trunk/Search/Processor.cpp

/* ============================================================================
 * Douglas Thrift's Search Engine License
 *
 * Copyright (C) 2002, 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
//
// Processor.cpp

#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, /*knowSize = page->getSize() > 0,
*/              follow = true, answer = true;
        unsigned startComment = 0, finishComment = 0;
        string line;
        while (http.good())
        {
                http.getline(line);
                string heading;

                unsigned begin = 0;
                while (begin < line.length())
                {
                        unsigned open = line.find('<', begin);
                        unsigned close = line.find('>', begin);

                        string next;
                        while (close == string::npos)
                        {
                                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)
                                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:	18
Committed:	2002-12-09T21:40:12-08:00 (22 years, 6 months ago) by douglas
File size:	10218 byte(s)
Log Message:	Implemented more HttpHandler stuff. Added news: protocol to those ignored by getLink().
#	Content
1	/* ============================================================================
2	* Douglas Thrift's Search Engine License
3	*
4	* Copyright (C) 2002, 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	// Processor.cpp
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, /*knowSize = page->getSize() > 0,
143	*/ follow = true, answer = true;
144	unsigned startComment = 0, finishComment = 0;
145	string line;
146	while (http.good())
147	{
148	http.getline(line);
149	string heading;
150
151	unsigned begin = 0;
152	while (begin < line.length())
153	{
154	unsigned open = line.find('<', begin);
155	unsigned close = line.find('>', begin);
156
157	string next;
158	while (close == string::npos)
159	{
160	http.getline(next);
161	line += '\n' + next;
162	close = line.find('>', begin);
163	}
164
165	// strangely this is necessary sometimes
166	if (open == string::npos) open = line.find('<', begin);
167
168	string between = line.substr(begin, open - begin);
169	string tag = getTag(line, open, close);
170	string lowerTag(tag.length(), ' ');
171
172	for (unsigned index = 0; index < tag.length(); index++)
173	{
174	lowerTag[index] = tolower(tag[index]);
175	}
176
177	if (inHtml && !inComment)
178	{
179	if (inHead && inTitle)
180	{
181	title += between + "\n";
182	}
183
184	if (inBody)
185	{
186	text += between + "\n";
187	}
188
189	if (inBody && inHeading)
190	{
191	heading += between + "\n";
192	}
193	if (((lowerTag.find("meta ") == 0) \|\| (lowerTag.find("meta\n")
194	== 0) \|\| (lowerTag.find("meta ") == 0)) && inHead)
195	{
196	if (lowerTag.find("name=robots") != string::npos \|\|
197	lowerTag.find("name=\"robots\"") != string::npos)
198	{
199	unsigned start = lowerTag.find("content=\"") + 9;
200	unsigned finish = lowerTag.find('\"', start);
201
202	string robots = lowerTag.substr(start, finish - start);
203
204	if ((robots.find("noindex") != string::npos &&
205	robots.find("nofollow") != string::npos) \|\|
206	robots.find("none") != string::npos)
207	{
208	answer = false;
209	follow = false;
210	links.clear();
211
212	return answer;
213	}
214	else if (robots.find("noindex") != string::npos)
215	{
216	answer = false;
217	}
218	else if (robots.find("nofollow") != string::npos)
219	{
220	follow = false;
221	links.clear();
222	}
223	}
224	else if (lowerTag.find("name=description") != string::npos
225	\|\| lowerTag.find("name=\"description\"") !=
