Fixed the extern C mismatches properly.

git-svn-id: https://tesseract-ocr.googlecode.com/svn/trunk@82 d0cd1f9f-072b-0410-8dd7-cf729c803f20

ccmain/tfacep.h

@@ -46,66 +46,8 @@
 typedef void (*TESS_TESTER) (TBLOB *, BOOL8, char *, INT32, LIST);
 typedef LIST (*TESS_MATCHER) (TBLOB *, TBLOB *, TBLOB *, void *, TEXTROW *);
 
-extern "C"
-{
-  /*
-  int							start_recog(				//Real main in C
-  int							argc,
-  char						*argv[]);
-  void						program_editup2(			//afterforking part
-  int							argc,
-  char**						argv);
-
-  int							end_recog(					//Real main in C
-  int							argc,
-  char						*argv[]);
-  void						set_interactive_pass();
-  void						set_pass1();
-  void						set_pass2();
-  //ARRAY						cc_recog(TWERD*,TESS_CHOICE*,TESS_CHOICE*,TESS_TESTER,
-  //										TESS_TESTER);*/
-  //void                                          wo_learn_blob(TBLOB*,TEXTROW*,char*,INT32);
-  //LIST                                  AdaptiveClassifier(TBLOB*,TBLOB*,TEXTROW*);
-  //void                                          LearnBlob(TBLOB*,TEXTROW*,char*,INT32);
-  //TWERD                                         *newword();
-  //TBLOB                                         *newblob();
-  //TESSLINE                                      *newoutline();
-  //EDGEPT                                                *newedgept();
-  //void                                          oldedgept(EDGEPT*);
-  //void                                          destroy_nodes(void*,void (*)(void*));
-  //TESS_LIST                                     *append_choice(TESS_LIST*,char*,double,double,char);
-  //void                                          fix_quotes (char*);
-  //void                                          record_certainty(double,int);
-  //int                                                   AcceptableResult(A_CHOICE*,A_CHOICE*);
-  //int                                                   AdaptableWord(TWERD*,const char*,const char*);
-  //void                                          delete_word(TWERD*);
-  //void                                          free_blob(TBLOB*);
-  //void                                          add_document_word(A_CHOICE*);
-  //void                                          AdaptToWord(TWERD*,TEXTROW*,const char*,const char*,const char*);
-  //void                                          SaveBadWord(const char*,double);
-  //void                                          free_choice(TESS_CHOICE*);
-  //TWERD                                         *newword();
-  //TBLOB                                         *newblob();
-  //void                                          free_blob(                                      //free a blob
-  //      TBLOB                                           *blob);                                         //blob to free
-
-  //int                                                   dict_word( const char* );
-
-  //extern int                                    tess_cn_matching;
-  //extern int                                    tess_bn_matching;
-  //extern int                                    last_word_on_line;
-  extern TEXTROW normalized_row;
-  //extern TESS_MATCHER                   blob_matchers[];
-  //extern FILE                                   *rawfile;
-  //extern FILE                                   *textfile;
-  //extern int                                    character_count;
-  //extern int                                    word_count;
-  //extern int                                    enable_assoc;
-  //extern int                                    chop_enable;
-  //extern int                                    permute_only_top;
-  extern int display_ratings;
-
-};
+extern TEXTROW normalized_row;
+extern int display_ratings;
 
 #if 0
 #define strsave(s)    \

ccutil/tessopt.cpp

@@ -23,8 +23,8 @@
 #include          "tessopt.h"
 #include          "notdll.h"     //must be last include
 
-int optind;
-char *optarg;
+int tessoptind;
+char *tessoptarg;
 
 /**********************************************************************
  * tessopt
@@ -37,22 +37,22 @@ INT32 argc,                      //arg count
 char *argv[],                    //args
 const char *arglist                    //string of arg chars
 ) {
-  char *arg;                     //arg char
+  const char *arg;                     //arg char
 
