dimus-taxamatch_rb 0.1.1 → 0.1.2
Sign up to get free protection for your applications and to get access to all the features.
- data/VERSION +1 -1
- data/lib/taxamatch_rb.rb +61 -428
- data/lib/taxamatch_rb/authormatch.rb +0 -0
- data/lib/taxamatch_rb/damerau_levenshtein_mod.rb +1 -2
- data/lib/taxamatch_rb/normalizer.rb +1 -1
- data/lib/taxamatch_rb/phonetizer.rb +0 -2
- data/spec/damerau_levenshtein_mod_test.txt +1 -1
- data/spec/spec_helper.rb +12 -8
- data/spec/taxamatch_rb_spec.rb +135 -6
- data/spec/taxamatch_test.txt +19 -0
- data/taxamatch_rb.gemspec +3 -1
- metadata +3 -1
data/VERSION
CHANGED
|
@@ -1 +1 @@
|
|
|
1
|
-
0.1.
|
|
1
|
+
0.1.2
|
data/lib/taxamatch_rb.rb
CHANGED
|
@@ -10,435 +10,68 @@ class Taxamatch
|
|
|
10
10
|
|
|
11
11
|
def initialize
|
|
12
12
|
@parser = Parser.new
|
|
13
|
+
@dlm = DamerauLevenshteinMod.new
|
|
14
|
+
end
|
|
15
|
+
|
|
16
|
+
|
|
17
|
+
#takes two scientific names and returns true if names match and false if they don't
|
|
18
|
+
def taxamatch(str1, str2)
|
|
19
|
+
parsed_data_1 = @parser.parse(str1)
|
|
20
|
+
parsed_data_2 = @parser.parse(str2)
|
|
21
|
+
taxamatch_parsed_data(parsed_data_1, parsed_data_2)[:match]
|
|
22
|
+
end
|
|
23
|
+
|
|
24
|
+
#takes two hashes of parsed scientific names, analyses them and returns back
|
|
25
|
+
#this function is useful when species strings are preparsed.
|
|
26
|
+
def taxamatch_parsed_data(parsed_data_1, parsed_data_2)
|
|
27
|
+
return match_uninomial(parsed_data_1, parsed_data_2) if parsed_data_1[:unicode] && parsed_data_2[:unicode]
|
|
28
|
+
return match_multinomial(parsed_data_1, parsed_data_2) if parsed_data_1[:genus] && parsed_data_2[:genus]
|
|
29
|
+
return false
|
|
30
|
+
end
|
|
31
|
+
|
|
32
|
+
def match_uninomial(parsed_data_1, parsed_data_2)
|
|
33
|
+
return false
|
|
13
34
|
end
|
|
14
|
-
#
|
|
15
|
-
#
|
|
16
|
-
# #takes two scientific names and returns true if names matched and false if they did not
|
|
17
|
-
# def taxamatch(str1, str2) {
|
|
18
|
-
# taxa1 = @parser.parse_prepare(str1)
|
|
19
|
-
# taxa2 = @parser.parse_prepare(str2)
|
|
20
|
-
#
|
|
21
|
-
# full_taxamatch(taxa1, taxa2)[:mached]
|
|
22
|
-
# }
|
|
23
|
-
#
|
|
24
|
-
# #takes two hashes of parsed scientific names, analyses them and returns back
|
|
25
|
-
# #this function is useful when species strings are preparsed.
|
|
26
|
-
# def full_taxamatch
|
|
27
|
-
# end
|
|
28
35
|
|
|
29
|
-
|
|
36
|
+
def match_multinomial(parsed_data_1, parsed_data_2)
|
|
37
|
+
gen_match = match_genera(parsed_data_1[:genus], parsed_data_2[:genus])
|
|
38
|
+
sp_match = match_species(parsed_data_1[:species], parsed_data_2[:species])
|
|
39
|
+
total_length = parsed_data_1[:genus][:epitheton].size + parsed_data_2[:genus][:epitheton].size + parsed_data_1[:species][:epitheton].size + parsed_data_2[:species][:epitheton].size
|
|
40
|
+
match = match_matches(gen_match, sp_match)
|
|
41
|
+
match.merge({:score => (1- match[:edit_distance]/(total_length/2))})
|
|
42
|
+
end
|
|
43
|
+
|
|
44
|
+
def match_genera(genus1, genus2)
|
|
45
|
+
genus1_length = genus1[:normalized].size
|
|
46
|
+
genus2_length = genus2[:normalized].size
|
|
47
|
+
match = false
|
|
48
|
+
ed = @dlm.distance(genus1[:normalized], genus2[:normalized],2,3)
|
|
49
|
+
return {:edit_distance => ed, :phonetic_match => true, :match => true} if genus1[:phonetized] == genus2[:phonetized]
|
|
50
|
+
|
|
51
|
+
match = true if ed <= 3 && ([genus1_length, genus2_length].min > ed * 2) && (ed < 2 || genus1[0] == genus2[0])
|
|
52
|
+
{:edit_distance => ed, :match => match, :phonetic_match => false}
|
|
53
|
+
end
|
|
30
54
|
|
|
55
|
+
def match_species(sp1, sp2)
|
|
56
|
+
sp1_length = sp1[:normalized].size
|
|
57
|
+
sp2_length = sp2[:normalized].size
|
|
58
|
+
sp1[:phonetized] = Phonetizer.normalize_ending sp1[:phonetized]
|
|
59
|
+
sp2[:phonetized] = Phonetizer.normalize_ending sp2[:phonetized]
|
|
60
|
+
match = false
|
|
61
|
+
ed = @dlm.distance(sp1[:normalized], sp2[:normalized], 4, 4)
|
|
62
|
+
return {:edit_distance => ed, :phonetic_match => true, :match => true} if sp1[:phonetized] == sp2[:phonetized]
|
|
63
|
+
|
|
64
|
+
match = true if ed <= 4 && ([sp1_length, sp2_length].min >= ed * 2) && (ed < 2 || sp1[:normalized][0] == sp2[:normalized][0]) && (ed < 4 || sp1[:normalized][0...3] == sp2[:normalized][0...3])
|
|
65
|
+
{:edit_distance => ed, :match => match, :phonetic_match => false}
|
|
66
|
+
end
|
|
67
|
+
|
|
68
|
+
def match_matches(genus_match, species_match, infraspecies_matches = [])
|
|
69
|
+
match = species_match
|
|
70
|
+
match[:edit_distance] += genus_match[:edit_distance]
|
|
71
|
+
match[:match] = false if match[:edit_distance] > 4
|
|
72
|
+
match[:match] &&= genus_match[:match]
|
|
73
|
+
match[:phonetic_match] &&= genus_match[:phonetic_match]
|
|
74
|
+
match
|
|
75
|
+
end
|
|
31
76
|
|
|
32
|
-
|
|
33
|
-
# public function name_strings_match($name_string1, $name_string2) {
|
|
34
|
-
# $info_1 = new Splitter(null,$name_string1);
|
|
35
|
-
# $info_2 = new Splitter(null,$name_string2);
|
|
36
|
-
# return $this->name_objects_match($info_1, $info_2);
|
|
37
|
-
# }
|
|
38
|
-
#
|
|
39
|
-
# public function name_objects_match($name_object_1, $name_object_2) {
|
|
40
|
-
# $genus_match = $this->match_genera($name_object_1->genus, $name_object_2->genus);
|
|
41
|
-
# $epithets_match = $this->match_species_epithets($name_object_1->species, $name_object_2->species);
|
|
42
|
-
# $total_length = strlen($name_object_1->genus) + strlen($name_object_1->species) + strlen($name_object_2->genus) + strlen($name_object_2->species);
