trunk/Search/Processor.cpp

/* ============================================================================
 * Douglas Thrift's Search Engine License
 *
 * Copyright (C) 2002-2004, Douglas Thrift. All Rights Reserved.
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions are met:
 *
 * 1. Redistributions of source code must retain the above copyright notice,
 *    this list of conditions and the following disclaimer.
 *
 * 2. Redistributions in binary form must reproduce the above copyright notice,
 *    this list of conditions and the following disclaimer in the documentation
 *    and/or other materials provided with the distribution.
 *
 * 3. The end-user documentation included with the redistribution, if any, must
 *    include the following acknowledgment:
 *
 *       "This product includes software developed by Douglas Thrift
 *       (http://computers.douglasthrift.net/searchengine/)."
 *
 *    Alternately, this acknowledgment may appear in the software itself, if
 *    and wherever such third-party acknowledgments normally appear.
 *
 * 4. The names "Douglas Thrift" and "Douglas Thrift's Search Engine" must not
 *    be used to endorse or promote products derived from this software without
 *    specific prior written permission.  For written permission, please visit
 *    http://www.douglasthrift.net/contact.cgi for contact information.
 *
 * 5. Products derived from this software may not be called "Douglas Thrift's
 *    Search Engine", nor may "Douglas Thrift's Search Engine" appear in their
 *    name, without prior written permission.
 *
 * THIS SOFTWARE IS PROVIDED "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES,
 * INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND
 * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
 * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
 * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA,
 * OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
 * LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
 * NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE,
 * EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 * ============================================================================
 */
// Douglas Thrift's Search Engine Processor
//
// Douglas Thrift
//
// $Id$

#include "Processor.hpp"

bool Processor::process(HttpHandler& http, URL& url)
{
        string title, description, text;
        vector<string> headings;

        if (http.contentType().find("text/html") == 0)
        {
                if (!process(http, url, title, description, text, headings)) return
                        false;

                entities(title, "&nbsp;", ' ');
                entities(title, "&lt;", '<');
                entities(title, "&gt;", '>');
                entities(title, "&quot;", '\"');
                entities(title, "&amp;", '&');
                entities(description, "&nbsp;", ' ');
                entities(description, "&lt;", '<');
                entities(description, "&gt;", '>');
                entities(description, "&quot;", '\"');
                entities(description, "&amp;", '&');
                entities(text, "&nbsp;", ' ');
                entities(text, "&lt;", '<');
                entities(text, "&gt;", '>');
                entities(text, "&quot;", '\"');
                entities(text, "&amp;", '&');

                for (size_t index(0); index < headings.size(); index++)
                {
                        entities(headings[index], "&nbsp;", ' ');
                        entities(headings[index], "&lt;", '<');
                        entities(headings[index], "&gt;", '>');
                        entities(headings[index], "&quot;", '\"');
                        entities(headings[index], "&amp;", '&');
                }

                normalize(title);
                normalize(description);
                normalize(text);

                for (size_t index0(0); index0 < headings.size(); index0++)
                {
                        normalize(headings[index0]);
                }
        }
        else
        {
                string line;

                while (http.good())
                {
                        http.getline(line);

                        text += line + "\n";
                }

                normalize(text);
        }

        page.setSize(http.contentLength());
        page.setURL(url);
        page.setTitle(title);
        page.setDescription(description);
        page.setText(text);
        page.setHeadings(headings);

        return true;
}

void Processor::reset()
{
        links.clear();

        page = Page();
}

bool Processor::process(HttpHandler& http, URL& url, string& title, string&
        description, string& text, vector<string>& headings)
{
        bool inHtml(false), inHead(false), inTitle(false), inBody(false),
                inHeading(false), inComment(false), follow(true), answer(true);
        string line, heading;

        while (http.good())
        {
                http.getline(line);

                unsigned begin(0), startComment(0);

                while (begin < line.length())
                {
                        unsigned open(line.find('<', begin)), close(line.find('>', begin));
                        string next;

                        while (close == string::npos && http.good())
                        {
                                http.getline(next);

                                line += '\n' + next;

                                close = line.find('>', begin);
                        }

                        // strangely this is necessary sometimes
                        if (open == string::npos) open = line.find('<', begin);