-  if (optind == 0)
-    optind = 1;
-  if (optind < argc && argv[optind][0] == '-') {
-    arg = strchr (arglist, argv[optind][1]);
+  if (tessoptind == 0)
+    tessoptind = 1;
+  if (tessoptind < argc && argv[tessoptind][0] == '-') {
+    arg = strchr (arglist, argv[tessoptind][1]);
     if (arg == NULL || *arg == ':')
       return '?';                //dud option
-    optind++;
-    optarg = argv[optind];
+    tessoptind++;
+    tessoptarg = argv[tessoptind];
     if (arg[1] == ':') {
-      if (argv[optind - 1][2] != '\0')
+      if (argv[tessoptind - 1][2] != '\0')
                                  //immediately after
-        optarg = argv[optind - 1] + 2;
+        tessoptarg = argv[tessoptind - 1] + 2;
       else
-        optind++;
+        tessoptind++;
     }
     return *arg;
   }

ccutil/tessopt.h

@@ -20,8 +20,8 @@
 #include          "host.h"
 #include          "notdll.h"     //must be last include
 
-extern int optind;
-extern char *optarg;
+extern int tessoptind;
+extern char *tessoptarg;
 
 int tessopt (                     //parse args
 INT32 argc,                      //arg count

classify/extern.h

@@ -28,7 +28,7 @@
  */
 
 #ifdef __cplusplus
-#define EXTERN extern "C"
+#define EXTERN extern
 #else
 #define EXTERN extern
 #endif

cutil/globals.h

@@ -39,14 +39,12 @@
 extern TBLOB *pageblobs;         /*first blob on page */
 extern TEXTBLOCK *pageblocks;    /*first block on page */
                                  /*class definitions */
-extern char classes[CLASSIZE][CLASSLENGTH];
+/* extern char classes[CLASSIZE][CLASSLENGTH]; */
 extern int resolution;           /*scanner res in dpi */
 extern int acts[MAXPROC];        /*action flags */
 extern int debugs[MAXPROC];      /*debug flags */
 extern int plots[MAXPROC];       /*plot flags */
 extern int corners[4];           /*corners of scan window */
-extern int optind;               /*option index */
-extern char *optarg;             /*option argument */
                                  /*image file name */
 extern char imagefile[FILENAMESIZE];
                                  /* main directory */

training/cnTraining.cpp

@@ -37,6 +37,7 @@
 #include <string.h>
 #include <stdio.h>
 #include <math.h>
+#include "unichar.h"
 
 #define MAXNAMESIZE	80
 #define MAX_NUM_SAMPLES	10000
@@ -219,21 +220,34 @@ int main (
 	ParseArguments (argc, argv);
 	while ((PageName = GetNextFilename()) != NULL)
 	{
-		printf ("\nReading %s ...", PageName);
+		printf ("Reading %s ...\n", PageName);
 		TrainingPage = Efopen (PageName, "r");
 		ReadTrainingSamples (TrainingPage, &CharList);
 		fclose (TrainingPage);
 		//WriteTrainingSamples (Directory, CharList);
 	}
+        printf("Clustering ...\n");
 	pCharList = CharList;
 	iterate(pCharList)
 	{
-		//Cluster
-		CharSample = (LABELEDLIST) first_node (pCharList);
-		printf ("\nClustering %s ...", CharSample->Label);
-		Clusterer = SetUpForClustering(CharSample);
-		ProtoList = ClusterSamples(Clusterer, &Config);
-		AddToNormProtosList(&NormProtoList, ProtoList, CharSample->Label);
+          //Cluster
+          CharSample = (LABELEDLIST) first_node (pCharList);
+          //printf ("\nClustering %s ...", CharSample->Label);
+          Clusterer = SetUpForClustering(CharSample);
+          float SavedMinSamples = Config.MinSamples;
+          while (Config.MinSamples > 0.001) {
+            ProtoList = ClusterSamples(Clusterer, &Config);
+            if (NumberOfProtos(ProtoList, 1, 0) > 0)
+              break;
+            else {
+              Config.MinSamples *= 0.95;
+              printf("0 significant protos for %s."
+                     " Retrying clustering with MinSamples = %f%%\n",
+                     CharSample->Label, Config.MinSamples);
+            }
+          }
+          Config.MinSamples = SavedMinSamples;
+          AddToNormProtosList(&NormProtoList, ProtoList, CharSample->Label);
 	}
 	FreeTrainingSamples (CharList);
 	WriteNormProtos (Directory, NormProtoList, Clusterer);
@@ -262,7 +276,7 @@ void ParseArguments(
 **		ShowSignificantProtos	flag controlling proto display
 **		ShowInsignificantProtos	flag controlling proto display
 **		Config			current clustering parameters
-**		optarg, optind		defined by tessopt sys call
+**		tessoptarg, tessoptind		defined by tessopt sys call
 **		Argc, Argv		global copies of argc and argv
 **	Operation:
 **		This routine parses the command line arguments that were
@@ -287,7 +301,6 @@ void ParseArguments(
 	int		Option;
 	int		ParametersRead;
 	BOOL8		Error;
-	extern char	*optarg;
 
