X-Git-Url: http://gitweb.fperrin.net/?a=blobdiff_plain;f=src%2Fcom%2Fhughes%2Fandroid%2Fdictionary%2Fengine%2FDictionaryTest.java;h=d1dcc2b36ebe05e4536aa23d2d7fe3d0c7ce910a;hb=929387f3294346accc6e3eaa514558b446a8e474;hp=bfceeffa515a57b5a22f61f9eb68210d3213d212;hpb=b256bde15d0f6e6cb5325fbe22c76496e40194eb;p=DictionaryPC.git diff --git a/src/com/hughes/android/dictionary/engine/DictionaryTest.java b/src/com/hughes/android/dictionary/engine/DictionaryTest.java index bfceeff..d1dcc2b 100644 --- a/src/com/hughes/android/dictionary/engine/DictionaryTest.java +++ b/src/com/hughes/android/dictionary/engine/DictionaryTest.java @@ -1,192 +1,397 @@ +// Copyright 2011 Google Inc. All Rights Reserved. +// +// Licensed 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 com.hughes.android.dictionary.engine; import java.io.IOException; import java.io.RandomAccessFile; -import java.util.ArrayList; import java.util.Arrays; -import java.util.Collections; import java.util.List; +import java.util.Random; import java.util.concurrent.atomic.AtomicBoolean; import junit.framework.TestCase; -import com.hughes.android.dictionary.engine.Index.SearchResult; +import com.hughes.android.dictionary.engine.Index.IndexEntry; +import com.hughes.util.CollectionUtil; public class DictionaryTest extends TestCase { - - RandomAccessFile raf; - Dictionary dict; - Index deIndex; - - @Override - public void setUp() { - try { - raf = new RandomAccessFile("testdata/de_en.dict", "r"); - dict = new Dictionary(raf); - } catch (IOException e) { - throw new RuntimeException(e); - } - - deIndex = dict.indices.get(0); -} - - @Override - public void tearDown() { - try { - raf.close(); - } catch (IOException e) { - throw new RuntimeException(e); - } - } - - - public void testGermanMetadata() throws IOException { - assertEquals("de", deIndex.shortName); - assertEquals("de->en", deIndex.longName); - } - - public void testGermanIndex() throws IOException { - for (final Index.IndexEntry indexEntry : deIndex.sortedIndexEntries) { - System.out.println("testing: " + indexEntry.token); - final Index.SearchResult searchResult = deIndex.findLongestSubstring(indexEntry.token, new AtomicBoolean( - false)); - assertEquals(indexEntry.token.toLowerCase(), searchResult.insertionPoint.token.toLowerCase()); - assertEquals(indexEntry.token.toLowerCase(), searchResult.longestPrefix.token.toLowerCase()); - } - - // TODO: maybe if user types capitalization, use it. - assertSearchResult("aaac", "aaac", deIndex.findLongestSubstring("aaac", new AtomicBoolean(false))); - assertSearchResult("aaac", "aaac", deIndex.findLongestSubstring("AAAC", new AtomicBoolean(false))); - assertSearchResult("aaac", "aaac", deIndex.findLongestSubstring("AAAc", new AtomicBoolean(false))); - assertSearchResult("aaac", "aaac", deIndex.findLongestSubstring("aAac", new AtomicBoolean(false))); - - // Before the beginning. - assertSearchResult("40", "40" /* special case */, deIndex.findLongestSubstring("__", new AtomicBoolean(false))); - - // After the end. - assertSearchResult("Zweckorientiertheit", "zählen", deIndex.findLongestSubstring("ZZZZZ", new AtomicBoolean(false))); - - assertSearchResult("ab", "aaac", deIndex.findLongestSubstring("aaaca", new AtomicBoolean(false))); - assertSearchResult("machen", "machen", deIndex.findLongestSubstring("m", new AtomicBoolean(false))); - - - assertSearchResult("überprüfe", "überprüfe", deIndex.findLongestSubstring("ueberprüfe", new AtomicBoolean(false))); - assertSearchResult("überprüfe", "überprüfe", deIndex.findLongestSubstring("ueberpruefe", new