Replace some tprintf by tesserr stream (fixes Windows compiler warnings)

Signed-off-by: Stefan Weil <sw@weilnetz.de>

src/ccmain/applybox.cpp

@@ -26,6 +26,7 @@
 #include <tesseract/unichar.h>
 #include "pageres.h"
 #include "tesseractclass.h"
+#include "tesserrstream.h"  // for tesserr
 #include "unicharset.h"
 
 #ifndef DISABLED_LEGACY_ENGINE
@@ -652,9 +653,10 @@ void Tesseract::SearchForText(const std::vector<BLOB_CHOICE_LIST *> *choices, in
     if (choices_pos + length == choices_length && text_index + 1 == target_text.size()) {
       // This is a complete match. If the rating is good record a new best.
       if (applybox_debug > 2) {
-        tprintf("Complete match, rating = %g, best=%g, seglength=%zu, best=%zu\n",
-                rating + choice_rating, *best_rating, segmentation->size(),
-                best_segmentation->size());
+        tesserr << "Complete match, rating = " << rating + choice_rating
+                << ", best=" << *best_rating
+                << ", seglength=" << segmentation->size()
+                << ", best=" << best_segmentation->size() << '\n';
       }
       if (best_segmentation->empty() || rating + choice_rating < *best_rating) {
         *best_segmentation = *segmentation;

src/ccmain/paragraphs.cpp

@@ -31,6 +31,7 @@
 #include "ratngs.h"              // for WERD_CHOICE
 #include "rect.h"                // for TBOX
 #include "statistc.h"            // for STATS
+#include "tesserrstream.h"       // for tesserr
 #include "tprintf.h"             // for tprintf
 #include "unicharset.h"          // for UNICHARSET
 #include "werd.h"                // for WERD, W_REP_CHAR
@@ -74,8 +75,8 @@ static bool AcceptableRowArgs(int debug_level, int min_num_rows, const char *fun
                               const std::vector<RowScratchRegisters> *rows, int row_start,
                               int row_end) {
   if (row_start < 0 || static_cast<size_t>(row_end) > rows->size() || row_start > row_end) {
-    tprintf("Invalid arguments rows[%d, %d) while rows is of size %zu.\n", row_start, row_end,
-            rows->size());
+    tesserr << "Invalid arguments rows[" << row_start << ", " << row_end
+            << ") while rows is of size " << rows->size() << ".\n";
     return false;
   }
   if (row_end - row_start < min_num_rows) {
@@ -915,10 +916,9 @@ struct GeometricClassifierState {
     tolerance = InterwordSpace(*r, r_start, r_end);
     CalculateTabStops(r, r_start, r_end, tolerance, &left_tabs, &right_tabs);
     if (debug_level >= 3) {
-      tprintf(
-          "Geometry: TabStop cluster tolerance = %d; "
-          "%zu left tabs; %zu right tabs\n",
-          tolerance, left_tabs.size(), right_tabs.size());
+      tesserr << "Geometry: TabStop cluster tolerance = " << tolerance << "; "
+              << left_tabs.size() << " left tabs; "
+              << right_tabs.size() << " right tabs\n";
     }
     ltr = (*r)[r_start].ri_->ltr;
   }

src/ccstruct/detlinefit.cpp

@@ -19,7 +19,7 @@
 #include "detlinefit.h"
 #include "helpers.h"        // for IntCastRounded
 #include "statistc.h"
-#include "tprintf.h"
+#include "tesserrstream.h"  // for tesserr
 
 #include <algorithm>
 #include <cfloat> // for FLT_MAX
@@ -143,13 +143,17 @@ double DetLineFit::ConstrainedFit(const FCOORD &direction, double min_dist, doub
   std::nth_element(distances_.begin(), distances_.begin() + median_index, distances_.end());
   *line_pt = distances_[median_index].data();
   if (debug) {
-    tprintf("Constrained fit to dir %g, %g = %d, %d :%zu distances:\n", direction.x(), direction.y(),
-            line_pt->x(), line_pt->y(), distances_.size());
+    tesserr << "Constrained fit to dir " << direction.x() << ", "
+            << direction.y() << " = "
+            << line_pt->x() << ", " << line_pt->y()
+            << " :" << distances_.size() << " distances:\n";
     for (unsigned i = 0; i < distances_.size(); ++i) {
-      tprintf("%d: %d, %d -> %g\n", i, distances_[i].data().x(), distances_[i].data().y(),
-              distances_[i].key());
+      tesserr << i << ": "
+              << distances_[i].data().x() << ", "
+              << distances_[i].data().y() << " -> "
+              << distances_[i].key() << '\n';
     }
-    tprintf("Result = %zu\n", median_index);
+    tesserr << "Result = " << median_index << '\n';
   }
   // Center distances on the fitted point.
   double dist_origin = direction * *line_pt;

src/ccstruct/imagedata.cpp

@@ -27,6 +27,7 @@
 #include "rect.h"       // for TBOX
 #include "scrollview.h" // for ScrollView, ScrollView::CYAN, ScrollView::NONE
 #include "tprintf.h"    // for tprintf
+#include "tesserrstream.h" // for tesserr
 
