LibPDF+PDFViewer+MacPDF: Don't draw hidden text by default

Text can be rendered in various ways in PDFs: Filled, stroked, both filled and stroked, set as clipping path, hidden, or some combinations thereof. We don't implement any of this at the moment except "filled". Hidden text is used in scanned documents: The image of the scan is drawn in the background, and then OCRd text is "drawn" as hidden on top of the scanned bitmap. That way, the (hidden) text can be selected and copied, and it looks like you're selecting text from the scanned bitmap. Find-in-page also works similarly. (We currently have neither text selection nor find-in-page, but one day we will.) Now that we have pretty good support for CCITT and are growing some support for JBIG2, we now draw both the scanned background image as well as the foreground text. They're not always perfectly aligned. This change makes it so that we don't render text that's marked as hidden. (We still do most of the coordinate math, which will probably come in handy at some point when we implement text selection.) This makes these scanned documents appear as they're supposed to appear (at least in documents where we manage to decode the background bitmap). This also adds a debug option to force rendering of hidden text.
nico · Mar 16, 2024 · 7fd79e2 · 7fd79e2
1 parent ead7428
commit 7fd79e2
Show file tree

Hide file tree

Showing 11 changed files with 51 additions and 4 deletions.
diff --git a/Meta/Lagom/Contrib/MacPDF/MacPDFView.h b/Meta/Lagom/Contrib/MacPDF/MacPDFView.h
@@ -31,5 +31,6 @@
 - (IBAction)toggleClipPaths:(id)sender;
 - (IBAction)toggleClipText:(id)sender;
 - (IBAction)toggleShowImages:(id)sender;
+- (IBAction)toggleShowHiddenText:(id)sender;
 
 @end
diff --git a/Meta/Lagom/Contrib/MacPDF/MacPDFView.mm b/Meta/Lagom/Contrib/MacPDF/MacPDFView.mm
@@ -185,6 +185,10 @@ - (BOOL)validateMenuItem:(NSMenuItem*)item
         [item setState:_preferences.show_images ? NSControlStateValueOn : NSControlStateValueOff];
         return _doc ? YES : NO;
     }
+    if ([item action] == @selector(toggleShowHiddenText:)) {
+        [item setState:_preferences.show_hidden_text ? NSControlStateValueOn : NSControlStateValueOff];
+        return _doc ? YES : NO;
+    }
     return NO;
 }
 
@@ -228,6 +232,14 @@ - (IBAction)toggleShowImages:(id)sender
     }
 }
 
+- (IBAction)toggleShowHiddenText:(id)sender
+{
+    if (_doc) {
+        _preferences.show_hidden_text = !_preferences.show_hidden_text;
+        [self invalidateCachedBitmap];
+    }
+}
+
 - (void)keyDown:(NSEvent*)event
 {
     // Calls moveLeft: or moveRight: below.

diff --git a/Meta/Lagom/Contrib/MacPDF/MacPDFWindowController.h b/Meta/Lagom/Contrib/MacPDF/MacPDFWindowController.h
@@ -26,6 +26,7 @@ NS_ASSUME_NONNULL_BEGIN
 - (IBAction)toggleClipPaths:(id)sender;
 - (IBAction)toggleClipText:(id)sender;
 - (IBAction)toggleShowImages:(id)sender;
+- (IBAction)toggleShowHiddenText:(id)sender;
 - (IBAction)showGoToPageDialog:(id)sender;
 
 - (void)pdfDidInitialize;

diff --git a/Meta/Lagom/Contrib/MacPDF/MacPDFWindowController.mm b/Meta/Lagom/Contrib/MacPDF/MacPDFWindowController.mm
@@ -164,6 +164,11 @@ - (IBAction)toggleShowImages:(id)sender
     [_pdfView toggleShowImages:sender];
 }
 
+- (IBAction)toggleShowHiddenText:(id)sender
+{
+    [_pdfView toggleShowHiddenText:sender];
+}
+
 - (IBAction)showGoToPageDialog:(id)sender
 {
     auto alert = [[NSAlert alloc] init];

diff --git a/Meta/Lagom/Contrib/MacPDF/MainMenu.xib b/Meta/Lagom/Contrib/MacPDF/MainMenu.xib
@@ -694,6 +694,12 @@
                                     <action selector="toggleShowImages:" target="-1" id="mNE-9J-Nle"/>
                                 </connections>
                             </menuItem>
+                            <menuItem title="Show Hidden Text" id="PhM-XC-ExK">
+                                <modifierMask key="keyEquivalentModifierMask"/>
+                                <connections>
+                                    <action selector="toggleShowHiddenText:" target="-1" id="7iT-L2-Jd1"/>
+                                </connections>
+                            </menuItem>
                         </items>
                     </menu>
                 </menuItem>

diff --git a/Userland/Applications/PDFViewer/PDFViewer.cpp b/Userland/Applications/PDFViewer/PDFViewer.cpp
@@ -48,6 +48,7 @@ PDFViewer::PDFViewer()
     m_page_view_mode = static_cast<PageViewMode>(Config::read_i32("PDFViewer"sv, "Display"sv, "PageMode"sv, 0));
     m_rendering_preferences.show_clipping_paths = Config::read_bool("PDFViewer"sv, "Rendering"sv, "ShowClippingPaths"sv, false);
     m_rendering_preferences.show_images = Config::read_bool("PDFViewer"sv, "Rendering"sv, "ShowImages"sv, true);
+    m_rendering_preferences.show_hidden_text = Config::read_bool("PDFViewer"sv, "Rendering"sv, "ShowHiddenText"sv, false);
     m_rendering_preferences.show_diagnostics = Config::read_bool("PDFViewer"sv, "Rendering"sv, "ShowDiagnostics"sv, false);
     m_rendering_preferences.clip_images = Config::read_bool("PDFViewer"sv, "Rendering"sv, "ClipImages"sv, true);
     m_rendering_preferences.clip_paths = Config::read_bool("PDFViewer"sv, "Rendering"sv, "ClipPaths"sv, true);
@@ -192,6 +193,13 @@ void PDFViewer::set_show_images(bool show_images)
     update();
 }
 
