C++ ROW類代碼示例

void BLOCK::compress() {  // squash it up
  #define           ROW_SPACING 5

  ROW_IT row_it(&rows);
  ROW *row;
  ICOORD row_spacing (0, ROW_SPACING);

  ICOORDELT_IT icoordelt_it;

  sort_rows();

  box = TBOX (box.topleft (), box.topleft ());
  box.move_bottom_edge (ROW_SPACING);
  for (row_it.mark_cycle_pt (); !row_it.cycled_list (); row_it.forward ()) {
    row = row_it.data ();
    row->move (box.botleft () - row_spacing -
      row->bounding_box ().topleft ());
    box += row->bounding_box ();
  }

  leftside.clear ();
  icoordelt_it.set_to_list (&leftside);
  icoordelt_it.add_to_end (new ICOORDELT (box.left (), box.bottom ()));
  icoordelt_it.add_to_end (new ICOORDELT (box.left (), box.top ()));
  rightside.clear ();
  icoordelt_it.set_to_list (&rightside);
  icoordelt_it.add_to_end (new ICOORDELT (box.right (), box.bottom ()));
  icoordelt_it.add_to_end (new ICOORDELT (box.right (), box.top ()));
}

/// Builds a PAGE_RES from the block_list in the way required for ApplyBoxes:
/// All fuzzy spaces are removed, and all the words are maximally chopped.
PAGE_RES* Tesseract::SetupApplyBoxes(const GenericVector<TBOX>& boxes,
                                     BLOCK_LIST *block_list) {
  PreenXHeights(block_list);
  // Strip all fuzzy space markers to simplify the PAGE_RES.
  BLOCK_IT b_it(block_list);
  for (b_it.mark_cycle_pt(); !b_it.cycled_list(); b_it.forward()) {
    BLOCK* block = b_it.data();
    ROW_IT r_it(block->row_list());
    for (r_it.mark_cycle_pt(); !r_it.cycled_list(); r_it.forward ()) {
      ROW* row = r_it.data();
      WERD_IT w_it(row->word_list());
      for (w_it.mark_cycle_pt(); !w_it.cycled_list(); w_it.forward()) {
        WERD* word = w_it.data();
        if (word->cblob_list()->empty()) {
          delete w_it.extract();
        } else {
          word->set_flag(W_FUZZY_SP, false);
          word->set_flag(W_FUZZY_NON, false);
        }
      }
    }
  }
  PAGE_RES* page_res = new PAGE_RES(false, block_list, NULL);
  PAGE_RES_IT pr_it(page_res);
  WERD_RES* word_res;
  while ((word_res = pr_it.word()) != NULL) {
    MaximallyChopWord(boxes, pr_it.block()->block,
                      pr_it.row()->row, word_res);
    pr_it.forward();
  }
  return page_res;
}

void LRURowStorage<ROW, V>::insert_inactive(int32_t row_id,
    const ROW<V>& new_row) {
  // Sanity check.
  int num_rows = inactive_list_.size() + active_list_.size();
  CHECK_LE(num_rows, this->capacity_)
    << "Number of rows exceed storage size.";

  // If necessary, make space...
  if (num_rows == this->capacity_) {
    // ...by purging the least-recently-used element (head of list).
    VLOG(1) << "Evicting row " << inactive_list_.right.begin()->second
      << " to insert row " << row_id;
    inactive_list_.right.erase(inactive_list_.right.begin());
  }

  // Create a new record from the key and the value
  // bimap's list_view defaults to inserting this at
  // the list tail (considered most-recently-used).
  int threshold_count = static_cast<int>(num_row_access_to_active_ *
    new_row.get_num_columns());  // Compute the threshold count.
  AccessStatsRow new_inactive_row = boost::tuples::make_tuple(new_row, 0,
      threshold_count);
  inactive_list_.insert(
    typename AccessCountMapList::value_type(row_id, new_inactive_row));
  VLOG(1) << "Inserted new row (row id = " << row_id << ") of size "
    << new_row.get_num_columns() << " with threshold count = "
    << threshold_count;
}

