1 | package org.greenstone.atea;
|
---|
2 |
|
---|
3 | import java.io.*;
|
---|
4 | import java.util.HashMap;
|
---|
5 | import java.util.Map;
|
---|
6 |
|
---|
7 | import org.apache.log4j.Logger;
|
---|
8 |
|
---|
9 |
|
---|
10 | public class TextDumpPage {
|
---|
11 | private static Logger logger = Logger.getLogger(org.greenstone.atea.TextDumpPage.class.getName());
|
---|
12 |
|
---|
13 | public static final String TEXT_START_MARKER="text:start:";
|
---|
14 | public static final String TEXT_END_MARKER="text:end:";
|
---|
15 |
|
---|
16 | private Map<String, String> tuples;
|
---|
17 |
|
---|
18 | public TextDumpPage(String siteID, String unparsedPageDump) {
|
---|
19 | tuples = new HashMap<String, String>();
|
---|
20 |
|
---|
21 | try (
|
---|
22 | BufferedReader reader = new BufferedReader(new StringReader(unparsedPageDump));
|
---|
23 | ) {
|
---|
24 |
|
---|
25 | String line = reader.readLine(); // should have at least first line
|
---|
26 |
|
---|
27 | // first line always has a "key:" somewhere after the pageURL
|
---|
28 | int endIndex = line.indexOf("key:");
|
---|
29 |
|
---|
30 | String pageURL = line.substring(0, endIndex);
|
---|
31 | //String pageURL = (endIndex == -1) ? line : line.substring(0, endIndex);
|
---|
32 |
|
---|
33 |
|
---|
34 | tuples.put("pageURL", pageURL.trim());
|
---|
35 |
|
---|
36 | //if(endIndex != -1) {
|
---|
37 | String key = line.substring(endIndex);
|
---|
38 | tuples.put("key", key.trim());
|
---|
39 | //} else {
|
---|
40 | //logger.debug("@@@@ no key for pageURL: " + pageURL);
|
---|
41 | //}
|
---|
42 | /*
|
---|
43 | if(pageURL.contains(TEXT_END_MARKER)) {
|
---|
44 | logger.debug("@@@@ TEXT_END_MARKER assigned to pageURL for page: ");
|
---|
45 | logger.debug("+++++++++");
|
---|
46 | logger.debug(unparsedPageDump);
|
---|
47 | logger.debug("+++++++++");
|
---|
48 | }
|
---|
49 | */
|
---|
50 |
|
---|
51 | boolean readingPageText = false;
|
---|
52 | StringBuilder pageText = null;
|
---|
53 |
|
---|
54 | // continue reading all other tuples for this page, if any
|
---|
55 | while((line = reader.readLine()) != null) {
|
---|
56 | line = line.trim();
|
---|
57 |
|
---|
58 | // check if we're dealing with metadata or start/end of page's text body
|
---|
59 | // or actual text body
|
---|
60 |
|
---|
61 | if(line.equals(TEXT_START_MARKER)) { // dealing with the page body text
|
---|
62 | pageText = new StringBuilder();
|
---|
63 | readingPageText = true;
|
---|
64 | }
|
---|
65 | else if(line.equals(TEXT_END_MARKER)) {
|
---|
66 | // finished with a page body
|
---|
67 | // Remove any FINAL artificial newline we introduced to a page's body text
|
---|
68 | tuples.put("pageText", pageText.toString().trim());
|
---|
69 | readingPageText = false;
|
---|
70 | pageText = null;
|
---|
71 | }
|
---|
72 | else {
|
---|
73 | if(readingPageText) { // So we're reading in the page text
|
---|
74 | pageText.append(line);
|
---|
75 | pageText.append("\n"); // there are no newlines within pageText
|
---|
76 | // but if there were newlines, add them back here as readLine() removes them
|
---|
77 | }
|
---|
78 | else { // dealing with the rest of the page dump's metadata
|
---|
79 | // QTODO: nutch's text dump output is problematic
|
---|
80 | // strange characters are in the stream and end up here
|
---|
81 | // and can't detect end of metadata or even end of line.
