// Evaluates the textlineiness of a ColPartition. Uses EvaluateBox below, // but uses the median top/bottom for horizontal and median left/right for // vertical instead of the bounding box edges. // Evaluates for both horizontal and vertical and returns the best result, // with a positive value for horizontal and a negative value for vertical. int TextlineProjection::EvaluateColPartition(const ColPartition& part, const DENORM* denorm, bool debug) const { if (part.IsSingleton()) return EvaluateBox(part.bounding_box(), denorm, debug); // Test vertical orientation. TBOX box = part.bounding_box(); // Use the partition median for left/right. box.set_left(part.median_left()); box.set_right(part.median_right()); int vresult = EvaluateBox(box, denorm, debug); // Test horizontal orientation. box = part.bounding_box(); // Use the partition median for top/bottom. box.set_top(part.median_top()); box.set_bottom(part.median_bottom()); int hresult = EvaluateBox(box, denorm, debug); if (debug) { tprintf("Partition hresult=%d, vresult=%d from:", hresult, vresult); part.bounding_box().print(); part.Print(); } return hresult >= -vresult ? hresult : vresult; }
// Provide debug output for this ColPartitionSet and all the ColPartitions. void ColPartitionSet::Print() { ColPartition_IT it(&parts_); tprintf("Partition set of %d parts, %d good, coverage=%d+%d" " (%d,%d)->(%d,%d)\n", it.length(), good_column_count_, good_coverage_, bad_coverage_, bounding_box_.left(), bounding_box_.bottom(), bounding_box_.right(), bounding_box_.top()); for (it.mark_cycle_pt(); !it.cycled_list(); it.forward()) { ColPartition* part = it.data(); part->Print(); } }
// Return true if the partitions in other are all compatible with the columns // in this. bool ColPartitionSet::CompatibleColumns(bool debug, ColPartitionSet* other, WidthCallback* cb) { if (debug) { tprintf("CompatibleColumns testing compatibility\n"); Print(); other->Print(); } if (other->parts_.empty()) { if (debug) tprintf("CompatibleColumns true due to empty other\n"); return true; } ColPartition_IT it(&other->parts_); for (it.mark_cycle_pt(); !it.cycled_list(); it.forward()) { ColPartition* part = it.data(); if (part->blob_type() < BRT_UNKNOWN) { if (debug) { tprintf("CompatibleColumns ignoring image partition\n"); part->Print(); } continue; // Image partitions are irrelevant to column compatibility. } int y = part->MidY(); int left = part->bounding_box().left(); int right = part->bounding_box().right(); ColPartition* left_col = ColumnContaining(left, y); ColPartition* right_col = ColumnContaining(right, y); if (right_col == NULL || left_col == NULL) { if (debug) { tprintf("CompatibleColumns false due to partition edge outside\n"); part->Print(); } return false; // A partition edge lies outside of all columns } if (right_col != left_col && cb->Run(right - left)) { if (debug) { tprintf("CompatibleColumns false due to good width in multiple cols\n"); part->Print(); } return false; // Partition with a good width must be in a single column. } ColPartition_IT it2= it; while (!it2.at_last()) { it2.forward(); ColPartition* next_part = it2.data(); if (!BLOBNBOX::IsTextType(next_part->blob_type())) continue; // Non-text partitions are irrelevant. int next_left = next_part->bounding_box().left(); if (next_left == right) { break; // They share the same edge, so one must be a pull-out. } // Search to see if right and next_left fall within a single column. ColPartition* next_left_col = ColumnContaining(next_left, y); if (right_col == next_left_col) { // There is a column break in this column. // This can be due to a figure caption within a column, a pull-out // block, or a simple broken textline that remains to be merged: // all allowed, or a change in column layout: not allowed. // If both partitions are of good width, then it is likely // a change in column layout, otherwise probably an allowed situation. if (part->good_width() && next_part->good_width()) { if (debug) { int next_right = next_part->bounding_box().right(); tprintf("CompatibleColumns false due to 2 parts of good width\n"); tprintf("part1 %d-%d, part2 %d-%d\n", left, right, next_left, next_right); right_col->Print(); } return false; } } break; } } if (debug) tprintf("CompatibleColumns true!\n"); return true; }