|
|
43
|
-
# $match = $this->match_binomials($genus_match, $epithets_match);
|
|
44
|
-
# return $this->match_response_to_float($match, $total_length);
|
|
45
|
-
# }
|
|
46
|
-
#
|
|
47
|
-
# public function match_response_to_float($match_response, $total_length_of_strings) {
|
|
48
|
-
# if(!$match_response['match']) return 0.0;
|
|
49
|
-
#
|
|
50
|
-
# return (1 - ($match_response['edit_distance'] / ($total_length_of_strings/2)));
|
|
51
|
-
# }
|
|
52
|
-
#
|
|
53
|
-
# public function match_genera($genus1, $genus2) {
|
|
54
|
-
# $match = $phonetic_match = false;
|
|
55
|
-
# $nm = new NearMatch();
|
|
56
|
-
# $genus1_phonetic = $nm->near_match($genus1);
|
|
57
|
-
# $genus2_phonetic = $nm->near_match($genus2);
|
|
58
|
-
# $genus1_length = strlen($genus1);
|
|
59
|
-
#
|
|
60
|
-
# $temp_genus_ED = $this->mdld($genus2, $genus1, 2);
|
|
61
|
-
# // add the genus post-filter
|
|
62
|
-
# // min. 51% "good" chars
|
|
63
|
-
# // first char must match for ED 2+
|
|
64
|
-
# if( ($temp_genus_ED <= 3 && ( min( strlen( $genus2 ), $genus1_length ) > ( $temp_genus_ED * 2 ))
|
|
65
|
-
# && ( $temp_genus_ED < 2 || ( substr($genus2,0,1) == substr($genus1,0,1) ) ) )
|
|
66
|
-
# || ($genus1_phonetic == $genus2_phonetic) ) {
|
|
67
|
-
# $match = true;
|
|
68
|
-
# // accept as exact or near match; append to genus results table
|
|
69
|
-
# $this->debug['process'][] = "6a (near_match_genus:$genus2_phonetic) (this_near_match_genus:$genus1_phonetic)";
|
|
70
|
-
#
|
|
71
|
-
# if($genus1_phonetic == $genus2_phonetic) $phonetic_match = true;
|
|
72
|
-
# }
|
|
73
|
-
# return array(
|
|
74
|
-
# 'match' => $match,
|
|
75
|
-
# 'phonetic_match' => $phonetic_match,
|
|
76
|
-
# 'edit_distance' => $temp_genus_ED);
|
|
77
|
-
# }
|
|
78
|
-
#
|
|
79
|
-
# public function match_species_epithets($species_epithet1, $species_epithet2) {
|
|
80
|
-
# $match = false;
|
|
81
|
-
# $phonetic_match = false;
|
|
82
|
-
# $epithet1_length = strlen($species_epithet1);
|
|
83
|
-
# $epithet2_length = strlen($species_epithet2);
|
|
84
|
-
#
|
|
85
|
-
# $nm = new NearMatch();
|
|
86
|
-
# $epithet1_phonetic = $nm->near_match($species_epithet1);
|
|
87
|
-
# $epithet2_phonetic = $nm->near_match($species_epithet2);
|
|
88
|
-
# $temp_species_ED = $this->mdld($species_epithet2, $species_epithet1, 4);
|
|
89
|
-
# // add the species post-filter
|
|
90
|
-
# // min. 50% "good" chars
|
|
91
|
-
# // first char must match for ED2+
|
|
92
|
-
# // first 3 chars must match for ED4
|
|
93
|
-
# if ($epithet2_phonetic == $epithet1_phonetic) $match = true;
|
|
94
|
-
# elseif( ($temp_species_ED <= 4 && min($epithet2_length, $epithet1_length) >= ($temp_species_ED*2)
|
|
95
|
-
# && ($temp_species_ED < 2 || strpos($species_epithet2 , substr($species_epithet1,0,1)) !== false)
|
|
96
|
-
# && ($temp_species_ED < 4 || strpos($species_epithet2 , substr($species_epithet1,0,3)) !== false))) $match = true;
|
|
97
|
-
#
|
|
98
|
-
# // if phonetic match, set relevant flag
|
|
99
|
-
# if ($epithet2_phonetic == $epithet1_phonetic) $phonetic_match = true;
|
|
100
|
-
#
|
|
101
|
-
# return array('match' => $match, 'phonetic_match' => $phonetic_match, 'edit_distance' => $temp_species_ED);
|
|
102
|
-
# }
|
|
103
|
-
#
|
|
104
|
-
#
|
|
105
|
-
# public function match_binomials($genus_match, $species_epithets_match) {
|
|
106
|
-
# $binomial_match = $species_epithets_match;
|
|
107
|
-
# $binomial_match['edit_distance'] = $genus_match["edit_distance"] + $species_epithets_match["edit_distance"];
|
|
108
|
-
#
|
|
109
|
-
# if(!$genus_match['match']) $binomial_match['match'] = false;
|
|
110
|
-
# if($binomial_match["edit_distance"] > 4) $binomial_match['match'] = false;
|
|
111
|
-
# if(!$genus_match['phonetic_match']) $binomial_match['phonetic_match'] = false;
|
|
112
|
-
#
|
|
113
|
-
#
|
|
114
|
-
# return $binomial_match;
|
