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"

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;
        string line, heading;

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

                unsigned begin = 0, startComment = 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)
                        {
                                close = line.find("-->", begin) + 3;
                                inComment = false;
                        }

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

        return answer;
}

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

        return tag;
}
Revision:	334
Committed:	2004-04-05T16:37:41-07:00 (21 years, 2 months ago) by Douglas Thrift
File size:	10138 byte(s)
Log Message:	Ah, I just love Subversion!
#	User	Rev	Content
1	douglas	1	/* ============================================================================
2			* Douglas Thrift's Search Engine License
3			*
4	douglas	312	* Copyright (C) 2002-2004, Douglas Thrift. All Rights Reserved.
5	douglas	1	* 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	Douglas Thrift	331	// $Id$
50	douglas	1
51	Douglas Thrift	334	#include "Processor.hpp"
52	douglas	1
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	douglas	17	if (http.contentType().find("text/html") == 0)
69	douglas	1	{
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			entities(description, " ", ' ');
79			entities(description, "<", '<');
80			entities(description, ">", '>');
81			entities(description, """, '\"');
82			entities(description, "&", '&');
83			entities(text, " ", ' ');
84			entities(text, "<", '<');
85			entities(text, ">", '>');
86			entities(text, """, '\"');
87			entities(text, "&", '&');
88
89			for (int index = 0; index < headings.size(); index++)
90			{
91			entities(headings[index], " ", ' ');
92			entities(headings[index], "<", '<');
93			entities(headings[index], ">", '>');
94			entities(headings[index], """, '\"');
95			entities(headings[index], "&", '&');
96			}
97
98			normalize(title);
99			normalize(description);
100			normalize(text);
101	douglas	316
102	douglas	1	for (int index0 = 0; index0 < headings.size(); index0++)
103			{
104			normalize(headings[index0]);
105			}
106			}
107			else
108			{
109			string line;
110	douglas	316
111	douglas	1	while (http.good())
112			{
113			http.getline(line);
114
115			text += line + "\n";
116			}
117
118			normalize(text);
119			}
120
121	douglas	17	page->setSize(http.contentLength());
122	douglas	1	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	douglas	316
135	douglas	1	delete page;
136	douglas	316
137	douglas	1	page = new Page();
138			}
139
140			bool Processor::process(HttpHandler& http, URL& url, string& title, string&
141			description, string& text, vector<string>& headings)
142			{
143			bool inHtml = false, inHead = false, inTitle = false, inBody = false,
144	douglas	19	inHeading = false, inComment = false, follow = true, answer = true;
145	douglas	315	string line, heading;
146
147	douglas	1	while (http.good())
148			{
149			http.getline(line);
150
151	douglas	316	unsigned begin = 0, startComment = 0;
152
153	douglas	1	while (begin < line.length())
154			{
155			unsigned open = line.find('<', begin);
156			unsigned close = line.find('>', begin);
157	douglas	316	string next;
158	douglas	1
159	douglas	21	while (close == string::npos && http.good())
160	douglas	1	{
161			http.getline(next);
162			line += '\n' + next;
163			close = line.find('>', begin);
164			}
165
166			// strangely this is necessary sometimes
167			if (open == string::npos) open = line.find('<', begin);
168
169			string between = line.substr(begin, open - begin);
170			string tag = getTag(line, open, close);
171			string lowerTag(tag.length(), ' ');
172
173			for (unsigned index = 0; index < tag.length(); index++)
174			{
175			lowerTag[index] = tolower(tag[index]);
176			}
177
178			if (inHtml && !inComment)
179			{
180			if (inHead && inTitle)
181			{
182			title += between + "\n";
183			}
184
185			if (inBody)
186			{
187			text += between + "\n";
188			}
189
190			if (inBody && inHeading)
191			{
192			heading += between + "\n";
193			}
194	douglas	316
195	douglas	1	if (((lowerTag.find("meta ") == 0) \|\| (lowerTag.find("meta\n")
196			== 0) \|\| (lowerTag.find("meta ") == 0)) && inHead)
197			{
198			if (lowerTag.find("name=robots") != string::npos \|\|
199			lowerTag.find("name=\"robots\"") != string::npos)
200			{
201			unsigned start = lowerTag.find("content=\"") + 9;
