/** Gets a word by tag and prints some of its attributes.
 *
 *  @param  tag     Tag.
 *
 *  @throws ModelException
 */
 
void getWordAndPrintAttributes (String tag)
    throws ModelException
{
    Word word = Word.get(tag);
    if (word == null) {
        System.out.println("There is no word with tag: " + tag);
        return;
    }
    System.out.println("tag = " + tag);
    System.out.println("corpus = " + word.getCorpus().getTitle());
    System.out.println("work = " + word.getWork().getTitle());
    System.out.println("work part = " + word.getWorkPart().getTitle());
    System.out.println("word parts:");
    for (WordPart wordPart : word.getWordParts()) {
        System.out.println("   lemma = " + wordPart.getLemma().getTag());
        System.out.println("   pos = " + wordPart.getPos().getTag());
    }
    System.out.println("spelling = " + word.getSpelling().getTag());
    System.out.println("punc before = $" + word.getPuncBefore() + "$");
    System.out.println("punc after = $" + word.getPuncAfter() + "$");
    System.out.println("work ordinal = " + word.getWorkOrdinal());
    System.out.println("collocation ordinal = " + word.getColOrdinal());
}

Example 2. Print a work part for "vertical" reading.

In this example we print the text for a work part "vertically," one word per line. On each line we print the word's spelling plus its lemmas and parts of speech.

/** Prints a work part for "vertical" reading.
 *
 *  @param  workPart        Work part.
 *
 *  @throws ModelException
 */
 
void printWorkPartVertical (WorkPart workPart)
    throws ModelException
{
    Word[] words = workPart.getWords();
    for (Word word : words) {
        System.out.print(word.getSpelling().getTag());
        for (WordPart wordPart : word.getWordParts())
            System.out.print("   " + wordPart.getLemma().getTag() + 
                "/" + wordPart.getPos().getTag());
        System.out.println();
    }
}

Note the call to workPart.getWords(). This convenience method finds all the words in a work part and returns them in order. It is implemented in the WorkPart class as follows:

/** Gets the words in the work part.
 *
 *  @return     Array of words in the work part, in order.
 *
 *  @throws ModelException
 *          <br>Could not get word list
 */
 
public Word[] getWords ()
    throws ModelException
{
    try {
        Collection<Word> words = 
            Word.find(new WorkPartCriterion(this));
        return Word.sort(words, Word.SortOption.WORK_ORDINAL_ASCENDING);
    } catch (Exception e) {
        throw new ModelException("Could not get word list", e);
    }
}

Example 3. Print an HTML concordance of all occurrences of a lemma in a corpus.

/** Prints an HTML concordance of all occurences of a lemma in a corpus.
 *
 *  @param  corpus      Corpus.
 *
 *  @param  lemma       Lemma.
 *
 *  @throws ModelException
 */
 
void printHtmlConcordance (Corpus corpus, Lemma lemma) 
    throws ModelException
{
    Collection<Word> words = Word.find(
        new CorpusCriterion(corpus),
        new LemmaCriterion(lemma));
    Collection<Concordance> result = Concordance.find(10, words);
    System.out.println("<table>");
    for (Concordance c : result) {
        String leftText = c.getLeftText(50);
        Word word = c.getWord();
        String rightText = c.getRightText(50);
        System.out.println("<tr>");
        System.out.println("<td align=\"right\">" + escape(leftText) + "</td>");
        System.out.println("<td align=\"left\">" + 
            "<b>" + escape(word.getSpelling().getTag()) + "</b>" +
            escape(rightText) + "</td>");
        System.out.println("</tr>");
    }
    System.out.println("</table>");
}

/** Escapes a string for HTML.
 *
 *  @param  str     String.
 *
 *  @return         Escaped string.
 */
 
String escape (String str) {
    str = str.replace("&", "&amp;");
    str = str.replace("<", "&lt;");
    str = str.replace(">", "&gt;");
    return str;
}

This example uses the Concordance class to build the concordance after doing the search. The call to Concordance.find requests that KWIC lines be constructed for 10 words to the left and the right of each word found in the search. The calls to getLeft, getWord, and getRight get the left text truncated to 50 characters, the word in the middle that was the search result, and the right text truncated to 50 characters. These items are output in HTML table format. The search result words are aligned vertically and displayed in boldface.

