Fix issue #86 by upgrade to MuPDF 1.24.4

.github/workflows/CI.yml

@@ -4,6 +4,7 @@ on:
   push:
     branches:
       - main
+      - dev
   pull_request:
   workflow_dispatch:
 
@@ -18,7 +19,7 @@ jobs:
     steps:
       - uses: actions/checkout@v3
         with:
-          submodules: 'recursive'
+          submodules: "recursive"
           fetch-depth: 500
       - uses: dtolnay/rust-toolchain@stable
       - run: cargo check
@@ -33,7 +34,7 @@ jobs:
     steps:
       - uses: actions/checkout@v3
         with:
-          submodules: 'recursive'
+          submodules: "recursive"
           fetch-depth: 500
       - uses: dtolnay/rust-toolchain@stable
       - name: Install LLVM
@@ -52,36 +53,41 @@ jobs:
       - name: Test package mupdf-sys
         if: matrix.os == 'ubuntu-latest'
         run: cargo package --manifest-path mupdf-sys/Cargo.toml
-  
+
   test-msys:
     name: Test Suite (MSYS2)
     runs-on: windows-2019
+    defaults:
+      run:
+        shell: msys2 {0}
     strategy:
       matrix:
         include:
-          - { sys: ucrt64,  env: ucrt-x86_64 }
+          - { sys: ucrt64, env: ucrt-x86_64 }
           - { sys: mingw64, env: x86_64 }
-          # - { sys: mingw32, env: i686 }
+          - { sys: mingw32, env: i686 }
           - { sys: clang64, env: clang-x86_64 }
     steps:
       - uses: msys2/setup-msys2@v2
         with:
           msystem: ${{matrix.sys}}
-          install: mingw-w64-${{matrix.env}}-rust base base-devel unzip git
+          install: mingw-w64-${{matrix.env}}-rust mingw-w64-${{matrix.env}}-clang base base-devel unzip git
 
       - uses: actions/checkout@v3
         with:
-          submodules: 'recursive'
+          submodules: "recursive"
           fetch-depth: 500
-      - run: cargo test
-
+      - run: |
+          rustc --version --verbose
+          cargo test
+
   asan:
     name: Address Sanitizer
     runs-on: ubuntu-latest
     steps:
       - uses: actions/checkout@v3
         with:
-          submodules: 'recursive'
+          submodules: "recursive"
           fetch-depth: 500
       - uses: dtolnay/rust-toolchain@nightly
         with:
@@ -98,7 +104,7 @@ jobs:
     steps:
       - uses: actions/checkout@v3
         with:
-          submodules: 'recursive'
+          submodules: "recursive"
           fetch-depth: 500
       - uses: dtolnay/rust-toolchain@stable
       - uses: taiki-e/install-action@valgrind

mupdf-sys/build.rs

@@ -91,7 +91,7 @@ fn build_libmupdf() {
     let mut make_flags = vec![
         "libs".to_owned(),
         format!("build={}", profile),
-        format!("OUT={}", build_dir.display()),
+        format!("OUT={}", &build_dir_str),
         #[cfg(feature = "sys-lib-freetype")]
         "USE_SYSTEM_FREETYPE=yes".to_owned(),
         #[cfg(feature = "sys-lib-gumbo")]
@@ -253,15 +253,15 @@ fn build_libmupdf() {
     };
     let output = Command::new(make)
         .args(&make_flags)
-        .current_dir(&build_dir)
+        .current_dir(&build_dir_str)
         .stdout(Stdio::inherit())
         .stderr(Stdio::inherit())
         .output()
         .expect("make failed");
     if !output.status.success() {
         panic!("Build error, exit code {}", output.status.code().unwrap());
     }
-    println!("cargo:rustc-link-search=native={}", build_dir.display());
+    println!("cargo:rustc-link-search=native={}", &build_dir_str);
     println!("cargo:rustc-link-lib=static=mupdf");
     // println!("cargo:rustc-link-lib=static=mupdf-pkcs7");
     println!("cargo:rustc-link-lib=static=mupdf-third");
@@ -351,6 +351,13 @@ fn build_libmupdf() {
                 profile
             );
         }
+
+        if profile == "Debug" {
+            println!("cargo:rustc-link-lib=dylib=ucrtd");
+            println!("cargo:rustc-link-lib=dylib=vcruntimed");
+            println!("cargo:rustc-link-lib=dylib=msvcrtd");
+        }
+
         println!("cargo:rustc-link-lib=dylib=libmupdf");
         println!("cargo:rustc-link-lib=dylib=libthirdparty");
     } else {

mupdf-sys/wrapper.c

@@ -3012,7 +3012,7 @@ void mupdf_pdf_page_set_crop_box(fz_context *ctx, pdf_page *page, fz_rect rect,
 {
     fz_try(ctx)
     {
-        fz_rect mediabox = pdf_bound_page(ctx, page);
+        fz_rect mediabox = pdf_bound_page(ctx, page, FZ_MEDIA_BOX);
         pdf_obj *obj = pdf_dict_get_inheritable(ctx, page->obj, PDF_NAME(MediaBox));
         if (obj)
         {

tests/test_issues.rs

@@ -80,3 +80,21 @@ fn test_issue_60_display_list() {
         })
         .collect();
 }
+
+#[test]
+fn test_issue_86_invalid_utf8() {
+    let doc = PdfDocument::open("tests/files/utf8-error-on-this-file.pdf").unwrap();
+    for (idx, page) in doc.pages().unwrap().enumerate() {
+        let page = page.unwrap();
+        let text = page.to_text();
+        assert!(text.is_ok());
+        println!("page: {idx}, text: {}", text.unwrap());
+
+        let json = page.stext_page_as_json_from_page(1.0);
+        assert!(json.is_ok());
+
+        // Validate JSON parsing
+        let parsed_json: Result<serde_json::Value, _> = serde_json::from_str(&json.unwrap());
+        assert!(parsed_json.is_ok());
+    }
+}