|
115
|
-
# }
|
|
116
|
-
#
|
|
117
|
-
# // public function match_species($genus1, $species_epithet1, $genus2, $species_epithet2, $genus_edit_distance) {
|
|
118
|
-
# // $match = false;
|
|
119
|
-
# // $phonetic_match = false;
|
|
120
|
-
# // $epithet1_length = strlen($species_epithet1);
|
|
121
|
-
# // $epithet2_length = strlen($species_epithet2);
|
|
122
|
-
# //
|
|
123
|
-
# // $nm = new NearMatch();
|
|
124
|
-
# // $genus1_phonetic = $nm->near_match(genus1);
|
|
125
|
-
# // $genus2_phonetic = $nm->near_match(genus2);
|
|
126
|
-
# // $epithet1_phonetic = $nm->near_match($species_epithet1);
|
|
127
|
-
# // $epithet2_phonetic = $nm->near_match($species_epithet2);
|
|
128
|
-
# // $temp_species_ED = $this->mdld($species2, $species1, 4);
|
|
129
|
-
# // // add the species post-filter
|
|
130
|
-
# // // min. 50% "good" chars
|
|
131
|
-
# // // first char must match for ED2+
|
|
132
|
-
# // // first 3 chars must match for ED4
|
|
133
|
-
# // if ( ($epithet2_phonetic == $epithet1_phonetic)
|
|
134
|
-
# // || ( ($genus_edit_distance + $temp_species_ED <= 4)
|
|
135
|
-
# // && ($temp_species_ED <= 4 && min(strlen($epithet2_length),$epithet1_length) >= ($temp_species_ED*2)
|
|
136
|
-
# // && ($temp_species_ED < 2 || strpos($species_epithet2 , substr($species_epithet1,1,1)) !== false)
|
|
137
|
-
# // && ($temp_species_ED < 4 || strpos($species_epithet2 , substr($species_epithet1,1,3)) !== false)
|
|
138
|
-
# // && ($genus_edit_distance + $temp_species_ED <= 4) ))) {
|
|
139
|
-
# // $match = true;
|
|
140
|
-
# // // accept as exact or near match, append to species results table
|
|
141
|
-
# // // if phonetic match, set relevant flag
|
|
142
|
-
# // if ( ($genus2_phonetic == $genus1_phonetic) && ($epithet2_phonetic == $epithet1_phonetic) ) $phonetic_match = true;
|
|
143
|
-
# // }
|
|
144
|
-
# // return array('match' => $match, 'phonetic_match' => $phonetic_match, 'edit_distance' => $temp_species_ED);
|
|
145
|
-
# // }
|
|
146
|
-
#
|
|
147
|
-
# /**
|
|
148
|
-
# * Function : process
|
|
149
|
-
# * Purpose: Perform exact and fuzzy matching on a species name, or single genus name
|
|
150
|
-
# * Input: - genus, genus+species, or genus+species+authority (in this version), as "searchtxt"
|
|
151
|
-
# * - "search_mode" to control search mode: currently normal (default) / rapid / no_shaping
|
|
152
|
-
# * - "debug" - print internal parameters used if not null
|
|
153
|
-
# * Outputs: list of genera and species that match (or near match) input terms, with associated
|
|
154
|
-
# * ancillary info as desired
|
|
155
|
-
# * Remarks:
|
|
156
|
-
# * (1) This demo version is configured to access base data in three tables:
|
|
157
|
-
# * - genlist_test1 (genus info); primary key (PK) is genus_id
|
|
158
|
-
# * - splist_test1 (species info); PK is species_id, has genus_id as foreign key (FK)
|
|
159
|
-
# * (= link to relevant row in genus table)
|
|
160
|
-
# * - auth_abbrev_test1 (authority abbreviations - required by subsidiary function
|
|
161
|
-
# * "normalize_auth". Refer README file for relevant minimum table definitions.
|
|
162
|
-
# * If authority comparisons are not required, calls to "normalize_auth" can be disabled and
|
|
163
|
-
# * relevant function commented out, removing need for third table.
|
|
164
|
-
# * (In a production system, table and column names can be varied as desired so long as
|
|
165
|
-
# * code is altered at relevant points, also could be re-configured to hold all genus+species info together in a single table with minor re-write).
|
|
166
|
-
# * (2) Writes to and reads back from pre-defined global temporary tables
|
|
167
|
-
# * "genus_id_matches" and "species_id_matches", new instances of these are automatically
|
|
168
|
-
# * created for each session (i.e., do not need clearing at procedure end). Refer
|
|
169
|
-
