Simplify function LoadTrainingData and fix mastertrainer_test

Signed-off-by: Stefan Weil <sw@weilnetz.de>
UB-Mannheim · Sep 17, 2021 · 6380451 · 6380451
1 parent 2b4c359
commit 6380451
Show file tree

Hide file tree

Showing 7 changed files with 10 additions and 37 deletions.
diff --git a/src/training/classifier_tester.cpp b/src/training/classifier_tester.cpp
@@ -102,7 +102,7 @@ int main(int argc, char **argv) {
   tesseract::CheckSharedLibraryVersion();
   ParseArguments(&argc, &argv);
   std::string file_prefix;
-  auto trainer = tesseract::LoadTrainingData(argc, argv, false, nullptr, file_prefix);
+  auto trainer = tesseract::LoadTrainingData(argv + 1, false, nullptr, file_prefix);
   tesseract::TessBaseAPI *api;
   // Decode the classifier string.
   tesseract::ShapeClassifier *shape_classifier =

diff --git a/src/training/cntraining.cpp b/src/training/cntraining.cpp
@@ -106,7 +106,6 @@ int main(int argc, char *argv[]) {
   // Set the global Config parameters before parsing the command line.
   Config = CNConfig;
 
-  const char *PageName;
   LIST CharList = NIL_LIST;
   CLUSTERER *Clusterer = nullptr;
   LIST ProtoList = NIL_LIST;
@@ -118,8 +117,7 @@ int main(int argc, char *argv[]) {
 
   ParseArguments(&argc, &argv);
   int num_fonts = 0;
-  int tessoptind = 1;
-  while ((PageName = GetNextFilename(argc, argv, tessoptind)) != nullptr) {
+  for (const char *PageName = *++argv; PageName != nullptr; PageName = *++argv) {
     printf("Reading %s ...\n", PageName);
     FILE *TrainingPage = fopen(PageName, "rb");
     ASSERT_HOST(TrainingPage);

diff --git a/src/training/common/commontraining.cpp b/src/training/common/commontraining.cpp
@@ -197,7 +197,7 @@ void WriteShapeTable(const std::string &file_prefix, const ShapeTable &shape_tab
  * If shape_table is not nullptr, but failed to load, make a fake flat one,
  * as shape clustering was not run.
  */
-std::unique_ptr<MasterTrainer> LoadTrainingData(int argc, const char *const *argv, bool replication,
+std::unique_ptr<MasterTrainer> LoadTrainingData(const char *const *filelist, bool replication,
                                                 ShapeTable **shape_table, std::string &file_prefix) {
   InitFeatureDefs(&feature_defs);
   InitIntegerFX();
@@ -236,10 +236,8 @@ std::unique_ptr<MasterTrainer> LoadTrainingData(int argc, const char *const *arg
     }
   }
   trainer->SetFeatureSpace(fs);
-  const char *page_name;
-  // Load training data from .tr files on the command line.
-  int tessoptind = 1;
-  while ((page_name = GetNextFilename(argc, argv, tessoptind)) != nullptr) {
+  // Load training data from .tr files in filelist (terminated by nullptr).
+  for (const char *page_name = *filelist++; page_name != nullptr; page_name = *filelist++) {
     tprintf("Reading %s ...\n", page_name);
     trainer->ReadTrainingSamples(page_name, feature_defs, false);
 
@@ -291,25 +289,6 @@ std::unique_ptr<MasterTrainer> LoadTrainingData(int argc, const char *const *arg
   return trainer;
 }
 
-/*---------------------------------------------------------------------------*/
-/**
- * This routine returns the next command line argument.  If
- * there are no remaining command line arguments, it returns
- * nullptr.  This routine should only be called after all option
- * arguments have been parsed and removed with ParseArguments.
- *
- * Globals:
- * - tessoptind defined by tessopt sys call
- * @return Next command line argument or nullptr.
- */
-const char *GetNextFilename(int argc, const char *const *argv, int &tessoptind) {
-  if (tessoptind < argc) {
-    return argv[tessoptind++];
-  } else {
-    return nullptr;
-  }
-} /* GetNextFilename */
-
 /*---------------------------------------------------------------------------*/
 /**
  * This routine searches through a list of labeled lists to find

diff --git a/src/training/common/commontraining.h b/src/training/common/commontraining.h
@@ -121,12 +121,9 @@ void WriteShapeTable(const std::string &file_prefix, const ShapeTable &shape_tab
 // If shape_table is not nullptr, but failed to load, make a fake flat one,
 // as shape clustering was not run.
 TESS_COMMON_TRAINING_API
-std::unique_ptr<MasterTrainer> LoadTrainingData(int argc, const char *const *argv, bool replication,
+std::unique_ptr<MasterTrainer> LoadTrainingData(const char *const *filelist, bool replication,
                                                 ShapeTable **shape_table, std::string &file_prefix);
 
-TESS_COMMON_TRAINING_API
-const char *GetNextFilename(int argc, const char *const *argv, int &tessoptind);
-
 LABELEDLIST FindList(tesseract::LIST List, const std::string &Label);
 
 TESS_COMMON_TRAINING_API

diff --git a/src/training/mftraining.cpp b/src/training/mftraining.cpp
@@ -198,7 +198,7 @@ int main(int argc, char **argv) {
   ShapeTable *shape_table = nullptr;
   std::string file_prefix;
   // Load the training data.
-  auto trainer = tesseract::LoadTrainingData(argc, argv, false, &shape_table, file_prefix);
+  auto trainer = tesseract::LoadTrainingData(argv + 1, false, &shape_table, file_prefix);
   if (trainer == nullptr) {
     return 1; // Failed.
   }

diff --git a/src/training/shapeclustering.cpp b/src/training/shapeclustering.cpp
@@ -47,7 +47,7 @@ int main(int argc, char **argv) {
   ParseArguments(&argc, &argv);
 
   std::string file_prefix;
-  auto trainer = tesseract::LoadTrainingData(argc, argv, false, nullptr, file_prefix);
+  auto trainer = tesseract::LoadTrainingData(argv + 1, false, nullptr, file_prefix);
 
   if (!trainer) {
     return 1;

diff --git a/unittest/mastertrainer_test.cc b/unittest/mastertrainer_test.cc
@@ -173,12 +173,11 @@ class MasterTrainerTest : public testing::Test {
     FLAGS_X = TestDataNameToPath("eng.xheights").c_str();
     FLAGS_U = TestDataNameToPath("eng.unicharset").c_str();
     std::string tr_file_name(TestDataNameToPath("eng.Arial.exp0.tr"));
-    const char *argv[] = {tr_file_name.c_str()};
-    int argc = 1;
+    const char *filelist[] = {tr_file_name.c_str(), nullptr};
     std::string file_prefix;
     delete shape_table_;
     shape_table_ = nullptr;
-    master_trainer_ = LoadTrainingData(argc, argv, false, &shape_table_, file_prefix);
+    master_trainer_ = LoadTrainingData(filelist, false, &shape_table_, file_prefix);
     EXPECT_TRUE(master_trainer_ != nullptr);
     EXPECT_TRUE(shape_table_ != nullptr);
   }