  return string[index];
+}
+
+function slice(string, start, end) {
+  var accumulator = "";
+  var character;
+  var stringIndex = 0;
+  var unicodeIndex = 0;
+  var length = string.length;
+
+  while (stringIndex < length) {
+    character = charAt(string, stringIndex);
+    if (unicodeIndex >= start && unicodeIndex < end) {
+      accumulator += character;
+    }
+    stringIndex += character.length;
+    unicodeIndex += 1;
+  }
+  return accumulator;
+}
+
+function toNumber(value, fallback) {
+  if (value === undefined) {
+    return fallback;
+  } else {
+    return Number(value);
+  }
+}
+
+module.exports = function (string, start, end) {
+  var realStart = toNumber(start, 0);
+  var realEnd = toNumber(end, string.length);
+  if (realEnd == realStart) {
+    return "";
+  } else if (realEnd > realStart) {
+    return slice(string, realStart, realEnd);
+  } else {
+    return slice(string, realEnd, realStart);
+  }
+};
+
 /***/ }),
 
 /***/ 5840:

package.json

@@ -28,12 +28,13 @@
   "homepage": "https://github.com/peter-evans/dockerhub-description#readme",
   "dependencies": {
     "@actions/core": "^1.10.0",
-    "node-fetch": "^2.6.9"
+    "node-fetch": "^2.6.9",
+    "unicode-substring": "^0.1.0"
   },
   "devDependencies": {
     "@types/jest": "^27.0.3",
-    "@types/node": "^16.18.12",
+    "@types/node": "^16.18.14",
-    "@typescript-eslint/parser": "^5.53.0",
+    "@typescript-eslint/parser": "^5.54.0",
     "@vercel/ncc": "^0.36.1",
     "eslint": "^8.35.0",
     "eslint-plugin-github": "^4.6.1",

src/readme-helper.ts

@@ -1,5 +1,6 @@
 import * as core from '@actions/core'
 import * as fs from 'fs'
+import unicodeSubstring = require('unicode-substring')
 
 export const README_FILEPATH_DEFAULT = './README.md'
 export const IMAGE_EXTENSIONS_DEFAULT = 'bmp,gif,jpg,jpeg,png,svg,webp'
@@ -10,6 +11,8 @@ const REPOSITORY_URL = `${process.env['GITHUB_SERVER_URL']}/${process.env['GITHU
 const BLOB_PREFIX = `${REPOSITORY_URL}/blob/${process.env['GITHUB_REF_NAME']}/`
 const RAW_PREFIX = `${REPOSITORY_URL}/raw/${process.env['GITHUB_REF_NAME']}/`
 
+const MAX_BYTES = 25000
+
 type Rule = {
   /**
    * all left of the relative url belonging to the markdown image/link
@@ -25,6 +28,14 @@ type Rule = {
   absUrlPrefix: string
 }
 
+export function truncateToBytes(s: string, n: number): string {
+  let len = n
+  while (Buffer.byteLength(s) > n) {
+    s = unicodeSubstring(s, 0, len--)
+  }
+  return s
+}
+
 export async function getReadmeContent(
   readmeFilepath: string,
   enableUrlCompletion: boolean,
@@ -42,7 +53,14 @@ export async function getReadmeContent(
     imageExtensions
   )
 
-  return readmeContent
+  const truncatedReadmeContent = truncateToBytes(readmeContent, MAX_BYTES)
+  if (truncatedReadmeContent.length !== readmeContent.length) {
+    core.warning(
+      `The README content exceeds DockerHub's limit and has been truncated to ${MAX_BYTES} bytes.`
+    )
+  }
+
+  return truncatedReadmeContent
 }
 
 export function completeRelativeUrls(