226	string::npos)
227	{
228	unsigned start = lowerTag.find("content=\"") + 9;
229	unsigned finish = lowerTag.find('\"', start);
230
231	description = tag.substr(start, finish - start);
232	}
233	}
234
235	if (((lowerTag.find("a ") == 0) \|\| (lowerTag.find("a\n") == 0)
236	\|\| (lowerTag.find("a ") == 0)) && inBody && follow)
237	{
238	if (lowerTag.find("href=\"") != string::npos)
239	{
240	unsigned start = lowerTag.find("href=\"") + 6;
241	unsigned finish = lowerTag.find('\"', start);
242
243	string link = getLink(tag.substr(start, finish -
244	start), url);
245
246	if (link != "") links.insert(link);
247	}
248	else if (lowerTag.find("href=") != string::npos)
249	{
250	unsigned start = lowerTag.find("href=") + 5;
251	unsigned finish = lowerTag.find(' ', start);
252
253	if (finish < close)
254	{
255	string link = getLink(tag.substr(start, finish -
256	start), url);
257
258	if (link != "") links.insert(link);
259	}
260	else
261	{
262	string link = getLink(tag.substr(start, close -
263	start), url);
264
265	if (link != "") links.insert(link);
266	}
267	}
268	}
269
270	if ((lowerTag.find("img ") == 0) \|\| (lowerTag.find("img\n") ==
271	0) \|\| (lowerTag.find("img ")) && inBody)
272	{
273	if (lowerTag.find("alt=\"") != string::npos)
274	{
275	unsigned start = lowerTag.find("alt=\"") + 5;
276	unsigned finish = lowerTag.find('\"', start);
277
278	text += tag.substr(start, finish - start) + ' ';
279	if (inHeading) heading += tag.substr(start, finish -
280	start) + ' ';
281	}
282	else if (lowerTag.find("alt=") != string::npos)
283	{
284	unsigned start = lowerTag.find("alt=") + 4;
285	unsigned finish = lowerTag.find(' ', start);
286
287	if (finish < close)
288	{
289	text += tag.substr(start, finish - start) + ' ';
290	if (inHeading) heading += tag.substr(start, finish
291	- start) + ' ';
292	}
293	else
294	{
295	text += tag.substr(start, close - start) + ' ';
296	if (inHeading) heading += tag.substr(start, close -
297	start) + ' ';
298	}
299	}
300	}
301	}
302
303	if (lowerTag.find("html") == 0) inHtml = true;
304	if (lowerTag.find("/html") == 0) inHtml = false;
305
306	if (lowerTag.find("head") == 0) inHead = true;
307	if (lowerTag.find("/head") == 0) inHead = false;
308
309	if (lowerTag.find("title") == 0) inTitle = true;
310	if (lowerTag.find("/title") == 0) inTitle = false;
311
312	if (lowerTag.find("body") == 0 \|\| lowerTag.find("noframes") == 0)
313	inBody = true;
314	if (lowerTag.find("/body") == 0 \|\| lowerTag.find("/noframes") == 0)
315	inBody = false;
316
317	if (lowerTag.find("h1") == 0 \|\| lowerTag.find("h2") == 0 \|\|
318	lowerTag.find("h3") == 0 \|\| lowerTag.find("h4") == 0 \|\|
319	lowerTag.find("h5") == 0 \|\| lowerTag.find("h6") == 0)
320	inHeading = true;
321	if (lowerTag.find("/h1") == 0 \|\| lowerTag.find("/h2") == 0 \|\|
322	lowerTag.find("/h3") == 0 \|\| lowerTag.find("/h4") == 0 \|\|
323	lowerTag.find("/h5") == 0 \|\| lowerTag.find("/h6") == 0)
324	{
325	if (heading != "") headings.push_back(heading);
326	inHeading = false;
327	}
328
329	if (lowerTag.find("!--") == 0)
330	{
331	startComment = open;
332	inComment = true;
333	}
334	if (line.find("-->", begin) >= startComment && line.find("-->",
335	begin) != string::npos)
336	{
337	finishComment = line.find("-->", begin) + 3;
338	inComment = false;
339	}
340
341	if (close == string::npos)
342	{
343	begin = close;
344	}
345	else
346	{
347	begin = close + 1;
348	}
349	}
350
351	startComment = 0;
352	finishComment = 0;
353	}
354
355	return answer;
356	}
357
358	string Processor::getTag(const string& line, unsigned open, unsigned close)
359	{
360	string tag = line.substr(open + 1, close - open - 1);
361
362	return tag;
363	}