void PrintSegmentationStats(BLOCK_LIST* block_list) {
  int num_blocks = 0;
  int num_rows = 0;
  int num_words = 0;
  int num_blobs = 0;
  BLOCK_IT block_it(block_list);
  for (block_it.mark_cycle_pt(); !block_it.cycled_list(); block_it.forward()) {
    BLOCK* block = block_it.data();
    ++num_blocks;
    ROW_IT row_it(block->row_list());
    for (row_it.mark_cycle_pt(); !row_it.cycled_list(); row_it.forward()) {
      ++num_rows;
      ROW* row = row_it.data();
      // Iterate over all werds in the row.
      WERD_IT werd_it(row->word_list());
      for (werd_it.mark_cycle_pt(); !werd_it.cycled_list(); werd_it.forward()) {
        WERD* werd = werd_it.data();
        ++num_words;
        num_blobs += werd->cblob_list()->length();
      }
    }
  }
  tprintf("Block list stats:\nBlocks = %d\nRows = %d\nWords = %d\nBlobs = %d\n",
          num_blocks, num_rows, num_words, num_blobs);
}

void apply_box_training(BLOCK_LIST *block_list) {
  BLOCK_IT block_it(block_list);
  ROW_IT row_it;
  ROW *row;
  WERD_IT word_it;
  WERD *word;
  WERD *bln_word;
  WERD copy_outword;             // copy to denorm
  PBLOB_IT blob_it;
  DENORM denorm;
  INT16 count = 0;
  char ch[2];

  ch[1] = '\0';

  tprintf ("Generating training data\n");
  for (block_it.mark_cycle_pt ();
  !block_it.cycled_list (); block_it.forward ()) {
    row_it.set_to_list (block_it.data ()->row_list ());
    for (row_it.mark_cycle_pt (); !row_it.cycled_list (); row_it.forward ()) {
      row = row_it.data ();
      word_it.set_to_list (row->word_list ());
      for (word_it.mark_cycle_pt ();
      !word_it.cycled_list (); word_it.forward ()) {
        word = word_it.data ();
        if ((strlen (word->text ()) == 1) &&
        (word->gblob_list ()->length () == 1)) {
          /* Here is a word with a single char label and a single blob so train on it */
          bln_word =
            make_bln_copy (word, row, row->x_height (), &denorm);
          blob_it.set_to_list (bln_word->blob_list ());
          ch[0] = *word->text ();
          tess_training_tester (blob_it.data (),
                                 //single blob
            &denorm, TRUE,       //correct
            ch,                  //correct ASCII char
            1,                   //ASCII length
            NULL);
          copy_outword = *(bln_word);
          copy_outword.baseline_denormalise (&denorm);
          blob_it.set_to_list (copy_outword.blob_list ());
          ch[0] = *word->text ();
          delete bln_word;
          count++;
        }
      }
    }
  }
  tprintf ("Generated training data for %d blobs\n", count);
}

void TABLE::tCut()

  {
  if(TABLE_ATTRIBUTES::AutoFill())
    return;

  DestroyAllEditControls();
  Copy();

  RUNTIMEINFO*  rtime = (RUNTIMEINFO*)AObjGetRuntimeInfo(theObject);
  ROW*          row = rtime->GetRow();
  row->DeleteSelectedRows();
  PostObjectChanged(CHANGED_NUMROWS);
  }

void TABLE::DestroyEditControl()

  {
  if(hEdit)
    {
    // Get the rectangle of the cell for intelligent update
    RECT  cellRect;
    GetWindowRect(hEdit, &cellRect);
    ScreenToClient(hTable, (POINT*)&cellRect);
    ScreenToClient(hTable, (POINT*)&cellRect.right);
    cellRect.left -= COLUMN::CellCharWidth(editCol);
    cellRect.top = cellRect.bottom - TABLE_ATTRIBUTES::CellHeight();
    cellRect.right += COLUMN::CellCharWidth(editCol);

