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: Processor.cpp,v 1.10 2004/02/21 02:58:09 douglas Exp $

#include "Processor.h"

Processor::Processor()
{
        page = new Page();
}

Processor::~Processor()
{
        delete page;
}

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 (int 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 (int 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();
        delete page;
        page = new 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;
        unsigned startComment = 0, finishComment = 0;
        string line, heading;

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

                unsigned begin = 0;
                while (begin < line.length())
                {
                        unsigned open = line.find('<', begin);
                        unsigned 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);
                        string tag = getTag(line, open, close);
                        string lowerTag(tag.length(), ' ');

                        for (unsigned index = 0; index < tag.length(); index++)
                        {
                                lowerTag[index] = tolower(tag[index]);
                        }

                        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;
                                                unsigned 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;
                                                unsigned 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;
                                                unsigned finish = lowerTag.find('\"', start);

                                                string link = getLink(tag.substr(start, finish -
                                                        start), url);

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

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

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

                                                        if (link != "") 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;
                                                unsigned 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;
                                                unsigned 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 = "";
                                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 != "") headings.push_back(heading);
                                inHeading = false;
                        }

                        if (lowerTag.find("!--") == 0)
                        {
                                startComment = open;
                                inComment = true;
                        }
                        if (line.find("-->", begin) >= startComment && line.find("-->",
                                begin) != string::npos)
                        {
                                finishComment = line.find("-->", begin) + 3;
                                inComment = false;
                        }

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

                startComment = 0;
                finishComment = 0;
        }

        return answer;
}

string Processor::getTag(const string& line, unsigned open, unsigned close)
{
        string tag = line.substr(open + 1, close - open - 1);

