[GitHub] [pinot] snleee commented on a diff in pull request #10856: Add the support for filling the default header if the header is missing

["snleee (via GitHub)" <gi...@apache.org>]

snleee commented on code in PR #10856:
URL: https://github.com/apache/pinot/pull/10856#discussion_r1220277626


##########
pinot-plugins/pinot-input-format/pinot-csv/src/main/java/org/apache/pinot/plugin/inputformat/csv/CSVDefaultHeaderUtil.java:
##########
@@ -0,0 +1,195 @@
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements.  See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership.  The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License.  You may obtain a copy of the License at
+ *
+ *   http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing,
+ * software distributed under the License is distributed on an
+ * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+ * KIND, either express or implied.  See the License for the
+ * specific language governing permissions and limitations
+ * under the License.
+ */
+package org.apache.pinot.plugin.inputformat.csv;
+
+import java.io.BufferedReader;
+import java.io.File;
+import java.io.IOException;
+import java.util.ArrayList;
+import java.util.HashMap;
+import java.util.Iterator;
+import java.util.List;
+import java.util.Map;
+import org.apache.commons.csv.CSVFormat;
+import org.apache.commons.csv.CSVParser;
+import org.apache.commons.csv.CSVRecord;
+import org.apache.commons.lang3.StringUtils;
+import org.apache.commons.lang3.math.NumberUtils;
+import org.apache.pinot.spi.data.readers.RecordReaderUtils;
+
+
+public class CSVDefaultHeaderUtil {
+
+  private CSVDefaultHeaderUtil() {
+  }
+
+  private static final String DEFAULT_CSV_COLUMN_PREFIX = "col_";
+  private static final int MAX_CSV_SAMPLE_ROWS = 100;
+
+  static class ColumnValueAttributes {
+    private Type _dataType;
+    private int _length;
+
+    enum Type {
+      STRING, NUMBER, ARRAY
+    }
+
+    ColumnValueAttributes(Object value, Character multiValueDelimiter) {
+      _dataType = Type.STRING;
+      if (multiValueDelimiter != null && value.toString().contains(String.valueOf(multiValueDelimiter))) {
+        _dataType = Type.ARRAY;
+        return;
+      } else if (NumberUtils.isParsable(value.toString())) {
+        _dataType = Type.NUMBER;
+      }
+      _length = value.toString().length();
+    }
+
+    public Type getDataType() {
+      return _dataType;
+    }
+
+    public int getLength() {
+      return _length;
+    }
+  }
+
+  /**
+   * Detects the types for each column based on the row values. If any column is of a single type (e.g. number), except
+   * for the first row, then the first row is presumed to be a header. If the type of the column is assumed to
+   * be a string, the length of the strings within the body is the determining factor. Specifically, if all the rows
+   * except the first are the same length, it's a header.
+   *
+   * If no header is detected, then the header property within the CSVRecordReaderConfig will be set with the following
+   * e.g. {col_0, col_1, col_2, ..}. If a header is detected within the file, then the header property will not
+   * be set and will default to use the first row of the file.
+   *
+   * Motivation for this logic is taken from: https://github.com/python/cpython/blob/main/Lib/csv.py
+   *
+   * Returns the default header when no header is detected. If the header is detected, then returns null.
+   */
+  public static String getDefaultHeader(File csvFile, CSVFormat format, Character delimiter,
+      Character multiValueDelimiter)
+      throws IOException {
+    // Assume there is no header and the default header based on number of columns. We need the default header or
+    // else the CSVRecordReader may not read the values correctly when there are columns with the same values.
+    int numColumns = 0;
+    int rowCounter = 0;
+    List<CSVRecord> records = new ArrayList<>();
+    String[] originalHeader = format.getHeader();
+    try (BufferedReader bufferedReader = RecordReaderUtils.getBufferedReader(csvFile)) {
+      format.withHeader(null);
+      CSVParser parser = format.parse(bufferedReader);