+void PDFViewer::set_show_hidden_text(bool show_hidden_text)
+{
+    m_rendering_preferences.show_hidden_text = show_hidden_text;
+    Config::write_bool("PDFViewer"sv, "Rendering"sv, "ShowHiddenText"sv, show_hidden_text);
+    update();
+}
+
 void PDFViewer::set_clip_images(bool clip_images)
 {
     m_rendering_preferences.clip_images = clip_images;

diff --git a/Userland/Applications/PDFViewer/PDFViewer.h b/Userland/Applications/PDFViewer/PDFViewer.h
@@ -68,6 +68,8 @@ class PDFViewer : public GUI::AbstractScrollableWidget {
     void set_show_clipping_paths(bool);
     bool show_images() const { return m_rendering_preferences.show_images; }
     void set_show_images(bool);
+    bool show_hidden_text() const { return m_rendering_preferences.show_hidden_text; }
+    void set_show_hidden_text(bool);
     bool clip_images() const { return m_rendering_preferences.clip_images; }
     void set_clip_images(bool);
     bool clip_paths() const { return m_rendering_preferences.clip_paths; }

diff --git a/Userland/Applications/PDFViewer/PDFViewerWidget.cpp b/Userland/Applications/PDFViewer/PDFViewerWidget.cpp
@@ -271,6 +271,11 @@ ErrorOr<void> PDFViewerWidget::initialize_menubar(GUI::Window& window)
     });
     toggle_show_images->set_checked(m_viewer->show_images());
     debug_menu->add_action(toggle_show_images);
+    auto toggle_show_hidden_text = GUI::Action::create_checkable("Show &Hidden Text", [&](auto& action) {
+        m_viewer->set_show_hidden_text(action.is_checked());
+    });
+    toggle_show_hidden_text->set_checked(m_viewer->show_hidden_text());
+    debug_menu->add_action(toggle_show_hidden_text);
     auto toggle_clip_images = GUI::Action::create_checkable("Clip I&mages", [&](auto& action) {
         m_viewer->set_clip_images(action.is_checked());
     });

diff --git a/Userland/Libraries/LibPDF/Fonts/SimpleFont.cpp b/Userland/Libraries/LibPDF/Fonts/SimpleFont.cpp
@@ -71,8 +71,10 @@ PDFErrorOr<Gfx::FloatPoint> SimpleFont::draw_string(Gfx::Painter& painter, Gfx::
         else
             glyph_width = font_size * m_missing_width * m_font_matrix.x_scale();
 
-        Gfx::FloatPoint glyph_render_position = text_rendering_matrix.map(glyph_position);
-        TRY(draw_glyph(painter, glyph_render_position, glyph_width, char_code, renderer));
+        if (renderer.text_state().rendering_mode != TextRenderingMode::Invisible || renderer.show_hidden_text()) {
+            Gfx::FloatPoint glyph_render_position = text_rendering_matrix.map(glyph_position);
+            TRY(draw_glyph(painter, glyph_render_position, glyph_width, char_code, renderer));
+        }
 
         // glyph_width is scaled by `text_rendering_matrix.x_scale() * renderer.text_state().font_size / horizontal_scaling`,
         // but it should only be scaled by `renderer.text_state().font_size`.

diff --git a/Userland/Libraries/LibPDF/Fonts/Type0Font.cpp b/Userland/Libraries/LibPDF/Fonts/Type0Font.cpp
@@ -478,8 +478,10 @@ PDFErrorOr<Gfx::FloatPoint> Type0Font::draw_string(Gfx::Painter& painter, Gfx::F
             }
         }
 
-        Gfx::FloatPoint glyph_render_position = text_rendering_matrix.map(glyph_position - Gfx::FloatPoint { position_vector_x, position_vector_y });
-        TRY(m_cid_font_type->draw_glyph(painter, glyph_render_position, glyph_width, cid, renderer));
+        if (renderer.text_state().rendering_mode != TextRenderingMode::Invisible || renderer.show_hidden_text()) {
+            Gfx::FloatPoint glyph_render_position = text_rendering_matrix.map(glyph_position - Gfx::FloatPoint { position_vector_x, position_vector_y });
+            TRY(m_cid_font_type->draw_glyph(painter, glyph_render_position, glyph_width, cid, renderer));
+        }
 
         // glyph_width is scaled by `text_rendering_matrix.x_scale() * renderer.text_state().font_size / horizontal_scaling`,
         // but it should only be scaled by `renderer.text_state().font_size`.

diff --git a/Userland/Libraries/LibPDF/Renderer.h b/Userland/Libraries/LibPDF/Renderer.h
@@ -89,6 +89,7 @@ struct GraphicsState {
 struct RenderingPreferences {
     bool show_clipping_paths { false };
     bool show_images { true };
+    bool show_hidden_text { false };
     bool show_diagnostics { false };
 
     bool clip_images { true };
@@ -121,6 +122,8 @@ class Renderer {
 
     PDFErrorOr<void> render_type3_glyph(Gfx::FloatPoint, StreamObject const&, Gfx::AffineTransform const&, Optional<NonnullRefPtr<DictObject>>);
 
+    bool show_hidden_text() const { return m_rendering_preferences.show_hidden_text; }
+
 private:
     Renderer(RefPtr<Document>, Page const&, RefPtr<Gfx::Bitmap>, Color background_color, RenderingPreferences);