Merge branch 'master' of github.com:egorpugin/tesseract

bbpatil · Oct 5, 2015 · a614edb · a614edb
2 parents 93d1c66 + f369585
commit a614edb
Show file tree

Hide file tree

Showing 83 changed files with 462 additions and 356 deletions.
diff --git a/.gitignore b/.gitignore
@@ -59,6 +59,8 @@ training/wordlist2dawg
 *.o
 *.Plo
 *.a
+*.class
+*.jar
 
 # tessdata
 *.cube.*

diff --git a/.travis.yml b/.travis.yml
@@ -0,0 +1,45 @@
+language: cpp
+
+notifications:
+  email: false
+
+sudo: required
+
+os:
+  - linux
+  - osx
+
+branches:
+  only:
+    - master
+
+addons:
+  apt:
+    sources:
+    - ubuntu-toolchain-r-test
+    packages:
+    - gcc-4.8
+    - g++-4.8
+
+before_install:
+  - if [[ $TRAVIS_OS_NAME == linux   ]]; then LINUX=true; fi
+  - if [[ $TRAVIS_OS_NAME == osx     ]]; then OSX=true; fi
+
+  - if [[ $OSX ]]; then brew update; fi
+
+install:
+  - if [[ $OSX ]]; then brew install icu4c pango; brew link --force gettext; fi
+  - if [[ $OSX ]]; then export ICU_ROOT=/usr/local/opt/icu4c ; fi
+  - wget http://www.cmake.org/files/v3.3/cmake-3.3.1-Linux-x86_64.sh
+  - sudo sh cmake-3.3.1-Linux-x86_64.sh --skip-license --prefix=/usr
+  - wget -O leptonica.zip https://github.com/egorpugin/leptonica/archive/master.zip
+  - unzip leptonica.zip -d .
+  - cmake -Hleptonica-master -Bleptonica-master/build
+  - make -C leptonica-master/build
+  - if [[ $LINUX && "$CXX" = "g++" ]]; then export CXX="g++-4.8" CC="gcc-4.8"; fi
+
+script:
+  - mkdir build
+  - cd build
+  - cmake .. -DLeptonica_DIR=leptonica-master/build
+  - make
diff --git a/CMakeLists.txt b/CMakeLists.txt
@@ -8,7 +8,7 @@
 #
 ###############################################################################
 
-cmake_minimum_required(VERSION 2.8.12)
+cmake_minimum_required(VERSION 2.8.11)
 
 # In-source builds are disabled.
 if (${CMAKE_SOURCE_DIR} STREQUAL ${CMAKE_BINARY_DIR})
@@ -47,15 +47,8 @@ set(VERSION_PLAIN ${VERSION_MAJOR}.${VERSION_MINOR})
 find_package(Leptonica 1.71 REQUIRED)
 
 find_package(ICU COMPONENTS uc i18n)
-find_package(PkgConfig QUIET)
-pkg_check_modules(Pango pango)
-pkg_check_modules(Cairo cairo)
-pkg_check_modules(PangoFt2 pangoft2)
-pkg_check_modules(PangoCairo pangocairo)
-pkg_check_modules(FontConfig fontconfig)
-
-include_directories(${Pango_INCLUDE_DIRS})
-include_directories(${Cairo_INCLUDE_DIRS})
+find_package(OpenCL QUIET)
+find_package(PkgConfig)
 
 ###############################################################################
 #
@@ -80,6 +73,10 @@ if (WIN32)
     set(LIB_Ws2_32 Ws2_32)
 endif()
 
+if (CYGWIN)
+    add_definitions(-D__CYGWIN__)
+endif()
+
 if (UNIX)
     set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} --std=c++11")
 