+      Iterator<CSVRecord> iterator = parser.iterator();
+      while (iterator.hasNext() && rowCounter < MAX_CSV_SAMPLE_ROWS + 1) {
+        CSVRecord row = iterator.next();
+        records.add(row);
+        int curNumColumns = row.size();
+        if (curNumColumns > numColumns) {
+          numColumns = curNumColumns;
+        }
+        rowCounter++;
+      }
+    }
+
+    if (records.size() == 0) {
+      throw new IllegalStateException("CSV file is empty. csvFile=" + csvFile.getAbsolutePath());
+    }
+
+    String[] defaultColumnNames = new String[numColumns];
+    for (int i = 0; i < numColumns; i++) {
+      defaultColumnNames[i] = DEFAULT_CSV_COLUMN_PREFIX + i;
+    }
+
+    HashMap<String, ColumnValueAttributes> csvHeaderAttributes = new HashMap<>(numColumns);
+    HashMap<String, ColumnValueAttributes> csvBodyAttributes = new HashMap<>(numColumns);
+    Iterator<CSVRecord> iterator = records.iterator();
+    if (iterator.hasNext()) {
+      CSVRecord header = iterator.next();
+      for (int i = 0; i < header.size(); i++) {
+        csvHeaderAttributes.put(defaultColumnNames[i], new ColumnValueAttributes(header.get(i), multiValueDelimiter));
+        // Initialize the body attributes with null values at first
+        csvBodyAttributes.put(defaultColumnNames[i], null);
+      }
+    }
+
+    while (iterator.hasNext() && rowCounter < MAX_CSV_SAMPLE_ROWS) {
+      CSVRecord row = iterator.next();
+      // Skip rows that have irregular number of columns
+      if (row.size() != numColumns) {
+        continue;
+      }
+
+      for (int i = 0; i < row.size(); i++) {
+        String fieldName = DEFAULT_CSV_COLUMN_PREFIX + i;
+        String columnValue = row.get(i);
+
+        if (!csvBodyAttributes.containsKey(fieldName) || columnValue == null) {
+          continue;
+        }
+        ColumnValueAttributes csvBodyAttribute = csvBodyAttributes.get(fieldName);
+        ColumnValueAttributes curColumnAttributes = new ColumnValueAttributes(columnValue, multiValueDelimiter);
+        if (csvBodyAttribute == null) {
+          csvBodyAttributes.put(fieldName, curColumnAttributes);
+        } else {
+          // If column type is String and length does not match, remove from consideration
+          if (csvBodyAttribute.getDataType().equals(ColumnValueAttributes.Type.STRING)
+              && csvBodyAttribute.getLength() != curColumnAttributes.getLength()) {
+            // Remove column from consideration
+            csvBodyAttributes.remove(fieldName);
+          } else {
+            // If data type does not match, remove from consideration
+            if (!csvBodyAttribute.getDataType().equals(curColumnAttributes.getDataType())) {
+              csvBodyAttributes.remove(fieldName);
+            }
+          }
+        }
+      }
+    }
+
+    int hasHeader = 0;
+    // Compare first row with the rest of the values
+    for (Map.Entry<String, ColumnValueAttributes> entry : csvBodyAttributes.entrySet()) {
+      ColumnValueAttributes headerAttributes = csvHeaderAttributes.get(entry.getKey());
+
+      if (entry.getValue().getDataType().equals(ColumnValueAttributes.Type.STRING)) {
+        // Compare lengths for String case
+        if (headerAttributes.getLength() != entry.getValue().getLength()) {
+          hasHeader++;
+        } else {
+          hasHeader--;
+        }
+      } else {
+        // Compare types
+        if (!headerAttributes.getDataType().equals(entry.getValue().getDataType())) {
+          hasHeader++;
+        } else {
+          hasHeader--;
+        }
+      }
+    }
+
+    if (hasHeader > 0) {
+      format.withHeader(originalHeader);
+      return null;
+    } else {
+      return StringUtils.join(defaultColumnNames, delimiter);

Review Comment:
   setting `null` would keep the backward compatibility. `withHeader()` <- this will ask csv library to read the header from the file instead of using the header directly passed via config.



-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: commits-unsubscribe@pinot.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@pinot.apache.org
For additional commands, e-mail: commits-help@pinot.apache.org