[pull] master from foliojs:master

CHANGELOG.md

@@ -2,6 +2,10 @@
 
 ### Unreleased
 
+### [v0.17.2] - 2025-08-30
+
+- Fix rendering lists that spans across pages
+
 ### [v0.17.1] - 2025-05-02
 
 - Fix null values in table cells rendering as `[object Object]`

jsconfig.json

@@ -0,0 +1,13 @@
+{
+  "compilerOptions": {
+    "target": "ES2020",
+    "module": "CommonJS",
+    "allowJs": true,
+    // Don't force external @types; our ambient file is included below
+    "baseUrl": "."
+  },
+  "include": [
+    "tests/**/*.js",
+    "types/**/*.d.ts"
+  ]
+}

lib/line_wrapper.js

@@ -285,6 +285,7 @@ class LineWrapper extends EventEmitter {
         // if we've reached the edge of the page,
         // continue on a new page or column
         if (PDFNumber(this.document.y + lh) > this.maxY) {
+          this.emit('sectionEnd', options, this);
           const shouldContinue = this.nextSection();
 
           // stop if we reached the maximum height
@@ -293,6 +294,7 @@ class LineWrapper extends EventEmitter {
             buffer = '';
             return false;
           }
+          this.emit('sectionStart', options, this);
         }
 
         // reset the space left and buffer
@@ -335,8 +337,6 @@ class LineWrapper extends EventEmitter {
   }
 
   nextSection(options) {
-    this.emit('sectionEnd', options, this);
-
     if (++this.column > this.columns) {
       // if a max height was specified by the user, we're done.
       // otherwise, the default is to make a new page at the bottom.
@@ -359,7 +359,6 @@ class LineWrapper extends EventEmitter {
       this.emit('columnBreak', options, this);
     }
 
-    this.emit('sectionStart', options, this);
     return true;
   }
 }

lib/mixins/text.js

@@ -4,6 +4,24 @@ import { cosine, sine } from '../utils';
 
 const { number } = PDFObject;
 
+/**
+ * Format a list label based on the list type
+ * @param {number} n
+ * @param {'numbered' | 'lettered'} listType
+ * @returns {string}
+ */
+function formatListLabel(n, listType) {
+  if (listType === 'numbered') {
+    return `${n}.`;
+  }
+
+  // lettered
+  var letter = String.fromCharCode(((n - 1) % 26) + 65);
+  var times = Math.floor((n - 1) / 26 + 1);
+  var text = Array(times + 1).join(letter);
+  return `${text}.`;
+}
+
 export default {
   initText() {
     this._line = this._line.bind(this);
@@ -252,7 +270,7 @@ export default {
     return height;
   },
 
-  list(list, x, y, options, wrapper) {
+  list(list, x, y, options) {
     options = this._initOptions(x, y, options);
 
     const listType = options.listType || 'bullet';
@@ -289,20 +307,8 @@ export default {
 
     flatten(list);
 
-    const label = function (n) {
-      switch (listType) {
-        case 'numbered':
-          return `${n}.`;
-        case 'lettered':
-          var letter = String.fromCharCode(((n - 1) % 26) + 65);
-          var times = Math.floor((n - 1) / 26 + 1);
-          var text = Array(times + 1).join(letter);
-          return `${text}.`;
-      }
-    };
-
     const drawListItem = function (listItem, i) {
-      wrapper = new LineWrapper(this, options);
+      const wrapper = new LineWrapper(this, options);
       wrapper.on('line', this._line);
 
       level = 1;
@@ -345,7 +351,7 @@ export default {
             break;
           case 'numbered':
           case 'lettered':
-            var text = label(numbers[i - 1]);
+            var text = formatListLabel(numbers[i - 1], listType);
             this._fragment(text, this.x - indent, this.y, options);
             break;
         }
@@ -435,12 +441,12 @@ export default {
 
   _line(text, options = {}, wrapper) {
     this._fragment(text, this.x, this.y, options);
-    const lineGap = options.lineGap || this._lineGap || 0;
 