    // un-subclass the control in preparation for destroy
    RemoveProp(hEdit, "loPtr");
    RemoveProp(hEdit, "hiPtr");
    RemoveProp(hEdit, "number");
    SetWindowLong(hEdit, GWL_WNDPROC, (long)oldEditControlProc);

    // get the text and update the cell as necessary
    WORD  len = (WORD)SendMessage(hEdit, WM_GETTEXTLENGTH, 0, 0L)+1;
    char* buffer = new char[len];
    SendMessage(hEdit, WM_GETTEXT, (WPARAM)(len+1), (LPARAM)buffer);
    RUNTIMEINFO*  rtime = (RUNTIMEINFO*)AObjGetRuntimeInfo(theObject);
    ROW*          row = rtime->GetRow();
    if(hstrcmp(row->GetCell(editRow, editCol, FALSE, NULL), buffer))
      {
      row->SetCell(editRow, editCol, buffer);
      TABLE_ATTRIBUTES::SetEditRow(editRow);
      TABLE_ATTRIBUTES::SetEditColumn(editCol);
      PostObjectChanged(CHANGED_FULLREDRAW);
      AEvtPostSignalAtMark(theObject, SIGNAL_TABLEEDITED);
      }
    delete buffer;

    // destroy the window
    DestroyWindow(hEdit);
    hEdit = NULL;
    DeleteObject(hEditFont);

    AEvtPostSignalAtMark(theObject, SIGNAL_STOPEDITING);

    // make sure the proper region is redrawn
    InvalidateRect(hTable, &cellRect, FALSE);
    UpdateWindow(hTable);

    // wipPreventThieves(ParentID, FALSE); SKJ need replacement
    }
  }

void Textord::cleanup_blocks(bool clean_noise, BLOCK_LIST* blocks) {
  BLOCK_IT block_it = blocks;    //iterator
  ROW_IT row_it;                 //row iterator

  int num_rows = 0;
  int num_rows_all = 0;
  int num_blocks = 0;
  int num_blocks_all = 0;
  for (block_it.mark_cycle_pt(); !block_it.cycled_list();
       block_it.forward()) {
    BLOCK* block = block_it.data();
    if (block->pdblk.poly_block() != nullptr && !block->pdblk.poly_block()->IsText()) {
      cleanup_nontext_block(block);
      continue;
    }
    num_rows = 0;
    num_rows_all = 0;
    if (clean_noise) {
      row_it.set_to_list(block->row_list());
      for (row_it.mark_cycle_pt(); !row_it.cycled_list(); row_it.forward()) {
        ROW* row = row_it.data();
        ++num_rows_all;
        clean_small_noise_from_words(row);
        if ((textord_noise_rejrows && !row->word_list()->empty() &&
             clean_noise_from_row(row)) ||
            row->word_list()->empty()) {
          delete row_it.extract();  // lose empty row.
        } else {
          if (textord_noise_rejwords)
            clean_noise_from_words(row_it.data());
          if (textord_blshift_maxshift >= 0)
            tweak_row_baseline(row, textord_blshift_maxshift,
                               textord_blshift_xfraction);
          ++num_rows;
        }
      }
    }
    if (block->row_list()->empty()) {
      delete block_it.extract();  // Lose empty text blocks.
    } else {
      ++num_blocks;
    }
    ++num_blocks_all;
    if (textord_noise_debug)
      tprintf("cleanup_blocks: # rows = %d / %d\n", num_rows, num_rows_all);
  }
  if (textord_noise_debug)
    tprintf("cleanup_blocks: # blocks = %d / %d\n", num_blocks, num_blocks_all);
}

void TABLE::tPaste()

  {
  if(TABLE_ATTRIBUTES::AutoFill())
    return;

  BOOL  available = IsClipboardFormatAvailable(CF_TEXT);
  if(!available)
    return;

  DestroyAllEditControls();

  OpenClipboard(hParentWnd);
  HGLOBAL hClipboard = GetClipboardData(CF_TEXT);
  if(!hClipboard)
    {
    CloseClipboard();
    return;
    }