|
---|
82 | endIndex = line.indexOf(":");
|
---|
83 | if(endIndex != -1) {
|
---|
84 | String k = line.substring(0, endIndex);
|
---|
85 | String v = line.substring(endIndex+1);
|
---|
86 | if(k.startsWith("metadata")) {
|
---|
87 | k = k.substring("metadata".length());
|
---|
88 | }
|
---|
89 |
|
---|
90 | tuples.put(k.trim(), v.trim());
|
---|
91 | } else {
|
---|
92 | if(NutchTextDumpProcessor.DEBUG_MODE) {
|
---|
93 | logger.error("No meta key for meta: " + line);
|
---|
94 | logger.error(unparsedPageDump);
|
---|
95 | }
|
---|
96 | }
|
---|
97 | }
|
---|
98 | }
|
---|
99 |
|
---|
100 | }
|
---|
101 |
|
---|
102 | // If the page had no pageText, add a "pageText" -> "" mapping
|
---|
103 | if(!tuples.containsKey("pageText")) {
|
---|
104 | tuples.put("pageText", "");
|
---|
105 | }
|
---|
106 |
|
---|
107 |
|
---|
108 | } catch (IOException ioe) {
|
---|
109 | logger.error("@@@@@@@@@ Error reading in txtdump of a page.", ioe);
|
---|
110 | }
|
---|
111 |
|
---|
112 |
|
---|
113 | // START DEBUG
|
---|
114 | debugTuples();
|
---|
115 | // END DEBUG
|
---|
116 |
|
---|
117 | }
|
---|
118 |
|
---|
119 | public void debugTuples() {
|
---|
120 | if(NutchTextDumpProcessor.DEBUG_MODE) {
|
---|
121 | logger.debug("__________________________________________");
|
---|
122 | for(Map.Entry<String, String> entry : tuples.entrySet()) {
|
---|
123 | String key = entry.getKey();
|
---|
124 | String value = entry.getValue();
|
---|
125 | logger.debug(key + " - " + value);
|
---|
126 | }
|
---|
127 | logger.debug("__________________________________________");
|
---|
128 | }
|
---|
129 | }
|
---|
130 |
|
---|
131 |
|
---|
132 | public String getPageURL() {
|
---|
133 | return tuples.get("pageURL");
|
---|
134 | }
|
---|
135 |
|
---|
136 | public String getPageText() {
|
---|
137 | return tuples.get("pageText");
|
---|
138 | }
|
---|
139 |
|
---|
140 | /* Dr Nichols suggested storing timestamp and char encoding. Not sure which timestamp
|
---|
141 | or encoding he meant, but storing 2 of several timestamps and selecting
|
---|
142 | original character encoding (presumably the char encoding of the page) out of 2
|
---|
143 | pieces of char encoding metadata to store. */
|
---|
144 | public String getModifiedTime() {
|
---|
145 | // is this the webpage's last mod time?
|
---|
146 | String time = tuples.get("modifiedTime");
|
---|
147 | time = time.equals("0") ? "" : time; // zero will be assumed to be epoch, rather than unset
|
---|
148 | return time;
|
---|
149 | }
|
---|
150 | public String getFetchTime() {
|
---|
151 | // is this the nutch crawl time
|
---|
152 | String time = tuples.get("fetchTime");
|
---|
153 | time = time.equals("0") ? "" : time; // zero will be assumed to be epoch, rather than unset
|
---|
154 | return time;
|
---|
155 |
|
---|
156 | }
|
---|
157 | public String getOriginalCharEncoding() {
|
---|
158 | // is this the web page's char-encoding?
|
---|
159 | return tuples.get("OriginalCharEncoding");
|
---|
160 | }
|
---|
161 |
|
---|
162 | public String get(String key) {
|
---|
163 | return tuples.get(key);
|
---|
164 | }
|
---|
165 |
|
---|
166 | public void add(String key, String value) {
|
---|
167 | tuples.put(key, value);
|
---|
168 | }
|
---|
169 |
|
---|
170 | public void addMRILanguageStatus(boolean status) {
|
---|
171 | if(status) {
|
---|
172 | tuples.put("isMRI", "true");
|
---|
173 | } else {
|
---|
174 | tuples.put("isMRI", "false");
|
---|
175 | }
|
---|
176 | }
|
---|
177 |
|
---|
178 | public boolean getMRILanguageStatus() {
|
---|
179 | String value = tuples.get("isMRI");
|
---|
180 | if(value == null) {
|
---|
181 | return false;
|
---|
182 | }
|
---|
183 | if(value.equals("true")) {
|
---|
184 | return true;
|
---|
185 | }
|
---|
186 | else {
|
---|
187 | return false;
|
---|
188 | }
|
---|
189 |
|
---|
190 | }
|
---|
191 |
|
---|
192 | }
|
---|