# * README file for relevant table definitions.
|
|
170
|
-
# * (3) When result shaping is on in this version, a relevant message displayed as required
|
|
171
|
-
# * for developer feedback, if more distant results are being masked (in producton version,
|
|
172
|
-
# * possibly would not do this)
|
|
173
|
-
# * (4) Requires the following subsidiary functions (supplied elsewhere in this package):
|
|
174
|
-
# * - normalize
|
|
175
|
-
# * - normalize_auth
|
|
176
|
-
# * - reduce_spaces
|
|
177
|
-
# * - ngram
|
|
178
|
-
# * - compare_auth
|
|
179
|
-
# * - near_match
|
|
180
|
-
# * - mdld
|
|
181
|
-
# * (5) Accepts "+" as input separator in place of space (e.g. "Homo+sapiens"), e.g. for calling
|
|
182
|
-
# * via a HTTP GET request as needed.
|
|
183
|
-
# * @param string $searchtxt : genus, genus+species, or genus+species+authority
|
|
184
|
-
# * @param string $search_mode : normal (default) / rapid / no_shaping
|
|
185
|
-
# * @param boolean $cache
|
|
186
|
-
# * @return boolean
|
|
187
|
-
# */
|
|
188
|
-
# public function process($searchtxt, $search_mode='normal', $cache = false) {
|
|
189
|
-
#
|
|
190
|
-
# $this->input = $searchtxt;
|
|
191
|
-
#
|
|
192
|
-
# $this->debug['process'][] = "1 (searchtxt:$searchtxt) (search_mode:$search_mode)";
|
|
193
|
-
# $this->searchtxt = $searchtxt;
|
|
194
|
-
# $this->search_mode=$search_mode;
|
|
195
|
-
#
|
|
196
|
-
# // accept "+" as separator if supplied, tranform to space
|
|
197
|
-
# if ( strpos($this->searchtxt,'+') !== false ) {
|
|
198
|
-
# $text_str = str_replace('+',' ',$this->searchtxt);
|
|
199
|
-
# } else {
|
|
200
|
-
# $text_str = $this->searchtxt;
|
|
201
|
-
# }
|
|
202
|
-
#
|
|
203
|
-
# $this->debug['process'][] = "1a (text_str:$text_str)";
|
|
204
|
-
#
|
|
205
|
-
# if ( is_null($text_str) || $text_str == '' ) {
|
|
206
|
-
# $this->debug['process'][] = "2 Return(false)";
|
|
207
|
-
# return false;
|
|
208
|
-
# }
|
|
209
|
-
#
|
|
210
|
-
# // Clearing the temporary tables
|
|
211
|
-
# $this->db->clearTempTables();
|
|
212
|
-
#
|
|
213
|
-
# // includes stripping of presumed non-relevant content including subgenera, comments, cf's, aff's, etc... to
|
|
214
|
-
#
|
|
215
|
-
# // Normalizing the search text
|
|
216
|
-
# $n = new Normalize($this->db);
|
|
217
|
-
#
|
|
218
|
-
# $this->debug['process'][] = "3 (text_str:$text_str)";
|
|
219
|
-
#
|
|
220
|
-
# if(!$this->chop_overload) {
|
|
221
|
-
# // leave presumed genus + species + authority (in this instance), with genus and species in uppercase
|
|
222
|
-
# $splitter = new Splitter($n,$text_str);
|
|
223
|
-
#
|
|
224
|
-
# $this->this_search_genus = $this_search_genus = $splitter->get('genus');
|
|
225
|
-
# $this->this_search_species = $this_search_species = $splitter->get('species');
|
|
226
|
-
# $this->this_authority = $this_authority = $splitter->get('author');
|
|
227
|
-
# }
|
|
228
|
-
#
|
|
229
|
-
# // cache_flag switch detemines if caching is allowed for the source
|
|
230
|
-
# if($this->cache_flag == true) {
|
|
231
|
-
#
|
|
232
|
-
# if ( $this_search_genus != '' && $this_search_species != '' && $this_authority != '' ) {
|
|
233
|
-
# $cache_key = $this_search_genus . '-' . $this_search_species . '-' . $this_authority . '_' . $search_mode;
|
|
234
|
-
# $cache_path = $this->cache_path . $this->db->source . "/authority/";
|
|
235
|
-
# } else if ( $this_search_genus != '' && $this_search_species != '' ) {
|
|
236
|
-
# $cache_key = $this_search_genus . '-' . $this_search_species . '_' . $search_mode;
|
|
237
|
-
# $cache_path = $this->cache_path . $this->db->source . "/species/";
|
|
238
|
-
# } else if ( $this_search_genus != '' ) {
|
|
239
|
-
# $cache_key = $this_search_genus . '_' . $search_mode;
|
|
240
|
-
# $cache_path = $this->cache_path . $this->db->source . "/genus/";
|
|
241
|
-
# }
|
|
242
|
-
#
|
|
243