 	Error = FALSE;
 	Argc = argc;
@@ -297,48 +310,48 @@ void ParseArguments(
 		switch ( Option )
 		{
 		case 'n':
-      sscanf(optarg,"%d", &ParametersRead);
+      sscanf(tessoptarg,"%d", &ParametersRead);
 			ShowInsignificantProtos = ParametersRead;
 			break;
 		case 'p':
-      sscanf(optarg,"%d", &ParametersRead);
+      sscanf(tessoptarg,"%d", &ParametersRead);
 			ShowSignificantProtos = ParametersRead;
 			break;
 		case 'd':
 			ShowAllSamples = FALSE;
 			break;
 		case 'C':
-			ParametersRead = sscanf( optarg, "%lf", &(Config.Confidence) );
+			ParametersRead = sscanf( tessoptarg, "%lf", &(Config.Confidence) );
 			if ( ParametersRead != 1 ) Error = TRUE;
 			else if ( Config.Confidence > 1 ) Config.Confidence = 1;
 			else if ( Config.Confidence < 0 ) Config.Confidence = 0;
 			break;
 		case 'I':
-			ParametersRead = sscanf( optarg, "%f", &(Config.Independence) );
+			ParametersRead = sscanf( tessoptarg, "%f", &(Config.Independence) );
 			if ( ParametersRead != 1 ) Error = TRUE;
 			else if ( Config.Independence > 1 ) Config.Independence = 1;
 			else if ( Config.Independence < 0 ) Config.Independence = 0;
 			break;
 		case 'M':
-			ParametersRead = sscanf( optarg, "%f", &(Config.MinSamples) );
+			ParametersRead = sscanf( tessoptarg, "%f", &(Config.MinSamples) );
 			if ( ParametersRead != 1 ) Error = TRUE;
 			else if ( Config.MinSamples > 1 ) Config.MinSamples = 1;
 			else if ( Config.MinSamples < 0 ) Config.MinSamples = 0;
 			break;
 		case 'B':
-			ParametersRead = sscanf( optarg, "%f", &(Config.MaxIllegal) );
+			ParametersRead = sscanf( tessoptarg, "%f", &(Config.MaxIllegal) );
 			if ( ParametersRead != 1 ) Error = TRUE;
 			else if ( Config.MaxIllegal > 1 ) Config.MaxIllegal = 1;
 			else if ( Config.MaxIllegal < 0 ) Config.MaxIllegal = 0;
 			break;
 		case 'R':
-			ParametersRead = sscanf( optarg, "%f", &RoundingAccuracy );
+			ParametersRead = sscanf( tessoptarg, "%f", &RoundingAccuracy );
 			if ( ParametersRead != 1 ) Error = TRUE;
 			else if ( RoundingAccuracy > 0.01 ) RoundingAccuracy = 0.01;
 			else if ( RoundingAccuracy < 0.0 ) RoundingAccuracy = 0.0;
 			break;
 		case 'S':
-			switch ( optarg[0] )
+			switch ( tessoptarg[0] )
 			{
 			case 's': Config.ProtoStyle = spherical; break;
 			case 'e': Config.ProtoStyle = elliptical; break;
@@ -348,10 +361,10 @@ void ParseArguments(
 			}
 			break;
 			case 'D':
-				Directory = optarg;
+				Directory = tessoptarg;
 				break;
 			case 'N':
-				if (sscanf (optarg, "%d", &MaxNumSamples) != 1 ||
+				if (sscanf (tessoptarg, "%d", &MaxNumSamples) != 1 ||
 					MaxNumSamples <= 0)
 					Error = TRUE;
 				break;
@@ -375,7 +388,7 @@ char *GetNextFilename ()
 /*
 **	Parameters: none
 **	Globals:
-**		optind			defined by tessopt sys call
+**		tessoptind			defined by tessopt sys call
 **		Argc, Argv		global copies of argc and argv
 **	Operation:
 **		This routine returns the next command line argument.  If
@@ -388,8 +401,8 @@ char *GetNextFilename ()
 */
 
