* Upgrade presets for Tesseract 5.5.0

saudet · saudet · commit d682f6df5c95 · 2024-11-12T19:34:51.000+09:00
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -9,7 +9,7 @@
  * Build FFmpeg with zimg to enable zscale filter ([pull #1481](https://github.com/bytedeco/javacpp-presets/pull/1481))
  * Enable PulseAudio support for FFmpeg on Linux ([pull #1472](https://github.com/bytedeco/javacpp-presets/pull/1472))
  * Virtualize `btCollisionWorld`, `btOverlapFilterCallback`, `btOverlapCallback` from Bullet Physics SDK ([pull #1475](https://github.com/bytedeco/javacpp-presets/pull/1475))
- * Upgrade presets for OpenCV 4.10.0, FFmpeg 7.1, Spinnaker 4.0.0.116 ([pull #1524](https://github.com/bytedeco/javacpp-presets/pull/1524)), MKL 2025.0, DNNL 3.6.1, OpenBLAS 0.3.28, CMINPACK 1.3.11, GSL 2.8, CPython 3.13.0, NumPy 2.1.3, SciPy 1.14.1, LLVM 19.1.3, LibRaw 0.21.2 ([pull #1520](https://github.com/bytedeco/javacpp-presets/pull/1520)), Leptonica 1.85.0, Tesseract 5.4.1, libffi 3.4.6, CUDA 12.6.2, cuDNN 9.5.1, NCCL 2.23.4, nvCOMP 4.1.0.6, OpenCL 3.0.17, NVIDIA Video Codec SDK 12.2.72, PyTorch 2.5.1 ([pull #1466](https://github.com/bytedeco/javacpp-presets/pull/1466)), SentencePiece 0.2.0, TensorFlow Lite 2.18.0, TensorRT 10.6.0.26, Triton Inference Server 2.51.0, ONNX 1.17.0, ONNX Runtime 1.20.0, TVM 0.18.0, and their dependencies
+ * Upgrade presets for OpenCV 4.10.0, FFmpeg 7.1, Spinnaker 4.0.0.116 ([pull #1524](https://github.com/bytedeco/javacpp-presets/pull/1524)), MKL 2025.0, DNNL 3.6.1, OpenBLAS 0.3.28, CMINPACK 1.3.11, GSL 2.8, CPython 3.13.0, NumPy 2.1.3, SciPy 1.14.1, LLVM 19.1.3, LibRaw 0.21.2 ([pull #1520](https://github.com/bytedeco/javacpp-presets/pull/1520)), Leptonica 1.85.0, Tesseract 5.5.0, libffi 3.4.6, CUDA 12.6.2, cuDNN 9.5.1, NCCL 2.23.4, nvCOMP 4.1.0.6, OpenCL 3.0.17, NVIDIA Video Codec SDK 12.2.72, PyTorch 2.5.1 ([pull #1466](https://github.com/bytedeco/javacpp-presets/pull/1466)), SentencePiece 0.2.0, TensorFlow Lite 2.18.0, TensorRT 10.6.0.26, Triton Inference Server 2.51.0, ONNX 1.17.0, ONNX Runtime 1.20.0, TVM 0.18.0, and their dependencies
 
 ### January 29, 2024 version 1.5.10
  * Introduce `macosx-arm64` builds for PyTorch ([pull #1463](https://github.com/bytedeco/javacpp-presets/pull/1463))
diff --git a/README.md b/README.md
@@ -213,7 +213,7 @@ Each child module in turn relies by default on the included [`cppbuild.sh` scrip
  * libpostal 1.1  https://github.com/openvenues/libpostal
  * LibRaw 0.21.x  https://www.libraw.org/download
  * Leptonica 1.85.x  http://www.leptonica.org/download.html
- * Tesseract 5.4.x  https://github.com/tesseract-ocr/tesseract
+ * Tesseract 5.5.x  https://github.com/tesseract-ocr/tesseract
  * Caffe 1.0  https://github.com/BVLC/caffe
  * OpenPose 1.7.0  https://github.com/CMU-Perceptual-Computing-Lab/openpose
  * CUDA 12.6.x  https://developer.nvidia.com/cuda-downloads
diff --git a/platform/pom.xml b/platform/pom.xml
@@ -257,7 +257,7 @@
     <dependency>
       <groupId>org.bytedeco</groupId>
       <artifactId>tesseract-platform</artifactId>
-      <version>5.4.1-${project.version}</version>
+      <version>5.5.0-${project.version}</version>
     </dependency>
 <!--    <dependency>-->
 <!--      <groupId>org.bytedeco</groupId>-->
diff --git a/tesseract/README.md b/tesseract/README.md
@@ -9,7 +9,7 @@ Introduction
 ------------
 This directory contains the JavaCPP Presets module for:
 
- * Tesseract 5.4.1  https://github.com/tesseract-ocr
+ * Tesseract 5.5.0  https://github.com/tesseract-ocr
 
 Please refer to the parent README.md file for more detailed information about the JavaCPP Presets.
 
