package com.hughes.android.dictionary.engine;
-import java.io.ByteArrayOutputStream;
import java.io.DataInput;
import java.io.DataInputStream;
import java.io.DataOutput;
import java.util.ArrayList;
import java.util.Collections;
import java.util.List;
-import java.util.zip.GZIPOutputStream;
import com.hughes.android.dictionary.DictionaryInfo;
import com.hughes.util.CachingList;
// persisted
final int dictFileVersion;
- private final long creationMillis;
+ public final long creationMillis;
public final String dictInfo;
public final List<PairEntry> pairEntries;
public final List<TextEntry> textEntries;
raf.writeUTF(END_OF_DICTIONARY);
}
- private void writev6Sources(RandomAccessFile out) throws IOException {
- out.writeInt(sources.size());
- long tocPos = out.getFilePointer();
- out.seek(tocPos + sources.size() * 8 + 8);
- for (EntrySource s : sources) {
- long dataPos = out.getFilePointer();
- out.seek(tocPos);
- out.writeLong(dataPos);
- tocPos += 8;
- out.seek(dataPos);
- out.writeUTF(s.getName());
- out.writeInt(s.getNumEntries());
- }
- long dataPos = out.getFilePointer();
- out.seek(tocPos);
- out.writeLong(dataPos);
- out.seek(dataPos);
- }
-
- private void writev6PairEntries(RandomAccessFile out) throws IOException {
- out.writeInt(pairEntries.size());
- long tocPos = out.getFilePointer();
- out.seek(tocPos + pairEntries.size() * 8 + 8);
- for (PairEntry pe : pairEntries) {
- long dataPos = out.getFilePointer();
- out.seek(tocPos);
- out.writeLong(dataPos);
- tocPos += 8;
- out.seek(dataPos);
- out.writeShort(pe.entrySource.index());
- out.writeInt(pe.pairs.size());
- for (PairEntry.Pair p : pe.pairs) {
- out.writeUTF(p.lang1);
- out.writeUTF(p.lang2);
- }
- }
- long dataPos = out.getFilePointer();
- out.seek(tocPos);
- out.writeLong(dataPos);
- out.seek(dataPos);
- }
-
- private void writev6TextEntries(RandomAccessFile out) throws IOException {
- out.writeInt(textEntries.size());
- long tocPos = out.getFilePointer();
- out.seek(tocPos + textEntries.size() * 8 + 8);
- for (TextEntry t : textEntries) {
- long dataPos = out.getFilePointer();
- out.seek(tocPos);
- out.writeLong(dataPos);
- tocPos += 8;
- out.seek(dataPos);
- out.writeShort(t.entrySource.index());
- out.writeUTF(t.text);
- }
- long dataPos = out.getFilePointer();
- out.seek(tocPos);
- out.writeLong(dataPos);
- out.seek(dataPos);
- }
-
- private void writev6EmptyList(RandomAccessFile out) throws IOException {
- out.writeInt(0);
- out.writeLong(out.getFilePointer() + 8);
- }
-
- private void writev6HtmlEntries(RandomAccessFile out) throws IOException {
- out.writeInt(htmlEntries.size());
- long tocPos = out.getFilePointer();
- out.seek(tocPos + htmlEntries.size() * 8 + 8);
- for (HtmlEntry h : htmlEntries) {
- long dataPos = out.getFilePointer();
- out.seek(tocPos);
- out.writeLong(dataPos);
- tocPos += 8;
- out.seek(dataPos);
- out.writeShort(h.entrySource.index());
- out.writeUTF(h.title);
- byte[] data = h.getHtml().getBytes(StandardCharsets.UTF_8);
- out.writeInt(data.length);
- ByteArrayOutputStream baos = new ByteArrayOutputStream();
- GZIPOutputStream gzout = new GZIPOutputStream(baos);
- gzout.write(data);
- gzout.close();
- out.writeInt(baos.size());
- out.write(baos.toByteArray());
- }
- long dataPos = out.getFilePointer();
- out.seek(tocPos);
- out.writeLong(dataPos);
- out.seek(dataPos);
- }
-
- private void writev6HtmlIndices(RandomAccessFile out, List<HtmlEntry> entries) throws IOException {
- out.writeInt(entries.size());
- long tocPos = out.getFilePointer();
- out.seek(tocPos + entries.size() * 8 + 8);
- for (HtmlEntry e : entries) {
- long dataPos = out.getFilePointer();
- out.seek(tocPos);
- out.writeLong(dataPos);
- tocPos += 8;
- out.seek(dataPos);
- out.writeInt(e.index());
- }
- long dataPos = out.getFilePointer();
- out.seek(tocPos);
- out.writeLong(dataPos);
- out.seek(dataPos);
- }
-
- private void writev6IndexEntries(RandomAccessFile out, List<Index.IndexEntry> entries, int[] prunedRowIdx) throws IOException {
- out.writeInt(entries.size());
- long tocPos = out.getFilePointer();