AtomicBoolean(false))); - - assertSearchResult("überprüfe", "überprüfe", deIndex.findLongestSubstring("ueberpBLEH", new AtomicBoolean(false))); - assertSearchResult("überprüfe", "überprüfe", deIndex.findLongestSubstring("überprBLEH", new AtomicBoolean(false))); - - assertSearchResult("überprüfen", "überprüfe", deIndex.findLongestSubstring("überprüfeBLEH", new AtomicBoolean(false))); - - } - - private void assertSearchResult(final String insertionPoint, final String longestPrefix, - final SearchResult actual) { - assertEquals(insertionPoint, actual.insertionPoint.token); - assertEquals(longestPrefix, actual.longestPrefix.token); - } - - public void testGermanTokenRows() { - // Pre-cache a few of these, just to make sure that's working. - for (int i = 0; i < deIndex.rows.size(); i += 7) { - deIndex.rows.get(i).getTokenRow(true); - } - - // Do the exhaustive searching. - TokenRow lastTokenRow = null; - for (final RowBase row : deIndex.rows) { - if (row instanceof TokenRow) { - lastTokenRow = (TokenRow) row; - } - assertEquals(lastTokenRow, row.getTokenRow(true)); - } - - // Now they're all cached, we shouldn't have to search. - for (final RowBase row : deIndex.rows) { - if (row instanceof TokenRow) { - lastTokenRow = (TokenRow) row; - } - // This will break if the Row cache isn't big enough. - assertEquals(lastTokenRow, row.getTokenRow(false)); - } - } - - public void testGermanSort() { - assertEquals("aüÄÄ", Language.de.textNorm("aueAeAE", false)); - final List words = Arrays.asList( - "er-ben", - "erben", - "Erben", - "Erbse", - "Erbsen", - "essen", - "Essen", - "Grosformat", - "Grosformats", - "Grossformat", - "Großformat", - "Grossformats", - "Großformats", - "Großpoo", - "Großpoos", - "hulle", - "Hulle", - "hülle", - "huelle", - "Hülle", - "Huelle", - "Hum" - ); - assertEquals(0, Language.de.sortComparator.compare("hülle", "huelle")); - assertEquals(0, Language.de.sortComparator.compare("huelle", "hülle")); - - assertEquals(-1, Language.de.sortComparator.compare("hülle", "Hülle")); - assertEquals(0, Language.de.findComparator.compare("hülle", "Hülle")); - assertEquals(-1, Language.de.findComparator.compare("hulle", "Hülle")); - - - for (final String s : words) { - System.out.println(s + "\t" + Language.de.textNorm(s, false)); - } - final List sorted = new ArrayList(words); -// Collections.shuffle(shuffled, new Random(0)); - Collections.sort(sorted, Language.de.sortComparator); - System.out.println(sorted.toString()); - for (int i = 0; i < words.size(); ++i) { - System.out.println(words.get(i) + "\t" + sorted.get(i)); - assertEquals(words.get(i), sorted.get(i)); - } - } - - @SuppressWarnings("unchecked") - public void testEnglishSort() { - - final List words = Arrays.asList( - "pre-print", - "preppie", - "preppy", - "preprocess"); - - final List sorted = new ArrayList(words); - Collections.sort(sorted, Language.en.getSortCollator()); - for (int i = 0; i < words.size(); ++i) { - if (i > 0) { - assertTrue(Language.en.getSortCollator().compare(words.get(i-1), words.get(i)) < 0); - } - System.out.println(words.get(i) + "\t" + sorted.get(i)); - assertEquals(words.get(i), sorted.get(i)); - } - - assertTrue(Language.en.getSortCollator().compare("pre-print", "preppy") < 0); - - } - - public void testLanguage() { - assertEquals(Language.de, Language.lookup("de")); - assertEquals(Language.en, Language.lookup("en")); - assertEquals("es", Language.lookup("es").getSymbol()); - } + static final String TEST_OUTPUTS = com.hughes.android.dictionary.engine.DictionaryBuilderTest.TEST_OUTPUTS; + public