Example 4. Do a full text search for patterns like "handsome, clever, and rich."

In this example we do a full text search for patterns exemplified by the phrase "handsome, clever, and rich," that is, patterns of the form "adjective, adjective, optional coordinating conjunction, adjective."

To be precise, we search the full sequence of word parts in a work for the following pattern:

part of speech syntax category = j or jp
part of speech syntax category = j or jp
optional part of speech word class = cc
part of speech syntax category = j or jp

Only phrases which do not cross sentence boundaries are printed.

/** Finds and prints patterns like "handsome, clever, and rich" in a work.
 *
 *  @param  work    Work.
 *
 *  @throws ModelException
 */
 
void handsomeCleverAndRich (Work work)
    throws ModelException
{
    Collection<WorkPart> workParts = work.getDescendants();
    SyntaxCategory j = SyntaxCategory.get("j");
    SyntaxCategory jp = SyntaxCategory.get("jp");
    WordClass cc = WordClass.get("cc");
    for (WorkPart workPart : workParts) {
        if (!workPart.hasWords()) continue;
        WordPart[] parts = workPart.getWordParts();
        int len = parts.length;
        for (int i = 0; i < len-2; i++) {
            WordPart part1 = parts[i];
            SyntaxCategory syntax = part1.getPos().getSyntaxCategory();
            if (syntax != j && syntax != jp) continue;
            WordPart part2 = parts[i+1];
            syntax = part2.getPos().getSyntaxCategory();
            if (syntax != j && syntax != jp) continue;
            WordPart part3 = parts[i+2];
            WordClass wordClass = part3.getPos().getWordClass();
            if (wordClass == cc) {
                if (i+3 >= len) continue;
                WordPart part4 = parts[i+3];
                syntax = part4.getPos().getSyntaxCategory();
                if (syntax != j && syntax != jp) continue;
                if (crossSentenceBoundary(part1, part2, part3, part4)) continue;
                printMatch(part1, part2, part3, part4);
            } else {
                syntax = part3.getPos().getSyntaxCategory();
                if (syntax != j && syntax != jp) continue;
                if (crossSentenceBoundary(part1, part2, part3)) continue;
                printMatch(part1, part2, part3);
            }
        }
    }
}
    
/** Returns true if a sequence of word parts crosses a sentence boundary.
 *
 *  @param  parts       Array of word parts.
 *
 *  @return             True if word parts cross a sentence boundary.
 */
 
private boolean crossSentenceBoundary (WordPart... parts) {
    int numParts = parts.length;
    Word lastWord = parts[numParts-1].getWord();
    long sentenceInWorkOrdinal = lastWord.getSentenceInWorkOrdinal();
    for (int i = 0; i < numParts-1; i++) {
        Word word = parts[i].getWord();
        if (word.getSentenceInWorkOrdinal() != sentenceInWorkOrdinal) return true;
    }
    return false;
}

/** Prints a match.
 *
 *  @param  parts       Word parts matching the pattern.
 */
 
void printMatch (WordPart... parts) {
    Word location = parts[0].getWord();
    String phrase = WordPart.formatAsString(parts);
    System.out.println(location.getTag() + ":   " + phrase);
}

Note the call to WordPart.formatAsString(parts). This convenience method formats a contiguous sequence of word parts as a string, with punctuation, and with line breaks represented by " / ". There's a similar method Word.formatAsString to format a contiguous sequence of words.

Also note that checking to see if a pattern match crosses a sentence boundary is a bit tricky, because we are matching sequences of word parts, but sentence ordinal is a property of words, not word parts.

Example 5. Find all occurrences of a lemma in a work which appear in speeches.

/** Finds all occurrences of a lemma in a work which appear in speeches.
 *
 *  @param  work        Work.
 *
 *  @param  lemma       Lemma.
 *
 *  @return             Collection of all the words in the work which have the
 *                      specified lemma and which occur inside a TEI "sp"
 *                      element.
 *
 *  @throws ModelException
 */
 
Collection<Word> findWordsInSpeeches (Work work, Lemma lemma) 
    throws ModelException
{
    return
        Word.find(
            new WorkCriterion(work),
            new LemmaCriterion(lemma),
            new ContextPatternCriterion("*/sp/*"));
}