VertexBatcher.java

/*
 * Python-bindings bridge: bulk vertex creation for GraphBatch.
 *
 * GraphBatch.create_vertices() from Python marshals an Object[][] property
 * matrix element-by-element (~19us/vertex measured vs 6.2us Java-native).
 * This helper accepts the rows as ONE JSON string (json.dumps costs
 * ~0.2us/vertex Python-side) and returns the created RIDs as ONE
 * semicolon-joined string, so the whole batch costs two bulk string copies.
 *
 * Property values are str/int/float/bool/null, nested lists and maps thereof,
 * and the tagged values DocumentBatcher.untag decodes (date-time, date,
 * BigDecimal, byte[]) sitting directly in a row; the Python side falls back to
 * the matrix path for anything else.
 */
package com.arcadedb.python;

import com.arcadedb.database.RID;
import com.arcadedb.graph.GraphBatch;
import com.arcadedb.serializer.json.JSONArray;
import com.arcadedb.serializer.json.JSONObject;

public final class VertexBatcher {

  private VertexBatcher() {
  }

  public static String createVerticesJson(final GraphBatch batch, final String typeName, final String jsonRows) {
    return createVerticesJson(batch, typeName, jsonRows, null, null);
  }

  /**
   * As {@link #createVerticesJson(GraphBatch, String, String)}, with float vectors that did not go through the JSON text:
   * {@code vectorNames[c]} is set on vertex i to row i of {@code vectors[c]}, a {@link FloatVectors.Matrix} with one row per
   * JSON row. Both are null when there are none.
   */
  public static String createVerticesJson(final GraphBatch batch, final String typeName, final String jsonRows,
      final String[] vectorNames, final Object[] vectors) {
    final JSONArray rows = new JSONArray(jsonRows);
    FloatVectors.check(vectorNames, vectors, rows.length());
    final int extra = vectorNames == null ? 0 : vectorNames.length;
    final Object[][] matrix = new Object[rows.length()][];
    for (int i = 0; i < rows.length(); i++) {
      final JSONObject row = rows.getJSONObject(i);
      final Object[] flat = new Object[(row.length() + extra) * 2];
      int j = 0;
      for (final String key : row.keySet()) {
        flat[j++] = key;
        flat[j++] = DocumentBatcher.value(row.isNull(key) ? null : row.get(key));
      }
      for (int c = 0; c < extra; c++) {
        flat[j++] = vectorNames[c];
        flat[j++] = ((FloatVectors.Matrix) vectors[c]).row(i);
      }
      matrix[i] = flat;
    }
    final RID[] rids = batch.createVertices(typeName, matrix);
    final StringBuilder sb = new StringBuilder(rids.length * 8);
    for (int i = 0; i < rids.length; i++) {
      if (i > 0)
        sb.append(';');
      sb.append(rids[i]);
    }
    return sb.toString();
  }
}