PDF Output: Add an option to break long words at the ends of lines.

Fixes #1773111 [add hyphenation to ePub-to-PDF conversion](https://bugs.launchpad.net/calibre/+bug/1773111)
2025-07-08 10:44:09 -04:00 · 2018-05-24 14:02:32 +05:30 · 2018-05-24 14:02:32 +05:30 · 4fc540bdaa
commit 4fc540bdaa
parent c6c689f9e3
4 changed files with 55 additions and 20 deletions
--- a/src/calibre/ebooks/conversion/plugins/pdf_output.py
+++ b/src/calibre/ebooks/conversion/plugins/pdf_output.py
@ -94,6 +94,8 @@ class PDFOutput(OutputFormatPlugin):
        OptionRecommendation(name='pdf_mono_font_size',
            recommended_value=16, help=_(
                'The default font size for monospaced text')),
+        OptionRecommendation(name='pdf_hyphenate', recommended_value=False,
+            help=_('Break long words at the end of lines. This can give the text at the right margin a more even appearance.')),
        OptionRecommendation(name='pdf_mark_links', recommended_value=False,
            help=_('Surround all links with a red box, useful for debugging.')),
        OptionRecommendation(name='uncompressed_pdf',
@ -184,8 +186,7 @@ class PDFOutput(OutputFormatPlugin):

    def get_cover_data(self):
        oeb = self.oeb
-        if (oeb.metadata.cover and
-                unicode(oeb.metadata.cover[0]) in oeb.manifest.ids):
+        if (oeb.metadata.cover and unicode(oeb.metadata.cover[0]) in oeb.manifest.ids):
            cover_id = unicode(oeb.metadata.cover[0])
            item = oeb.manifest.ids[cover_id]
            self.cover_data = item.data
--- a/src/calibre/ebooks/pdf/render/from_html.py
+++ b/src/calibre/ebooks/pdf/render/from_html.py
@ -24,6 +24,7 @@ from calibre.ebooks.pdf.render.common import (inch, cm, mm, pica, cicero,
                                              didot, PAPER_SIZES, current_log)
 from calibre.ebooks.pdf.render.engine import PdfDevice
 from calibre.ptempfile import PersistentTemporaryFile
+from calibre.utils.resources import load_hyphenator_dicts


 def get_page_size(opts, for_comic=False):  # {{{
@ -213,6 +214,7 @@ class PDFWriter(QObject):
        self.margin_top, self.margin_bottom = map(lambda x:int(floor(x)), (mt, mb))

        self.painter = QPainter(self.doc)
+        self.book_language = pdf_metadata.mi.languages[0]
        self.doc.set_metadata(title=pdf_metadata.title,
                              author=pdf_metadata.author,
                              tags=pdf_metadata.tags, mi=pdf_metadata.mi)
@ -344,6 +346,25 @@ class PDFWriter(QObject):

        return sections

+    def hyphenate(self, evaljs):
+        evaljs(u'''\
+        Hyphenator.config(
+            {
+            'minwordlength'    : 6,
+            // 'hyphenchar'     : '|',
+            'displaytogglebox' : false,
+            'remoteloading'    : false,
+            'doframes'         : true,
+            'defaultlanguage'  : 'en',
+            'storagetype'      : 'session',
+            'onerrorhandler'   : function (e) {
+                                    console.log(e);
+                                }
+            });
+        Hyphenator.hyphenate(document.body, "%s");
+        ''' % self.hyphenate_lang
+        )
+
    def do_paged_render(self):
        if self.paged_js is None:
            import uuid
@ -352,6 +373,10 @@ class PDFWriter(QObject):
            self.paged_js += cc('ebooks.oeb.display.indexing')
            self.paged_js += cc('ebooks.oeb.display.paged')
            self.paged_js += cc('ebooks.oeb.display.mathjax')
+            if self.opts.pdf_hyphenate:
+                self.paged_js += P('viewer/hyphenate/Hyphenator.js', data=True).decode('utf-8')
+                hjs, self.hyphenate_lang = load_hyphenator_dicts({}, self.book_language)
+                self.paged_js += hjs
            self.hf_uuid = str(uuid.uuid4()).replace('-', '')

        self.view.page().mainFrame().addToJavaScriptWindowObject("py_bridge", self)
@ -359,6 +384,8 @@ class PDFWriter(QObject):
        evaljs = self.view.page().mainFrame().evaluateJavaScript
        evaljs(self.paged_js)
        self.load_mathjax()
+        if self.opts.pdf_hyphenate:
+            self.hyphenate(evaljs)

        amap = json.loads(evaljs('''
        document.body.style.backgroundColor = "white";
--- a/src/calibre/gui2/convert/pdf_output.py
+++ b/src/calibre/gui2/convert/pdf_output.py
@ -24,7 +24,7 @@ class PluginWidget(Widget, Ui_Form):

