Fixed issues 1081-1090

git-svn-id: https://tesseract-ocr.googlecode.com/svn/trunk@1046 d0cd1f9f-072b-0410-8dd7-cf729c803f20

api/baseapi.cpp

@@ -1428,10 +1428,10 @@ char* TessBaseAPI::GetHOCRText(int page_number) {
         break;
     }
     hocr_str += ">";
-    const char *font_name;
     bool bold, italic, underlined, monospace, serif, smallcaps;
     int pointsize, font_id;
-    font_name = res_it->WordFontAttributes(&bold, &italic, &underlined,
+    // TODO(rays): Is hOCR interested in the font name?
+    (void) res_it->WordFontAttributes(&bold, &italic, &underlined,
                                       &monospace, &serif, &smallcaps,
                                       &pointsize, &font_id);
     bool last_word_in_line = res_it->IsAtFinalElement(RIL_TEXTLINE, RIL_WORD);

api/baseapi.h

@@ -659,12 +659,6 @@ class TESS_API TessBaseAPI {
    */
   void SetProbabilityInContextFunc(ProbabilityInContextFunc f);
 
-
-  /** Sets Dict::params_model_classify_ function to point to the given
-   * function.
-   */
-  void SetParamsModelClassifyFunc(ParamsModelClassifyFunc f);
-
   /** Sets Wordrec::fill_lattice_ function to point to the given function. */
   void SetFillLatticeFunc(FillLatticeFunc f);
 
@@ -704,9 +698,6 @@ class TESS_API TessBaseAPI {
   /** Return the number of dawgs loaded into tesseract_ object. */
   int NumDawgs() const;
 
-  /** Return the language used in the last valid initialization. */
-  const char* GetLastInitLanguage() const;
-
   /** Returns a ROW object created from the input row specification. */
   static ROW *MakeTessOCRRow(float baseline, float xheight,
                              float descender, float ascender);

ccmain/docqual.cpp

@@ -322,7 +322,7 @@ void Tesseract::doc_and_block_rejection(  //reject big chunks
 
         /* Walk rows in block testing for row rejection */
         row_no = 0;
-        while ((word = page_res_it.word()) != NULL &&
+        while (page_res_it.word() != NULL &&
                page_res_it.block() == current_block) {
           current_row = page_res_it.row();
           row_no++;

ccmain/osdetect.h

@@ -96,7 +96,6 @@ class ScriptDetector {
   ScriptDetector(const GenericVector<int>* allowed_scripts,
                  OSResults* osr, tesseract::Tesseract* tess);
   void detect_blob(BLOB_CHOICE_LIST* scores);
-  void get_script() ;
   bool must_stop(int orientation);
  private:
   OSResults* osr_;

ccmain/pgedit.cpp

@@ -728,15 +728,14 @@ BOOL8 Tesseract:: word_blank_and_set_display(BLOCK* block, ROW* row,
  * Normalize word and display in word window
  */
 BOOL8 Tesseract::word_bln_display(BLOCK* block, ROW* row, WERD_RES* word_res) {
-  TWERD *bln_word = word_res->chopped_word;
+  if (word_res->chopped_word == NULL) {
-  if (bln_word == NULL) {
+    // Setup word normalization parameters.
     word_res->SetupForRecognition(unicharset, this, BestPix(),
                                   tessedit_ocr_engine_mode, NULL,
                                   classify_bln_numeric_mode,
                                   textord_use_cjk_fp_model,
                                   poly_allow_detailed_fx,
                                   row, block);
-    bln_word = word_res->chopped_word;
   }
   bln_word_window_handle()->Clear();
   display_bln_lines(bln_word_window_handle(), ScrollView::CYAN,

ccstruct/boxread.cpp

@@ -130,7 +130,8 @@ bool ParseBoxFileStr(const char* boxfile_str, int* page_number,
   // Test for long space-delimited string label.
   if (strcmp(uch, kMultiBlobLabelCode) == 0 &&
       (buffptr = strchr(buffptr, '#')) != NULL) {
-    strncpy(uch, buffptr + 1, kBoxReadBufSize);
+    strncpy(uch, buffptr + 1, kBoxReadBufSize - 1);
+    uch[kBoxReadBufSize - 1] = '\0';  // Prevent buffer overrun.
     chomp_string(uch);
     uch_len = strlen(uch);
   }