                        string between(line.substr(begin, open - begin)), tag(getTag(line,
                                open, close)), lowerTag(tolower(tag));

                        if (inHtml && !inComment)
                        {
                                if (inHead && inTitle)
                                {
                                        title += between + '\n';
                                }

                                if (inBody)
                                {
                                        text += between + '\n';
                                }

                                if (inBody && inHeading)
                                {
                                        heading += between + '\n';
                                }

                                if (((lowerTag.find("meta ") == 0) || (lowerTag.find("meta\n")
                                        == 0) || (lowerTag.find("meta   ") == 0)) && inHead)
                                {
                                        if (lowerTag.find("name=robots") != string::npos ||
                                                lowerTag.find("name=\"robots\"") != string::npos)
                                        {
                                                unsigned start(lowerTag.find("content=\"") + 9),
                                                        finish(lowerTag.find('\"', start));
                                                string robots(lowerTag.substr(start, finish - start));

                                                if ((robots.find("noindex") != string::npos &&
                                                        robots.find("nofollow") != string::npos) ||
                                                        robots.find("none") != string::npos)
                                                {
                                                        answer = false;
                                                        follow = false;

                                                        links.clear();

                                                        return answer;
                                                }
                                                else if (robots.find("noindex") != string::npos)
                                                {
                                                        answer = false;
                                                }
                                                else if (robots.find("nofollow") != string::npos)
                                                {
                                                        follow = false;

                                                        links.clear();
                                                }
                                        }
                                        else if (lowerTag.find("name=description") != string::npos
                                                || lowerTag.find("name=\"description\"") !=
                                                string::npos)
                                        {
                                                unsigned start(lowerTag.find("content=\"") + 9),
                                                        finish(lowerTag.find('\"', start));

                                                description = tag.substr(start, finish - start);
                                        }
                                }

                                if (((lowerTag.find("a ") == 0) || (lowerTag.find("a\n") == 0)
                                        || (lowerTag.find("a    ") == 0)) && inBody && follow)
                                {
                                        if (lowerTag.find("href=\"") != string::npos)
                                        {
                                                unsigned start(lowerTag.find("href=\"") + 6),
                                                        finish(lowerTag.find('\"', start));
                                                string link(getLink(tag.substr(start, finish -
                                                        start), url));

                                                if (!link.empty()) links.insert(link);
                                        }
                                        else if (lowerTag.find("href=") != string::npos)
                                        {
                                                unsigned start(lowerTag.find("href=") + 5),
                                                        finish(lowerTag.find(' ', start));

                                                if (finish < close)
                                                {
                                                        string link(getLink(tag.substr(start, finish -
                                                                start), url));

                                                        if (!link.empty()) links.insert(link);
                                                }
                                                else
                                                {
                                                        string link(getLink(tag.substr(start, close -
                                                                start), url));

                                                        if (!link.empty()) links.insert(link);
                                                }
                                        }
                                }

                                if ((lowerTag.find("img ") == 0) || (lowerTag.find("img\n") ==
                                        0) || (lowerTag.find("img       ")) && inBody)
                                {
                                        if (lowerTag.find("alt=\"") != string::npos)
                                        {
                                                unsigned start(lowerTag.find("alt=\"") + 5),
                                                        finish(lowerTag.find('\"', start));

                                                text += tag.substr(start, finish - start) + ' ';

                                                if (inHeading) heading += tag.substr(start, finish -
                                                        start) + ' ';
                                        }
                                        else if (lowerTag.find("alt=") != string::npos)
                                        {
                                                unsigned start(lowerTag.find("alt=") + 4),
                                                        finish(lowerTag.find(' ', start));

                                                if (finish < close)
                                                {
                                                        text += tag.substr(start, finish - start) + ' ';

                                                        if (inHeading) heading += tag.substr(start, finish
                                                                - start) + ' ';
                                                }
                                                else
                                                {
                                                        text += tag.substr(start, close - start) + ' ';

                                                        if (inHeading) heading += tag.substr(start, close -
                                                                start) + ' ';
                                                }
                                        }
                                }
                        }

