* Upgrade presets for Tesseract 4.0.0-beta.1 (issue #385)

CHANGELOG.md

@@ -4,7 +4,7 @@
  * Update build for FFmpeg on Raspbian Stretch ([pull #548](https://github.com/bytedeco/javacpp-presets/pull/548))
  * Add presets for TensorRT
  * Fix build for FFmpeg on `android-x86` and `android-x86_64` platforms ([issue bytedeco/javacv#945](https://github.com/bytedeco/javacv/issues/945))
- * Upgrade presets for HDF5 1.10.2, LLVM 6.0.0, TensorFlow 1.7.0
+ * Upgrade presets for HDF5 1.10.2, LLVM 6.0.0, Tesseract 4.0.0-beta.1 ([issue #385](https://github.com/bytedeco/javacpp-presets/issues/385)), TensorFlow 1.7.0
 
 ### March 29, 2018 version 1.4.1
  * Disable unneeded error messages from LibTIFF in presets for Leptonica ([issue deeplearning4j/DataVec#518](https://github.com/deeplearning4j/DataVec/pull/518))

README.md

@@ -124,7 +124,7 @@ Each child module in turn relies by default on the included [`cppbuild.sh` scrip
  * LLVM 6.0.0  http://llvm.org/releases/download.html
  * libpostal 1.1-alpha  https://github.com/openvenues/libpostal
  * Leptonica 1.75.3  http://www.leptonica.org/download.html
- * Tesseract 3.05.01  https://github.com/tesseract-ocr/tesseract
+ * Tesseract 4.0.0-beta.1  https://github.com/tesseract-ocr/tesseract
  * Caffe 1.0  https://github.com/BVLC/caffe
  * CUDA 9.1  https://developer.nvidia.com/cuda-downloads
    * cuDNN 7.1  https://developer.nvidia.com/cudnn

ci/build.cmd

@@ -29,7 +29,7 @@ if "%APPVEYOR_PULL_REQUEST_NUMBER%" == "" if "%APPVEYOR_REPO_BRANCH%" == "releas
 
 rem C:\msys64\usr\bin\bash -lc "pacman -Syu --noconfirm"
 rem C:\msys64\usr\bin\bash -lc "pacman -Su --noconfirm"
-C:\msys64\usr\bin\bash -lc "pacman -S --needed --noconfirm base-devel git tar nasm yasm pkg-config unzip autoconf automake libtool make patch gnupg"
+C:\msys64\usr\bin\bash -lc "pacman -S --needed --noconfirm base-devel git tar nasm yasm pkg-config unzip autoconf autoconf-archive automake libtool make patch gnupg"
 C:\msys64\usr\bin\bash -lc "pacman -S --needed --noconfirm mingw-w64-x86_64-toolchain mingw-w64-x86_64-libtool mingw-w64-x86_64-cmake mingw-w64-x86_64-gcc mingw-w64-i686-gcc mingw-w64-x86_64-gcc-fortran mingw-w64-i686-gcc-fortran mingw-w64-x86_64-libwinpthread-git mingw-w64-i686-libwinpthread-git mingw-w64-x86_64-SDL mingw-w64-i686-SDL"
 
 C:\msys64\usr\bin\bash -lc "/c/projects/javacpp-presets/ci/install-windows.sh %PROJ%"

ci/install-ppc.sh

@@ -36,7 +36,7 @@ docker exec -ti $DOCKER_CONTAINER_ID /bin/bash -xec 'echo "deb [arch=ppc64el] ht
 docker exec -ti $DOCKER_CONTAINER_ID /bin/bash -xec "sed -i 's/deb http/deb [arch=i386,amd64] http/g' /etc/apt/sources.list"
 docker exec -ti $DOCKER_CONTAINER_ID /bin/bash -xec "apt-get update > /dev/null"
 docker exec -ti $DOCKER_CONTAINER_ID /bin/bash -xec "apt-get -y install python python2.7 python-minimal python2.7-minimal libgtk2.0-dev:ppc64el libasound2-dev:ppc64el libusb-dev:ppc64el libusb-1.0-0-dev:ppc64el zlib1g-dev:ppc64el libxcb1-dev:ppc64el > /dev/null"
-docker exec -ti $DOCKER_CONTAINER_ID /bin/bash -xec "apt-get -y install pkg-config gcc-powerpc64le-linux-gnu g++-powerpc64le-linux-gnu gfortran-powerpc64le-linux-gnu linux-libc-dev-ppc64el-cross binutils-multiarch default-jdk ant maven python python-dev python-numpy swig git file wget unzip tar bzip2 patch autogen automake make libtool perl nasm yasm curl cmake3 > /dev/null"
+docker exec -ti $DOCKER_CONTAINER_ID /bin/bash -xec "apt-get -y install pkg-config gcc-powerpc64le-linux-gnu g++-powerpc64le-linux-gnu gfortran-powerpc64le-linux-gnu linux-libc-dev-ppc64el-cross binutils-multiarch default-jdk ant maven python python-dev python-numpy swig git file wget unzip tar bzip2 patch autoconf-archive autogen automake make libtool perl nasm yasm curl cmake3 > /dev/null"
 docker exec -ti $DOCKER_CONTAINER_ID /bin/bash -xec "apt-get -y remove libxdmcp-dev libx11-dev libxcb1-dev libxt-dev > /dev/null"
 