  HGLOBAL hCopy = GlobalAlloc(GHND, GlobalSize(hClipboard));
  if(!hCopy)
    {
    CloseClipboard();
    return;
    }

  char* copy = (char*)GlobalLock(hCopy);
  char* clip = (char*)GlobalLock(hClipboard);
  hmemcpy(copy, clip, GlobalSize(hClipboard));
  GlobalUnlock(hClipboard);
  CloseClipboard();

  RUNTIMEINFO*  rtime = (RUNTIMEINFO*)AObjGetRuntimeInfo(theObject);
  ROW*          row = rtime->GetRow();
  long          first = row->FirstSelection();
  char*         token = (char*)myStrtok((char huge*)copy);
  while(token)
    {
    row->AddRow(first, token, TABLE_ATTRIBUTES::NumColumns(), FALSE);
    token = (char*)myStrtok(NULL);
    first++;
    }
  GlobalUnlock(hCopy);
  GlobalFree(hCopy);

  row->DeleteSelectedRows();
  PostObjectChanged(CHANGED_NUMROWS);
  }

void TABLE::tCopy()

  {
  RUNTIMEINFO*  rtime = (RUNTIMEINFO*)AObjGetRuntimeInfo(theObject);
  ROW*          row = rtime->GetRow();
  long          numRows = row->NumSelections();
  long          length = numRows;
  BOOL          dbAccessed = FALSE;

  if(numRows<1 || numRows>row->NumRows())
    return;

  for(long i=0; i<numRows; i++)
    length += row->RowLength(row->GetSelection(i), TRUE, &dbAccessed);

  HGLOBAL hClip = GlobalAlloc(GHND, length);
  if(!hClip)
    {
    if(dbAccessed)
      AEvtPostSignalAtMark(theObject, SIGNAL_TABLEAUTOFILLED);

    return;
    }

  char huge*  clip = (char huge*)GlobalLock(hClip);
  long        sel;
  long        offset=0;
  for(i=0; i<numRows; i++)
    {
    sel = row->GetSelection(i);
    length = row->RowLength(sel, TRUE, &dbAccessed);
    row->GetRow(sel, clip+offset, length, TRUE, TRUE, &dbAccessed);
    if(i!=(numRows-1))
      {
      // give the cr-lf combination
      *(clip+(offset + length - 1)) = '\r';
      *(clip+(offset + length++)) = '\n';
      }
    else
      *(clip+(offset + length - 1)) = 0;

    offset+=length;
    }
  GlobalUnlock(hClip);

  OpenClipboard(hParentWnd);
  EmptyClipboard();
  SetClipboardData(CF_TEXT, hClip);
  CloseClipboard();

  if(dbAccessed)
    AEvtPostSignalAtMark(theObject, SIGNAL_TABLEAUTOFILLED);
  }

//----------------------------------------------------------------------
MAT EEMD::eemdf90( ROW input, float noise_amplitude, int num_imfs, int num_ensembles ) {

    // Tanspose the Row vector into a Column vector
    VEC col_input = input.t();
    MAT col_imfs = eemdf90(col_input, noise_amplitude, num_imfs, num_ensembles);
    MAT row_imfs = col_imfs.t();
    
    return row_imfs;