                        if (lowerTag.find("html") == 0) inHtml = true;
                        if (lowerTag.find("/html") == 0) inHtml = false;
                        if (lowerTag.find("head") == 0) inHead = true;
                        if (lowerTag.find("/head") == 0) inHead = false;
                        if (lowerTag.find("title") == 0) inTitle = true;
                        if (lowerTag.find("/title") == 0) inTitle = false;
                        if (lowerTag.find("body") == 0 || lowerTag.find("noframes") == 0)
                                inBody = true;
                        if (lowerTag.find("/body") == 0 || lowerTag.find("/noframes") == 0)
                                inBody = false;

                        if (lowerTag.find("h1") == 0 || lowerTag.find("h2") == 0 ||
                                lowerTag.find("h3") == 0 || lowerTag.find("h4") == 0 ||
                                lowerTag.find("h5") == 0 || lowerTag.find("h6") == 0)
                        {
                                heading.erase();

                                inHeading = true;
                        }

                        if (lowerTag.find("/h1") == 0 || lowerTag.find("/h2") == 0 ||
                                lowerTag.find("/h3") == 0 || lowerTag.find("/h4") == 0 ||
                                lowerTag.find("/h5") == 0 || lowerTag.find("/h6") == 0)
                        {
                                if (!heading.empty()) headings.push_back(heading);

                                inHeading = false;
                        }

                        if (lowerTag.find("!--") == 0)
                        {
                                startComment = open;
                                inComment = true;
                        }

                        if (line.find("-->", begin) >= startComment && line.find("-->",
                                begin) != string::npos)
                        {
                                close = line.find("-->", begin) + 2;
                                inComment = false;
                        }

                        if (close == string::npos)
                        {
                                begin = close;
                        }
                        else
                        {
                                begin = close + 1;
                        }
                }
        }

