@polotno/pdf-export 0.1.27 → 0.1.29

package/lib/text.d.ts

@@ -36,10 +36,14 @@ export interface TextSegment {
  * while preserving inline formatting tags.
  */
 declare function normalizeRichText(text: string): string;
+/**
+ * Parse HTML text into styled segments
+ */
+declare function parseHTMLToSegments(html: string, baseElement: TextElement): TextSegment[];
 export declare function getGoogleFontPath(fontFamily: string, fontWeight?: string, italic?: boolean): Promise<string>;
 export declare function loadFontIfNeeded(doc: any, element: TextElement, fonts: Record<string, boolean>): Promise<string>;
 /**
  * Main text rendering function
  */
 export declare function renderText(doc: PDFKit.PDFDocument, element: TextElement, fonts: Record<string, boolean>, attrs?: RenderAttrs): Promise<void>;
-export { normalizeRichText as __normalizeRichTextForTests };
+export { normalizeRichText as __normalizeRichTextForTests, parseHTMLToSegments as __parseHTMLToSegmentsForTests, };

package/lib/text.js

@@ -2,6 +2,14 @@ import { parseColor, srcToBuffer } from './utils.js';
 import getUrls from 'get-urls';
 import fetch from 'node-fetch';
 import { stripHtml } from 'string-strip-html';
+import { decode as decodeEntities } from 'html-entities';
+function decodeHtmlEntities(text) {
+    if (!text) {
+        return text;
+    }
+    const decoded = decodeEntities(text);
+    return decoded.replace(/\t/g, ' ');
+}
 /**
  * Check if text contains HTML tags
  */
@@ -18,6 +26,8 @@ function normalizeRichText(text) {
         return text;
     }
     let normalized = text.replace(/\r\n/g, '\n').replace(/\r/g, '\n');
+    // Normalize tab characters into 8 spaces
+    normalized = normalized.replace(/\t/g, ' '.repeat(8));
     // Convert explicit HTML break tags into newline characters
     normalized = normalized.replace(/<br\s*\/?>/gi, '\n');
     // Treat paragraph boundaries as newlines and drop opening tags
@@ -29,6 +39,8 @@ function normalizeRichText(text) {
     normalized = normalized.replace(/^\n+/, '').replace(/\n+$/, '');
     // Decode common HTML non-breaking space entities into their unicode counterpart
     normalized = normalized.replace(/&(nbsp|#160|#xA0);/gi, '\u00A0');
+    // Strip zero-width characters that can create missing-glyph boxes in PDF output
+    normalized = normalized.replace(/[\u200B\u200C\u200D\uFEFF\u2060]/g, '');
     return normalized;
 }
 /**
@@ -43,7 +55,7 @@ function parseHTMLToSegments(html, baseElement) {
     while ((match = regex.exec(html)) !== null) {
         if (match[4]) {
             // Text content
-            const text = match[4];
+            const text = decodeHtmlEntities(match[4]);
             // Calculate current styles from tag stack
             let bold = false;
             let italic = false;
@@ -177,9 +189,11 @@ function tokenizeHTML(html) {
     while ((match = regex.exec(html)) !== null) {
         if (match[4]) {
             // Text content
+            const decodedContent = decodeHtmlEntities(match[4]);
             tokens.push({
                 type: 'text',
                 content: match[4],
+                decodedContent,
             });
         }
         else {
@@ -539,7 +553,7 @@ function splitTextIntoLines(doc, element, props) {
         // Extract plain text for width calculation
         const plainText = tokens
             .filter((t) => t.type === 'text')
-            .map((t) => t.content)
+            .map((t) => t.decodedContent ?? decodeHtmlEntities(t.content))
             .join('');
         const baseMeta = paragraph.listMeta
             ? createListLineMeta(doc, element, props, paragraph.listMeta)
@@ -563,7 +577,7 @@ function splitTextIntoLines(doc, element, props) {
         }
         else {
             // Need to split paragraph into multiple lines
-            let currentLine = '';
+            let currentLineDecoded = '';
             let currentWidth = 0;
             let currentTokens = [];
             let openTags = [];
@@ -573,34 +587,32 @@ function splitTextIntoLines(doc, element, props) {
                     continue;
                 }
                 // Text token - split by words
-                const textWords = token.content.split(' ');
-                for (let i = 0; i < textWords.length; i++) {
-                    const word = textWords[i];
-                    const testLine = currentLine
