package com.hughes.android.dictionary.parser;
import java.io.BufferedReader;
+import java.io.BufferedInputStream;
+import java.io.DataInputStream;
import java.io.File;
import java.io.FileInputStream;
+import java.io.FileWriter;
import java.io.IOException;
import java.io.InputStreamReader;
import java.nio.charset.Charset;
import java.util.regex.Matcher;
import java.util.regex.Pattern;
+import com.hughes.android.dictionary.engine.AbstractEntry;
import com.hughes.android.dictionary.engine.DictionaryBuilder;
import com.hughes.android.dictionary.engine.EntrySource;
import com.hughes.android.dictionary.engine.EntryTypeName;
+import com.hughes.android.dictionary.engine.HtmlEntry;
import com.hughes.android.dictionary.engine.IndexBuilder;
+import com.hughes.android.dictionary.engine.IndexBuilder.TokenData;
import com.hughes.android.dictionary.engine.IndexedEntry;
import com.hughes.android.dictionary.engine.Language;
import com.hughes.android.dictionary.engine.PairEntry;
+import com.hughes.util.StringUtil;
public class DictFileParser implements Parser {
static final Logger logger = Logger.getLogger(DictFileParser.class.getName());
// Dictcc
- public static final Pattern TAB = Pattern.compile("\\t");
+ public static final String TAB = "\t";
// Chemnitz
- public static final Pattern DOUBLE_COLON = Pattern.compile(" :: ");
- public static final Pattern PIPE = Pattern.compile("\\|");
+ public static final String DOUBLE_COLON = " :: ";
+ public static final String PIPE = "|";
static final Pattern SPACES = Pattern.compile("\\s+");
- static final Pattern BRACKETED = Pattern.compile("\\[([^]]+)\\]");
- static final Pattern PARENTHESIZED = Pattern.compile("\\(([^)]+)\\)");
static final Pattern CURLY_BRACED = Pattern.compile("\\{([^}]+)\\}");
// http://www.regular-expressions.info/unicode.html
final Charset charset;
final boolean flipCols;
- final Pattern fieldSplit;
- final Pattern subfieldSplit;
+ final String fieldSplit;
+ final String subfieldSplit;
+
+ final boolean singleLang;
final DictionaryBuilder dictBuilder;
- final IndexBuilder[] langIndexBuilders;
- final IndexBuilder bothIndexBuilder;
EntrySource entrySource;
// final Set<String> alreadyDone = new HashSet<String>();
public DictFileParser(final Charset charset, boolean flipCols,
- final Pattern fieldSplit, final Pattern subfieldSplit,
- final DictionaryBuilder dictBuilder, final IndexBuilder[] langIndexBuilders,
- final IndexBuilder bothIndexBuilder) {
+ final String fieldSplit, final String subfieldSplit,
+ final boolean singleLang,
+ final DictionaryBuilder dictBuilder) {
this.charset = charset;
this.flipCols = flipCols;
this.fieldSplit = fieldSplit;
this.subfieldSplit = subfieldSplit;
+ this.singleLang = singleLang;
this.dictBuilder = dictBuilder;
- this.langIndexBuilders = langIndexBuilders;
- this.bothIndexBuilder = bothIndexBuilder;
}
@Override
public void parse(final File file, final EntrySource entrySouce, final int pageLimit) throws IOException {
this.entrySource = entrySouce;
- final BufferedReader reader = new BufferedReader(new InputStreamReader(new FileInputStream(file), charset));
+ final BufferedReader reader = new BufferedReader(new InputStreamReader(new FileInputStream(file)));
String line;
int count = 0;
while ((line = reader.readLine()) != null) {
}
private void parseLine(final String line) {
- if (line.startsWith("#") || line.length() == 0) {
+ if (line.startsWith("#") || line.isEmpty()) {
logger.info("Skipping comment line: " + line);
return;
}
- final String[] fields = fieldSplit.split(line);
+ final String[] fields = StringUtil.split(line, fieldSplit);
if (fields.length < 2 || fields.length > 4) {
logger.warning("Malformed line, expected 3 or 4 fields, got " + fields.length + ": " + line);
return;
final String[][] subfields = new String[2][];
if (subfieldSplit != null) {
- subfields[0] = subfieldSplit.split(fields[0]);
- subfields[1] = subfieldSplit.split(fields[1]);
+ subfields[0] = StringUtil.split(fields[0], subfieldSplit);