|
-
# $this->mkdir_recursive($cache_path);
|
|
244
|
-
# $this->_cache = new Cache( $cache_path );
|
|
245
|
-
# $this->_cache->setKey($cache_key);
|
|
246
|
-
#
|
|
247
|
-
# }
|
|
248
|
-
#
|
|
249
|
-
# $cache_loop_flag = false;
|
|
250
|
-
# if($cache == true && $this->cache_flag == true) {
|
|
251
|
-
# if($this->_cache->cache_exists()) $cache_loop_flag = true;
|
|
252
|
-
# }
|
|
253
|
-
#
|
|
254
|
-
# if(!$cache_loop_flag) {
|
|
255
|
-
#
|
|
256
|
-
# $this->debug['process'][] = "3a (this_search_genus:$this_search_genus) (this_search_species:$this_search_species) (this_authority:$this_authority)";
|
|
257
|
-
#
|
|
258
|
-
# $nm = new NearMatch();
|
|
259
|
-
# $this_near_match_genus = $nm->near_match($this_search_genus);
|
|
260
|
-
#
|
|
261
|
-
# $this->debug['process'][] = "3b (this_near_match_genus:$this_near_match_genus)";
|
|
262
|
-
# //TODO refactor inside of a method
|
|
263
|
-
# $this_genus_start = substr($this_search_genus,0,3);
|
|
264
|
-
# $this_genus_end = substr($this_search_genus,-3);
|
|
265
|
-
# $this_genus_length = strlen($this_search_genus);
|
|
266
|
-
# //TODO_END
|
|
267
|
-
# $this->debug['process'][] = "3c (this_search_genus,$this_search_genus) (this_genus_start:$this_genus_start) (this_genus_end:$this_genus_end) (this_genus_length:$this_genus_length)";
|
|
268
|
-
#
|
|
269
|
-
# if ($this_search_species != '') {
|
|
270
|
-
# $this_near_match_species = $nm->near_match($this_search_species, 'epithet_only');
|
|
271
|
-
# $this_species_length = strlen($this_search_species);
|
|
272
|
-
# $this->debug['process'][] = "4 (this_search_species:$this_search_species) (this_near_match_species:$this_near_match_species) (this_species_length:$this_species_length)";
|
|
273
|
-
# }
|
|
274
|
-
#
|
|
275
|
-
#
|
|
276
|
-
# // now look for exact or near matches on genus first select candidate genera for edit distance (MDLD) test
|
|
277
|
-
#
|
|
278
|
-
# // for drec in genus_cur loop -- includes the genus pre-filter (main portion)
|
|
279
|
-
# $genus_res = $this->db->genus_cur($this->search_mode, $this_near_match_genus, $this_near_match_species, $this_genus_length,$this_genus_start,$this_genus_end);
|
|
280
|
-
#
|
|
281
|
-
# $this->debug['process'][] = "5 (genus_res:$genus_res)";
|
|
282
|
-
#
|
|
283
|
-
# if(count($genus_res)) {
|
|
284
|
-
# foreach($genus_res as $drec) {
|
|
285
|
-
# // test candidate genera for edit distance, keep if satisfies post-test criteria
|
|
286
|
-
# $this->genera_tested++;
|
|
287
|
-
# // do the genus edit distance test
|
|
288
|
-
#
|
|
289
|
-
# $genus_match = $this->match_genera($this_search_genus, $drec->search_genus_name);
|
|
290
|
-
# if ($genus_match['match']) {
|
|
291
|
-
# $phonetic_flag = $genus_match['phonetic_match'] ? 'Y' : null;
|
|
292
|
-
# $this->db->saveGenusMatches($drec->genus_id, $drec->genus, $genus_match['edit_distance'], $phonetic_flag);
|
|
293
|
-
#
|
|
294
|
-
# if ( ($this_search_species != null) && ($this_search_species != '') ) {
|
|
295
|
-
# $species_res = $this->db->species_cur($drec->genus_id, $this_species_length );
|
|
296
|
-
#
|
|
297
|
-
# if(count($species_res)) {
|
|
298
|
-
# foreach($species_res as $drec1) {
|
|
299
|
-
# $this->species_tested++;
|
|
300
|
-
#
|
|
301
|
-
# // do the species edit distance test
|
|
302
|
-
# $species_epithets_match = $this->match_species_epithets($this_search_species, $drec1->search_species_name);
|
|
303
|
-
# $binomials_match = $this->match_binomials($genus_match, $species_epithets_match);
|
|
304
|
-
# if ($binomials_match['match']) {
|
|
305
|
-
#
|
|
306
|
-
# $bionial_phonetic_flag = $binomials_match['phonetic_match'] ? 'Y' : null;
|
|
307
|
-
# $this->db->saveSpeciesMatches($drec1->species_id, $drec1->genus_species, $genus_match['edit_distance'], $temp_species_ED, $binomials_match['edit_distance'], $bionial_phonetic_flag);