202			unsigned finish = lowerTag.find('\"', start);
203
204			string robots = lowerTag.substr(start, finish - start);
205
206			if ((robots.find("noindex") != string::npos &&
207			robots.find("nofollow") != string::npos) \|\|
208			robots.find("none") != string::npos)
209			{
210			answer = false;
211			follow = false;
212			links.clear();
213
214			return answer;
215			}
216			else if (robots.find("noindex") != string::npos)
217			{
218			answer = false;
219			}
220			else if (robots.find("nofollow") != string::npos)
221			{
222			follow = false;
223			links.clear();
224			}
225			}
226			else if (lowerTag.find("name=description") != string::npos
227			\|\| lowerTag.find("name=\"description\"") !=
228			string::npos)
229			{
230			unsigned start = lowerTag.find("content=\"") + 9;
231			unsigned finish = lowerTag.find('\"', start);
232
233			description = tag.substr(start, finish - start);
234			}
235			}
236
237			if (((lowerTag.find("a ") == 0) \|\| (lowerTag.find("a\n") == 0)
238			\|\| (lowerTag.find("a ") == 0)) && inBody && follow)
239			{
240			if (lowerTag.find("href=\"") != string::npos)
241			{
242			unsigned start = lowerTag.find("href=\"") + 6;
243			unsigned finish = lowerTag.find('\"', start);
244
245	douglas	15	string link = getLink(tag.substr(start, finish -
246			start), url);
247	douglas	1
248	douglas	17	if (link != "") links.insert(link);
249	douglas	1	}
250			else if (lowerTag.find("href=") != string::npos)
251			{
252			unsigned start = lowerTag.find("href=") + 5;
253			unsigned finish = lowerTag.find(' ', start);
254
255			if (finish < close)
256			{
257	douglas	15	string link = getLink(tag.substr(start, finish -
258			start), url);
259	douglas	1
260	douglas	17	if (link != "") links.insert(link);
261	douglas	1	}
262			else
263			{
264	douglas	15	string link = getLink(tag.substr(start, close -
265			start), url);
266	douglas	1
267	douglas	17	if (link != "") links.insert(link);
268	douglas	1	}
269			}
270			}
271
272			if ((lowerTag.find("img ") == 0) \|\| (lowerTag.find("img\n") ==
273			0) \|\| (lowerTag.find("img ")) && inBody)
274			{
275			if (lowerTag.find("alt=\"") != string::npos)
276			{
277			unsigned start = lowerTag.find("alt=\"") + 5;
278			unsigned finish = lowerTag.find('\"', start);
279
280			text += tag.substr(start, finish - start) + ' ';
281			if (inHeading) heading += tag.substr(start, finish -
282			start) + ' ';
283			}
284			else if (lowerTag.find("alt=") != string::npos)
285			{
286			unsigned start = lowerTag.find("alt=") + 4;
287			unsigned finish = lowerTag.find(' ', start);
288
289			if (finish < close)
290			{
291			text += tag.substr(start, finish - start) + ' ';
292			if (inHeading) heading += tag.substr(start, finish
293			- start) + ' ';
294			}
295			else
296			{
297			text += tag.substr(start, close - start) + ' ';
298			if (inHeading) heading += tag.substr(start, close -
299			start) + ' ';
300			}
301			}
302			}
303			}
304
305			if (lowerTag.find("html") == 0) inHtml = true;
306			if (lowerTag.find("/html") == 0) inHtml = false;
307			if (lowerTag.find("head") == 0) inHead = true;
308			if (lowerTag.find("/head") == 0) inHead = false;
309			if (lowerTag.find("title") == 0) inTitle = true;
310			if (lowerTag.find("/title") == 0) inTitle = false;
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	douglas	315	{
320			heading = "";
321	douglas	1	inHeading = true;
322	douglas	315	}
323	douglas	316
324	douglas	1	if (lowerTag.find("/h1") == 0 \|\| lowerTag.find("/h2") == 0 \|\|
325			lowerTag.find("/h3") == 0 \|\| lowerTag.find("/h4") == 0 \|\|
326			lowerTag.find("/h5") == 0 \|\| lowerTag.find("/h6") == 0)
327			{
328			if (heading != "") headings.push_back(heading);
329	douglas	316
330	douglas	1	inHeading = false;
331			}
332
333			if (lowerTag.find("!--") == 0)
334			{
335			startComment = open;
336			inComment = true;
337			}
338	douglas	316
339	douglas	1	if (line.find("-->", begin) >= startComment && line.find("-->",
340			begin) != string::npos)
341			{
342	douglas	316	close = line.find("-->", begin) + 3;
343	douglas	1	inComment = false;
344			}
345
346			if (close == string::npos)
347			{
348			begin = close;
349			}
350			else
351			{
352			begin = close + 1;
353			}
354			}
355			}
356
357			return answer;
358			}
359
360	douglas	15	string Processor::getTag(const string& line, unsigned open, unsigned close)
361	douglas	1	{
362			string tag = line.substr(open + 1, close - open - 1);
363
364			return tag;
365			}