static final String OUTPUTS = "data/outputs/"; + + @Override + protected void setUp() { + while (!TransliteratorManager.init(null, null)) { + try { + Thread.sleep(10); + } catch (InterruptedException e) { + e.printStackTrace(); + } + } + } + + public void testURLFormatting() { + } + + public void testEnItWiktionary() throws IOException { + final RandomAccessFile raf = new RandomAccessFile(OUTPUTS + "EN-IT.quickdic", "r"); + final Dictionary dict = new Dictionary(raf.getChannel()); + final Index enIndex = dict.indices.get(0); + + final RowBase row = enIndex.rows.get(4); + assertEquals("-ical", row.getRawText(false)); + + final Index itIndex = dict.indices.get(1); + { + final List rows = itIndex.multiWordSearch("come mai", Arrays.asList("come", "mai"), new AtomicBoolean(false)); + System.out.println(CollectionUtil.join(rows, "\n ")); + assertTrue(rows.toString(), rows.size() > 0); + assertTrue(rows.get(0).toString().startsWith("come mai@")); + assertTrue(rows.get(0) instanceof TokenRow); + assertTrue(!((TokenRow)rows.get(0)).getIndexEntry().htmlEntries.isEmpty()); + } + + { + final List rows = itIndex.multiWordSearch("buon g", Arrays.asList("buon", "g"), new AtomicBoolean(false)); + System.out.println(CollectionUtil.join(rows, "\n ")); + assertTrue(rows.toString(), rows.size() > 0); + assertTrue(rows.get(0).toString().startsWith("buon giorno@")); + assertTrue(rows.get(0) instanceof TokenRow); + assertTrue(!((TokenRow)rows.get(0)).getIndexEntry().htmlEntries.isEmpty()); + } + + { + final IndexEntry searchResult = itIndex.findInsertionPoint("azzurro", new AtomicBoolean( + false)); + HtmlEntry htmlEntry = searchResult.htmlEntries.get(0); + System.out.println("azzurro:\n" + htmlEntry.getHtml()); + } + + raf.close(); + } + +// public void testFr() throws IOException { +// final RandomAccessFile raf = new RandomAccessFile(OUTPUTS + "FR.quickdic", "r"); +// final Dictionary dict = new Dictionary(raf); +// final Index frIndex = dict.indices.get(0); +// +// // Now they're all cached, we shouldn't have to search. +// for (final IndexEntry indexEntry : frIndex.sortedIndexEntries) { +// System.out.println(indexEntry.token); +// } +// +// raf.close(); +// } + + + public void testDeEnWiktionary() throws IOException { + final RandomAccessFile raf = new RandomAccessFile(OUTPUTS + "DE-EN.quickdic", "r"); + final Dictionary dict = new Dictionary(raf.getChannel()); + + final Index deIndex = dict.indices.get(0); + + { + final IndexEntry searchResult = deIndex.findInsertionPoint("rot", new AtomicBoolean( + false)); + HtmlEntry htmlEntry = searchResult.htmlEntries.get(0); + System.out.println("rot:\n" + htmlEntry.getHtml()); + } + + raf.close(); + } + + public void testGermanMetadata() throws IOException { + final RandomAccessFile raf = new RandomAccessFile(TEST_OUTPUTS + "de-en.quickdic", "r"); + final Dictionary dict = new Dictionary(raf.getChannel()); + final Index deIndex = dict.indices.get(0); + + assertEquals("DE", deIndex.shortName); + assertEquals("DE->EN", deIndex.longName); + + assertEquals(2, dict.sources.size()); + assertEquals("chemnitz", dict.sources.get(0).name); + assertEquals("dictcc", dict.sources.get(1).name); + + assertEquals("dictcc", dict.pairEntries.get(0).entrySource.name); + assertEquals("chemnitz", dict.pairEntries.get(1).entrySource.name); + + raf.close(); + } + + public void testGermanIndex() throws IOException { + final RandomAccessFile raf = new RandomAccessFile(TEST_OUTPUTS + "de-en.quickdic", "r"); + final Dictionary dict = new Dictionary(raf.getChannel()); + final Index deIndex = dict.indices.get(0); + + for (final Index.IndexEntry indexEntry : deIndex.sortedIndexEntries) { + System.out.println("testing: " + indexEntry.token); + final IndexEntry searchResult = deIndex.findInsertionPoint(indexEntry.token, new AtomicBoolean( + false)); + assertEquals("Looked up: " + indexEntry.token, indexEntry.token.toLowerCase(), searchResult.token.toLowerCase()); + } + + // TODO: maybe if user types capitalization, use it. + assertSearchResult("aaac", "aaac", deIndex.findInsertionPoint("aaac", new AtomicBoolean(false))); + assertSearchResult("aaac", "aaac", deIndex.findInsertionPoint("AAAC", new AtomicBoolean(false))); + assertSearchResult("aaac", "aaac", deIndex.findInsertionPoint("AAAc", new AtomicBoolean(false))); + assertSearchResult("aaac", "aaac", deIndex.findInsertionPoint("aAac", new AtomicBoolean(false))); + + // Before the beginning. + assertSearchResult("40", "40" /* special case */, deIndex.findInsertionPoint("", new AtomicBoolean(false))); + assertSearchResult("40", "40" /* special case */, deIndex.findInsertionPoint("__", new AtomicBoolean(false))); + + // After the end. + assertSearchResult("Zweckorientiertheit", "zählen", deIndex.findInsertionPoint("ZZZZZ", new AtomicBoolean(false))); + + assertSearchResult("ab", "aaac", deIndex.findInsertionPoint("aaaca", new AtomicBoolean(false))); + assertSearchResult("machen", "machen", deIndex.findInsertionPoint("m", new AtomicBoolean(false))); + assertSearchResult("machen", "machen", deIndex.findInsertionPoint("macdddd", new AtomicBoolean(false))); + + + assertSearchResult("überprüfe", "überprüfe", deIndex.findInsertionPoint("ueberprüfe", new AtomicBoolean(false))); + assertSearchResult("überprüfe", "überprüfe", deIndex.findInsertionPoint("ueberpruefe", new AtomicBoolean(false))); + + assertSearchResult("überprüfe", "überprüfe", deIndex.findInsertionPoint("ueberpBLEH", new AtomicBoolean(false))); + assertSearchResult("überprüfe", "überprüfe", deIndex.findInsertionPoint("überprBLEH", new AtomicBoolean(false))); + + assertSearchResult("überprüfen", "überprüfe", deIndex.findInsertionPoint("überprüfeBLEH", new AtomicBoolean(false))); + + // Check that search in lowercase works. + assertSearchResult("Alibi", "Alibi", deIndex.findInsertionPoint("alib", new AtomicBoolean(false))); + System.out.println(deIndex.findInsertionPoint("alib", new AtomicBoolean(false)).toString()); + + raf.close(); + } + + private void assertSearchResult(final String insertionPoint, final String longestPrefix, + final IndexEntry actual) { + assertEquals(insertionPoint, actual.token); + } + + public void testGermanTokenRows() throws IOException { + final RandomAccessFile raf = new RandomAccessFile(TEST_OUTPUTS + "de-en.quickdic", "r"); + final Dictionary dict = new Dictionary(raf.getChannel()); + final Index deIndex = dict.indices.get(0); + + // Pre-cache a few of these, just to make sure that's working. + for (int i = 0; i < deIndex.rows.size(); i += 7) { + deIndex.rows.get(i).getTokenRow(true); + } + + // Do the exhaustive searching. + TokenRow lastTokenRow = null; + for (final RowBase row : deIndex.rows) { + if (row instanceof TokenRow) { + lastTokenRow = (TokenRow) row; + } + assertEquals(lastTokenRow, row.getTokenRow(true)); + } + + // Now they're all cached, we shouldn't have to search. + for (final RowBase row : deIndex.rows) { + if (row instanceof TokenRow) { + lastTokenRow = (TokenRow) row; + } + // This will break if the Row cache isn't big enough. + assertEquals(lastTokenRow, row.getTokenRow(false)); + } + + raf.close(); + } + + public void