|
|
308
|
-
# } //
|
|
309
|
-
# } // End foreach species_res
|
|
310
|
-
# } // End If elements exist for species_res
|
|
311
|
-
# } // End Search Species Exist
|
|
312
|
-
# }
|
|
313
|
-
# }
|
|
314
|
-
# }
|
|
315
|
-
# } // End Cache Loop Flag
|
|
316
|
-
# return true;
|
|
317
|
-
# }
|
|
318
|
-
#
|
|
319
|
-
# /**
|
|
320
|
-
# * generateResponse
|
|
321
|
-
# * Result generation section (including ranking, result shaping,
|
|
322
|
-
# * and authority comparison) - for demo purposes only
|
|
323
|
-
# * NB, in a production system this would be replaced by something
|
|
324
|
-
# * more appropriate, e.g. write to a file or database table,
|
|
325
|
-
# * generate a HTML page for web display,
|
|
326
|
-
# * generate XML response, etc. etc.
|
|
327
|
-
# * @param boolean $cache
|
|
328
|
-
# * @return boolean
|
|
329
|
-
# */
|
|
330
|
-
# public function generateResponse($cache) {
|
|
331
|
-
#
|
|
332
|
-
# $cache_loop_flag = false;
|
|
333
|
-
# if($cache == true && $this->cache_flag == true) {
|
|
334
|
-
# if($this->_cache->cache_exists()) $cache_loop_flag = true;
|
|
335
|
-
# }
|
|
336
|
-
#
|
|
337
|
-
# // if($cache == true && $this->_cache->cache_exists() && $this->cache_flag == true) {
|
|
338
|
-
# if($cache_loop_flag) {
|
|
339
|
-
#
|
|
340
|
-
# $this->data = $this->_cache->fetch();
|
|
341
|
-
# $data_array = json_decode($this->data,true);
|
|
342
|
-
# $data_array['cache'] = $cache;
|
|
343
|
-
# $this->data = json_encode($data_array);
|
|
344
|
-
#
|
|
345
|
-
# } else {
|
|
346
|
-
#
|
|
347
|
-
# // genus exact, phonetic, and other near matches
|
|
348
|
-
# $this->output['input'] = $this->searchtxt;
|
|
349
|
-
# $this->debug['generateResponse'][] = "1 (input:" . $this->searchtxt . ")";
|
|
350
|
-
#
|
|
351
|
-
# // Genus Exact
|
|
352
|
-
# $this->debug['generateResponse'][] = "1a (getGenusAuthority:exact)";
|
|
353
|
-
# $this->getGenusAuthority(0,'exact');
|
|
354
|
-
# // Genus Phonetic
|
|
355
|
-
# $this->debug['generateResponse'][] = "1b (getGenusAuthority:phonetic)";
|
|
356
|
-
# $this->getGenusAuthority('P','phonetic');
|
|
357
|
-
# // Genus near matches
|
|
358
|
-
# $this->debug['generateResponse'][] = "1c (getGenusAuthority:near_1)";
|
|
359
|
-
# $this->getGenusAuthority(1,'near_1');
|
|
360
|
-
# $this->debug['generateResponse'][] = "1d (getGenusAuthority:near_2)";
|
|
361
|
-
# $this->getGenusAuthority(2,'near_2');
|
|
362
|
-
#
|
|
363
|
-
# if(!is_array($this->output['genus']) && $this->this_search_genus != '') {$this->output['genus'] = array();}
|
|
364
|
-
#
|
|
365
|
-
# if ( !is_null($this->this_search_species) ) {
|
|
366
|
-
# // species exact, phonetic, and other near matches
|
|
367
|
-
#
|
|
368
|
-
# $this->debug['generateResponse'][] = "2a (getSpeciesAuthority:exact) ($this->this_authority)";
|
|
369
|
-
# $this->getSpeciesAuthority( 0, 'exact', $this->this_authority );
|
|
370
|
-
# $this->debug['generateResponse'][] = "2b (getSpeciesAuthority:phonetic) ($this->this_authority)";
|
|
371
|
-
# $this->getSpeciesAuthority( 'P', 'phonetic', $this->this_authority );
|
|
372
|
-
# $this->debug['generateResponse'][] = "2c (getSpeciesAuthority:near_1) ($this->this_authority)";
|
|
373
|
-
# $this->getSpeciesAuthority( 1, 'near_1', $this->this_authority );
|
|
374
|
-
# $this->debug['generateResponse'][] = "2d (getSpeciesAuthority:near_2) ($this->this_authority)";
|
|
375
|
-
# $this->getSpeciesAuthority( 2, 'near_2', $this->this_authority );
|
|
376
|
-
#
|
|
377
|
-
# // -- Here is the result shaping section (only show ED 3 if no ED 1,2 or phonetic matches, only
|
|
378
|
-
# // -- show ED 4 if no ED 1,2,3 or phonetic matches). By default shaping is on, unless disabled
|