-    if (!wrapper) {
-      this.x += this.widthOfString(text, options);
-    } else {
+    if (wrapper) {
+      const lineGap = options.lineGap || this._lineGap || 0;
       this.y += this.currentLineHeight(true) + lineGap;
+    } else {
+      this.x += this.widthOfString(text, options);
     }
   },
 

package.json

@@ -9,7 +9,7 @@
     "document",
     "vector"
   ],
-  "version": "0.17.1",
+  "version": "0.17.2",
   "homepage": "http://pdfkit.org/",
   "author": {
     "name": "Devon Govett",
@@ -39,7 +39,6 @@
     "gh-pages": "^6.2.0",
     "globals": "^15.14.0",
     "jest": "^29.7.0",
-    "jest-environment-jsdom": "^29.7.0",
     "jest-image-snapshot": "^6.4.0",
     "markdown": "~0.5.0",
     "pdfjs-dist": "^2.14.305",
@@ -66,9 +65,9 @@
     "docs": "npm run pdf-guide && npm run website && npm run browserify-example",
     "lint": "eslint {lib,tests}/**/*.js",
     "prettier": "prettier lib tests docs",
-    "test": "jest -i --env=node",
-    "test:visual": "jest visual/ -i --env=node",
-    "test:unit": "jest unit/ --env=node"
+    "test": "jest",
+    "test:visual": "jest visual/",
+    "test:unit": "jest unit/"
   },
   "main": "js/pdfkit.js",
   "module": "js/pdfkit.es.js",
@@ -81,7 +80,7 @@
     "node >= v18.0.0"
   ],
   "jest": {
-    "testEnvironment": "jest-environment-jsdom",
+    "testEnvironment": "node",
     "testPathIgnorePatterns": [
       "/node_modules/",
       "<rootDir>/examples/"
@@ -90,4 +89,4 @@
       "<rootDir>/tests/unit/setupTests.js"
     ]
   }
-}
+}

tests/unit/helpers.js

@@ -7,6 +7,8 @@
  * @property {string} text
  * @property {string} font
  * @property {number} fontSize
+ * @property {number} x
+ * @property {number} y
  *
  * @typedef {string | Buffer} PDFDataItem
  * @typedef {Array<PDFDataItem>} PDFData
@@ -68,54 +70,99 @@ function getObjects(data) {
 }
 
 /**
- * @param {Buffer} textStream
- * @return {TextStream | undefined}
+ * Parse all text objects (multiple TJ) in a decoded stream.
+ * @param {string} decodedStream
+ * @return {TextStream[]}
  */
-function parseTextStream(textStream) {
-  const decodedStream = textStream.toString('utf8');
+function parseTextStreams(decodedStream) {
+  const tjRegex = /\[([^\]]+)\]\s+TJ/g;
+  const fontRegex = /\/([A-Za-z0-9]+)\s+(\d+(?:\.\d+)?)\s+Tf/g;
+  const tmRegex =
+    /([+-]?\d+(?:\.\d+)?)\s+([+-]?\d+(?:\.\d+)?)\s+([+-]?\d+(?:\.\d+)?)\s+([+-]?\d+(?:\.\d+)?)\s+([+-]?\d+(?:\.\d+)?)\s+([+-]?\d+(?:\.\d+)?)\s+Tm/g;
+  const cmRegex =
+    /([+-]?\d+(?:\.\d+)?)\s+([+-]?\d+(?:\.\d+)?)\s+([+-]?\d+(?:\.\d+)?)\s+([+-]?\d+(?:\.\d+)?)\s+([+-]?\d+(?:\.\d+)?)\s+([+-]?\d+(?:\.\d+)?)\s+cm/g;
+  /** @type {TextStream[]} */
+  const results = [];
+  let tjMatch;
 