        return answer;
}

string Processor::getTag(const string& line, unsigned open, unsigned close)
{
        return line.substr(open + 1, close - open - 1);
}
Revision:	360
Committed:	2004-08-19T20:09:52-07:00 (20 years, 10 months ago) by Douglas Thrift
File size:	9803 byte(s)
Log Message:	Did a lot of stuff, including fixing a bug.
#	Content
1	/* ============================================================================
2	* Douglas Thrift's Search Engine License
3	*
4	* Copyright (C) 2002-2004, Douglas Thrift. All Rights Reserved.
5	* Redistribution and use in source and binary forms, with or without
6	* modification, are permitted provided that the following conditions are met:
7	*
8	* 1. Redistributions of source code must retain the above copyright notice,
9	* this list of conditions and the following disclaimer.
10	*
11	* 2. Redistributions in binary form must reproduce the above copyright notice,
12	* this list of conditions and the following disclaimer in the documentation
13	* and/or other materials provided with the distribution.
14	*
15	* 3. The end-user documentation included with the redistribution, if any, must
16	* include the following acknowledgment:
17	*
18	* "This product includes software developed by Douglas Thrift
19	* (http://computers.douglasthrift.net/searchengine/)."
20	*
21	* Alternately, this acknowledgment may appear in the software itself, if
22	* and wherever such third-party acknowledgments normally appear.
23	*
24	* 4. The names "Douglas Thrift" and "Douglas Thrift's Search Engine" must not
25	* be used to endorse or promote products derived from this software without
26	* specific prior written permission. For written permission, please visit
27	* http://www.douglasthrift.net/contact.cgi for contact information.
28	*
29	* 5. Products derived from this software may not be called "Douglas Thrift's
30	* Search Engine", nor may "Douglas Thrift's Search Engine" appear in their
31	* name, without prior written permission.
32	*
33	* THIS SOFTWARE IS PROVIDED "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES,
34	* INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND
35	* FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
36	* COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
37	* INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
38	* LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA,
39	* OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
40	* LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
41	* NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE,
42	* EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
43	* ============================================================================
44	*/
45	// Douglas Thrift's Search Engine Processor
46	//
47	// Douglas Thrift
48	//
49	// $Id$
50
51	#include "Processor.hpp"
52
53	bool Processor::process(HttpHandler& http, URL& url)
54	{
55	string title, description, text;
56	vector<string> headings;
57
58	if (http.contentType().find("text/html") == 0)
59	{
60	if (!process(http, url, title, description, text, headings)) return
61	false;
62
63	entities(title, " ", ' ');
64	entities(title, "<", '<');
65	entities(title, ">", '>');
66	entities(title, """, '\"');
67	entities(title, "&", '&');
68	entities(description, " ", ' ');
69	entities(description, "<", '<');
70	entities(description, ">", '>');
71	entities(description, """, '\"');
72	entities(description, "&", '&');
73	entities(text, " ", ' ');
74	entities(text, "<", '<');
75	entities(text, ">", '>');
76	entities(text, """, '\"');
77	entities(text, "&", '&');
78
79	for (size_t index(0); index < headings.size(); index++)
80	{
81	entities(headings[index], " ", ' ');
82	entities(headings[index], "<", '<');
83	entities(headings[index], ">", '>');
84	entities(headings[index], """, '\"');
85	entities(headings[index], "&", '&');
86	}
87
88	normalize(title);
89	normalize(description);
90	normalize(text);
91
92	for (size_t index0(0); index0 < headings.size(); index0++)
93	{
94	normalize(headings[index0]);
95	}
96	}
97	else
98	{
99	string line;
100
101	while (http.good())
102	{
103	http.getline(line);
104
105	text += line + "\n";
106	}
107
108	normalize(text);
109	}
110
111	page.setSize(http.contentLength());
112	page.setURL(url);
113	page.setTitle(title);
114	page.setDescription(description);
115	page.setText(text);
116	page.setHeadings(headings);
117
118	return true;
119	}
120
121	void Processor::reset()
122	{
123	links.clear();
124
125	page = Page();
126	}
127
128	bool Processor::process(HttpHandler& http, URL& url, string& title, string&
129	description, string& text, vector<string>& headings)
130	{
131	bool inHtml(false), inHead(false), inTitle(false), inBody(false),
132	inHeading(false), inComment(false), follow(true), answer(true);
133	string line, heading;
134
135	while (http.good())
136	{
137	http.getline(line);
138
139	unsigned begin(0), startComment(0);
140
141	while (begin < line.length())
142	{
143	unsigned open(line.find('<', begin)), close(line.find('>', begin));
144	string next;
145
146	while (close == string::npos && http.good())
147	{
