whatever.cpp

// Feeling overwhelmed by "find the words in a grid" tasks for preschoolers?
// Look no further!
#include <leptonica/allheaders.h>
#include <tesseract/baseapi.h>
#include <fstream>
#include <iostream>
#include <sstream>
#include <string>
#include <vector>
using namespace std;
bool search2D(vector<string> grid, int row, int col, string word) {
  int m = grid.size();
  int n = grid[0].size();
  if (grid[row][col] != word[0]) return false;
  int len = word.size();
  vector<int> x = { -1, -1, -1, 0, 0, 1, 1, 1 };
  vector<int> y = { -1, 0, 1, -1, 1, -1, 0, 1 };
  for (int dir = 0; dir < 8; dir++) {
    int k, currX = row + x[dir], currY = col + y[dir];
    for (k = 1; k < len; k++) {
      if (currX >= m || currX < 0 || currY >= n || currY < 0) break;
      if (grid[currX][currY] != word[k]) break;
      currX += x[dir], currY += y[dir];
    }
    if (k == len) return true;
  }
  return false;
}
vector<vector<int>> searchWord(vector<string> grid, string word) {
  int m = grid.size();
  int n = grid[0].size();
  vector<vector<int>> ans;
  for (int i = 0; i < m; i++)
    for (int j = 0; j < n; j++)
      if (search2D(grid, i, j, word))
        ans.push_back({ i, j });
  return ans;
}
void displayMatrix(vector<string> grid, int row, int col) {
  int m = grid.size();
  int n = grid[0].size();
  for (int i = 0; i < m; i++) {
    for (int j = 0; j < n; j++)
      if (i == row && j == col) cout << "\033[1;31m" << grid[i][j] << "\033[0m";
      else cout << grid[i][j];
    cout << endl;
  }
}
int main(int argc, char ** argv) {
  if (argc < 2) { cerr << "Give one argument." << endl; return 1; }
  const char * imagePath = argv[1];
  tesseract::TessBaseAPI tess;
  if (tess.Init(NULL, "eng", tesseract::OEM_LSTM_ONLY) != 0) {
    cerr << "Could not initialize tesseract." << endl;
    return 1;
  }
  Pix * image = pixRead(imagePath);
  if (!image) {
    cerr << "Could not open image file: " << imagePath << endl;
    return 1;
  }
  tess.SetImage(image);
  tess.SetVariable("tessedit_char_blacklist", "0123456789!@#$%^&*()_+-=[]{};:'\"\\|,.<>/?");
  tess.SetVariable("load_system_dawg", "0");
  tess.SetVariable("load_freq_dawg", "0");
  tess.SetPageSegMode(tesseract::PSM_SINGLE_BLOCK);
  char * ocrResult = tess.GetUTF8Text();
  if (!ocrResult) {
    cerr << "OCR failed." << endl;
    pixDestroy(&image);
    tess.End();
    return 1;
  }
  cout << "OCR Result:\n" << ocrResult << endl;
  cout << "Due to OCR fuckery kindly please paste the cleaned up word matrix and put the words to search on the last "
          "line: "
       << endl;
  vector<string> lines;
  string line;
  while (getline(cin, line))
    lines.push_back(line);
  vector<string> words_to_search;
  istringstream iss(lines.back());
  string word;
  while (iss >> word)
    words_to_search.push_back(word);
  lines.pop_back();
  vector<vector<int>> ans;
  for (auto word : words_to_search) {
    ans = searchWord(lines, word);
    if (ans.size() > 0) {
      cout << "Word " << word << " found at: ";
      for (auto ele : ans) {
        cout << "(" << ele[0] << ", " << ele[1] << ") " << endl;
        displayMatrix(lines, ele[0], ele[1]);
      }
      cout << endl;
    } else cout << "Word " << word << " not found." << endl;
  }
}