 #include "helpers.h"  // for IntCastRounded, TRand, ClipToRange, Modulo
 #include "serialis.h" // for TFile
@@ -618,9 +619,10 @@ bool DocumentData::ReCachePages() {
     pages_.clear();
   } else if (loaded_pages > 1) {
     // Avoid lots of messages for training with single line images.
-    tprintf("Loaded %zu/%d lines (%d-%zu) of document %s\n", pages_.size(),
-            loaded_pages, pages_offset_ + 1, pages_offset_ + pages_.size(),
-            document_name_.c_str());
+    tesserr << "Loaded " << pages_.size() << '/' << loaded_pages << " lines ("
+            << pages_offset_ + 1 << '-'
+            << pages_offset_ + pages_.size() << ") of document "
+            << document_name_ << '\n';
   }
   set_total_pages(loaded_pages);
   return !pages_.empty();

src/classify/shapetable.h

@@ -26,6 +26,7 @@
 #include "fontinfo.h"
 #include "genericheap.h"
 #include "intmatcher.h"
+#include "tesserrstream.h"  // for tesserr
 
 namespace tesseract {
 
@@ -41,10 +42,10 @@ struct UnicharRating {
 
   // Print debug info.
   void Print() const {
-    tprintf(
-        "Unichar-id=%d, rating=%g, adapted=%d, config=%d, misses=%u,"
-        " %zu fonts\n",
-        unichar_id, static_cast<double>(rating), adapted, config, feature_misses, fonts.size());
+    tesserr << "Unichar-id=" << unichar_id << ", rating=" << rating
+            << ", adapted=" << adapted << ", config=" << config
+            << ", misses=" << feature_misses << ", "
+            << fonts.size() << " fonts\n";
   }
 
   // Helper function to get the index of the first result with the required

src/dict/dict.cpp

@@ -18,6 +18,7 @@
 
 #include "dict.h"
 
+#include "tesserrstream.h"  // for tesserr
 #include "tprintf.h"
 
 #include <cstdio>
@@ -410,10 +411,10 @@ int Dict::def_letter_is_okay(void *void_dawg_args, const UNICHARSET &unicharset,
   ASSERT_HOST(unicharset.contains_unichar_id(unichar_id));
 
   if (dawg_debug_level >= 3) {
-    tprintf(
-        "def_letter_is_okay: current unichar=%s word_end=%d"
-        " num active dawgs=%zu\n",
-        getUnicharset().debug_str(unichar_id).c_str(), word_end, dawg_args->active_dawgs->size());
+    tesserr << "def_letter_is_okay: current unichar="
+            << getUnicharset().debug_str(unichar_id)
+            << " word_end=" << word_end
+            << " num active dawgs=" << dawg_args->active_dawgs->size() << '\n';
   }
 
   // Do not accept words that contain kPatternUnicharID.

src/lstm/series.cpp

@@ -20,6 +20,7 @@
 #include "fullyconnected.h"
 #include "networkscratch.h"
 #include "scrollview.h"
+#include "tesserrstream.h"  // for tesserr
 #include "tprintf.h"
 
 namespace tesseract {
@@ -164,7 +165,8 @@ void Series::SplitAt(unsigned last_start, Series **start, Series **end) {
   *start = nullptr;
   *end = nullptr;
   if (last_start >= stack_.size()) {
-    tprintf("Invalid split index %u must be in range [0,%zu]!\n", last_start, stack_.size() - 1);
+    tesserr << "Invalid split index " << last_start
+            << " must be in range [0," << stack_.size() - 1 << "]!\n";
     return;
   }
   auto *master_series = new Series("MasterSeries");

src/textord/baselinedetect.cpp

@@ -34,6 +34,7 @@
 #include "helpers.h"
 #include "linlsq.h"
 #include "makerow.h"
+#include "tesserrstream.h"  // for tesserr
 #include "textord.h"
 #include "tprintf.h"
 #include "underlin.h"
@@ -671,9 +672,10 @@ bool BaselineBlock::ComputeLineSpacing() {
     }
   }
   if (debug_level_ > 0) {
-    tprintf("Spacing %g, in %zu rows, %d gaps fitted out of %d non-trivial\n",
-            line_spacing_, row_positions.size(), fitting_gaps,
-            non_trivial_gaps);
+    tesserr << "Spacing " << line_spacing_ << ", in "
+            << row_positions.size() << " rows, "
+            << fitting_gaps << " gaps fitted out of "
+            << non_trivial_gaps << " non-trivial\n";
   }
   return fitting_gaps > non_trivial_gaps * kMinFittingLinespacings;
 }