 if [[ "$PROJ" =~ cuda ]]; then

ci/install-travis.sh

@@ -37,7 +37,7 @@ if [[ "$OS" == "linux-x86" ]] || [[ "$OS" == "linux-x86_64" ]] || [[ "$OS" =~ an
   DOCKER_CONTAINER_ID=$(docker ps | grep centos | awk '{print $1}')
   echo "Container id is $DOCKER_CONTAINER_ID please wait while updates applied"
   docker exec -ti $DOCKER_CONTAINER_ID /bin/bash -xec "yum -y install centos-release-scl-rh epel-release" > /dev/null
-  docker exec -ti $DOCKER_CONTAINER_ID /bin/bash -xec "yum -y install devtoolset-4-toolchain rh-java-common-ant rh-maven33 python27 clang gcc-c++ gcc-gfortran java-1.8.0-openjdk-devel ant maven python numpy swig git file which wget unzip tar bzip2 gzip xz patch make cmake3 libtool perl nasm yasm alsa-lib-devel freeglut-devel glfw-devel gtk2-devel libusb-devel libusb1-devel zlib-devel SDL-devel libva-devel" > /dev/null
+  docker exec -ti $DOCKER_CONTAINER_ID /bin/bash -xec "yum -y install devtoolset-4-toolchain rh-java-common-ant rh-maven33 python27 clang gcc-c++ gcc-gfortran java-1.8.0-openjdk-devel ant maven python numpy swig git file which wget unzip tar bzip2 gzip xz patch make cmake3 autoconf-archive libtool perl nasm yasm alsa-lib-devel freeglut-devel glfw-devel gtk2-devel libusb-devel libusb1-devel zlib-devel SDL-devel libva-devel" > /dev/null
   if [ "$OS" == "linux-x86" ]; then
     docker exec -ti $DOCKER_CONTAINER_ID /bin/bash -xec "rpm -qa | sed s/.x86_64$/.i686/ | xargs yum -y install > /dev/null"
   fi
@@ -93,7 +93,7 @@ if [ "$OS" == "linux-armhf" ]; then
 	sudo dpkg --add-architecture i386
 	sudo apt-get update
 	sudo apt-get -y install libc6:i386 libncurses5:i386 libstdc++6:i386 lib32z1
-	sudo apt-get -y install clang git file wget unzip tar bzip2 gzip patch autogen automake libtool perl nasm yasm libasound2-dev freeglut3-dev libglfw3-dev libgtk2.0-dev libusb-dev zlib1g
+	sudo apt-get -y install clang git file wget unzip tar bzip2 gzip patch autoconf-archive autogen automake libtool perl nasm yasm libasound2-dev freeglut3-dev libglfw3-dev libgtk2.0-dev libusb-dev zlib1g
 	git -C $HOME clone https://github.com/raspberrypi/tools
 	git -C $HOME clone https://github.com/raspberrypi/userland
 	export PATH=$PATH:$HOME/tools/arm-bcm2708/gcc-linaro-arm-linux-gnueabihf-raspbian/bin
@@ -118,7 +118,7 @@ if [ "$TRAVIS_OS_NAME" == "osx" ]; then
    echo "performing brew update and install of dependencies, please wait.."
    brew update > /dev/null
    brew upgrade maven
-   brew install gcc swig libtool libusb nasm yasm xz sdl gpg1
+   brew install gcc swig autoconf-archive libtool libusb nasm yasm xz sdl gpg1
    brew link --overwrite gcc
    export PATH=/usr/local/opt/gpg1/libexec/gpgbin/:$PATH
 

platform/pom.xml

@@ -163,7 +163,7 @@
     <dependency>
       <groupId>org.bytedeco.javacpp-presets</groupId>
       <artifactId>tesseract-platform</artifactId>
-      <version>3.05.01-${project.version}</version>
+      <version>4.0.0-beta.1-${project.version}</version>
     </dependency>
     <dependency>
       <groupId>org.bytedeco.javacpp-presets</groupId>

tesseract/COPYING

@@ -1,5 +1,5 @@
 This package contains the Tesseract Open Source OCR Engine.
-Orignally developed at Hewlett Packard Laboratories Bristol and
+Originally developed at Hewlett Packard Laboratories Bristol and
 at Hewlett Packard Co, Greeley Colorado, all the code
 in this distribution is now licensed under the Apache License:
 