        return tag;
}
Revision:	315
Committed:	2004-02-20T18:58:09-08:00 (21 years, 4 months ago) by douglas
File size:	10266 byte(s)
Log Message:	Fixed a bug in Processor that caused headings to be saved incorrectly.
#	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: Processor.cpp,v 1.10 2004/02/21 02:58:09 douglas Exp $
50
51	#include "Processor.h"
52
53	Processor::Processor()
54	{
55	page = new Page();
56	}
57
58	Processor::~Processor()
59	{
60	delete page;
61	}
62
63	bool Processor::process(HttpHandler& http, URL& url)
64	{
65	string title, description, text;
66	vector<string> headings;
67
68	if (http.contentType().find("text/html") == 0)
69	{
70	if (!process(http, url, title, description, text, headings)) return
71	false;
72
73	entities(title, " ", ' ');
74	entities(title, "<", '<');
75	entities(title, ">", '>');
76	entities(title, """, '\"');
77	entities(title, "&", '&');
78
79	entities(description, " ", ' ');
80	entities(description, "<", '<');
81	entities(description, ">", '>');
82	entities(description, """, '\"');
83	entities(description, "&", '&');
84
85	entities(text, " ", ' ');
86	entities(text, "<", '<');
87	entities(text, ">", '>');
88	entities(text, """, '\"');
89	entities(text, "&", '&');
90
91	for (int index = 0; index < headings.size(); index++)
92	{
93	entities(headings[index], " ", ' ');
94	entities(headings[index], "<", '<');
95	entities(headings[index], ">", '>');
96	entities(headings[index], """, '\"');
97	entities(headings[index], "&", '&');
98	}
99
100	normalize(title);
101	normalize(description);
102	normalize(text);
103	for (int index0 = 0; index0 < headings.size(); index0++)
104	{
105	normalize(headings[index0]);
106	}
107	}
108	else
109	{
110	string line;
111	while (http.good())
112	{
113	http.getline(line);
114
115	text += line + "\n";
116	}
117
118	normalize(text);
119	}
120
121	page->setSize(http.contentLength());
122	page->setURL(url);
123	page->setTitle(title);
124	page->setDescription(description);
125	page->setText(text);
126	page->setHeadings(headings);
127
128	return true;
129	}
130
131	void Processor::reset()
132	{
133	links.clear();
134	delete page;
135	page = new Page();
136	}
137
138	bool Processor::process(HttpHandler& http, URL& url, string& title, string&
139	description, string& text, vector<string>& headings)
140	{
141	bool inHtml = false, inHead = false, inTitle = false, inBody = false,
142	inHeading = false, inComment = false, follow = true, answer = true;
143	unsigned startComment = 0, finishComment = 0;
144	string line, heading;
145
146	while (http.good())
147	{
148	http.getline(line);
149
150	unsigned begin = 0;
151	while (begin < line.length())
152	{
153	unsigned open = line.find('<', begin);
154	unsigned close = line.find('>', begin);
155
156	string next;
157	while (close == string::npos && http.good())
158	{
159	http.getline(next);
160	line += '\n' + next;
161	close = line.find('>', begin);
162	}
163
164	// strangely this is necessary sometimes
165	if (open == string::npos) open = line.find('<', begin);
166
167	string between = line.substr(begin, open - begin);
168	string tag = getTag(line, open, close);
169	string lowerTag(tag.length(), ' ');
170
171	for (unsigned index = 0; index < tag.length(); index++)
172	{
173	lowerTag[index] = tolower(tag[index]);
174	}
175
176	if (inHtml && !inComment)
177	{
178	if (inHead && inTitle)
179	{
180	title += between + "\n";
181	}
182
183	if (inBody)
184	{
185	text += between + "\n";
186	}
187
188	if (inBody && inHeading)
189	{
190	heading += between + "\n";
191	}
192	if (((lowerTag.find("meta ") == 0) \|\| (lowerTag.find("meta\n")
193	== 0) \|\| (lowerTag.find("meta ") == 0)) && inHead)
194	{
195	if (lowerTag.find("name=robots") != string::npos \|\|
196	lowerTag.find("name=\"robots\"") != string::npos)
197	{
198	unsigned start = lowerTag.find("content=\"") + 9;
199	unsigned finish = lowerTag.find('\"', start);
200
201	string robots = lowerTag.substr(start, finish - start);
202
203	if ((robots.find("noindex") != string::npos &&
204	robots.find("nofollow") != string::npos) \|\|
205	robots.find("none") != string::npos)
206	{
207	answer = false;
208	follow = false;
209	links.clear();
210
211	return answer;
212	}
213	else if (robots.find("noindex") != string::npos)
214	{
215	answer = false;
216	}
217	else if (robots.find("nofollow") != string::npos)
218	{
219	follow = false;
220	links.clear();
221	}
222	}
223	else if (lowerTag.find("name=description") != string::npos
224	\|\| lowerTag.find("name=\"description\"") !=
225	string::npos)
226	{
227	unsigned start = lowerTag.find("content=\"") + 9;
228	unsigned finish = lowerTag.find('\"', start);
229
230	description = tag.substr(start, finish - start);
231	}
232	}
233
234	if (((lowerTag.find("a ") == 0) \|\| (lowerTag.find("a\n") == 0)
235	\|\| (lowerTag.find("a ") == 0)) && inBody && follow)
236	{
237	if (lowerTag.find("href=\"") != string::npos)
238	{
239	unsigned start = lowerTag.find("href=\"") + 6;
240	unsigned finish = lowerTag.find('\"', start);
241
242	string link = getLink(tag.substr(start, finish -
243	start), url);
244
245	if (link != "") links.insert(link);
246	}
247	else if (lowerTag.find("href=") != string::npos)
248	{
249	unsigned start = lowerTag.find("href=") + 5;
250	unsigned finish = lowerTag.find(' ', start);
251
252	if (finish < close)
253	{
254	string link = getLink(tag.substr(start, finish -
255	start), url);
256
257	if (link != "") links.insert(link);
258	}
259	else
260	{
261	string link = getLink(tag.substr(start, close -
262	start), url);
263
264	if (link != "") links.insert(link);
265	}
266	}
267	}
268
269	if ((lowerTag.find("img ") == 0) \|\| (lowerTag.find("img\n") ==
270	0) \|\| (lowerTag.find("img ")) && inBody)
271	{
272	if (lowerTag.find("alt=\"") != string::npos)
273	{
274	unsigned start = lowerTag.find("alt=\"") + 5;
275	unsigned finish = lowerTag.find('\"', start);
276
277	text += tag.substr(start, finish - start) + ' ';
278	if (inHeading) heading += tag.substr(start, finish -
279	start) + ' ';
280	}
281	else if (lowerTag.find("alt=") != string::npos)
282	{
283	unsigned start = lowerTag.find("alt=") + 4;
284	unsigned finish = lowerTag.find(' ', start);
285
286	if (finish < close)
287	{
288	text += tag.substr(start, finish - start) + ' ';
289	if (inHeading) heading += tag.substr(start, finish
290	- start) + ' ';
291	}
292	else
293	{
294	text += tag.substr(start, close - start) + ' ';
295	if (inHeading) heading += tag.substr(start, close -
296	start) + ' ';
297	}
298	}
299	}
300	}
301
302	if (lowerTag.find("html") == 0) inHtml = true;
303	if (lowerTag.find("/html") == 0) inHtml = false;
304
305	if (lowerTag.find("head") == 0) inHead = true;
306	if (lowerTag.find("/head") == 0) inHead = false;
307
308	if (lowerTag.find("title") == 0) inTitle = true;
309	if (lowerTag.find("/title") == 0) inTitle = false;
310
311	if (lowerTag.find("body") == 0 \|\| lowerTag.find("noframes") == 0)
312	inBody = true;
313	if (lowerTag.find("/body") == 0 \|\| lowerTag.find("/noframes") == 0)
314	inBody = false;
315
316	if (lowerTag.find("h1") == 0 \|\| lowerTag.find("h2") == 0 \|\|
317	lowerTag.find("h3") == 0 \|\| lowerTag.find("h4") == 0 \|\|
318	lowerTag.find("h5") == 0 \|\| lowerTag.find("h6") == 0)
319	{
320	heading = "";
321	inHeading = true;
322	}
323	if (lowerTag.find("/h1") == 0 \|\| lowerTag.find("/h2") == 0 \|\|
324	lowerTag.find("/h3") == 0 \|\| lowerTag.find("/h4") == 0 \|\|
325	lowerTag.find("/h5") == 0 \|\| lowerTag.find("/h6") == 0)
326	{
327	if (heading != "") headings.push_back(heading);
328	inHeading = false;
329	}
330
331	if (lowerTag.find("!--") == 0)
332	{
333	startComment = open;
334	inComment = true;
335	}
336	if (line.find("-->", begin) >= startComment && line.find("-->",
337	begin) != string::npos)
338	{
339	finishComment = line.find("-->", begin) + 3;
340	inComment = false;
341	}
342
343	if (close == string::npos)
344	{
345	begin = close;
346	}
347	else
348	{
349	begin = close + 1;
350	}
351	}
352
353	startComment = 0;
354	finishComment = 0;
355	}
356
357	return answer;
358	}
359
360	string Processor::getTag(const string& line, unsigned open, unsigned close)
361	{
362	string tag = line.substr(open + 1, close - open - 1);
363
364	return tag;
365	}