-                        ? `${currentLine}${i > 0 ? ' ' : ''}${word}`
-                        : word;
-                    const testWidth = doc.widthOfString(testLine, props);
+                const rawWords = token.content.split(' ');
+                const decodedWords = (token.decodedContent ?? decodeHtmlEntities(token.content)).split(' ');
+                for (let i = 0; i < rawWords.length; i++) {
+                    const rawWord = rawWords[i];
+                    const decodedWord = decodedWords[i] ?? decodeHtmlEntities(rawWord);
+                    const separator = i > 0 ? ' ' : '';
+                    const hasCurrentLine = currentLineDecoded.length > 0;
+                    const testLineDecoded = hasCurrentLine
+                        ? `${currentLineDecoded}${separator}${decodedWord}`
+                        : decodedWord;
+                    const testWidth = doc.widthOfString(testLineDecoded, props);
                     if (testWidth <= availableWidth) {
-                        currentLine = testLine;
+                        currentLineDecoded = testLineDecoded;
                         currentWidth = testWidth;
                         // Add text token (with space if not first word in token)
-                        if (i > 0 || currentTokens.length > 0) {
-                            let content = (i > 0 ? ' ' : '') + word;
-                            currentTokens.push({
-                                type: 'text',
-                                content: content,
-                            });
-                        }
-                        else {
-                            currentTokens.push({
-                                type: 'text',
-                                content: word,
-                            });
-                        }
+                        const rawContent = separator.length > 0 ? `${separator}${rawWord}` : rawWord;
+                        const decodedContent = separator.length > 0 ? `${separator}${decodedWord}` : decodedWord;
+                        currentTokens.push({
+                            type: 'text',
+                            content: rawContent,
+                            decodedContent,
+                        });
                     }
                     else {
                         // Line is too long, save current line and start new one
-                        if (currentLine) {
+                        if (currentLineDecoded.length > 0) {
                             const result = tokensToHTML(currentTokens, openTags);
                             const listMeta = cloneListMetaForLine(baseMeta, showMarkerForLine);
                             lines.push({
@@ -613,17 +625,18 @@ function splitTextIntoLines(doc, element, props) {
                             currentTokens = [];
                             showMarkerForLine = false;
                         }
-                        currentLine = word;
-                        currentWidth = doc.widthOfString(word, props);
+                        currentLineDecoded = decodedWord;
+                        currentWidth = doc.widthOfString(decodedWord, props);
                         currentTokens.push({
                             type: 'text',
-                            content: word,
+                            content: rawWord,
+                            decodedContent: decodedWord,
                         });
                     }
                 }
             }
             // Add the last line
-            if (currentLine) {
+            if (currentLineDecoded.length > 0) {
                 const result = tokensToHTML(currentTokens, openTags);
                 const listMeta = cloneListMetaForLine(baseMeta, showMarkerForLine);
                 lines.push({
@@ -1034,4 +1047,4 @@ export async function renderText(doc, element, fonts, attrs = {}) {
     }
 }
 // Internal exports for testing
-export { normalizeRichText as __normalizeRichTextForTests };
+export { normalizeRichText as __normalizeRichTextForTests, parseHTMLToSegments as __parseHTMLToSegmentsForTests, };

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@polotno/pdf-export",
-  "version": "0.1.27",
+  "version": "0.1.29",
   "description": "Convert Polotno JSON into vector PDF",
   "type": "module",
   "main": "./lib/index.js",