@@ -17,6 +17,5 @@ in this distribution is now licensed under the Apache License:
 Other Dependencies and Licenses:
 ================================
 
-Tesseract use Leptonica library (http://leptonica.com/) with a very weakly
-restricted copyright license (http://leptonica.com/about-the-license.html)
-
+Tesseract uses Leptonica library (http://leptonica.com/) which essentially
+uses a BSD 2-clause license. (http://leptonica.com/about-the-license.html)

tesseract/README.md

@@ -5,7 +5,7 @@ Introduction
 ------------
 This directory contains the JavaCPP Presets module for:
 
- * Tesseract 3.05.01  https://github.com/tesseract-ocr
+ * Tesseract 4.0.0-beta.1  https://github.com/tesseract-ocr
 
 Please refer to the parent README.md file for more detailed information about the JavaCPP Presets.
 
@@ -35,15 +35,15 @@ We can use [Maven 3](http://maven.apache.org/) to download and install automatic
     <modelVersion>4.0.0</modelVersion>
     <groupId>org.bytedeco.javacpp-presets.tesseract</groupId>
     <artifactId>BasicExample</artifactId>
-    <version>1.4.1</version>
+    <version>1.4.2-SNAPSHOT</version>
     <properties>
         <exec.mainClass>BasicExample</exec.mainClass>
     </properties>
     <dependencies>
         <dependency>
             <groupId>org.bytedeco.javacpp-presets</groupId>
             <artifactId>tesseract-platform</artifactId>
-            <version>3.05.01-1.4.1</version>
+            <version>4.0.0-beta.1-1.4.2-SNAPSHOT</version>
         </dependency>
     </dependencies>
 </project>

tesseract/cppbuild.sh

@@ -7,7 +7,7 @@ if [[ -z "$PLATFORM" ]]; then
     exit
 fi
 
-TESSERACT_VERSION=3.05.01
+TESSERACT_VERSION=4.0.0-beta.1
 download https://github.com/tesseract-ocr/tesseract/archive/$TESSERACT_VERSION.tar.gz tesseract-$TESSERACT_VERSION.tar.gz
 
 mkdir -p $PLATFORM
@@ -36,6 +36,8 @@ fi
 
 LEPTONICA_PATH="${LEPTONICA_PATH//\\//}"
 
+sedinplace 's/static string/static std::string/g' ccutil/unichar.h
+
 case $PLATFORM in
     android-arm)
 #        ANDROID_ROOT=${ANDROID_ROOT//14/21}
@@ -123,14 +125,14 @@ case $PLATFORM in
         make install-strip
         ;;
     windows-x86)
-        patch -Np1 < ../../../tesseract-windows.patch
+        #patch -Np1 < ../../../tesseract-windows.patch
         cp vs2010/port/* ccutil/
         ./configure --prefix=$INSTALL_PATH --host="i686-w64-mingw32" CC="gcc -m32" CXX="g++ -m32 -fpermissive" LEPTONICA_CFLAGS="-I$LEPTONICA_PATH/include/leptonica/" LEPTONICA_LIBS="-L$LEPTONICA_PATH/lib/ -llept" CPPFLAGS="-I$LEPTONICA_PATH/include/" LDFLAGS="-L$LEPTONICA_PATH/lib/" LIBS="-llept"
         make -j $MAKEJ
         make install-strip
         ;;
     windows-x86_64)
-        patch -Np1 < ../../../tesseract-windows.patch
+        #patch -Np1 < ../../../tesseract-windows.patch
         cp vs2010/port/* ccutil/
         ./configure --prefix=$INSTALL_PATH --host="x86_64-w64-mingw32" CC="gcc -m64" CXX="g++ -m64 -fpermissive" LEPTONICA_CFLAGS="-I$LEPTONICA_PATH/include/leptonica/" LEPTONICA_LIBS="-L$LEPTONICA_PATH/lib/ -llept" CPPFLAGS="-I$LEPTONICA_PATH/include/" LDFLAGS="-L$LEPTONICA_PATH/lib/" LIBS="-llept"
         make -j $MAKEJ

tesseract/platform/pom.xml

@@ -12,7 +12,7 @@
 
   <groupId>org.bytedeco.javacpp-presets</groupId>
   <artifactId>tesseract-platform</artifactId>
-  <version>3.05.01-${project.parent.version}</version>
+  <version>4.0.0-beta.1-${project.parent.version}</version>
   <name>JavaCPP Presets Platform for Tesseract</name>
 