@@ -190,7 +187,9 @@ set(tesseract_src ${tesseract_src}
 )
 
 add_library                     (tesseract ${LIBRARY_TYPE} ${tesseract_src} ${tesseract_hdr})
+if (NOT STATIC)
 target_compile_definitions      (tesseract PUBLIC -DTESS_EXPORTS)
+endif()
 target_link_libraries           (tesseract ${Leptonica_LIBRARIES} ${LIB_Ws2_32} ${LIB_pthread})
 set_target_properties           (tesseract PROPERTIES OUTPUT_NAME tesseract${VERSION_MAJOR}${VERSION_MINOR})
 set_target_properties           (tesseract PROPERTIES DEBUG_OUTPUT_NAME tesseract${VERSION_MAJOR}${VERSION_MINOR}d)

diff --git a/COPYING b/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:
 

diff --git a/README.md b/README.md
@@ -1,3 +1,6 @@
+[![Build Status](https://travis-ci.org/egorpugin/tesseract.svg?branch=master)](https://travis-ci.org/egorpugin/tesseract)
+[![Build status](https://ci.appveyor.com/api/projects/status/34s8gu4md3i9s93k?svg=true)](https://ci.appveyor.com/project/egorpugin/tesseract)
+
 Note that this is possibly out-of-date version of the wiki ReadMe,
 which is located at:
 
@@ -97,7 +100,7 @@ find its data directory. You must either:
     ./autogen.sh
     ./configure
     make
-    make install
+    sudo make install
     sudo ldconfig
 
 to move the data files to the standard place, or:

diff --git a/api/baseapi.cpp b/api/baseapi.cpp
@@ -1660,7 +1660,7 @@ char* TessBaseAPI::GetUNLVText() {
             word->word->space() > 0 &&
             !word->word->flag(W_FUZZY_NON) &&
             !word->word->flag(W_FUZZY_SP)) {
-          /* Write a space to separate from preceeding good text */
+          /* Write a space to separate from preceding good text */
           *ptr++ = ' ';
           last_char_was_tilde = false;
         }

diff --git a/api/pdfrenderer.cpp b/api/pdfrenderer.cpp
@@ -178,7 +178,7 @@ void TessPDFRenderer::AppendPDFObject(const char *data) {
   AppendString((const char *)data);
 }
 
-// Helper function to prevent us from accidentaly writing
+// Helper function to prevent us from accidentally writing
 // scientific notation to an HOCR or PDF file. Besides, three
 // decimal points are all you really need.
 double prec(double x) {

diff --git a/api/tesseractmain.cpp b/api/tesseractmain.cpp
@@ -227,7 +227,7 @@ int main(int argc, char **argv) {
   }
 
   // We have 2 possible sources of pagesegmode: a config file and
-  // the command line. For backwards compatability reasons, the
+  // the command line. For backwards compatibility reasons, the
   // default in tesseract is tesseract::PSM_SINGLE_BLOCK, but the
   // default for this program is tesseract::PSM_AUTO. We will let
   // the config file take priority, so the command-line default

diff --git a/appveyor.yml b/appveyor.yml
@@ -0,0 +1,24 @@
+os: Visual Studio 2015
+
+platform:
+  - Win32
+  - Win64
+
+configuration:
+  - Debug
+
+before_build:
+  - if %platform%==Win32 set generator=Visual Studio 14
+  - if %platform%==Win64 set generator=Visual Studio 14 Win64
+  - if %platform%==Win32 set vcplatform=Win32
+  - if %platform%==Win64 set vcplatform=x64
+  - ps: Start-FileDownload 'https://github.com/egorpugin/leptonica/archive/master.zip' -FileName leptonica.zip
+  - 7z x leptonica.zip
+  - cmake -Hleptonica-master -Bleptonica-master/build -G "%generator%"
+  - msbuild leptonica-master/build/leptonica.sln /p:Platform=%vcplatform% /logger:"C:\Program Files\AppVeyor\BuildAgent\Appveyor.MSBuildLogger.dll"
+
+build_script:
+  - mkdir build
+  - cd build
+  - cmake .. -G "%generator%" -DLeptonica_DIR=leptonica-master/build -DSTATIC=1
+  - msbuild tesseract.sln /p:Platform=%vcplatform% /logger:"C:\Program Files\AppVeyor\BuildAgent\Appveyor.MSBuildLogger.dll"
diff --git a/ccmain/control.cpp b/ccmain/control.cpp
@@ -1556,7 +1556,7 @@ void Tesseract::match_word_pass_n(int pass_n, WERD_RES *word,
        word->fix_quotes();
       if (tessedit_fix_hyphens)
         word->fix_hyphens();
-      /* Dont trust fix_quotes! - though I think I've fixed the bug */
+      /* Don't trust fix_quotes! - though I think I've fixed the bug */
       if (word->best_choice->length() != word->box_word->length()) {
         tprintf("POST FIX_QUOTES FAIL String:\"%s\"; Strlen=%d;"
                 " #Blobs=%d\n",
@@ -1694,7 +1694,7 @@ ACCEPTABLE_WERD_TYPE Tesseract::acceptable_word_string(
       goto not_a_word;
     /*
     Allow a single hyphen in a lower case word
-    - dont trust upper case - I've seen several cases of "H" -> "I-I"
+    - don't trust upper case - I've seen several cases of "H" -> "I-I"
     */
     if (lengths[i] == 1 && s[offset] == '-') {
       hyphen_pos = i;