|
379
|
-
# // -- via the input parameter "search_mode" set to 'no_shaping'.
|
|
380
|
-
# // -- In this demo we supplement any actual shaping with a message to show that it has been invoked,
|
|
381
|
-
# // -- to show the system operates correctly.
|
|
382
|
-
# if ($this->species_found == 'Y') {
|
|
383
|
-
# $temp_species_count = $this->db->countSpeciesMatches(3);
|
|
384
|
-
# $this->debug['generateResponse'][] = "3 (temp_species_count:$temp_species_count)";
|
|
385
|
-
# }
|
|
386
|
-
#
|
|
387
|
-
# if( $temp_species_count > 0 && $this->search_mode == 'no_shaping' ) {
|
|
388
|
-
# $this->debug['generateResponse'][] = "4 (getSpeciesAuthority:near_3) ($this->this_authority)";
|
|
389
|
-
# $this->getSpeciesAuthority( 3, 'near_3', $this->this_authority );
|
|
390
|
-
#
|
|
391
|
-
# if( $this->species_found == 'Y' ) {
|
|
392
|
-
# $temp_species_count = $this->db->countSpeciesMatches(4);
|
|
393
|
-
# }
|
|
394
|
-
#
|
|
395
|
-
# if( $temp_species_count > 0 && $this->search_mode == 'no_shaping') {
|
|
396
|
-
# $this->debug['generateResponse'][] = "4 (getSpeciesAuthority:near_4) ($this->this_authority)";
|
|
397
|
-
# $this->getSpeciesAuthority( 4, 'near_4', $this->this_authority );
|
|
398
|
-
# }
|
|
399
|
-
# } // END temp_species_count > 0 and "no_shaping"
|
|
400
|
-
#
|
|
401
|
-
# } // END If this_search_species
|
|
402
|
-
#
|
|
403
|
-
# if(!is_array($this->output['species']) && $this->this_search_species != '') {$this->output['species'] = array();}
|
|
404
|
-
# if($this->output_type == 'rest') {
|
|
405
|
-
# if($this->debug_flag) {
|
|
406
|
-
# $this->data = json_encode( array( 'success' => true, 'cache' => $cache, 'data' => $this->output, 'debug' => $this->debug ) );
|
|
407
|
-
# } else {
|
|
408
|
-
# $this->data = json_encode( array( 'success' => true, 'cache' => $cache, 'data' => $this->output));
|
|
409
|
-
# }
|
|
410
|
-
# } else {
|
|
411
|
-
# $this->data = $this->output;
|
|
412
|
-
# }
|
|
413
|
-
#
|
|
414
|
-
# if($this->cache_flag == true) {
|
|
415
|
-
# if( ! $this->_cache->cache_exists()) {
|
|
416
|
-
# if($this->debug_flag) {
|
|
417
|
-
# $op_array = array (
|
|
418
|
-
# 'success' => true
|
|
419
|
-
# , 'cache_date' => date('Y-m-d')
|
|
420
|
-
# , 'data' => $this->output
|
|
421
|
-
# , 'debug' => $this->debug
|
|
422
|
-
# );
|
|
423
|
-
# } else {
|
|
424
|
-
# $op_array = array (
|
|
425
|
-
# 'success' => true
|
|
426
|
-
# , 'cache_date' => date('Y-m-d')
|
|
427
|
-
# , 'data' => $this->output
|
|
428
|
-
# );
|
|
429
|
-
#
|
|
430
|
-
# }
|
|
431
|
-
# $op = json_encode($op_array);
|
|
432
|
-
# $this->_cache->update($op);
|
|
433
|
-
# $tmp_cache_key = $this->_cache->getKey();
|
|
434
|
-
# $this->_cache->setKey($tmp_cache_key . '_debug');
|
|
435
|
-
# $dbg = @json_encode($this->debug);
|
|
436
|
-
# $this->_cache->update($dbg);
|
|
437
|
-
# $this->_cache->setKey($tmp_cache_key);
|
|
438
|
-
# }
|
|
439
|
-
# }
|
|
440
|
-
# }
|
|
441
|
-
#
|
|
442
|
-
# return true;
|
|
443
|
-
#
|
|
444
|
-
# }
|
|
77
|
+
end
|
|
File without changes
|
|
@@ -6,8 +6,7 @@ require 'time'
|
|
|
6
6
|
class DamerauLevenshteinMod
|
|
7
7
|
def distance(str1, str2, block_size=2, max_distance=10)
|
|
8
8
|
# puts str1.unpack("U*");
|
|
9
|
-
|
|
10
|
-
(res > max_distance) ? nil : res
|
|
9
|
+
distance_utf(str1.unpack("U*"), str2.unpack("U*"), block_size, max_distance)
|
|
11
10
|
end
|
|
12
11
|
|
|
13
12
|
inline do |builder|
|
data/spec/spec_helper.rb
CHANGED
|
@@ -9,16 +9,20 @@ end
|
|
|
9
9
|
$:.unshift(File.dirname(__FILE__) + '/../lib')
|
|
10
10
|
require 'taxamatch_rb'
|
|
11
11
|
|
|
12
|
-
def read_test_file
|
|
13
|
-
f = open(
|
|
12
|
+