@@ -47,7 +47,7 @@ We can use [Maven 3](http://maven.apache.org/) to download and install automatic
         <dependency>
             <groupId>org.bytedeco</groupId>
             <artifactId>tesseract-platform</artifactId>
-            <version>5.4.1-1.5.11-SNAPSHOT</version>
+            <version>5.5.0-1.5.11-SNAPSHOT</version>
         </dependency>
     </dependencies>
     <build>
diff --git a/tesseract/cppbuild.sh b/tesseract/cppbuild.sh
@@ -7,7 +7,7 @@ if [[ -z "$PLATFORM" ]]; then
     exit
 fi
 
-TESSERACT_VERSION=5.4.1
+TESSERACT_VERSION=5.5.0
 download https://github.com/tesseract-ocr/tesseract/archive/$TESSERACT_VERSION.tar.gz tesseract-$TESSERACT_VERSION.tar.gz
 
 mkdir -p $PLATFORM
@@ -48,6 +48,8 @@ export PKG_CONFIG_PATH=$INSTALL_PATH/lib/pkgconfig/
 
 CMAKE_CONFIG="-DCMAKE_BUILD_TYPE=Release -DCMAKE_PREFIX_PATH=$LEPTONICA_PATH -DCMAKE_INSTALL_PREFIX=$INSTALL_PATH -DCMAKE_INSTALL_LIBDIR=$INSTALL_PATH/lib -DDISABLE_ARCHIVE=ON -DDISABLE_CURL=ON -DMARCH_NATIVE_OPT=OFF -DOPENMP_BUILD=OFF -DBUILD_SHARED_LIBS=ON -DBUILD_TRAINING_TOOLS=OFF -DLEPT_TIFF_RESULT=1"
 
+patch -RNp1 < ../../../tesseract.patch
+
 case $PLATFORM in
     android-arm)
         patch -Np1 < ../../../tesseract-android.patch
diff --git a/tesseract/platform/pom.xml b/tesseract/platform/pom.xml
@@ -12,7 +12,7 @@
 
   <groupId>org.bytedeco</groupId>
   <artifactId>tesseract-platform</artifactId>
-  <version>5.4.1-${project.parent.version}</version>
+  <version>5.5.0-${project.parent.version}</version>
   <name>JavaCPP Presets Platform for Tesseract</name>
 
   <properties>
diff --git a/tesseract/pom.xml b/tesseract/pom.xml
@@ -11,7 +11,7 @@
 
   <groupId>org.bytedeco</groupId>
   <artifactId>tesseract</artifactId>
-  <version>5.4.1-${project.parent.version}</version>
+  <version>5.5.0-${project.parent.version}</version>
   <name>JavaCPP Presets for Tesseract</name>
 
   <dependencies>
diff --git a/tesseract/samples/pom.xml b/tesseract/samples/pom.xml
@@ -12,7 +12,7 @@
         <dependency>
             <groupId>org.bytedeco</groupId>
             <artifactId>tesseract-platform</artifactId>
-            <version>5.4.1-1.5.11-SNAPSHOT</version>
+            <version>5.5.0-1.5.11-SNAPSHOT</version>
         </dependency>
     </dependencies>
     <build>
diff --git a/tesseract/src/gen/java/org/bytedeco/tesseract/global/tesseract.java b/tesseract/src/gen/java/org/bytedeco/tesseract/global/tesseract.java
@@ -145,15 +145,15 @@ public class tesseract extends org.bytedeco.tesseract.presets.tesseract {
 // clang-format off
 
 public static final int TESSERACT_MAJOR_VERSION = 5;
-public static final int TESSERACT_MINOR_VERSION = 4;
-public static final int TESSERACT_MICRO_VERSION = 1;
+public static final int TESSERACT_MINOR_VERSION = 5;
+public static final int TESSERACT_MICRO_VERSION = 0;
 
 public static final int TESSERACT_VERSION =          
   (TESSERACT_MAJOR_VERSION << 16 | 
    TESSERACT_MINOR_VERSION <<  8 | 
    TESSERACT_MICRO_VERSION);
 