diff --git a/ccmain/docqual.cpp b/ccmain/docqual.cpp
@@ -129,7 +129,7 @@ inT16 Tesseract::count_outline_errs(char c, inT16 outline_count) {
   int expected_outline_count;
 
   if (STRING (outlines_odd).contains (c))
-    return 0;                    //Dont use this char
+    return 0;                    //Don't use this char
   else if (STRING (outlines_2).contains (c))
     expected_outline_count = 2;
   else
@@ -157,7 +157,7 @@ void Tesseract::quality_based_rejection(PAGE_RES_IT &page_res_it,
  *    - Word segmentation is the same as the original image
  *		- All characters have the expected number of outlines
  * NOTE - the rejection counts are recalculated after unrejection
- *      - CANT do it in a single pass without a bit of fiddling
+ *      - CAN'T do it in a single pass without a bit of fiddling
  *		- keep it simple but inefficient
  *************************************************************************/
 void Tesseract::unrej_good_quality_words(  //unreject potential
@@ -403,7 +403,7 @@ void Tesseract::doc_and_block_rejection(  //reject big chunks
 
 /*************************************************************************
  * reject_whole_page()
- * Dont believe any of it - set the reject map to 00..00 in all words
+ * Don't believe any of it - set the reject map to 00..00 in all words
  *
  *************************************************************************/
 

diff --git a/ccmain/fixspace.cpp b/ccmain/fixspace.cpp
@@ -55,7 +55,7 @@ void Tesseract::fix_fuzzy_spaces(ETEXT_DESC *monitor,
   WERD_RES *word_res;
   WERD_RES_LIST fuzzy_space_words;
   inT16 new_length;
-  BOOL8 prevent_null_wd_fixsp;   // DONT process blobless wds
+  BOOL8 prevent_null_wd_fixsp;   // DON'T process blobless wds
   inT32 word_index;              // current word
 
   block_res_it.set_to_list(&page_res->block_res_list);
@@ -222,7 +222,7 @@ void Tesseract::match_current_words(WERD_RES_LIST &words, ROW *row,
  * fuzzy spaces. The problem with the basic measure is that "561 63" would score
  * the same as "56163", though given our knowledge that the space is fuzzy, and
  * that there is a "1" next to the fuzzy space, we need to ensure that "56163"
- * is prefered.
+ * is preferred.
  *
  * The solution is to NOT COUNT the score of any word which has a digit at one
  * end and a "1Il" as the character the other side of the space.
@@ -272,8 +272,8 @@ inT16 Tesseract::eval_word_spacing(WERD_RES_LIST &word_res_list) {
     } else {
       /*
         Can we add the prev word score and potentially count this word?
-        Yes IF it didnt end in a 1 when the first char of this word is a digit
-          AND it didnt end in a digit when the first char of this word is a 1
+        Yes IF it didn't end in a 1 when the first char of this word is a digit
+          AND it didn't end in a digit when the first char of this word is a 1
       */
       word_len = word->reject_map.length();
       current_word_ok_so_far = FALSE;
@@ -507,7 +507,7 @@ BOOL8 Tesseract::fixspace_thinks_word_done(WERD_RES *word) {
 