src/training/combine_lang_model.cpp

@@ -19,6 +19,7 @@
 #include "commandlineflags.h"
 #include "commontraining.h" // CheckSharedLibraryVersion
 #include "lang_model_helpers.h"
+#include "tesserrstream.h"  // for tesserr
 #include "tprintf.h"
 #include "unicharset_training_utils.h"
 
@@ -60,8 +61,8 @@ int main(int argc, char **argv) {
     tprintf("Failed to load unicharset from %s\n", FLAGS_input_unicharset.c_str());
     return EXIT_FAILURE;
   }
-  tprintf("Loaded unicharset of size %zu from file %s\n", unicharset.size(),
-          FLAGS_input_unicharset.c_str());
+  tesserr << "Loaded unicharset of size " << unicharset.size()
+          << " from file " << FLAGS_input_unicharset.c_str() << '\n';
 
   // Set unichar properties
   tprintf("Setting unichar properties\n");

src/training/common/trainingsampleset.cpp

@@ -28,6 +28,7 @@
 #include "intfeaturemap.h"
 #include "intfeaturespace.h"
 #include "shapetable.h"
+#include "tesserrstream.h"  // for tesserr
 #include "trainingsample.h"
 #include "trainingsampleset.h"
 #include "unicity_table.h"
@@ -566,8 +567,9 @@ void TrainingSampleSet::OrganizeByFontAndClass() {
     int font_id = samples_[s]->font_id();
     int class_id = samples_[s]->class_id();
     if (font_id < 0 || font_id >= font_id_map_.SparseSize()) {
-      tprintf("Font id = %d/%d, class id = %d/%d on sample %zu\n", font_id,
-              font_id_map_.SparseSize(), class_id, unicharset_size_, s);
+      tesserr << "Font id = " << font_id << '/' << font_id_map_.SparseSize()
+              << ", class id = " << class_id << '/' << unicharset_size_
+              << " on sample " << s << '\n';
     }
     ASSERT_HOST(font_id >= 0 && font_id < font_id_map_.SparseSize());
     ASSERT_HOST(class_id >= 0 && class_id < unicharset_size_);

src/training/pango/boxchar.cpp

@@ -23,6 +23,7 @@
 
 #include "fileio.h"
 #include "normstrngs.h"
+#include "tesserrstream.h"  // for tesserr
 #include "tprintf.h"
 #include "unicharset.h"
 #include "unicode/uchar.h" // from libicu
@@ -100,7 +101,7 @@ void BoxChar::PrepareToWrite(std::vector<BoxChar *> *boxes) {
   InsertSpaces(rtl_rules, vertical_rules, boxes);
   for (size_t i = 0; i < boxes->size(); ++i) {
     if ((*boxes)[i]->box_ == nullptr) {
-      tprintf("Null box at index %zu\n", i);
+      tesserr << "Null box at index " << i << '\n';
     }
   }
   if (rtl_rules) {

src/training/unicharset/unicharset_training_utils.cpp

@@ -2,7 +2,6 @@
 // File:        unicharset_training_utils.cpp
 // Description: Training utilities for UNICHARSET.
 // Author:      Ray Smith
-// Created:     Fri Oct 17 17:09:01 PDT 2014
 //
 // (C) Copyright 2014, Google Inc.
 // Licensed under the Apache License, Version 2.0 (the "License");
@@ -29,6 +28,7 @@
 #include "icuerrorcode.h"
 #include "normstrngs.h"
 #include "statistc.h"
+#include "tesserrstream.h"   // for tesserr
 #include "unicharset.h"
 #include "unicode/uchar.h"   // from libicu
 #include "unicode/uscript.h" // from libicu
@@ -189,8 +189,8 @@ void SetPropertiesForInputFile(const std::string &script_dir,
 
   // Load the input unicharset
   unicharset.load_from_file(input_unicharset_file.c_str());
-  tprintf("Loaded unicharset of size %zu from file %s\n", unicharset.size(),
-          input_unicharset_file.c_str());
+  tesserr << "Loaded unicharset of size " << unicharset.size()
+          << " from file " << input_unicharset_file << '\n';
 
   // Set unichar properties
   tprintf("Setting unichar properties\n");