-public static final String TESSERACT_VERSION_STR = "5.4.1";
+public static final String TESSERACT_VERSION_STR = "5.5.0";
 
 // clang-format on
 
diff --git a/tesseract/src/main/java/org/bytedeco/tesseract/presets/tesseract.java b/tesseract/src/main/java/org/bytedeco/tesseract/presets/tesseract.java
@@ -42,9 +42,9 @@
     @Platform(define = "TESS_CAPI_INCLUDE_BASEAPI", include = {"tesseract/export.h", /*"tesseract/osdetect.h",*/ "tesseract/unichar.h",
         "tesseract/version.h", "tesseract/publictypes.h", "tesseract/pageiterator.h", "tesseract/ocrclass.h", "tesseract/ltrresultiterator.h",
         "tesseract/renderer.h", "tesseract/resultiterator.h", "tesseract/baseapi.h", "tesseract/capi.h", "locale.h"},
-        compiler = "cpp14", link = "tesseract@.5.4.1"/*, resource = {"include", "lib"}*/),
+        compiler = "cpp14", link = "tesseract@.5.5"/*, resource = {"include", "lib"}*/),
     @Platform(value = "android", link = "tesseract"),
-    @Platform(value = "windows", link = "tesseract54", preload = "libtesseract54") })
+    @Platform(value = "windows", link = "tesseract55", preload = "libtesseract55") })
 public class tesseract implements InfoMapper {
     static { Loader.checkVersion("org.bytedeco", "tesseract"); }
 