   /*
     Use all the standard pass 2 conditions for mode 5 in set_done() in
-    reject.c BUT DONT REJECT IF THE WERD IS AMBIGUOUS - FOR SPACING WE DONT
+    reject.c BUT DON'T REJECT IF THE WERD IS AMBIGUOUS - FOR SPACING WE DON'T
     CARE WHETHER WE HAVE of/at on/an etc.
   */
   if (fixsp_done_mode > 0 &&

diff --git a/ccmain/output.cpp b/ccmain/output.cpp
@@ -297,7 +297,7 @@ UNICHAR_ID Tesseract::get_rep_char(WERD_RES *word) {  // what char is repeated?
 /*************************************************************************
  * SUSPECT LEVELS
  *
- * 0 - dont reject ANYTHING
+ * 0 - don't reject ANYTHING
  * 1,2 - partial rejection
  * 3 - BEST
  *
@@ -337,7 +337,7 @@ void Tesseract::set_unlv_suspects(WERD_RES *word_res) {
   rating_per_ch = word.rating() / word_res->reject_map.length();
 
   if (rating_per_ch >= suspect_rating_per_ch)
-    return;                      //Dont touch bad ratings
+    return;                      //Don't touch bad ratings
 
   if ((word_res->tess_accepted) || (rating_per_ch < suspect_accept_rating)) {
     /* Unreject any Tess Acceptable word - but NOT tess reject chs*/

diff --git a/ccmain/paramsd.cpp b/ccmain/paramsd.cpp
@@ -329,13 +329,13 @@ void ParamsEditor::WriteParams(char *filename,
     fclose(fp);
     sprintf (msg_str, "Overwrite file " "%s" "? (Y/N)", filename);
     int a = sv_window_->ShowYesNoDialog(msg_str);
-    if (a == 'n') { return; }  // dont write
+    if (a == 'n') { return; }  // don't write
   }
 
 
   fp = fopen (filename, "wb");  // can we write to it?
   if (fp == NULL) {
-    sv_window_->AddMessage("Cant write to file " "%s" "", filename);
+    sv_window_->AddMessage("Can't write to file " "%s" "", filename);
     return;
   }
 

diff --git a/ccmain/reject.cpp b/ccmain/reject.cpp
@@ -521,7 +521,7 @@ BOOL8 Tesseract::word_contains_non_1_digit(const char *word,
 
 /*************************************************************************
  * dont_allow_1Il()
- * Dont unreject LONE accepted 1Il conflict set chars
+ * Don't unreject LONE accepted 1Il conflict set chars
  *************************************************************************/
 void Tesseract::dont_allow_1Il(WERD_RES *word) {
   int i = 0;
@@ -633,7 +633,7 @@ void Tesseract::flip_hyphens(WERD_RES *word_res) {
       next_left = 9999;
     else
       next_left = word_res->rebuild_word->blobs[i + 1]->bounding_box().left();
-    // Dont touch small or touching blobs - it is too dangerous.
+    // Don't touch small or touching blobs - it is too dangerous.
     if ((out_box.width() > 8 * word_res->denorm.x_scale()) &&
         (out_box.left() > prev_right) && (out_box.right() < next_left)) {
       aspect_ratio = out_box.width() / (float) out_box.height();

