com.monitorjbl.xlsx.sst.BufferedStringsTable Maven / Gradle / Ivy

Go to download
Show more of this group Show more artifacts with this name
Show all versions of xlsx-streamer Show documentation
Streaming Excel reader
There is a newer version: 2.2.0
package com.monitorjbl.xlsx.sst;

import org.apache.poi.openxml4j.opc.OPCPackage;
import org.apache.poi.openxml4j.opc.PackagePart;
import org.apache.poi.ss.usermodel.RichTextString;
import org.apache.poi.util.StaxHelper;
import org.apache.poi.xssf.model.SharedStringsTable;
import org.apache.poi.xssf.usermodel.XSSFRelation;
import org.apache.poi.xssf.usermodel.XSSFRichTextString;
import org.openxmlformats.schemas.spreadsheetml.x2006.main.CTRst;

import javax.xml.stream.XMLEventReader;
import javax.xml.stream.XMLStreamException;
import javax.xml.stream.events.XMLEvent;
import java.io.File;
import java.io.IOException;
import java.io.InputStream;
import java.util.List;

public class BufferedStringsTable extends SharedStringsTable implements AutoCloseable {
  private final FileBackedList list;

  public static BufferedStringsTable getSharedStringsTable(File tmp, int cacheSize, OPCPackage pkg)
      throws IOException {
    List parts = pkg.getPartsByContentType(XSSFRelation.SHARED_STRINGS.getContentType());
    return parts.size() == 0 ? null : new BufferedStringsTable(parts.get(0), tmp, cacheSize);
  }

  private BufferedStringsTable(PackagePart part, File file, int cacheSize) throws IOException {
    this.list = new FileBackedList(file, cacheSize);
    readFrom(part.getInputStream());
  }

  @Override
  public void readFrom(InputStream is) throws IOException {
    try {
      XMLEventReader xmlEventReader = StaxHelper.newXMLInputFactory().createXMLEventReader(is);

      while(xmlEventReader.hasNext()) {
        XMLEvent xmlEvent = xmlEventReader.nextEvent();

        if(xmlEvent.isStartElement() && xmlEvent.asStartElement().getName().getLocalPart().equals("si")) {
          list.add(parseCT_Rst(xmlEventReader));
        }
      }
    } catch(XMLStreamException e) {
      throw new IOException(e);
    }
  }

  /**
   * Parses a {@code } String Item. Returns just the text and drops the formatting. See xmlschema
   * type {@code CT_Rst}.
   */
  private String parseCT_Rst(XMLEventReader xmlEventReader) throws XMLStreamException {
    // Precondition: pointing to ;  Post condition: pointing to 
    StringBuilder buf = new StringBuilder();
    XMLEvent xmlEvent;
    while((xmlEvent = xmlEventReader.nextTag()).isStartElement()) {
      switch(xmlEvent.asStartElement().getName().getLocalPart()) {
        case "t": // Text
          buf.append(xmlEventReader.getElementText());
          break;
        case "r": // Rich Text Run
          parseCT_RElt(xmlEventReader, buf);
          break;
        case "rPh": // Phonetic Run
        case "phoneticPr": // Phonetic Properties
          skipElement(xmlEventReader);
          break;
        default:
          throw new IllegalArgumentException(xmlEvent.asStartElement().getName().getLocalPart());
      }
    }
    return buf.length() > 0 ? buf.toString() : null;
  }

  /**
   * Parses a {@code } Rich Text Run. Returns just the text and drops the formatting. See xmlschema
   * type {@code CT_RElt}.
   */
  private void parseCT_RElt(XMLEventReader xmlEventReader, StringBuilder buf) throws XMLStreamException {
    // Precondition: pointing to ;  Post condition: pointing to 
    XMLEvent xmlEvent;
    while((xmlEvent = xmlEventReader.nextTag()).isStartElement()) {
      switch(xmlEvent.asStartElement().getName().getLocalPart()) {
        case "t": // Text
          buf.append(xmlEventReader.getElementText());
          break;
        case "rPr": // Run Properties
          skipElement(xmlEventReader);
          break;
        default:
          throw new IllegalArgumentException(xmlEvent.asStartElement().getName().getLocalPart());
      }
    }
  }

  private void skipElement(XMLEventReader xmlEventReader) throws XMLStreamException {
    // Precondition: pointing to start element;  Post condition: pointing to end element
    while(xmlEventReader.nextTag().isStartElement()) {
      skipElement(xmlEventReader); // recursively skip over child
    }
  }

  @Override
  public RichTextString getItemAt(int idx) {
    return new XSSFRichTextString(list.getAt(idx));
  }

  @Override
  public CTRst getEntryAt(int idx) {
    return ((XSSFRichTextString)getItemAt(idx)).getCTRst();
  }

  @Override
  public void close() throws IOException {
    super.close();
    list.close();
  }
}