 {
-	if (optind < Argc)
-		return (Argv [optind++]);
+	if (tessoptind < Argc)
+		return (Argv [tessoptind++]);
 	else
 		return (NULL);
 
@@ -417,32 +430,32 @@ void ReadTrainingSamples (
 */
 
 {
-	char			CharName[MAXNAMESIZE];
+	char		unichar[UNICHAR_LEN + 1];
 	LABELEDLIST	CharSample;
 	FEATURE_SET	FeatureSamples;
-	CHAR_DESC		CharDesc;
-	int			Type, i;
-
-	while (fscanf (File, "%s %s", FontName, CharName) == 2) {
-		CharSample = FindList (*TrainingSamples, CharName);
-		if (CharSample == NULL) {
-			CharSample = NewLabeledList (CharName);
-			*TrainingSamples = push (*TrainingSamples, CharSample);
-		}
-		CharDesc = ReadCharDescription (File);
-		Type = ShortNameToFeatureType(PROGRAM_FEATURE_TYPE);
-		FeatureSamples = FeaturesOfType(CharDesc, Type);
-    for (int feature = 0; feature < FeatureSamples->NumFeatures; ++feature) {
-      FEATURE f = FeatureSamples->Features[feature];
-      for (int dim =0; dim < f->Type->NumParams; ++dim)
-        f->Params[dim] += UniformRandomNumber(-MINSD, MINSD);
-    }
-		CharSample->List = push (CharSample->List, FeatureSamples);
-		for (i = 0; i < NumFeatureSetsIn (CharDesc); i++)
-			if (Type != i)
-				FreeFeatureSet (FeaturesOfType (CharDesc, i));
-		free (CharDesc);
-    }
+	CHAR_DESC	CharDesc;
+	int		Type, i;
+
+	while (fscanf (File, "%s %s", FontName, unichar) == 2) {
+          CharSample = FindList (*TrainingSamples, unichar);
+          if (CharSample == NULL) {
+            CharSample = NewLabeledList (unichar);
+            *TrainingSamples = push (*TrainingSamples, CharSample);
+          }
+          CharDesc = ReadCharDescription (File);
+          Type = ShortNameToFeatureType(PROGRAM_FEATURE_TYPE);
+          FeatureSamples = FeaturesOfType(CharDesc, Type);
+          for (int feature = 0; feature < FeatureSamples->NumFeatures; ++feature) {
+            FEATURE f = FeatureSamples->Features[feature];
+            for (int dim =0; dim < f->Type->NumParams; ++dim)
+              f->Params[dim] += UniformRandomNumber(-MINSD, MINSD);
+          }
+          CharSample->List = push (CharSample->List, FeatureSamples);
+          for (i = 0; i < NumFeatureSetsIn (CharDesc); i++)
+            if (Type != i)
+              FreeFeatureSet (FeaturesOfType (CharDesc, i));
+          free (CharDesc);
+        }
 }	// ReadTrainingSamples
 