diff --git a/tesseract/tesseract.patch b/tesseract/tesseract.patch
@@ -0,0 +1,276 @@
+diff --git a/src/api/baseapi.cpp b/src/api/baseapi.cpp
+index 72503636c0..bae30ab8bb 100644
+--- a/src/api/baseapi.cpp
++++ b/src/api/baseapi.cpp
+@@ -64,6 +64,7 @@
+ #include <cmath>    // for round, M_PI
+ #include <cstdint>  // for int32_t
+ #include <cstring>  // for strcmp, strcpy
++#include <filesystem> // for std::filesystem
+ #include <fstream>  // for size_t
+ #include <iostream> // for std::cin
+ #include <locale>   // for std::locale::classic
+@@ -82,15 +83,9 @@
+ #endif
+ 
+ #if defined(_WIN32)
+-#  include <fcntl.h>
+-#  include <io.h>
+-#else
+-#  include <dirent.h> // for closedir, opendir, readdir, DIR, dirent
+-#  include <libgen.h>
+-#  include <sys/stat.h> // for stat, S_IFDIR
+-#  include <sys/types.h>
+-#  include <unistd.h>
+-#endif // _WIN32
++#  include <fcntl.h> // for _O_BINARY
++#  include <io.h>    // for _setmode
++#endif
+ 
+ namespace tesseract {
+ 
+@@ -149,61 +144,18 @@ static void ExtractFontName(const char* filename, std::string* fontname) {
+ 
+ /* Add all available languages recursively.
+  */
+-static void addAvailableLanguages(const std::string &datadir, const std::string &base,
++static void addAvailableLanguages(const std::string &datadir,
+                                   std::vector<std::string> *langs) {
+-  auto base2 = base;
+-  if (!base2.empty()) {
+-    base2 += "/";
+-  }
+-  const size_t extlen = sizeof(kTrainedDataSuffix);
+-#ifdef _WIN32
+-  WIN32_FIND_DATA data;
+-  HANDLE handle = FindFirstFile((datadir + base2 + "*").c_str(), &data);
+-  if (handle != INVALID_HANDLE_VALUE) {
+-    BOOL result = TRUE;
+-    for (; result;) {
+-      char *name = data.cFileName;
+-      // Skip '.', '..', and hidden files
+-      if (name[0] != '.') {
+-        if ((data.dwFileAttributes & FILE_ATTRIBUTE_DIRECTORY) == FILE_ATTRIBUTE_DIRECTORY) {
+-          addAvailableLanguages(datadir, base2 + name, langs);
+-        } else {
+-          size_t len = strlen(name);
+-          if (len > extlen && name[len - extlen] == '.' &&
+-              strcmp(&name[len - extlen + 1], kTrainedDataSuffix) == 0) {
+-            name[len - extlen] = '\0';
+-            langs->push_back(base2 + name);
+-          }
+-        }
+-      }
+-      result = FindNextFile(handle, &data);
+-    }
+-    FindClose(handle);
+-  }
+-#else // _WIN32
+-  DIR *dir = opendir((datadir + base).c_str());
+-  if (dir != nullptr) {
+-    dirent *de;
+-    while ((de = readdir(dir))) {
+-      char *name = de->d_name;
+-      // Skip '.', '..', and hidden files
+-      if (name[0] != '.') {
+-        struct stat st;
+-        if (stat((datadir + base2 + name).c_str(), &st) == 0 && (st.st_mode & S_IFDIR) == S_IFDIR) {
+-          addAvailableLanguages(datadir, base2 + name, langs);
+-        } else {
+-          size_t len = strlen(name);
+-          if (len > extlen && name[len - extlen] == '.' &&
+-              strcmp(&name[len - extlen + 1], kTrainedDataSuffix) == 0) {
+-            name[len - extlen] = '\0';
+-            langs->push_back(base2 + name);
+-          }
+-        }
+-      }
++  for (const auto& entry :
++       std::filesystem::recursive_directory_iterator(datadir,
++         std::filesystem::directory_options::follow_directory_symlink |
++         std::filesystem::directory_options::skip_permission_denied)) {
++    auto path = entry.path().lexically_relative(datadir).string();
++    auto extPos = path.rfind(".traineddata");
++    if (extPos != std::string::npos) {
++      langs->push_back(path.substr(0, extPos));
+     }
+-    closedir(dir);
+   }
+-#endif
+ }
+ 
+ TessBaseAPI::TessBaseAPI()
+@@ -444,7 +396,7 @@ void TessBaseAPI::GetLoadedLanguagesAsVector(std::vector<std::string> *langs) co
+ void TessBaseAPI::GetAvailableLanguagesAsVector(std::vector<std::string> *langs) const {
+   langs->clear();
+   if (tesseract_ != nullptr) {
+-    addAvailableLanguages(tesseract_->datadir, "", langs);
++    addAvailableLanguages(tesseract_->datadir, langs);
+     std::sort(langs->begin(), langs->end());
+   }
+ }
+diff --git a/src/ccutil/ccutil.cpp b/src/ccutil/ccutil.cpp
+index 5e93e70866..930aa2636e 100644
+--- a/src/ccutil/ccutil.cpp
++++ b/src/ccutil/ccutil.cpp
+@@ -10,10 +10,6 @@
+ // See the License for the specific language governing permissions and
+ // limitations under the License.
+ 
+-#if defined(_WIN32)
+-#  include <io.h> // for _access
+-#endif
+-
+ #include "ccutil.h"
+ #include "tprintf.h"  // for tprintf
+ 
+@@ -63,7 +59,7 @@ void CCUtil::main_setup(const std::string &argv0, const std::string &basename) {
+     /* Use tessdata prefix from the environment. */
+     datadir = tessdata_prefix;
+ #if defined(_WIN32)
+-  } else if (datadir.empty() || _access(datadir.c_str(), 0) != 0) {
++  } else if (datadir.empty() || !std::filesystem::exists(datadir)) {
+     /* Look for tessdata in directory of executable. */
+     char path[_MAX_PATH];