- out.seek(tocPos + entries.size() * 8 + 8);
- for (Index.IndexEntry e : entries) {
- long dataPos = out.getFilePointer();
- out.seek(tocPos);
- out.writeLong(dataPos);
- tocPos += 8;
- out.seek(dataPos);
- out.writeUTF(e.token);
-
- int startRow = e.startRow;
- int numRows = e.numRows;
- if (prunedRowIdx != null) {
- // note: the start row will always be a TokenRow
- // and thus never be pruned
- int newNumRows = 1;
- for (int i = 1; i < numRows; i++) {
- if (prunedRowIdx[startRow + i] >= 0) newNumRows++;
- }
- startRow = prunedRowIdx[startRow];
- numRows = newNumRows;
- }
-
- out.writeInt(startRow);
- out.writeInt(numRows);
- final boolean hasNormalizedForm = !e.token.equals(e.normalizedToken());
- out.writeBoolean(hasNormalizedForm);
- if (hasNormalizedForm) out.writeUTF(e.normalizedToken());
- writev6HtmlIndices(out, prunedRowIdx == null ? e.htmlEntries : Collections.<HtmlEntry>emptyList());
- }
- long dataPos = out.getFilePointer();
- out.seek(tocPos);
- out.writeLong(dataPos);
- out.seek(dataPos);
- }
-
- private void writev6Index(RandomAccessFile out, boolean skipHtml) throws IOException {
- out.writeInt(indices.size());
- long tocPos = out.getFilePointer();
- out.seek(tocPos + indices.size() * 8 + 8);
- for (Index idx : indices) {
- // create pruned index for skipHtml feature
- int[] prunedRowIdx = null;
- int prunedSize = 0;
- if (skipHtml) {
- prunedRowIdx = new int[idx.rows.size()];
- for (int i = 0; i < idx.rows.size(); i++) {
- final RowBase r = idx.rows.get(i);
- // prune Html entries
- boolean pruned = r instanceof HtmlEntry.Row;
- prunedRowIdx[i] = pruned ? -1 : prunedSize;
- if (!pruned) prunedSize++;
- }
- }
-
- long dataPos = out.getFilePointer();
- out.seek(tocPos);
- out.writeLong(dataPos);
- tocPos += 8;
- out.seek(dataPos);
- out.writeUTF(idx.shortName);
- out.writeUTF(idx.longName);
- out.writeUTF(idx.sortLanguage.getIsoCode());
- out.writeUTF(idx.normalizerRules);
- out.writeBoolean(idx.swapPairEntries);
- out.writeInt(idx.mainTokenCount);
- writev6IndexEntries(out, idx.sortedIndexEntries, prunedRowIdx);
-
- // write stoplist, serializing the whole Set *shudder*
- final ByteArrayOutputStream baos = new ByteArrayOutputStream();
- final ObjectOutputStream oos = new ObjectOutputStream(baos);
- oos.writeObject(idx.stoplist);
- oos.close();
- final byte[] bytes = baos.toByteArray();
- out.writeInt(bytes.length);
- out.write(bytes);
-
- out.writeInt(skipHtml ? prunedSize : idx.rows.size());
- out.writeInt(5);
- for (RowBase r : idx.rows) {
- int type = 0;
- if (r instanceof PairEntry.Row) {
- type = 0;
- } else if (r instanceof TokenRow) {
- final TokenRow tokenRow = (TokenRow)r;
- type = tokenRow.hasMainEntry ? 1 : 3;
- } else if (r instanceof TextEntry.Row) {
- type = 2;
- } else if (r instanceof HtmlEntry.Row) {
- type = 4;
- if (skipHtml) continue;
- } else {
- throw new RuntimeException("Row type not supported for v6");
- }
- out.writeByte(type);
- out.writeInt(r.referenceIndex);
- }
- }
- long dataPos = out.getFilePointer();
- out.seek(tocPos);
- out.writeLong(dataPos);
- out.seek(dataPos);
- }
-
- public void writev6(DataOutput out, boolean skipHtml) throws IOException {
- RandomAccessFile raf = (RandomAccessFile)out;
- raf.writeInt(6);
- raf.writeLong(creationMillis);
- raf.writeUTF(dictInfo);
- System.out.println("sources start: " + raf.getFilePointer());
- writev6Sources(raf);
- System.out.println("pair start: " + raf.getFilePointer());
- writev6PairEntries(raf);
- System.out.println("text start: " + raf.getFilePointer());
- writev6TextEntries(raf);
- System.out.println("html index start: " + raf.getFilePointer());
- if (skipHtml) writev6EmptyList(raf);
- else writev6HtmlEntries(raf);
- System.out.println("indices start: " + raf.getFilePointer());
- writev6Index(raf, skipHtml);
- System.out.println("end: " + raf.getFilePointer());
- raf.writeUTF(END_OF_DICTIONARY);
- }
-
private final class IndexSerializer implements RAFListSerializer<Index> {
private final FileChannel ch;
projects / Dictionary.git / blobdiff