testChemnitz() throws IOException { + final RandomAccessFile raf = new RandomAccessFile(TEST_OUTPUTS + "de-en.quickdic", "r"); + final Dictionary dict = new Dictionary(raf.getChannel()); + final Index deIndex = dict.indices.get(0); + + assertSearchResult("Höschen", "Hos", deIndex.findInsertionPoint("Hos", new AtomicBoolean(false))); + assertSearchResult("Höschen", "hos", deIndex.findInsertionPoint("hos", new AtomicBoolean(false))); + + raf.close(); + } + + public void testMultiSearch() throws IOException { + final RandomAccessFile raf = new RandomAccessFile(TEST_OUTPUTS + "de-en.quickdic", "r"); + final Dictionary dict = new Dictionary(raf.getChannel()); + final Index deIndex = dict.indices.get(0); + + { + final List rows = deIndex.multiWordSearch("aaa aaab", Arrays.asList("aaa", "aaab"), new AtomicBoolean(false)); + System.out.println(CollectionUtil.join(rows, "\n ")); + assertTrue(rows.toString(), rows.size() > 0); + } + + raf.close(); + } + + public void testMultiSearchIt() throws IOException { + final RandomAccessFile raf = new RandomAccessFile(OUTPUTS + "IT.quickdic", "r"); + final Dictionary dict = new Dictionary(raf.getChannel()); + final Index index = dict.indices.get(0); + + { + final List rows = index.multiWordSearch("fare centro", + Arrays.asList("fare", "centro"), new AtomicBoolean(false)); + System.out.println(CollectionUtil.join(rows, "\n ")); + assertTrue(rows.toString(), rows.size() > 0); + assertTrue(rows.get(0).toString().startsWith("fare centro@")); + } + } + + public void testMultiSearchDeBig() throws IOException { + final RandomAccessFile raf = new RandomAccessFile(OUTPUTS + "DE-EN.quickdic", "r"); + final Dictionary dict = new Dictionary(raf.getChannel()); + final Index enIndex = dict.indices.get(1); + + { + final List rows = enIndex.multiWordSearch("train station", Arrays.asList("train", "station"), new AtomicBoolean(false)); + System.out.println(CollectionUtil.join(rows, "\n ")); + assertTrue(rows.toString(), rows.size() > 0); + assertTrue(rows.get(0).toString().startsWith("train station@")); + } + + { + final List rows = enIndex.multiWordSearch("a train station", Arrays.asList("a", "train", "station"), new AtomicBoolean(false)); + System.out.println(CollectionUtil.join(rows, "\n ")); + assertTrue(rows.toString(), rows.size() > 0); + assertEquals("Bahnhofsuhr {{de-noun|g=f|plural=Bahnhofsuhren}}\tstation clock (at a train station)", rows.get(0).toString()); + } + + { + final List rows = enIndex.multiWordSearch("a station", Arrays.asList("a", "station"), new AtomicBoolean(false)); + System.out.println(CollectionUtil.join(rows, "\n ")); + assertTrue(rows.toString(), rows.size() > 0); + assertEquals("Abfahrthalle {en-noun}\tDeparture room of a station.", rows.get(0).toString()); + } + + { + // Should print: Giving up, too many words with prefix: p + final List rows = enIndex.multiWordSearch("p eat", Arrays.asList("p", "eat"), new AtomicBoolean(false)); + System.out.println(CollectionUtil.join(rows, "\n ")); + assertTrue(rows.toString(), rows.size() > 0); + assertTrue(rows.toString().contains("verschlingen; verputzen\tto dispatch (eat)")); + } + + { + // Should print: Giving up, too many words with prefix: p + final List rows = enIndex.multiWordSearch("p p", Arrays.asList("p", "p"), new AtomicBoolean(false)); + assertTrue(rows.size() >= 1000); + } + + { + // Should print: Giving up, too many words with prefix: a + final List rows = enIndex.multiWordSearch("a a", Arrays.asList("a", "a"), new AtomicBoolean(false)); + assertTrue(rows.size() >= 1000); + } + + { + // Should print: Giving up, too many words with prefix: a + final List rows = enIndex.multiWordSearch("b ba", Arrays.asList("b", "ba"), new AtomicBoolean(false)); + assertTrue(rows.size() >= 1000); + } + + { + // Should print: Giving up, too many words with prefix: a + final List rows = enIndex.multiWordSearch("b ba", Arrays.asList("b", "ba"), new AtomicBoolean(false)); + assertTrue(rows.size() >= 1000); + } + + raf.close(); + } + + public void testMultiSearchBigAF() throws IOException { + final RandomAccessFile raf = new RandomAccessFile(OUTPUTS + "AF-EN.quickdic", "r"); + final Dictionary dict = new Dictionary(raf.getChannel()); + final Index enIndex = dict.indices.get(1); + + { + final List rows = enIndex.multiWordSearch("pig eats", Arrays.asList("pig", "eats"), new AtomicBoolean(false)); + System.out.println(CollectionUtil.join(rows, "\n ")); + assertTrue(rows.toString(), rows.size() > 0); + assertEquals("vark\tpig (someone who overeats or eats rapidly) (noun)", rows.get(0).toString()); + } + + { + final List rows = enIndex.multiWordSearch("pig eat", Arrays.asList("pig", "eat"), new AtomicBoolean(false)); + System.out.println(CollectionUtil.join(rows, "\n ")); + assertTrue(rows.toString(), rows.size() > 0); + assertEquals("vark\tpig (someone who overeats or eats rapidly) (noun)", rows.get(0).toString()); + } + + { + final List rows = enIndex.multiWordSearch("pi ea", Arrays.asList("pi", "ea"), new AtomicBoolean(false)); + System.out.println(CollectionUtil.join(rows, "\n ")); + assertTrue(rows.toString(), rows.size() > 0); + assertTrue(rows.toString().contains("vark\tpig (someone who overeats or eats rapidly) (noun)")); + } + + { + final List rows = enIndex.multiWordSearch("p eat", Arrays.asList("p", "eat"), new AtomicBoolean(false)); + System.out.println(CollectionUtil.join(rows, "\n ")); + assertTrue(rows.toString(), rows.size() > 0); + assertTrue(rows.toString().contains("vark\tpig (someone who overeats or eats rapidly) (noun)")); + } + + + raf.close(); + } + + + public void testExactSearch() throws IOException { + final RandomAccessFile raf = new RandomAccessFile(OUTPUTS + "EN-cmn.quickdic", "r"); + final Dictionary dict = new Dictionary(raf.getChannel()); + final Index cmnIndex = dict.indices.get(1); + + final Random random = new Random(10); + + for (int i = 0; i < 1000; ++i) { + final int ii = random.nextInt(cmnIndex.sortedIndexEntries.size()); + final IndexEntry indexEntry = cmnIndex.sortedIndexEntries.get(ii); + final IndexEntry found = cmnIndex.findExact(indexEntry.token); + assertNotNull(found); + assertEquals(indexEntry.token, found.token); + assertEquals(indexEntry, found); // Test of caching.... + } + + raf.close(); + } + + public void testThai() throws IOException { + final RandomAccessFile raf = new RandomAccessFile(OUTPUTS + "EN-TH.quickdic", "r"); + final Dictionary dict = new Dictionary(raf.getChannel()); + final Index thIndex = dict.indices.get(1); + + final IndexEntry entry = thIndex.findInsertionPoint("ดี", new AtomicBoolean(false)); + assertEquals("di", entry.token); + + raf.close(); + } + + public void testNorwegian() throws IOException { + final RandomAccessFile raf = new RandomAccessFile(OUTPUTS + "EN-NL.quickdic", "r"); + final Dictionary dict = new Dictionary(raf.getChannel()); + final Index nlIndex = dict.indices.get(1); + + IndexEntry entry = nlIndex.findInsertionPoint("Xhosa", new AtomicBoolean(false)); + assertEquals("Xhosa", entry.token); + + entry = nlIndex.findInsertionPoint("Zyne", new AtomicBoolean(false)); + assertEquals("Zyne", entry.token); + + raf.close(); + } }