+     DWORD length = GetModuleFileName(nullptr, path, sizeof(path));
+@@ -73,7 +69,7 @@ void CCUtil::main_setup(const std::string &argv0, const std::string &basename) {
+         *separator = '\0';
+         std::string subdir = path;
+         subdir += "/tessdata";
+-        if (_access(subdir.c_str(), 0) == 0) {
++        if (std::filesystem::exists(subdir)) {
+           datadir = subdir;
+         }
+       }
+diff --git a/unittest/pagesegmode_test.cc b/unittest/pagesegmode_test.cc
+index 9689e407e1..781e67d3f9 100644
+--- a/unittest/pagesegmode_test.cc
++++ b/unittest/pagesegmode_test.cc
+@@ -9,13 +9,9 @@
+ // See the License for the specific language governing permissions and
+ // limitations under the License.
+ 
+-#if defined(_WIN32)
+-#  include <io.h> // for _access
+-#else
+-#  include <unistd.h> // for access
+-#endif
+ #include <allheaders.h>
+ #include <tesseract/baseapi.h>
++#include <filesystem>
+ #include <string>
+ #include "helpers.h"
+ #include "include_gunit.h"
+@@ -24,15 +20,6 @@
+ 
+ namespace tesseract {
+ 
+-// Replacement for std::filesystem::exists (C++-17)
+-static bool file_exists(const char *filename) {
+-#if defined(_WIN32)
+-  return _access(filename, 0) == 0;
+-#else
+-  return access(filename, 0) == 0;
+-#endif
+-}
+-
+ // The fixture for testing Tesseract.
+ class PageSegModeTest : public testing::Test {
+ protected:
+@@ -86,7 +73,7 @@ class PageSegModeTest : public testing::Test {
+ // and differently to line and block mode.
+ TEST_F(PageSegModeTest, WordTest) {
+   std::string filename = file::JoinPath(TESTING_DIR, "segmodeimg.tif");
+-  if (!file_exists(filename.c_str())) {
++  if (!std::filesystem::exists(filename)) {
+     LOG(INFO) << "Skip test because of missing " << filename << '\n';
+     GTEST_SKIP();
+   } else {
+diff --git a/unittest/tatweel_test.cc b/unittest/tatweel_test.cc
+index d0d8f2ae6f..10e673b217 100644
+--- a/unittest/tatweel_test.cc
++++ b/unittest/tatweel_test.cc
+@@ -9,12 +9,7 @@
+ // See the License for the specific language governing permissions and
+ // limitations under the License.
+ 
+-#if defined(_WIN32)
+-#  include <io.h> // for _access
+-#else
+-#  include <unistd.h> // for access
+-#endif
+-
++#include <filesystem>
+ #include "dawg.h"
+ #include "include_gunit.h"
+ #include "trie.h"
+@@ -23,15 +18,6 @@
+ 
+ namespace tesseract {
+ 
+-// Replacement for std::filesystem::exists (C++-17)
+-static bool file_exists(const char *filename) {
+-#if defined(_WIN32)
+-  return _access(filename, 0) == 0;
+-#else
+-  return access(filename, 0) == 0;
+-#endif
+-}
+-
+ class TatweelTest : public ::testing::Test {
+ protected:
+   void SetUp() override {
+@@ -41,7 +27,7 @@ class TatweelTest : public ::testing::Test {
+ 
+   TatweelTest() {
+     std::string filename = TestDataNameToPath("ara.wordlist");
+-    if (file_exists(filename.c_str())) {
++    if (std::filesystem::exists(filename)) {
+       std::string wordlist("\u0640");
+       CHECK_OK(file::GetContents(filename, &wordlist, file::Defaults()));
+       // Put all the unicodes in the unicharset_.
+@@ -77,7 +63,7 @@ TEST_F(TatweelTest, DictIgnoresTatweel) {
+   // This test verifies that the dictionary ignores the Tatweel character.
+   tesseract::Trie trie(tesseract::DAWG_TYPE_WORD, "ara", SYSTEM_DAWG_PERM, unicharset_.size(), 0);
+   std::string filename = TestDataNameToPath("ara.wordlist");
+-  if (!file_exists(filename.c_str())) {
++  if (!std::filesystem::exists(filename)) {
+     LOG(INFO) << "Skip test because of missing " << filename;
+     GTEST_SKIP();
+   } else {
+@@ -91,7 +77,7 @@ TEST_F(TatweelTest, UnicharsetLoadKeepsTatweel) {
+   // This test verifies that a load of an existing unicharset keeps any
+   // existing tatweel for backwards compatibility.
+   std::string filename = TestDataNameToPath("ara.unicharset");
+-  if (!file_exists(filename.c_str())) {
++  if (!std::filesystem::exists(filename)) {
+     LOG(INFO) << "Skip test because of missing " << filename;
+     GTEST_SKIP();
+   } else {
+diff --git a/src/ccutil/ccutil.cpp b/src/ccutil/ccutil.cpp
+index 7cf57f2ee9..483d5bc1ee 100644
+--- a/src/ccutil/ccutil.cpp
++++ b/src/ccutil/ccutil.cpp
+@@ -17,7 +17,8 @@
+ #include "ccutil.h"
+ 
+ #include <cstdlib>
+-#include <cstring> // for std::strrchr
++#include <cstring>    // for std::strrchrA
++#include <filesystem> // for std::filesystem
+ 
+ namespace tesseract {
+ 
+@@ -48,6 +49,12 @@ void CCUtil::main_setup(const std::string &argv0, const std::string &basename) {
+ 
+   const char *tessdata_prefix = getenv("TESSDATA_PREFIX");
+ 
++  // Ignore TESSDATA_PREFIX if there is no matching filesystem entry.
++  if (tessdata_prefix != nullptr && !std::filesystem::exists(tessdata_prefix)) {
++    tprintf("Warning: TESSDATA_PREFIX %s does not exist, ignore it\n", tessdata_prefix);
++    tessdata_prefix = nullptr;
++  }
++
+   if (!argv0.empty()) {
+     /* Use tessdata prefix from the command line. */
+     datadir = argv0;