148	http.getline(next);
149
150	line += '\n' + next;
151
152	close = line.find('>', begin);
153	}
154
155	// strangely this is necessary sometimes
156	if (open == string::npos) open = line.find('<', begin);
157
158	string between(line.substr(begin, open - begin)), tag(getTag(line,
159	open, close)), lowerTag(tolower(tag));
160
161	if (inHtml && !inComment)
162	{
163	if (inHead && inTitle)
164	{
165	title += between + '\n';
166	}
167
168	if (inBody)
169	{
170	text += between + '\n';
171	}
172
173	if (inBody && inHeading)
174	{
175	heading += between + '\n';
176	}
177
178	if (((lowerTag.find("meta ") == 0) \|\| (lowerTag.find("meta\n")
179	== 0) \|\| (lowerTag.find("meta ") == 0)) && inHead)
180	{
181	if (lowerTag.find("name=robots") != string::npos \|\|
182	lowerTag.find("name=\"robots\"") != string::npos)
183	{
184	unsigned start(lowerTag.find("content=\"") + 9),
185	finish(lowerTag.find('\"', start));
186	string robots(lowerTag.substr(start, finish - start));
187
188	if ((robots.find("noindex") != string::npos &&
189	robots.find("nofollow") != string::npos) \|\|
190	robots.find("none") != string::npos)
191	{
192	answer = false;
193	follow = false;
194
195	links.clear();
196
197	return answer;
198	}
199	else if (robots.find("noindex") != string::npos)
200	{
201	answer = false;
202	}
203	else if (robots.find("nofollow") != string::npos)
204	{
205	follow = false;
206
207	links.clear();
208	}
209	}
210	else if (lowerTag.find("name=description") != string::npos
211	\|\| lowerTag.find("name=\"description\"") !=
212	string::npos)
213	{
214	unsigned start(lowerTag.find("content=\"") + 9),
215	finish(lowerTag.find('\"', start));
216
217	description = tag.substr(start, finish - start);
218	}
219	}
220
221	if (((lowerTag.find("a ") == 0) \|\| (lowerTag.find("a\n") == 0)
222	\|\| (lowerTag.find("a ") == 0)) && inBody && follow)
223	{
224	if (lowerTag.find("href=\"") != string::npos)
225	{
226	unsigned start(lowerTag.find("href=\"") + 6),
227	finish(lowerTag.find('\"', start));
228	string link(getLink(tag.substr(start, finish -
229	start), url));
230
231	if (!link.empty()) links.insert(link);
232	}
233	else if (lowerTag.find("href=") != string::npos)
234	{
235	unsigned start(lowerTag.find("href=") + 5),
236	finish(lowerTag.find(' ', start));
237
238	if (finish < close)
239	{
240	string link(getLink(tag.substr(start, finish -
241	start), url));
242
243	if (!link.empty()) links.insert(link);
244	}
245	else
246	{
247	string link(getLink(tag.substr(start, close -
248	start), url));
249
250	if (!link.empty()) links.insert(link);
251	}
252	}
253	}
254
255	if ((lowerTag.find("img ") == 0) \|\| (lowerTag.find("img\n") ==
256	0) \|\| (lowerTag.find("img ")) && inBody)
257	{
258	if (lowerTag.find("alt=\"") != string::npos)
259	{
260	unsigned start(lowerTag.find("alt=\"") + 5),
261	finish(lowerTag.find('\"', start));
262
263	text += tag.substr(start, finish - start) + ' ';
264
265	if (inHeading) heading += tag.substr(start, finish -
266	start) + ' ';
267	}
268	else if (lowerTag.find("alt=") != string::npos)
269	{
270	unsigned start(lowerTag.find("alt=") + 4),
271	finish(lowerTag.find(' ', start));
272
273	if (finish < close)
274	{
275	text += tag.substr(start, finish - start) + ' ';
276
277	if (inHeading) heading += tag.substr(start, finish
278	- start) + ' ';
279	}
280	else
281	{
282	text += tag.substr(start, close - start) + ' ';
283
284	if (inHeading) heading += tag.substr(start, close -
285	start) + ' ';
286	}
287	}
288	}
289	}
290
291	if (lowerTag.find("html") == 0) inHtml = true;
292	if (lowerTag.find("/html") == 0) inHtml = false;
293	if (lowerTag.find("head") == 0) inHead = true;
294	if (lowerTag.find("/head") == 0) inHead = false;
295	if (lowerTag.find("title") == 0) inTitle = true;
296	if (lowerTag.find("/title") == 0) inTitle = false;
297	if (lowerTag.find("body") == 0 \|\| lowerTag.find("noframes") == 0)
298	inBody = true;
299	if (lowerTag.find("/body") == 0 \|\| lowerTag.find("/noframes") == 0)
300	inBody = false;
301
302	if (lowerTag.find("h1") == 0 \|\| lowerTag.find("h2") == 0 \|\|
303	lowerTag.find("h3") == 0 \|\| lowerTag.find("h4") == 0 \|\|
304	lowerTag.find("h5") == 0 \|\| lowerTag.find("h6") == 0)
305	{
306	heading.erase();
307
308	inHeading = true;
309	}
310
311	if (lowerTag.find("/h1") == 0 \|\| lowerTag.find("/h2") == 0 \|\|
312	lowerTag.find("/h3") == 0 \|\| lowerTag.find("/h4") == 0 \|\|
313	lowerTag.find("/h5") == 0 \|\| lowerTag.find("/h6") == 0)
314	{
315	if (!heading.empty()) headings.push_back(heading);
316
317	inHeading = false;
318	}
319
320	if (lowerTag.find("!--") == 0)
321	{
322	startComment = open;
323	inComment = true;
324	}
325
326	if (line.find("-->", begin) >= startComment && line.find("-->",
327	begin) != string::npos)
328	{
329	close = line.find("-->", begin) + 2;
330	inComment = false;
331	}
332
333	if (close == string::npos)
334	{
335	begin = close;
336	}
337	else
338	{
339	begin = close + 1;
340	}
341	}
342	}
343
344	return answer;
345	}
346
347	string Processor::getTag(const string& line, unsigned open, unsigned close)
348	{
349	return line.substr(open + 1, close - open - 1);
350	}