Merge branch 'tesseract-ocr:main' into 3871-disappearing-word

.github/workflows/autotools-macos.yml

@@ -14,11 +14,11 @@ jobs:
       fail-fast: false
       matrix:
         config:
-          - { name: macos-12-clang-13-autotools, os: macos-12, cxx: clang++ }
-          - { name: macos-12-gcc-11-autotools, os: macos-12, cxx: g++-11 }
+          - { name: macos-12-clang-14-autotools, os: macos-12, cxx: clang++ }
+          #- { name: macos-12-gcc-11-autotools, os: macos-12, cxx: g++-11 }
 
     steps:
-    - uses: actions/checkout@v2
+    - uses: actions/checkout@v3
       with:
         submodules: recursive
 
@@ -115,10 +115,10 @@ jobs:
       fail-fast: false
       matrix:
         config:
-          - { name: macos-10.15-clang-12-autotools, os: macos-10.15, cxx: clang++ }
+          - { name: macos-12-clang-14-autotools, os: macos-12, cxx: clang++ }
 
     steps:
-    - uses: actions/checkout@v2
+    - uses: actions/checkout@v3
       with:
         submodules: recursive
 

.github/workflows/autotools-openmp.yml

@@ -17,7 +17,7 @@ jobs:
           - { name: 22.04-openmp, os: ubuntu-22.04 }
 
     steps:
-    - uses: actions/checkout@v2
+    - uses: actions/checkout@v3
       with:
         submodules: recursive
 

.github/workflows/autotools.yml

@@ -24,7 +24,7 @@ jobs:
           - { name: ubuntu-20.04-gcc-10-autotools, os: ubuntu-20.04, cxx: g++-10 } #installed
 
     steps:
-    - uses: actions/checkout@v2
+    - uses: actions/checkout@v3
       with:
         submodules: recursive
 

.github/workflows/cifuzz.yml

@@ -26,7 +26,7 @@ jobs:
         fuzz-seconds: 600
         dry-run: false
     - name: Upload Crash
-      uses: actions/upload-artifact@v1
+      uses: actions/upload-artifact@v3
       if: failure() && steps.build.outcome == 'success'
       with:
         name: artifacts

.github/workflows/cmake-win64.yml

@@ -151,7 +151,7 @@ jobs:
           tesseract test/testing/phototest.tif -
           
       - name: Upload Build Results
-        uses: actions/upload-artifact@v2
+        uses: actions/upload-artifact@v3
         with:
           name: tesseract-${{ steps.get_version.outputs.VERSION }}-VS2019_win64
           path: ${{env.ILOC}}

.github/workflows/cmake.yml

@@ -61,7 +61,7 @@ jobs:
         if: runner.os == 'macOS'
 
       - name: Checkout Source
-        uses: actions/checkout@v2
+        uses: actions/checkout@v3
         with:
              submodules: recursive
 

.github/workflows/codeql-analysis.yml

@@ -55,7 +55,7 @@ jobs:
 
     steps:
     - name: Checkout repository
-      uses: actions/checkout@v2
+      uses: actions/checkout@v3
 
     - name: Install dependencies
       run: |

.github/workflows/msys2.yml

@@ -19,7 +19,7 @@ jobs:
       run:
         shell: msys2 {0}
     steps:
-    - uses: actions/checkout@v2
+    - uses: actions/checkout@v3
       with:
         submodules: recursive
     - uses: msys2/setup-msys2@v2

.github/workflows/sw.yml

@@ -28,7 +28,7 @@ jobs:
         os: [windows-2022, windows-2019, ubuntu-22.04, ubuntu-20.04, macos-12]
 
     steps:
-    - uses: actions/checkout@v2
+    - uses: actions/checkout@v3
       with:
         submodules: recursive
     - uses: egorpugin/sw-action@master
@@ -71,7 +71,7 @@ jobs:
 
     - name: Upload Unit Test Results
       if: always() && matrix.os != 'windows-2022' && matrix.os != 'windows-2019'
-      uses: actions/upload-artifact@v2
+      uses: actions/upload-artifact@v3
       with:
         name: Test Results (${{ matrix.os }})
         path: .sw/test/results.xml

.github/workflows/unittest-disablelegacy.yml

@@ -17,7 +17,7 @@ jobs:
         os: [ ubuntu-20.04 ]
 
     steps:
-    - uses: actions/checkout@v2
+    - uses: actions/checkout@v3
       with:
         submodules: recursive
 

.github/workflows/unittest-macos.yml

@@ -17,7 +17,7 @@ jobs:
           - { name: macos-12-gcc-unittest, os: macos-12, cxx: g++ }
 
     steps:
-    - uses: actions/checkout@v2
+    - uses: actions/checkout@v3
       with:
         submodules: recursive
 

.github/workflows/unittest.yml

@@ -17,7 +17,7 @@ jobs:
           - { name: ubuntu-20.04-gcc-unittest, os: ubuntu-20.04, cxx: g++ }
           - { name: ubuntu-22.04-clang-unittest, os: ubuntu-22.04, cxx: clang++ }
     steps:
-    - uses: actions/checkout@v2
+    - uses: actions/checkout@v3
       with:
         submodules: recursive
 

.github/workflows/vcpkg-4.1.1.yml

@@ -17,7 +17,7 @@ jobs:
 
     steps:
       - name: Checkout Tesseract Source (for test images)
-        uses: actions/checkout@v2
+        uses: actions/checkout@v3
         with:
           submodules: recursive
 

.github/workflows/vcpkg.yml

@@ -16,7 +16,7 @@ jobs:
 
     steps:
       - name: Checkout Tesseract Source (--head from main branch)
-        uses: actions/checkout@v2
+        uses: actions/checkout@v3
         with:
           submodules: recursive
 

src/ccmain/thresholder.cpp

@@ -287,15 +287,26 @@ bool ImageThresholder::ThresholdToPix(Image *pix) {
     tprintf("Image too large: (%d, %d)\n", image_width_, image_height_);
     return false;
   }
+  Image original = GetPixRect();
   if (pix_channels_ == 0) {
     // We have a binary image, but it still has to be copied, as this API
     // allows the caller to modify the output.
-    Image original = GetPixRect();
     *pix = original.copy();
-    original.destroy();
   } else {
-    OtsuThresholdRectToPix(pix_, pix);
+    if (pixGetColormap(original)) {
+      Image without_cmap =
+          pixRemoveColormap(original, REMOVE_CMAP_BASED_ON_SRC);
+      int depth = pixGetDepth(without_cmap);
+      if (depth > 1 && depth < 8) {
+        original = pixConvertTo8(without_cmap, false);
+      } else {
+        original = without_cmap.copy();
+      }
+      without_cmap.destroy();
+    }
+    OtsuThresholdRectToPix(original, pix);
   }
+  original.destroy();
   return true;
 }
 
@@ -353,7 +364,7 @@ Image ImageThresholder::GetPixRect() {
 Image ImageThresholder::GetPixRectGrey() {
   auto pix = GetPixRect(); // May have to be reduced to grey.
   int depth = pixGetDepth(pix);
-  if (depth != 8) {
+  if (depth != 8 || pixGetColormap(pix)) {
     if (depth == 24) {
       auto tmp = pixConvert24To32(pix);
       pix.destroy();