}

void RefreshWordBlobsFromNewBlobs(BLOCK_LIST* block_list,
                                  C_BLOB_LIST* new_blobs,
                                  C_BLOB_LIST* not_found_blobs) {
  // Now iterate over all the blobs in the segmentation_block_list_, and just
  // replace the corresponding c-blobs inside the werds.
  BLOCK_IT block_it(block_list);
  for (block_it.mark_cycle_pt(); !block_it.cycled_list(); block_it.forward()) {
    BLOCK* block = block_it.data();
    if (block->poly_block() != NULL && !block->poly_block()->IsText())
      continue;  // Don't touch non-text blocks.
    // Iterate over all rows in the block.
    ROW_IT row_it(block->row_list());
    for (row_it.mark_cycle_pt(); !row_it.cycled_list(); row_it.forward()) {
      ROW* row = row_it.data();
      // Iterate over all werds in the row.
      WERD_IT werd_it(row->word_list());
      WERD_LIST new_words;
      WERD_IT new_words_it(&new_words);
      for (werd_it.mark_cycle_pt(); !werd_it.cycled_list(); werd_it.forward()) {
        WERD* werd = werd_it.extract();
        WERD* new_werd = werd->ConstructWerdWithNewBlobs(new_blobs,
                                                         not_found_blobs);
        if (new_werd) {
          // Insert this new werd into the actual row's werd-list. Remove the
          // existing one.
          new_words_it.add_after_then_move(new_werd);
          delete werd;
        } else {
          // Reinsert the older word back, for lack of better options.
          // This is critical since dropping the words messes up segmentation:
          // eg. 1st word in the row might otherwise have W_FUZZY_NON turned on.
          new_words_it.add_after_then_move(werd);
        }
      }
      // Get rid of the old word list & replace it with the new one.
      row->word_list()->clear();
      werd_it.move_to_first();
      werd_it.add_list_after(&new_words);
    }
  }
}

/// Any row xheight that is significantly different from the median is set
/// to the median.
void Tesseract::PreenXHeights(BLOCK_LIST *block_list) {
  double median_xheight = MedianXHeight(block_list);
  double max_deviation = kMaxXHeightDeviationFraction * median_xheight;
  // Strip all fuzzy space markers to simplify the PAGE_RES.
  BLOCK_IT b_it(block_list);
  for (b_it.mark_cycle_pt(); !b_it.cycled_list(); b_it.forward()) {
    BLOCK* block = b_it.data();
    ROW_IT r_it(block->row_list());
    for (r_it.mark_cycle_pt(); !r_it.cycled_list(); r_it.forward ()) {
      ROW* row = r_it.data();
      float diff = fabs(row->x_height() - median_xheight);
      if (diff > max_deviation) {
        if (applybox_debug) {
          tprintf("row xheight=%g, but median xheight = %g\n",
                  row->x_height(), median_xheight);
        }
        row->set_x_height(static_cast<float>(median_xheight));
      }
    }
  }
}

const FP_LIB_TABLE::ROW* FP_LIB_TABLE::FindRow( const wxString& aNickname )
    throw( IO_ERROR )
{
    ROW* row = findRow( aNickname );

    if( !row )
    {
        wxString msg = wxString::Format(
            _( "fp-lib-table files contain no lib with nickname '%s'" ),
            GetChars( aNickname ) );

        THROW_IO_ERROR( msg );
    }

    // We've been 'lazy' up until now, but it cannot be deferred any longer,
    // instantiate a PLUGIN of the proper kind if it is not already in this ROW.
    if( !row->plugin )
        row->setPlugin( IO_MGR::PluginFind( row->type ) );

    return row;
}

/**
 * Returns the baseline of the current object at the given level.
 * The baseline is the line that passes through (x1, y1) and (x2, y2).
 * WARNING: with vertical text, baselines may be vertical!
 */
bool PageIterator::Baseline(PageIteratorLevel level,
                            int* x1, int* y1, int* x2, int* y2) const {
  if (it_->word() == NULL) return false;  // Already at the end!
  ROW* row = it_->row()->row;
  WERD* word = it_->word()->word;
  TBOX box = (level == RIL_WORD || level == RIL_SYMBOL)
           ? word->bounding_box()
           : row->bounding_box();
  int left = box.left();
  ICOORD startpt(left, static_cast<inT16>(row->base_line(left) + 0.5));
  int right = box.right();
  ICOORD endpt(right, static_cast<inT16>(row->base_line(right) + 0.5));
  // Rotate to image coordinates and convert to global image coords.
  startpt.rotate(it_->block()->block->re_rotation());
  endpt.rotate(it_->block()->block->re_rotation());
  *x1 = startpt.x() / scale_ + rect_left_;
  *y1 = (rect_height_ - startpt.y()) / scale_ + rect_top_;
  *x2 = endpt.x() / scale_ + rect_left_;
  *y2 = (rect_height_ - endpt.y()) / scale_ + rect_top_;
  return true;
}