    def __init__(self, parent, get_option, get_help, db=None, book_id=None):
        Widget.__init__(self, parent, [
-            'use_profile_size', 'paper_size', 'custom_size',
+            'use_profile_size', 'paper_size', 'custom_size', 'pdf_hyphenate',
            'preserve_cover_aspect_ratio', 'pdf_serif_family', 'unit',
            'pdf_sans_family', 'pdf_mono_family', 'pdf_standard_font',
            'pdf_default_font_size', 'pdf_mono_font_size', 'pdf_page_numbers',
--- a/src/calibre/gui2/convert/pdf_output.ui
+++ b/src/calibre/gui2/convert/pdf_output.ui
@ -91,14 +91,14 @@
     </property>
    </widget>
   </item>
-   <item row="6" column="0" colspan="2">
+   <item row="7" column="0" colspan="2">
    <widget class="QCheckBox" name="opt_pdf_add_toc">
     <property name="text">
      <string>Add a printable &amp;Table of Contents at the end</string>
     </property>
    </widget>
   </item>
-   <item row="7" column="0">
+   <item row="8" column="0">
    <widget class="QLabel" name="label_14">
     <property name="text">
      <string>&amp;Title for ToC:</string>
@ -108,10 +108,10 @@
     </property>
    </widget>
   </item>
-   <item row="7" column="1">
+   <item row="8" column="1">
    <widget class="QLineEdit" name="opt_toc_title"/>
   </item>
-   <item row="9" column="0">
+   <item row="10" column="0">
    <widget class="QLabel" name="label_4">
     <property name="text">
      <string>Serif famil&amp;y:</string>
@ -121,10 +121,10 @@
     </property>
    </widget>
   </item>
-   <item row="9" column="1">
+   <item row="10" column="1">
    <widget class="QFontComboBox" name="opt_pdf_serif_family"/>
   </item>
-   <item row="10" column="0">
+   <item row="11" column="0">
    <widget class="QLabel" name="label_5">
     <property name="text">
      <string>Sans fami&amp;ly:</string>
@ -134,10 +134,10 @@
     </property>
    </widget>
   </item>
-   <item row="10" column="1">
+   <item row="11" column="1">
    <widget class="QFontComboBox" name="opt_pdf_sans_family"/>
   </item>
-   <item row="11" column="0">
+   <item row="12" column="0">
    <widget class="QLabel" name="label_6">
     <property name="text">
      <string>&amp;Monospace family:</string>
@ -147,10 +147,10 @@
     </property>
    </widget>
   </item>
-   <item row="11" column="1">
+   <item row="12" column="1">
    <widget class="QFontComboBox" name="opt_pdf_mono_family"/>
   </item>
-   <item row="12" column="0">
+   <item row="13" column="0">
    <widget class="QLabel" name="label_7">
     <property name="text">
      <string>S&amp;tandard font:</string>
@ -160,10 +160,10 @@
     </property>
    </widget>
   </item>
-   <item row="12" column="1">
+   <item row="13" column="1">
    <widget class="QComboBox" name="opt_pdf_standard_font"/>
   </item>
-   <item row="13" column="0">
+   <item row="14" column="0">
    <widget class="QLabel" name="label_8">
     <property name="text">
      <string>Default font si&amp;ze:</string>
@ -173,14 +173,14 @@
     </property>
    </widget>
   </item>
-   <item row="13" column="1">
+   <item row="14" column="1">
    <widget class="QSpinBox" name="opt_pdf_default_font_size">
     <property name="suffix">
      <string> px</string>
     </property>
    </widget>
   </item>
-   <item row="14" column="0">
+   <item row="15" column="0">
    <widget class="QLabel" name="label_9">
     <property name="text">
      <string>Monospace &amp;font size:</string>
@ -190,21 +190,21 @@
     </property>
    </widget>
   </item>
-   <item row="14" column="1">
+   <item row="15" column="1">
    <widget class="QSpinBox" name="opt_pdf_mono_font_size">
     <property name="suffix">
      <string> px</string>
     </property>
    </widget>
   </item>
-   <item row="15" column="0" colspan="2">
+   <item row="16" column="0" colspan="2">
    <widget class="QGroupBox" name="page_margins_box">
     <property name="title">
      <string>Page  margins</string>
     </property>
    </widget>
   </item>
-   <item row="16" column="0" colspan="2">
+   <item row="17" column="0" colspan="2">
    <widget class="QGroupBox" name="template_box">
     <property name="title">
      <string>Page headers and footers</string>
@ -252,6 +252,13 @@
     </layout>
    </widget>
   </item>
+   <item row="6" column="0" colspan="2">
+    <widget class="QCheckBox" name="opt_pdf_hyphenate">
+     <property name="text">
+      <string>&amp;Break long words at the end of lines</string>
+     </property>
+    </widget>
+   </item>
  </layout>
 </widget>
 <resources/>