 /*---------------------------------------------------------------------------*/
@@ -606,7 +619,6 @@ void WriteNormProtos (
 	char		Filename[MAXNAMESIZE];
 	LABELEDLIST LabeledProto;
 	int N;
-	char Label;
 
 	strcpy (Filename, "");
 	if (Directory != NULL)
@@ -623,9 +635,17 @@ void WriteNormProtos (
 	{
 		LabeledProto = (LABELEDLIST) first_node (LabeledProtoList);
 		N = NumberOfProtos(LabeledProto->List,
-			ShowSignificantProtos, ShowInsignificantProtos);
-		Label = NameToChar(LabeledProto->Label);
-		fprintf(File, "\n%c %d\n", Label, N);
+		ShowSignificantProtos, ShowInsignificantProtos);
+                if (N < 1) {
+                  printf ("\nError! Not enough protos for %s: %d protos"
+                          " (%d significant protos"
+                          ", %d insignificant protos)\n",
+                          LabeledProto->Label, N,
+                          NumberOfProtos(LabeledProto->List, 1, 0),
+                          NumberOfProtos(LabeledProto->List, 0, 1));
+                  exit(1);
+                }
+		fprintf(File, "\n%s %d\n", LabeledProto->Label, N);
 		WriteProtos(File, Clusterer->SampleSize, LabeledProto->List,
 			ShowSignificantProtos, ShowInsignificantProtos);
 	}

training/mfTraining.cpp

@@ -44,6 +44,9 @@
 #include "intproto.h"
 #include "variables.h"
 #include "freelist.h"
+#include "efio.h"
+#include "danerror.h"
+#include "globals.h"
 
 #include <string.h>
 #include <stdio.h>
@@ -73,7 +76,6 @@ typedef MERGE_CLASS_NODE* MERGE_CLASS;
 
 #define round(x,frag)(floor(x/frag+.5)*frag)
 
-
 /**----------------------------------------------------------------------------
 					Public Function Prototypes
 ----------------------------------------------------------------------------**/
@@ -164,21 +166,7 @@ void Normalize (
 void SetUpForFloat2Int(
 	LIST LabeledClassList);
 
-void WritePFFMTable(INT_TEMPLATES Templates, const char* filename) {
-  FILE* fp = Efopen(filename, "wb");
-  /* then write out each class */
-  for (int i = 0; i < NumClassesIn (Templates); i++) {
-    int MaxLength = 0;
-    INT_CLASS Class = ClassForIndex (Templates, i);
-    for (int ConfigId = 0; ConfigId < NumIntConfigsIn (Class); ConfigId++) {
-      if (LengthForConfigId (Class, ConfigId) > MaxLength)
-        MaxLength = LengthForConfigId (Class, ConfigId);
-    }
-    fprintf(fp, "%c %d\n", ClassIdForIndex(Templates, i), MaxLength);
-  }
-  fclose(fp);
-}
-
+void WritePFFMTable(INT_TEMPLATES Templates, const char* filename);
 
 //--------------Global Data Definitions and Declarations--------------
 static char FontName[MAXNAMESIZE];
@@ -200,6 +188,9 @@ static CLUSTERCONFIG Config =
 
 static FLOAT32 RoundingAccuracy = 0.0;
 
+// The unicharset used during mftraining
+static UNICHARSET unicharset_mftraining;
+
 /*----------------------------------------------------------------------------
 						Public Code
 -----------------------------------------------------------------------------*/
@@ -260,12 +251,17 @@ int main (
 	LIST pCharList, pProtoList;
 	char Filename[MAXNAMESIZE];
 