+ subfields[1] = StringUtil.split(fields[1], subfieldSplit);
if (subfields[0].length != subfields[1].length) {
logger.warning("Number of subfields doesn't match: " + line);
return;
subfields[1] = new String[] { fields[1] };
}
- final PairEntry pairEntry = new PairEntry(entrySource);
for (int i = 0; i < subfields[0].length; ++i) {
subfields[0][i] = subfields[0][i].trim();
subfields[1][i] = subfields[1][i].trim();
- if (subfields[0][i].length() == 0 && subfields[1][i].length() == 0) {
+ if (subfields[0][i].isEmpty() && subfields[1][i].isEmpty()) {
logger.warning("Empty pair: " + line);
continue;
}
- if (subfields[0][i].length() == 0) {
+ if (subfields[0][i].isEmpty()) {
subfields[0][i] = "__";
}
- if (subfields[1][i].length() == 0) {
+ if (subfields[1][i].isEmpty()) {
subfields[1][i] = "__";
}
- pairEntry.pairs.add(new PairEntry.Pair(subfields[0][i], subfields[1][i]));
}
- final IndexedEntry entryData = new IndexedEntry(pairEntry);
- entryData.isValid = true;
-
- for (int l = 0; l < 2; ++l) {
- // alreadyDone.clear();
-
- for (int j = 0; j < subfields[l].length; ++j) {
- String subfield = subfields[l][j];
- final IndexBuilder indexBuilder = langIndexBuilders[l];
- if (indexBuilder.index.sortLanguage == Language.de) {
- subfield = parseField_DE(indexBuilder, subfield, entryData, j);
- } else if (indexBuilder.index.sortLanguage == Language.en) {
- subfield = parseField_EN(indexBuilder, subfield, entryData, j);
+
+ if (singleLang) {
+ HtmlEntry htmlEntry = new HtmlEntry(entrySource, fields[0]);
+ htmlEntry.html = StringUtil.escapeUnicodeToPureHtml(fields[1]);
+
+ final IndexBuilder titleIndexBuilder = dictBuilder.indexBuilders.get(0);
+ htmlEntry.addToDictionary(titleIndexBuilder.index.dict);
+
+ TokenData tokenData = titleIndexBuilder.getOrCreateTokenData(fields[0]);
+ tokenData.hasMainEntry = true;
+ tokenData.htmlEntries.add(0, htmlEntry);
+
+ final String[] tokens = NON_CHAR.split(fields[0]);
+ if (tokens.length > 1) {
+ for (final String token : tokens) {
+ assert token.length() >= 1;
+ assert token.indexOf("-") == -1;
+ if (/*!alreadyDone.contains(token) && */!token.isEmpty()) {
+ tokenData = titleIndexBuilder.getOrCreateTokenData(token);
+ tokenData.htmlEntries.add(htmlEntry);
+ }
+ }
+ }
+ final IndexedEntry entryData = new IndexedEntry(htmlEntry);
+ entryData.isValid = true;
+
+ } else {
+ PairEntry pairEntry = new PairEntry(entrySource);
+ for (int i = 0; i < subfields[0].length; ++i) {
+ pairEntry.pairs.add(new PairEntry.Pair(subfields[0][i], subfields[1][i]));
+ }
+ final IndexedEntry entryData = new IndexedEntry(pairEntry);
+ entryData.isValid = true;
+ for (int l = 0; l < 2; ++l) {
+ // alreadyDone.clear();
+
+ final IndexBuilder indexBuilder = dictBuilder.indexBuilders.get(l);
+ for (int j = 0; j < subfields[l].length; ++j) {
+ String subfield = subfields[l][j];
+ if (indexBuilder.index.sortLanguage == Language.de) {
+ subfield = parseField_DE(indexBuilder, subfield, entryData, j);
+ } else if (indexBuilder.index.sortLanguage == Language.en) {
+ subfield = parseField_EN(indexBuilder, subfield, entryData, j);
+ }
+ parseFieldGeneric(indexBuilder, subfield, entryData, j, subfields[l].length);
}
- parseFieldGeneric(indexBuilder, subfield, entryData, j, subfields[l].length);
}
}
+
+ }
+
+ private StringBuilder extractParenthesized(StringBuilder in, String startChar, String endChar) {
+ StringBuilder res = new StringBuilder();
+ int pos = 0;
+ while ((pos = in.indexOf(startChar, pos)) != -1) {
+ int end = in.indexOf(endChar, pos + 1);
+ if (end == -1) break;
+ res.append(in, pos + 1, end).append(" ");
+ in.replace(pos, end + 1, " ");
+ pos++; // skip the just appended space
+ }
+ return res;
}
private void parseFieldGeneric(final IndexBuilder indexBuilder, String field,
final IndexedEntry entryData, final int subfieldIdx, final int numSubFields) {
+ final StringBuilder fieldsb = new StringBuilder(field);
// remove bracketed and parenthesized stuff.
