From 27dfc3354ebc4ddf291f191e577b079a45f5323b Mon Sep 17 00:00:00 2001
From: Stefan Weil <sw@weilnetz.de>
Date: Tue, 29 Jun 2021 17:48:06 +0200
Subject: [PATCH] Fix vector resize with init for all elements (issue #3473)

Fixes: 9710bc0465fc0df1a5bc0e71177151e8c5901204
Signed-off-by: Stefan Weil <sw@weilnetz.de>
---
 src/ccmain/control.cpp    | 21 ++++++++++++++-------
 src/ccmain/linerec.cpp    |  3 +--
 src/ccmain/paragraphs.cpp |  4 ++--
 3 files changed, 17 insertions(+), 11 deletions(-)

diff --git a/src/ccmain/control.cpp b/src/ccmain/control.cpp
index b439aa649a..6d4cf4827c 100644
--- a/src/ccmain/control.cpp
+++ b/src/ccmain/control.cpp
@@ -982,9 +982,12 @@ void Tesseract::AssignDiacriticsToOverlappingBlobs(const std::vector<C_OUTLINE *
                                                    std::vector<bool> *overlapped_any_blob,
                                                    std::vector<C_BLOB *> *target_blobs) {
   std::vector<bool> blob_wanted;
-  word_wanted->resize(outlines.size(), false);
-  overlapped_any_blob->resize(outlines.size(), false);
-  target_blobs->resize(outlines.size(), nullptr);
+  word_wanted->clear();
+  word_wanted->resize(outlines.size());
+  overlapped_any_blob->clear();
+  overlapped_any_blob->resize(outlines.size());
+  target_blobs->clear();
+  target_blobs->resize(outlines.size());
   // For each real blob, find the outlines that seriously overlap it.
   // A single blob could be several merged characters, so there can be quite
   // a few outlines overlapping, and the full engine needs to be used to chop
@@ -993,7 +996,8 @@ void Tesseract::AssignDiacriticsToOverlappingBlobs(const std::vector<C_OUTLINE *
   for (blob_it.mark_cycle_pt(); !blob_it.cycled_list(); blob_it.forward()) {
     C_BLOB *blob = blob_it.data();
     const TBOX blob_box = blob->bounding_box();
-    blob_wanted.resize(outlines.size(), false);
+    blob_wanted.clear();
+    blob_wanted.resize(outlines.size());
     int num_blob_outlines = 0;
     for (unsigned i = 0; i < outlines.size(); ++i) {
       if (blob_box.major_x_overlap(outlines[i]->bounding_box()) && !(*word_wanted)[i]) {
@@ -1032,15 +1036,18 @@ void Tesseract::AssignDiacriticsToNewBlobs(const std::vector<C_OUTLINE *> &outli
                                            std::vector<bool> *word_wanted,
                                            std::vector<C_BLOB *> *target_blobs) {
   std::vector<bool> blob_wanted;
-  word_wanted->resize(outlines.size(), false);
-  target_blobs->resize(outlines.size(), nullptr);
+  word_wanted->clear();
+  word_wanted->resize(outlines.size());
+  target_blobs->clear();
+  target_blobs->resize(outlines.size());
   // Check for outlines that need to be turned into stand-alone blobs.
   for (unsigned i = 0; i < outlines.size(); ++i) {
     if (outlines[i] == nullptr) {
       continue;
     }
     // Get a set of adjacent outlines that don't overlap any existing blob.
-    blob_wanted.resize(outlines.size(), false);
+    blob_wanted.clear();
+    blob_wanted.resize(outlines.size());
     int num_blob_outlines = 0;
     TBOX total_ol_box(outlines[i]->bounding_box());
     while (i < outlines.size() && outlines[i] != nullptr) {
diff --git a/src/ccmain/linerec.cpp b/src/ccmain/linerec.cpp
index 4f85fd670e..f1deced486 100644
--- a/src/ccmain/linerec.cpp
+++ b/src/ccmain/linerec.cpp
@@ -151,8 +151,7 @@ ImageData *Tesseract::GetLineData(const TBOX &line_box, const std::vector<TBOX>
     line_boxes.push_back(box);
     line_texts.push_back(texts[b]);
   }
-  std::vector<int> page_numbers;
-  page_numbers.resize(line_boxes.size(), applybox_page);
+  std::vector<int> page_numbers(line_boxes.size(), applybox_page);
   image_data->AddBoxes(line_boxes, line_texts, page_numbers);
   return image_data;
 }
diff --git a/src/ccmain/paragraphs.cpp b/src/ccmain/paragraphs.cpp
index f817bafdd4..b9525b9950 100644
--- a/src/ccmain/paragraphs.cpp
+++ b/src/ccmain/paragraphs.cpp
@@ -2314,14 +2314,14 @@ void CanonicalizeDetectionResults(std::vector<PARA *> *row_owners, PARA_LIST *pa
 void DetectParagraphs(int debug_level, std::vector<RowInfo> *row_infos,
                       std::vector<PARA *> *row_owners, PARA_LIST *paragraphs,
                       std::vector<ParagraphModel *> *models) {
-  std::vector<RowScratchRegisters> rows;
   ParagraphTheory theory(models);
 
   // Initialize row_owners to be a bunch of nullptr pointers.
+  row_owners->clear();
   row_owners->resize(row_infos->size());
 
   // Set up row scratch registers for the main algorithm.
-  rows.resize(row_infos->size(), RowScratchRegisters());
+  std::vector<RowScratchRegisters> rows(row_infos->size());
   for (unsigned i = 0; i < row_infos->size(); i++) {
     rows[i].Init((*row_infos)[i]);
   }