Resolves #11

Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "pdfium-render"
-version = "0.5.4"
+version = "0.5.5"
 edition = "2018"
 publish = true
 description = "A high-level idiomatic Rust wrapper around Pdfium, the C++ PDF library used by the Google Chromium project."

README.md

@@ -4,7 +4,7 @@
 Pdfium exposed by the excellent `pdfium-sys` crate.
 
 ```
-    // Exports each page in the given test PDF file as a separate JPEG image.
+    // Renders each page in the given test PDF file to a separate JPEG file.
 
     use pdfium_render::{pdfium::Pdfium, bitmap::PdfBitmapRotation, bitmap_config::PdfBitmapConfig};
     use image::ImageFormat;
@@ -120,6 +120,9 @@ If you need a function that is not currently exposed, just raise an issue.
 
 ## Version history
 
+* 0.5.5: fixes two bugs in the WASM implementation, one to do with colors, one to do with text extraction.
+  See <https://github.com/ajrcarey/pdfium-render/issues/9> and <https://github.com/ajrcarey/pdfium-render/issues/11>
+  for more information.
 * 0.5.4: changes default setting of `PdfBitmapConfig::set_reverse_byte_order()` to `true` to switch from
   Pdfium's default BGRA8 pixel format to RGBA8. This is necessary since the `image` crate dropped
   support for BGRA8 in version 0.24. See <https://github.com/ajrcarey/pdfium-render/issues/9> for more information.

examples/README.md

@@ -12,7 +12,7 @@ Simple examples demonstrating how to use `pdfium-render` on both native and WASM
 
 Since `pdfium-render` does not include Pdfium itself, an external pre-packaged WASM build of `pdfium` is required. Suitable builds are available from https://github.com/paulo-coutinho/pdfium-lib/releases.
 
-* Build the WASM module for the sample: `cargo install wasm-pack && wasm-pack build examples/ --target no-modules`. This creates a WASM module and supporting Javascript files in `examples/pkg`.  
+* Build the WASM module for the sample: `cargo install wasm-pack && wasm-pack build examples/ --target no-modules`. This creates a WASM module and supporting Javascript files in `examples/pkg`.
 * Copy the `pdfium_render_wasm_example.js` and `pdfium_render_wasm_example_bg.wasm` files from `examples/pkg/` into a release folder.
 * Download a pre-packaged WASM build from https://github.com/paulo-coutinho/pdfium-lib/releases and extract the `release/node/pdfium.js` and `release/node/pdfium.wasm` files into your release folder.
 * Copy the `index.html` file from `examples` into your release folder.

examples/objects.rs

@@ -2,21 +2,23 @@ use pdfium_render::page_object::PdfPageObjectCommon;
 use pdfium_render::pdfium::Pdfium;
 
 pub fn main() {
-    // This example differs from export.rs in that we iterate over every page object in each page,
-    // outputting information about each to the console. Comments that would duplicate
-    // those in export.rs have been removed; only code that substantively differs is commented.
+    // For general comments about pdfium-render and binding to Pdfium, see comments in export.rs.
 
     let bindings = Pdfium::bind_to_library(Pdfium::pdfium_platform_library_name_at_path("./"))
         .or_else(|_| Pdfium::bind_to_system_library());
 
     match bindings {
         Ok(bindings) => {
+            // For every page in our sample file...
+
             Pdfium::new(bindings)
-                .load_pdf_from_file("test/export-test.pdf", None) // Load the sample file...
+                .load_pdf_from_file("test/export-test.pdf", None)
                 .unwrap()
                 .pages()
                 .iter()
                 .for_each(|page| {
+                    // ... output information about every object on the page to the console.
+
                     println!("=============== Page {} ===============", page.index());
 
                     page.objects().iter().for_each(|object| {
@@ -27,6 +29,9 @@ pub fn main() {
                             object.object_type()
                         );
 
+                        // For text objects, we take the extra step of outputting the text
+                        // contained by the object.
+
                         if let Some(object) = object.as_text_object() {
                             println!(
                                 "{} {}-pt: {}",
@@ -35,6 +40,11 @@ pub fn main() {
                                 object.text()
                             );
                         }
+
+                        // If we wanted to extract _all_ the text contained by all the text objects
+                        // on the page, an easier way would be to simply use
+
+                        // page.text().unwrap().all()
                     });
                 });
         }

examples/serve/index.html

@@ -0,0 +1,76 @@
+<!DOCTYPE html>
+<html>
+    <head>
+        <meta charset="utf-8" />
+        <script src="pdfium.js"></script>
+        <script src="pdfium_render_wasm_example.js"></script>
+    </head>
+
+    <body>
+        <canvas id="canvas" style="max-width: 100%; height: auto; border: 1px solid black;"></canvas>
+
+        <script>
+            // Initialize Pdfium's emscripten-wrapped WASM module.
+
+            Module.onRuntimeInitialized = async _ => {
+                // Pdfium's WASM module has now been loaded and initialized. We can now
+                // bind to any functions exported from our Rust application via the
+                // #[wasm_bindgen] declaration. In example/wasm.rs there are two such
+                // functions: log_page_metrics_to_console() and get_image_data_for_page().
+
+                // In addition to any functions exported by our Rust application, pdfium-render will
+                // always export an initialization function, initialize_pdfium_render(),
+                // which must be called from Javascript prior to the use of any Pdfium functionality.
+
+                const {
+                    initialize_pdfium_render,
+                    log_page_metrics_to_console,
+                    get_image_data_for_page
+                } = wasm_bindgen;
+
+                // Load the wasm_bindgen-wrapped WASM module containing our Rust application.
+
+                wasm_bindgen('pdfium_render_wasm_example_bg.wasm').then(function () {
+                    // The functions exported by our Rust application are now available.
+
+                    // First, we call pdfium-render's exported initialize_pdfium_render() function,
+                    // passing in the Pdfium WASM module and, optionally, a debug flag.
+                    // pdfium-render will bind to the functions it needs, returning a boolean value
+                    // indicating success or failure.
+
+                    const debug = false; // Set this to true to see more debugging information in the Javascript console.
+
+                    console.assert(
+                        initialize_pdfium_render(Module, debug),
+                        "Initialization of pdfium-render failed!"
+                    );
+
+                    // Now we can call the Rust functions we exported in example/wasm.rs.
+                    // The first function dumps sizing metrics for each page to the console.
+
+                    log_page_metrics_to_console();
+
+                    // The second function generates the raw bitmap byte data for a single page in
+                    // a sample PDF file. We can use this byte data to render the page to an HTML canvas.
+
+                    const pageIndex = 0;
+                    const width = 1414;
+                    const height = 1999;
+
+                    const image_bytes = get_image_data_for_page(pageIndex, width, height);
+
+                    const canvas = document.getElementById("canvas");
+
+                    canvas.width = width;
+                    canvas.height = height;
+
+                    const context = canvas.getContext("2d");
+                    const image = context.createImageData(width, height);
+
+                    image.data.set(image_bytes);
+                    context.putImageData(image, 0, 0);
+                });
+            };
+        </script>
+    </body>
+</html>