diff --git a/ccmain/tesseractclass.cpp b/ccmain/tesseractclass.cpp
@@ -136,7 +136,7 @@ Tesseract::Tesseract()
       BOOL_MEMBER(tessedit_fix_fuzzy_spaces, true,
                   "Try to improve fuzzy spaces", this->params()),
       BOOL_MEMBER(tessedit_unrej_any_wd, false,
-                  "Dont bother with word plausibility", this->params()),
+                  "Don't bother with word plausibility", this->params()),
       BOOL_MEMBER(tessedit_fix_hyphens, true, "Crunch double hyphens?",
                   this->params()),
       BOOL_MEMBER(tessedit_redo_xheight, true, "Check/Correct x-height",
@@ -310,19 +310,19 @@ Tesseract::Tesseract()
                  this->params()),
       INT_MEMBER(crunch_pot_indicators, 1,
                  "How many potential indicators needed", this->params()),
-      BOOL_MEMBER(crunch_leave_ok_strings, true, "Dont touch sensible strings",
+      BOOL_MEMBER(crunch_leave_ok_strings, true, "Don't touch sensible strings",
                   this->params()),
       BOOL_MEMBER(crunch_accept_ok, true, "Use acceptability in okstring",
                   this->params()),
       BOOL_MEMBER(crunch_leave_accept_strings, false,
-                  "Dont pot crunch sensible strings", this->params()),
+                  "Don't pot crunch sensible strings", this->params()),
       BOOL_MEMBER(crunch_include_numerals, false, "Fiddle alpha figures",
                   this->params()),
       INT_MEMBER(crunch_leave_lc_strings, 4,
-                 "Dont crunch words with long lower case strings",
+                 "Don't crunch words with long lower case strings",
                  this->params()),
       INT_MEMBER(crunch_leave_uc_strings, 4,
-                 "Dont crunch words with long lower case strings",
+                 "Don't crunch words with long lower case strings",
                  this->params()),
       INT_MEMBER(crunch_long_repetitions, 3,
                  "Crunch words with long repetitions", this->params()),
@@ -393,21 +393,21 @@ Tesseract::Tesseract()
       INT_MEMBER(suspect_space_level, 100,
                  "Min suspect level for rejecting spaces", this->params()),
       INT_MEMBER(suspect_short_words, 2,
-                 "Dont Suspect dict wds longer than this", this->params()),
+                 "Don't suspect dict wds longer than this", this->params()),
       BOOL_MEMBER(suspect_constrain_1Il, false, "UNLV keep 1Il chars rejected",
                   this->params()),
-      double_MEMBER(suspect_rating_per_ch, 999.9, "Dont touch bad rating limit",
+      double_MEMBER(suspect_rating_per_ch, 999.9, "Don't touch bad rating limit",
                     this->params()),
       double_MEMBER(suspect_accept_rating, -999.9, "Accept good rating limit",
                     this->params()),
       BOOL_MEMBER(tessedit_minimal_rejection, false,
                   "Only reject tess failures", this->params()),
-      BOOL_MEMBER(tessedit_zero_rejection, false, "Dont reject ANYTHING",
+      BOOL_MEMBER(tessedit_zero_rejection, false, "Don't reject ANYTHING",
                   this->params()),
       BOOL_MEMBER(tessedit_word_for_word, false,
                   "Make output have exactly one word per WERD", this->params()),
       BOOL_MEMBER(tessedit_zero_kelvin_rejection, false,
-                  "Dont reject ANYTHING AT ALL", this->params()),
+                  "Don't reject ANYTHING AT ALL", this->params()),
       BOOL_MEMBER(tessedit_consistent_reps, true,
                   "Force all rep chars the same", this->params()),
       INT_MEMBER(tessedit_reject_mode, 0, "Rejection algorithm",
@@ -424,7 +424,7 @@ Tesseract::Tesseract()
                   "Use DOC dawg in 11l conf. detector", this->params()),
       BOOL_MEMBER(rej_1Il_use_dict_word, false, "Use dictword test",
                   this->params()),
-      BOOL_MEMBER(rej_1Il_trust_permuter_type, true, "Dont double check",
+      BOOL_MEMBER(rej_1Il_trust_permuter_type, true, "Don't double check",
                   this->params()),
       BOOL_MEMBER(rej_use_tess_accepted, true, "Individual rejection control",
                   this->params()),
-Original file line number
+Diff line change
@@ Expand Up / @@ -59,6 +59,8 @@ training/wordlist2dawg @@
     *.o
     *.Plo
     *.a
+    *.class
+    *.jar
     # tessdata
     *.cube.*
@@ Expand Down @@