classify/adaptmatch.cpp

@@ -1146,15 +1146,12 @@ void Classify::ExpandShapesAndApplyCorrections(
   if (classes != NULL) {
     // Adapted result.
     fontinfo_id = GetFontinfoId(classes[class_id], int_result.Config);
-    if (int_result.Config2 >= 0)
     fontinfo_id2 = GetFontinfoId(classes[class_id], int_result.Config2);
   } else {
     // Pre-trained result.
     fontinfo_id = ClassAndConfigIDToFontOrShapeID(class_id, int_result.Config);
-    if (int_result.Config2 >= 0) {
     fontinfo_id2 = ClassAndConfigIDToFontOrShapeID(class_id,
                                                    int_result.Config2);
-    }
     if (shape_table_ != NULL) {
       // Actually fontinfo_id is an index into the shape_table_ and it
       // contains a list of unchar_id/font_id pairs.

classify/intmatcher.h

@@ -42,6 +42,8 @@ struct INT_RESULT_STRUCT {
   INT_RESULT_STRUCT() : Rating(0.0f), Config(0), Config2(0), FeatureMisses(0) {}
 
   FLOAT32 Rating;
+  // TODO(rays) It might be desirable for these to be able to represent a
+  // null config.
   uinT8 Config;
   uinT8 Config2;
   uinT16 FeatureMisses;

classify/mastertrainer.cpp

@@ -462,6 +462,7 @@ bool MasterTrainer::AddSpacingInfo(const char *filename) {
       if (fscanf(fontinfo_file, "%s %d", kerned_uch, &x_gap) != 2) {
         tprintf("Bad format of font spacing file %s\n", filename);
         fclose(fontinfo_file);
+        delete spacing;
         return false;
       }
       if (!valid || !unicharset_.contains_unichar(kerned_uch)) continue;

cube/bmp_8.cpp

@@ -223,6 +223,7 @@ Bmp8 * Bmp8::FromCharDumpFile(CachedFile *fp) {
 
   if (bmp_obj->LoadFromCharDumpFile(fp) == false) {
     delete bmp_obj;
+    return NULL;
   }
 
   return bmp_obj;
@@ -320,6 +321,7 @@ Bmp8 * Bmp8::FromCharDumpFile(FILE *fp) {
 
   if (bmp_obj->LoadFromCharDumpFile(fp) == false) {
     delete bmp_obj;
+    return NULL;
   }
 
   return bmp_obj;

cube/char_bigrams.cpp

@@ -83,6 +83,7 @@ CharBigrams *CharBigrams::Create(const string &data_file_path,
     if (sscanf(str_vec[big].c_str(), "%d %x %x", &cnt, &ch1, &ch2) != 3) {
       fprintf(stderr, "Cube ERROR (CharBigrams::Create): invalid format "
               "reading line: %s\n", str_vec[big].c_str());
+      delete char_bigrams_obj;
       return NULL;
     }
 
@@ -117,6 +118,7 @@ CharBigrams *CharBigrams::Create(const string &data_file_path,
       if (bigram == NULL) {
         fprintf(stderr, "Cube ERROR (CharBigrams::Create): error allocating "
                 "memory for bigram.\n");
+        delete char_bigrams_obj;
         return NULL;
       }
 

textord/tablerecog.cpp

@@ -280,7 +280,11 @@ double StructuredTable::CalculateCellFilledPercentage(int row, int column) {
     if (text->IsTextType())
       area_covered += text->bounding_box().intersection(kCellBox).area();
   }
-  return MIN(1.0, area_covered / kCellBox.area());
+  const inT32 current_area = kCellBox.area();
+  if (current_area == 0) {
+    return 1.0;
+  }
+  return MIN(1.0, area_covered / current_area);
 }
 
 void StructuredTable::Display(ScrollView* window, ScrollView::Color color) {

textord/topitch.cpp

@@ -1037,6 +1037,7 @@ BOOL8 fixed_pitch_row(TO_ROW *row,       // row to do
         break;
       case PITCH_MAYBE_FIXED:
         res_string = "MF";
+        break;
       default:
         res_string = "??";
     }