- final StringBuilder bracketed = new StringBuilder();
- final StringBuilder parenthesized = new StringBuilder();
-
- Matcher matcher;
- while ((matcher = BRACKETED.matcher(field)).find()) {
- bracketed.append(matcher.group(1)).append(" ");
- field = matcher.replaceFirst(" ");
- }
+ final StringBuilder bracketed = extractParenthesized(fieldsb, "[", "]");
+ final StringBuilder parenthesized = extractParenthesized(fieldsb, "(", ")");
- while ((matcher = PARENTHESIZED.matcher(field)).find()) {
- parenthesized.append(matcher.group(1)).append(" ");
- field = matcher.replaceFirst(" ");
- }
-
- field = SPACES.matcher(field).replaceAll(" ").trim();
+ field = fieldsb.toString().trim();
// split words on non -A-z0-9, do them.
final String[] tokens = NON_CHAR_DASH.split(field);
for (String token : tokens) {
token = TRIM_PUNC.matcher(token).replaceAll("");
- if (/*!alreadyDone.contains(token) && */token.length() > 0) {
+ if (/*!alreadyDone.contains(token) && */!token.isEmpty()) {
indexBuilder.addEntryWithTokens(entryData, Collections.singleton(token), entryTypeName);
// alreadyDone.add(token);
// also split words on dashes, do them, too.
- if (token.contains("-")) {
- final String[] dashed = token.split("-");
+ if (token.indexOf('-') != -1) {
+ final String[] dashed = StringUtil.split(token, "-");
for (final String dashedToken : dashed) {
- if (/*!alreadyDone.contains(dashedToken) && */dashedToken.length() > 0) {
+ if (/*!alreadyDone.contains(dashedToken) && */!dashedToken.isEmpty()) {
indexBuilder.addEntryWithTokens(entryData, Collections.singleton(dashedToken), EntryTypeName.PART_OF_HYPHENATED);
}
}
} // for (final String token : tokens) {
// process bracketed stuff (split on spaces and dashes always)
- final String[] bracketedTokens = NON_CHAR.split(bracketed.toString());
- for (final String token : bracketedTokens) {
- assert !token.contains("-");
- if (/*!alreadyDone.contains(token) && */token.length() > 0) {
- indexBuilder.addEntryWithTokens(entryData, Collections.singleton(token), EntryTypeName.BRACKETED);
+ if (bracketed.length() > 0) {
+ final String[] bracketedTokens = NON_CHAR.split(bracketed.toString());
+ for (final String token : bracketedTokens) {
+ assert token.indexOf("-") == -1;
+ if (/*!alreadyDone.contains(token) && */!token.isEmpty()) {
+ indexBuilder.addEntryWithTokens(entryData, Collections.singleton(token), EntryTypeName.BRACKETED);
+ }
}
}
// process paren stuff
- final String[] parenTokens = NON_CHAR.split(parenthesized.toString());
- for (final String token : parenTokens) {
- assert !token.contains("-");
- if (/*!alreadyDone.contains(token) && */token.length() > 0) {
- indexBuilder.addEntryWithTokens(entryData, Collections.singleton(token), EntryTypeName.PARENTHESIZED);
+ if (parenthesized.length() > 0) {
+ final String[] parenTokens = NON_CHAR.split(parenthesized.toString());
+ for (final String token : parenTokens) {
+ assert token.indexOf("-") == -1;
+ if (/*!alreadyDone.contains(token) && */!token.isEmpty()) {
+ indexBuilder.addEntryWithTokens(entryData, Collections.singleton(token), EntryTypeName.PARENTHESIZED);
+ }
}
}
// }
// }
+ if (field.indexOf('{') == -1) return field;
+
// In English, curly braces are used for different tenses.
field = CURLY_BRACED.matcher(field).replaceAll(" ");