   <properties>

tesseract/pom.xml

@@ -11,7 +11,7 @@
 
   <groupId>org.bytedeco.javacpp-presets</groupId>
   <artifactId>tesseract</artifactId>
-  <version>3.05.01-${project.parent.version}</version>
+  <version>4.0.0-beta.1-${project.parent.version}</version>
   <name>JavaCPP Presets for Tesseract</name>
 
   <properties>

tesseract/src/main/java/org/bytedeco/javacpp/presets/tesseract.java

@@ -1,5 +1,5 @@
 /*
- * Copyright (C) 2014-2015 Samuel Audet
+ * Copyright (C) 2014-2018 Samuel Audet
  *
  * Licensed either under the Apache License, Version 2.0, or (at your option)
  * under the terms of the GNU General Public License as published by
@@ -32,13 +32,13 @@
  *
  * @author Samuel Audet
  */
-@Properties(target="org.bytedeco.javacpp.tesseract", inherit=lept.class, value={
-    @Platform(define="TESS_CAPI_INCLUDE_BASEAPI", include={"tesseract/platform.h", "tesseract/apitypes.h", "tesseract/unichar.h", "tesseract/host.h",
+@Properties(target = "org.bytedeco.javacpp.tesseract", inherit = lept.class, value = {
+    @Platform(define = "TESS_CAPI_INCLUDE_BASEAPI", include = {"tesseract/platform.h", "tesseract/apitypes.h", "tesseract/unichar.h", "tesseract/host.h",
         "tesseract/tesscallback.h", "tesseract/publictypes.h", "tesseract/thresholder.h", "tesseract/pageiterator.h", "tesseract/ltrresultiterator.h",
         "tesseract/resultiterator.h", "tesseract/strngs.h", "tesseract/genericvector.h", "tesseract/baseapi.h", "tesseract/capi.h"},
-        link="tesseract@.3"/*, resource = {"include", "lib"}*/),
-    @Platform(value="android", link="tesseract"),
-    @Platform(value="windows", link="libtesseract", preload="libtesseract-3") })
+        compiler = "cpp11", link = "tesseract@.4"/*, resource = {"include", "lib"}*/),
+    @Platform(value = "android", link = "tesseract"),
+    @Platform(value = "windows", link = "libtesseract", preload = "libtesseract-4") })
 public class tesseract implements InfoMapper {
     public void map(InfoMap infoMap) {
         infoMap.put(new Info("__NATIVE__", "ultoa", "snprintf", "vsnprintf", "SIGNED",
@@ -65,14 +65,15 @@ public void map(InfoMap infoMap) {
                .put(new Info("TessCallback1<STRING>").pointerTypes("StringClearCallback").define().virtualize())
                .put(new Info("TessResultCallback2<bool,const STRING&,const STRING&>").pointerTypes("StringCompareCallback").define().virtualize())
                .put(new Info("TessResultCallback2<bool,FILE*,const STRING&>").pointerTypes("StringWriteCallback").define().virtualize())
-               .put(new Info("TessResultCallback3<bool,FILE*,STRING*,bool>").pointerTypes("StringReadCallback").define().virtualize())
+               .put(new Info("TessResultCallback2<bool,tesseract::TFile*,char*>").pointerTypes("CharReadCallback").define().virtualize())
+               .put(new Info("TessResultCallback2<bool,tesseract::TFile*,STRING*>").pointerTypes("StringReadCallback").define().virtualize())
                .put(new Info("GenericVector<STRING>").pointerTypes("StringGenericVector").define())
                .put(new Info("GenericVector<STRING>::WithinBounds").skip())
 
                .put(new Info("TessCallback1<int>").pointerTypes("IntClearCallback").define().virtualize())
                .put(new Info("TessResultCallback2<bool,const int&,const int&>").pointerTypes("IntCompareCallback").define().virtualize())
                .put(new Info("TessResultCallback2<bool,FILE*,const int&>").pointerTypes("IntWriteCallback").define().virtualize())
-               .put(new Info("TessResultCallback3<bool,FILE*,int*,bool>").pointerTypes("IntReadCallback").define().virtualize())
+               .put(new Info("TessResultCallback2<bool,tesseract::TFile*,int*>").pointerTypes("IntReadCallback").define().virtualize())
                .put(new Info("GenericVector<int>").pointerTypes("IntGenericVector").define())
                .put(new Info("GenericVector<StrongScriptDirection>").cast().pointerTypes("IntGenericVector"))
                .put(new Info("GenericVectorEqEq<int>").pointerTypes("IntGenericVectorEqEq").define())