+        // Clean the unichar set
+        unicharset_mftraining.clear();
+        // Space character needed to represent NIL classification
+        unicharset_mftraining.unichar_insert(" ");
+
 	ParseArguments (argc, argv);
 	InitFastTrainerVars ();
 	InitSubfeatureVars ();
 	while ((PageName = GetNextFilename()) != NULL)
 	{
-		printf ("\nReading %s ...", PageName);
+		printf ("Reading %s ...\n", PageName);
 		TrainingPage = Efopen (PageName, "r");
 		CharList = ReadTrainingSamples (TrainingPage);
 		fclose (TrainingPage);
@@ -275,7 +271,7 @@ int main (
 		{
 			//Cluster
 			CharSample = (LABELEDLIST) first_node (pCharList);
-			printf ("\nClustering %s ...", CharSample->Label);
+// 			printf ("\nClustering %s ...", CharSample->Label);
 			Clusterer = SetUpForClustering(CharSample);
 			ProtoList = ClusterSamples(Clusterer, &Config);
 			//WriteClusteredTrainingSamples (Directory, ProtoList, Clusterer, CharSample);
@@ -320,14 +316,13 @@ int main (
 			FreeProtoList (&ProtoList);
 		}
 		FreeTrainingSamples (CharList);
-		printf ("\n");
 	}
 	//WriteMergedTrainingSamples(Directory,ClassList);
 	WriteMicrofeat(Directory, ClassList);
 	InitIntProtoVars ();
 	InitPrototypes ();
 	SetUpForFloat2Int(ClassList);
-	IntTemplates = CreateIntTemplates(TrainingData);
+	IntTemplates = CreateIntTemplates(TrainingData, unicharset_mftraining);
 	strcpy (Filename, "");
 	if (Directory != NULL)
 	{
@@ -340,11 +335,18 @@ int main (
 #else
 	OutFile = Efopen (Filename, "wb");
 #endif
-	WriteIntTemplates(OutFile, IntTemplates);
+	WriteIntTemplates(OutFile, IntTemplates, unicharset_mftraining);
 	fclose (OutFile);
-  // Now create pffmtable.
-  WritePFFMTable(IntTemplates, "pffmtable");
-	printf ("\nDone!\n"); /**/
+	strcpy (Filename, "");
+	if (Directory != NULL)
+	{
+		strcat (Filename, Directory);
+		strcat (Filename, "/");
+	}
+	strcat (Filename, "pffmtable");
+        // Now create pffmtable.
+        WritePFFMTable(IntTemplates, Filename);
+	printf ("Done!\n"); /**/
 	FreeLabeledClassList (ClassList);
   return 0;
 }	/* main */
@@ -367,7 +369,7 @@ char	**argv)
 **		ShowSignificantProtos	flag controlling proto display
 **		ShowInsignificantProtos	flag controlling proto display
 **		Config			current clustering parameters
-**		optarg, optind		defined by tessopt sys call
+**		tessoptarg, tessoptind		defined by tessopt sys call
 **		Argc, Argv		global copies of argc and argv
 **	Operation:
 **		This routine parses the command line arguments that were
@@ -392,7 +394,6 @@ char	**argv)
 	int		Option;
 	int		ParametersRead;
 	BOOL8		Error;
-	extern char	*optarg;
 
 	Error = FALSE;
 	Argc = argc;
@@ -411,37 +412,37 @@ char	**argv)
 			ShowAllSamples = FALSE;
 			break;
 		case 'C':
-			ParametersRead = sscanf( optarg, "%lf", &(Config.Confidence) );
+			ParametersRead = sscanf( tessoptarg, "%lf", &(Config.Confidence) );
 			if ( ParametersRead != 1 ) Error = TRUE;
 			else if ( Config.Confidence > 1 ) Config.Confidence = 1;
 			else if ( Config.Confidence < 0 ) Config.Confidence = 0;
 			break;
 		case 'I':
-			ParametersRead = sscanf( optarg, "%f", &(Config.Independence) );
+			ParametersRead = sscanf( tessoptarg, "%f", &(Config.Independence) );
 			if ( ParametersRead != 1 ) Error = TRUE;
 			else if ( Config.Independence > 1 ) Config.Independence = 1;
 			else if ( Config.Independence < 0 ) Config.Independence = 0;
 			break;
 		case 'M':
-			ParametersRead = sscanf( optarg, "%f", &(Config.MinSamples) );
+			ParametersRead = sscanf( tessoptarg, "%f", &(Config.MinSamples) );
 			if ( ParametersRead != 1 ) Error = TRUE;
 			else if ( Config.MinSamples > 1 ) Config.MinSamples = 1;
 			else if ( Config.MinSamples < 0 ) Config.MinSamples = 0;
 			break;
 		case 'B':
-			ParametersRead = sscanf( optarg, "%f", &(Config.MaxIllegal) );
+			ParametersRead = sscanf( tessoptarg, "%f", &(Config.MaxIllegal) );
 			if ( ParametersRead != 1 ) Error = TRUE;
 			else if ( Config.MaxIllegal > 1 ) Config.MaxIllegal = 1;
 			else if ( Config.MaxIllegal < 0 ) Config.MaxIllegal = 0;
 			break;
 		case 'R':
-			ParametersRead = sscanf( optarg, "%f", &RoundingAccuracy );
+			ParametersRead = sscanf( tessoptarg, "%f", &RoundingAccuracy );
 			if ( ParametersRead != 1 ) Error = TRUE;
 			else if ( RoundingAccuracy > 0.01 ) RoundingAccuracy = 0.01;
 			else if ( RoundingAccuracy < 0.0 ) RoundingAccuracy = 0.0;
 			break;
 		case 'S':
-			switch ( optarg[0] )
+			switch ( tessoptarg[0] )
 			{
 			case 's': Config.ProtoStyle = spherical; break;
 			case 'e': Config.ProtoStyle = elliptical; break;
@@ -451,10 +452,10 @@ char	**argv)
 			}
 			break;
 			case 'D':
-				Directory = optarg;
+				Directory = tessoptarg;
 				break;
 			case 'N':
-				if (sscanf (optarg, "%d", &MaxNumSamples) != 1 ||
+				if (sscanf (tessoptarg, "%d", &MaxNumSamples) != 1 ||
 					MaxNumSamples <= 0)
 					Error = TRUE;
 				break;
@@ -478,7 +479,7 @@ char *GetNextFilename ()
 /*
 **	Parameters: none
 **	Globals:
-**		optind			defined by tessopt sys call
+**		tessoptind			defined by tessopt sys call
 **		Argc, Argv		global copies of argc and argv
 **	Operation:
 **		This routine returns the next command line argument.  If
@@ -491,8 +492,8 @@ char *GetNextFilename ()
 */
 