-  // Extract font and font size
-  const fontMatch = decodedStream.match(/\/([A-Za-z0-9]+)\s+(\d+)\s+Tf/);
-
-  if (!fontMatch) {
-    return undefined;
-  }
-
-  const font = fontMatch[1];
-  const fontSize = parseInt(fontMatch[2], 10);
+  while ((tjMatch = tjRegex.exec(decodedStream)) !== null) {
+    const tjIndex = tjMatch.index;
+    let fMatch;
+    let lastFontName;
+    let lastFontSize;
+    fontRegex.lastIndex = 0;
+    while (
+      (fMatch = fontRegex.exec(decodedStream)) !== null &&
+      fMatch.index < tjIndex
+    ) {
+      lastFontName = fMatch[1];
+      lastFontSize = parseFloat(fMatch[2]);
+    }
+    if (!lastFontName || !lastFontSize) continue;
 
-  // Extract hex strings inside TJ array
-  const tjMatch = decodedStream.match(/\[([^\]]+)\]\s+TJ/);
-  if (!tjMatch) {
-    return undefined;
-  }
-  let text = '';
+    // Find the nearest preceding text matrix (Tm) and current transformation (cm)
+    let tmMatch;
+    let lastTm = undefined;
+    tmRegex.lastIndex = 0;
+    while (
+      (tmMatch = tmRegex.exec(decodedStream)) !== null &&
+      tmMatch.index < tjIndex
+    ) {
+      lastTm = tmMatch;
+    }
+    // Default to origin if no Tm found
+    let tx = 0;
+    let ty = 0;
+    if (lastTm) {
+      tx = parseFloat(lastTm[5]);
+      ty = parseFloat(lastTm[6]);
+    }
 
-  // this is a simplified version
-  // the correct way is to retrieve the encoding from /Resources /Font dictionary and decode using it
-  // https://stackoverflow.com/a/29468049/5724645
+    // Find the nearest preceding cm (CTM)
+    let cmMatch;
+    let lastCm = undefined;
+    cmRegex.lastIndex = 0;
+    while (
+      (cmMatch = cmRegex.exec(decodedStream)) !== null &&
+      cmMatch.index < tjIndex
+    ) {
+      lastCm = cmMatch;
+    }
+    // Apply transform: [a b c d e f] to point (tx, ty)
+    let x = tx;
+    let y = ty;
+    if (lastCm) {
+      const a = parseFloat(lastCm[1]);
+      const b = parseFloat(lastCm[2]);
+      const c = parseFloat(lastCm[3]);
+      const d = parseFloat(lastCm[4]);
+      const e = parseFloat(lastCm[5]);
+      const f = parseFloat(lastCm[6]);
+      x = a * tx + c * ty + e;
+      y = b * tx + d * ty + f;
+    }
 
-  // Match all hex strings like <...>
-  const hexMatches = [...tjMatch[1].matchAll(/<([0-9a-fA-F]+)>/g)];
-  for (const m of hexMatches) {
-    // Convert hex to string
-    const hex = m[1];
-    for (let i = 0; i < hex.length; i += 2) {
-      const code = parseInt(hex.substr(i, 2), 16);
-      let char = String.fromCharCode(code);
-      // Handle special cases
-      if (code === 0x0a) {
-        char = '\n'; // Newline
-      } else if (code === 0x0d) {
-        char = '\r'; // Carriage return
-      } else if (code === 0x85) {
-        char = '...';
+    const arrayContent = tjMatch[1];
+    let text = '';
+    const hexMatches = [...arrayContent.matchAll(/<([0-9a-fA-F]+)>/g)];
+    for (const m of hexMatches) {
+      const hex = m[1];
+      for (let i = 0; i < hex.length; i += 2) {
+        // this is a simplified version
+        // the correct way is to retrieve the encoding from /Resources /Font dictionary and decode using it
+        // https://stackoverflow.com/a/29468049/5724645
+        const code = parseInt(hex.substring(i, i + 2), 16);
+        let char = String.fromCharCode(code);
+        if (code === 0x0a) char = '\n';
+        else if (code === 0x0d) char = '\r';
+        else if (code === 0x85) char = '...';
+        text += char;
       }
-      text += char;
     }
+    results.push({ text, font: lastFontName, fontSize: lastFontSize, x, y });
   }
 
-  return { text, font, fontSize };
+  return results;
 }
 
-export { logData, joinTokens, parseTextStream, getObjects };
+export { logData, joinTokens, parseTextStreams, getObjects };