Squid url redirector

Check-in [7224844efa]
anonymous

Check-in [7224844efa]

Overview
Comment:* instead of fuzzy logic with merging many tags splitted some functionality from mark() which only works now with single tag, for many tags there is a set() function; * added option to drop site table.
Downloads: Tarball | ZIP archive | SQL archive
Timelines: family | ancestors | descendants | both | master | trunk
Files: files | file ages | folders
SHA3-256: 7224844efa8f0750b5354bc106903fa948e2fb8ee1c2a207037404d654f9c5aa
User & Date: c.kworr@d4daf22a-8aaf-11de-a64d-234b64dd91b4 on 2010-03-18 15:55:11.000
Other Links: branch diff | manifest | tags
Context
2010-03-22
15:32
added sorting by site check-in: c27c7eb208 user: c.kworr@d4daf22a-8aaf-11de-a64d-234b64dd91b4 tags: master, trunk
2010-03-18
15:55
* instead of fuzzy logic with merging many tags splitted some functionality from mark() which only works now with single tag, for many tags there is a set() function; * added option to drop site table. check-in: 7224844efa user: c.kworr@d4daf22a-8aaf-11de-a64d-234b64dd91b4 tags: master, trunk
15:26
* reworked indexes * added scripts to dump/load db check-in: f1bafd194a user: c.kworr@d4daf22a-8aaf-11de-a64d-234b64dd91b4 tags: master, trunk
Changes
Side-by-Side Diff Ignore Whitespace Patch
Modified database.sql from [83b3f67d5d] to [bb384fd21a].
︙
132
133
134
135
136
137
138
139
140
141



142


143

144
145
146
147
148



149
















150

151
152
153
154
155
156
157

132
133
134
135
136
137
138



139
140
141
142
143
144

145
146
147



148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167

168
169
170
171
172
173
174
175








-
-
-
+
+
+

+
+
-
+


-
-
-
+
+
+

+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
-
+








-- this function adds tag to domain
CREATE or replace FUNCTION mark(domain text, new_tag text) RETURNS integer
	LANGUAGE sql immutable STRICT
	AS $$
select mark(get_site($1), $2) as result;
$$;

-- this function adds tag to domain with regexp
CREATE or replace FUNCTION mark(domain text, tags text[], regexp text) RETURNS integer
	LANGUAGE sql immutable STRICT
-- this function sets tags for site/regexp pair
CREATE or replace FUNCTION set(my_id_site integer, my_id_tag integer) RETURNS integer
	LANGUAGE sql STRICT
	AS $$
delete from urls where $1 = id_site and regexp is null;
insert into urls (id_site, id_tag) values ($1, $2);
select mark(get_site($1), get_tag($2), $3) as result;
select $1;
$$;

-- this function adds tag to domain with regexp
CREATE or replace FUNCTION mark(domain text, tags text[]) RETURNS integer
	LANGUAGE sql immutable STRICT
-- this function sets tags for site/regexp pair
CREATE or replace FUNCTION set(my_id_site integer, my_id_tag integer, my_regexp text) RETURNS integer
	LANGUAGE sql STRICT
	AS $$
delete from urls where $1 = id_site and $3 = regexp;
insert into urls (id_site, id_tag, regexp) values ($1, $2, $3);
select $1;
$$;

-- this function stores new data for site/regexp pair
create or replace function set(domain text, tags text, regexp text) returns integer
	language sql immutable strict
	as $$
select set(get_site($1), get_tag($2::text[]), $3);
$$;

-- this function stores new data for site/regexp pair
create or replace function set(domain text, tags text) returns integer
	language sql immutable strict
	as $$
select mark(get_site($1), get_tag($2), NULL) as result;
select set(get_site($1), get_tag($2::text[]));
$$;

-- this function returns id of tag array
create or replace function get_tag(my_tag text[]) returns integer
	language plpgsql strict
	as $$
declare
︙
Modified st-load.py from [3246bbaf0b] to [02168859d1].
︙
35
36
37
38
39
40
41
42

43
44
45
46
47



48
49
50
51
52
53
54

35
36
37
38
39
40
41

42
43
44
45


46
47
48
49
50
51
52
53
54
55








-
+



-
-
+
+
+








				config['user'],
				config['password'],
				config['host'],
				config['database'],
		) )

	def load(self, csv_data):
		insert = self._db.prepare("select mark($1, array[$2], $3)")
		insert = self._db.prepare("select set($1, $2, $3)")
		with self._db.xact():
			config.section('loader')
			if config['drop_database']:
				self._db.execute('delete from urls; delete from site;');
				#print('dropped', config['drop_database'])
				self._db.execute('delete from urls;')
				if config['drop_site']:
					self._db.execute('delete from site;');
			for row in csv_data:
				insert(row[0], row[1], row[2])
		self._db.execute('vacuum analyze site;')
		self._db.execute('vacuum analyze urls;')

# this classes processes config file and substitutes default values
class Config:
︙
64
65
66
67
68
69
70

71
72
73
74
75
76
77
78
79
80
81



82
83
84
85
86
87



88
89
90
91
92
93
94

65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102








+











+
+
+






+
+
+








			'user': 'squidTag',
			'password': 'password',
			'host': 'localhost',
			'database': 'squidTag',
		},
		'loader': {
			'drop_database': False,
			'drop_site': False,
	},}

	# function to read in config file
	def __init__(self):
		parser = optparse.OptionParser()
		parser.add_option('-c', '--config', dest = 'config',
			help = 'config file location', metavar = 'FILE',
			default = '/usr/local/etc/squid-tagger.conf')
		parser.add_option('-d', '--drop-database', dest = 'drop_database',
			help = 'signals loader to drop previous database',
			action = 'store_true')
		parser.add_option('-D', '--drop-site', dest = 'drop_site',
			help = 'signals loader to drop not only url definitions but site index too',
			action = 'store_true')

		(options, args) = parser.parse_args()

		if options.drop_database:
			self._default['loader']['drop_database'] = True

		if options.drop_site:
			self._default['loader']['drop_site'] = True

		if not os.access(options.config, os.R_OK):
			print("Can't read {}: exitting".format(options.config))
			sys.exit(2)

		self._config = configparser.ConfigParser()
		self._config.readfp(open(options.config))
︙