 {
-	if (optind < Argc)
-		return (Argv [optind++]);
+	if (tessoptind < Argc)
+		return (Argv [tessoptind++]);
 	else
 		return (NULL);
 
@@ -519,33 +520,41 @@ LIST ReadTrainingSamples (
 */
 
 {
-	char			CharName[MAXNAMESIZE];
-	LABELEDLIST	CharSample;
-  FEATURE_SET FeatureSamples;
+	char			unichar[UNICHAR_LEN + 1];
+	LABELEDLIST             CharSample;
+        FEATURE_SET             FeatureSamples;
 	LIST			TrainingSamples = NIL;
 	CHAR_DESC		CharDesc;
 	int			Type, i;
 
-	while (fscanf (File, "%s %s", FontName, CharName) == 2) {
-		CharSample = FindList (TrainingSamples, CharName);
+	while (fscanf (File, "%s %s", FontName, unichar) == 2) {
+          if (!unicharset_mftraining.contains_unichar(unichar)) {
+            unicharset_mftraining.unichar_insert(unichar);
+            if (unicharset_mftraining.size() > MAX_NUM_CLASSES) {
+              cprintf("Error: Size of unicharset of mftraining is "
+                      "greater than MAX_NUM_CLASSES\n");
+              exit(1);
+            }
+          }
+		CharSample = FindList (TrainingSamples, unichar);
 		if (CharSample == NULL) {
-			CharSample = NewLabeledList (CharName);
+			CharSample = NewLabeledList (unichar);
 			TrainingSamples = push (TrainingSamples, CharSample);
 		}
 		CharDesc = ReadCharDescription (File);
 		Type = ShortNameToFeatureType(PROGRAM_FEATURE_TYPE);
 		FeatureSamples = FeaturesOfType(CharDesc, Type);
-    for (int feature = 0; feature < FeatureSamples->NumFeatures; ++feature) {
-      FEATURE f = FeatureSamples->Features[feature];
-      for (int dim =0; dim < f->Type->NumParams; ++dim)
-        f->Params[dim] += UniformRandomNumber(-MINSD, MINSD);
-    }
+                for (int feature = 0; feature < FeatureSamples->NumFeatures; ++feature) {
+                  FEATURE f = FeatureSamples->Features[feature];
+                  for (int dim =0; dim < f->Type->NumParams; ++dim)
+                    f->Params[dim] += UniformRandomNumber(-MINSD, MINSD);
+                }
 		CharSample->List = push (CharSample->List, FeatureSamples);
 		for (i = 0; i < NumFeatureSetsIn (CharDesc); i++)
-			if (Type != i)
-				FreeFeatureSet (FeaturesOfType (CharDesc, i));
+                  if (Type != i)
+                    FreeFeatureSet (FeaturesOfType (CharDesc, i));
 		free (CharDesc);
-    }
+        }
 	return (TrainingSamples);
 
