Coverage Report

Coverage Report - org.galagosearch.core.parse.TrecTextParser

Classes in this File

Line Coverage

Branch Coverage

Complexity

TrecTextParser

0/41

0/28

4.75

 // BSD License (http://www.galagosearch.org/license)
 package org.galagosearch.core.parse;
 
 import java.io.BufferedReader;
 import java.io.FileNotFoundException;
 import java.io.IOException;
 
 /**
  *
  * @author trevor
  */
 public class TrecTextParser implements DocumentStreamParser {
     BufferedReader reader;
 
     /** Creates a new instance of TrecTextParser */
     public TrecTextParser(BufferedReader reader) throws FileNotFoundException, IOException {
         this.reader = reader;
     }
 
     public String waitFor(String tag) throws IOException {
         String line;
 
         while ((line = reader.readLine()) != null) {
             if (line.startsWith(tag)) {
                 return line;
             }
         }
 
         return null;
     }
 
     public String parseDocNumber() throws IOException {
         String allText = waitFor("<DOCNO>");
 
         while (allText.contains("</DOCNO>") == false) {
             String line = reader.readLine();
             if (line == null) {
                 break;
             }
             allText += line;
         }
 
         int start = allText.indexOf("<DOCNO>") + 7;
         int end = allText.indexOf("</DOCNO>");
 
         return new String(allText.substring(start, end).trim());
     }
 
     public Document nextDocument() throws IOException {
         String line;
 
         if (null == waitFor("<DOC>")) {
             return null;
         }
         String identifier = parseDocNumber();
         StringBuffer buffer = new StringBuffer();
 
         String[] startTags = {"<TEXT>", "<HEADLINE>", "<TITLE>", "<HL>", "<HEAD>",
             "<TTL>", "<DD>", "<DATE>", "<LP>", "<LEADPARA>"
         };
         String[] endTags = {"</TEXT>", "</HEADLINE>", "</TITLE>", "</HL>", "</HEAD>",
             "</TTL>", "</DD>", "</DATE>", "</LP>", "</LEADPARA>"
         };
 
         int inTag = -1;
 
         while ((line = reader.readLine()) != null) {
             if (line.startsWith("</DOC>")) {
                 break;
             }
             if (line.startsWith("<")) {
                 if (inTag >= 0 && line.startsWith(endTags[inTag])) {
                     inTag = -1;
 
                     buffer.append(line);
                     buffer.append(' ');
                 } else if (inTag < 0) {
                     for (int i = 0; i < startTags.length; i++) {
                         if (line.startsWith(startTags[i])) {
                             inTag = i;
                             break;
                         }
                     }
                 }
             }
 
             if (inTag >= 0) {
                 buffer.append(line);
                 buffer.append(' ');
             }
         }
 
         return new Document(identifier, buffer.toString());
     }
 }

1		// BSD License (http://www.galagosearch.org/license)
2		package org.galagosearch.core.parse;
3
4		import java.io.BufferedReader;
5		import java.io.FileNotFoundException;
6		import java.io.IOException;
7
8		/**
9		*
10		* @author trevor
11		*/
12		public class TrecTextParser implements DocumentStreamParser {
13		BufferedReader reader;
14
15		/** Creates a new instance of TrecTextParser */
16	0	public TrecTextParser(BufferedReader reader) throws FileNotFoundException, IOException {
17	0	this.reader = reader;
18	0	}
19
20		public String waitFor(String tag) throws IOException {
21		String line;
22
23	0	while ((line = reader.readLine()) != null) {
24	0	if (line.startsWith(tag)) {
25	0	return line;
26		}
27		}
28
29	0	return null;
30		}
31
32		public String parseDocNumber() throws IOException {
33	0	String allText = waitFor("<DOCNO>");
34
35	0	while (allText.contains("</DOCNO>") == false) {
36	0	String line = reader.readLine();
37	0	if (line == null) {
38	0	break;
39		}
40	0	allText += line;
41	0	}
42
43	0	int start = allText.indexOf("<DOCNO>") + 7;
44	0	int end = allText.indexOf("</DOCNO>");
45
46	0	return new String(allText.substring(start, end).trim());
47		}
48
49		public Document nextDocument() throws IOException {
50		String line;
51
52	0	if (null == waitFor("<DOC>")) {
53	0	return null;
54		}
55	0	String identifier = parseDocNumber();
56	0	StringBuffer buffer = new StringBuffer();
57
58	0	String[] startTags = {"<TEXT>", "<HEADLINE>", "<TITLE>", "<HL>", "<HEAD>",
59		"<TTL>", "<DD>", "<DATE>", "<LP>", "<LEADPARA>"
60		};
61	0	String[] endTags = {"</TEXT>", "</HEADLINE>", "</TITLE>", "</HL>", "</HEAD>",
62		"</TTL>", "</DD>", "</DATE>", "</LP>", "</LEADPARA>"
63		};
64
65	0	int inTag = -1;
66
67	0	while ((line = reader.readLine()) != null) {
68	0	if (line.startsWith("</DOC>")) {
69	0	break;
70		}
71	0	if (line.startsWith("<")) {
72	0	if (inTag >= 0 && line.startsWith(endTags[inTag])) {
73	0	inTag = -1;
74
75	0	buffer.append(line);
76	0	buffer.append(' ');
77	0	} else if (inTag < 0) {
78	0	for (int i = 0; i < startTags.length; i++) {
79	0	if (line.startsWith(startTags[i])) {
80	0	inTag = i;
81	0	break;
82		}
83		}
84		}
85		}
86
87	0	if (inTag >= 0) {
88	0	buffer.append(line);
89	0	buffer.append(' ');
90		}
91		}
92
93	0	return new Document(identifier, buffer.toString());
94		}
95		}