 }	/* ReadTrainingSamples */
@@ -843,7 +852,7 @@ void WriteProtos(
 	int i;
 	PROTO Proto;
 
-	fprintf(File, "%c\n", NameToChar(MergeClass->Label));
+	fprintf(File, "%s\n", MergeClass->Label);
 	fprintf(File, "%d\n", NumProtosIn(MergeClass->Class));
 	for(i=0; i < NumProtosIn(MergeClass->Class); i++)
 	{
@@ -900,7 +909,7 @@ void FreeTrainingSamples (
 	LIST		FeatureList;
 
 
-	printf ("\nFreeTrainingSamples...");
+// 	printf ("FreeTrainingSamples...\n");
 	iterate (CharList) 		/* iterate thru all of the fonts */
 	{
 		CharSample = (LABELEDLIST) first_node (CharList);
@@ -1161,12 +1170,13 @@ void SetUpForFloat2Int(
 	BIT_VECTOR		NewConfig;
 	BIT_VECTOR		OldConfig;
 
-	printf("Float2Int ...");
+// 	printf("Float2Int ...\n");
 
 	iterate(LabeledClassList)
 	{
 		MergeClass = (MERGE_CLASS) first_node (LabeledClassList);
-		Class = &TrainingData[NameToChar(MergeClass->Label)];
+		Class = &TrainingData[unicharset_mftraining.unichar_to_id(
+                                          MergeClass->Label)];
 		NumProtos = NumProtosIn(MergeClass->Class);
 		NumConfigs = NumConfigsIn(MergeClass->Class);
 
@@ -1204,3 +1214,20 @@ void SetUpForFloat2Int(
 		}
 	}
 } // SetUpForFloat2Int
+
+/*--------------------------------------------------------------------------*/
+void WritePFFMTable(INT_TEMPLATES Templates, const char* filename) {
+  FILE* fp = Efopen(filename, "wb");
+  /* then write out each class */
+  for (int i = 0; i < NumClassesIn (Templates); i++) {
+    int MaxLength = 0;
+    INT_CLASS Class = ClassForIndex (Templates, i);
+    for (int ConfigId = 0; ConfigId < NumIntConfigsIn (Class); ConfigId++) {
+      if (LengthForConfigId (Class, ConfigId) > MaxLength)
+        MaxLength = LengthForConfigId (Class, ConfigId);
+    }
+    fprintf(fp, "%s %d\n", unicharset_mftraining.id_to_unichar(
+                ClassIdForIndex(Templates, i)), MaxLength);
+  }
+  fclose(fp);
+} // WritePFFMTable

training/unicharset_extractor.cpp

@@ -52,8 +52,8 @@ int main(int argc, char** argv) {
   while ((option = tessopt(argc, argv, "D" )) != EOF) {
     switch (option) {
       case 'D':
-        output_directory = optarg;
-        ++optind;
+        output_directory = tessoptarg;
+        ++tessoptind;
         break;
     }
   }
@@ -64,12 +64,12 @@ int main(int argc, char** argv) {
   unicharset_file_name += kUnicharsetFileName;
 
   // Load box files
-  for (; optind < argc; ++optind) {
-    printf("Extracting unicharset from %s\n", argv[optind]);
+  for (; tessoptind < argc; ++tessoptind) {
+    printf("Extracting unicharset from %s\n", argv[tessoptind]);
 
-    FILE* box_file = fopen(argv[optind], "r");
+    FILE* box_file = fopen(argv[tessoptind], "r");
     if (box_file == NULL) {
-      printf("Cannot open box file %s\n", argv[optind]);
+      printf("Cannot open box file %s\n", argv[tessoptind]);
       return -1;
     }