def read_test_file(file, fields_num)
|
|
13
|
+
f = open(file)
|
|
14
14
|
f.each do |line|
|
|
15
|
-
|
|
16
|
-
if line.match(/^\s*#/) == nil &&
|
|
17
|
-
|
|
18
|
-
|
|
19
|
-
yield({:str1 => str1, :str2 => str2, :max_dist => max_dist.to_i, :block_size => block_size.to_i, :distance => distance})
|
|
15
|
+
fields = line.split("|")
|
|
16
|
+
if line.match(/^\s*#/) == nil && fields.size == fields_num
|
|
17
|
+
fields[-1] = fields[-1].split('#')[0].strip
|
|
18
|
+
yield(fields)
|
|
20
19
|
else
|
|
21
|
-
yield(
|
|
20
|
+
yield(nil)
|
|
22
21
|
end
|
|
23
22
|
end
|
|
23
|
+
end
|
|
24
|
+
|
|
25
|
+
def make_taxamatch_hash(string)
|
|
26
|
+
normalized = Normalizer.normalize(string)
|
|
27
|
+
{:epitheton => string, :normalized => normalized, :phonetized => Phonetizer.near_match(normalized)}
|
|
24
28
|
end
|
data/spec/taxamatch_rb_spec.rb
CHANGED
|
@@ -3,11 +3,12 @@ require File.dirname(__FILE__) + '/spec_helper.rb'
|
|
|
3
3
|
|
|
4
4
|
describe 'DamerauLevensteinMod' do
|
|
5
5
|
it 'should get tests' do
|
|
6
|
-
read_test_file do |y|
|
|
6
|
+
read_test_file(File.expand_path(File.dirname(__FILE__)) + '/damerau_levenshtein_mod_test.txt', 5) do |y|
|
|
7
7
|
dl = DamerauLevenshteinMod.new
|
|
8
|
-
|
|
9
|
-
|
|
10
|
-
|
|
8
|
+
if y
|
|
9
|
+
res = dl.distance(y[0], y[1], y[3].to_i, y[2].to_i)
|
|
10
|
+
puts y if res != y[4].to_i
|
|
11
|
+
res.should == y[4].to_i
|
|
11
12
|
end
|
|
12
13
|
end
|
|
13
14
|
end
|
|
@@ -45,6 +46,134 @@ describe 'Normalizer' do
|
|
|
45
46
|
end
|
|
46
47
|
|
|
47
48
|
it 'should normalize words' do
|
|
48
|
-
Normalizer.normalize_word('
|
|
49
|
+
Normalizer.normalize_word('L-3eœ|pt[ura$').should == 'L-3EOEPTURA'
|
|
49
50
|
end
|
|
50
|
-
end
|
|
51
|
+
end
|
|
52
|
+
|
|
53
|
+
describe 'Taxamatch' do
|
|
54
|
+
before(:all) do
|
|
55
|
+
@tm = Taxamatch.new
|
|
56
|
+
end
|
|
57
|
+
|
|
58
|
+
it 'should get txt tests' do
|
|
59
|
+
dl = DamerauLevenshteinMod.new
|
|
60
|
+
read_test_file(File.expand_path(File.dirname(__FILE__)) + '/taxamatch_test.txt', 3) do |y|
|
|
61
|
+
if y
|
|
62
|
+
y[2] = y[2] == 'true' ? true : false
|
|
63
|
+
res = @tm.taxamatch(y[0], y[1])
|
|
64
|
+
puts "%s, %s, %s" % [y[0], y[1], y[2]] if res != y[2]
|
|
65
|
+
res.should == y[2]
|
|
66
|
+
end
|
|
67
|
+
end
|
|
68
|
+
end
|
|
69
|
+
|
|
70
|
+
|
|
71
|
+
it 'should compare genera' do
|
|
72
|
+
#edit distance 1 always match
|
|
73
|
+
g1 = make_taxamatch_hash 'Plantago'
|
|
74
|
+
g2 = make_taxamatch_hash 'Plantagon'
|
|
75
|
+
@tm.match_genera(g1, g2).should == {:phonetic_match=>false, :edit_distance=>1, :match=>true}
|
|
76
|
+
#edit_distance above threshold does not math
|
|
77
|
+
g1 = make_taxamatch_hash 'Plantago'
|
|
78
|
+
g2 = make_taxamatch_hash 'This shouldnt match'
|
|
79
|
+
@tm.match_genera(g1, g2).should == {:phonetic_match=>false, :match=>false, :edit_distance=>4}
|
|
80
|
+
#phonetic_match matches
|
|
81
|
+
g1 = make_taxamatch_hash 'Plantagi'
|
|
82
|
+
g2 = make_taxamatch_hash 'Plantagy'
|
|
83
|
+
@tm.match_genera(g1, g2).should == {:phonetic_match=>true, :edit_distance=>1, :match=>true}
|
|
84
|
+
#distance 1 in first letter also matches
|
|
85
|
+
g1 = make_taxamatch_hash 'Xantheri'
|
|
86
|
+
g2 = make_taxamatch_hash 'Pantheri'
|
|
87
|
+
@tm.match_genera(g1, g2).should == {:phonetic_match=>false, :edit_distance=>1, :match=>true}
|
|
88
|
+
#phonetic match tramps everything
|
|
89
|
+
g1 = make_taxamatch_hash 'Xantheriiiiiiiiiiiiiii'
|
|
90
|
+
g2 = make_taxamatch_hash 'Zanthery'
|
|
91
|
+
@tm.match_genera(g1, g2).should == {:phonetic_match=>true, :edit_distance=>4, :match=>true}
|
|
92
|
+
#same first letter and distance 2 should match
|
|
93
|
+
g1 = make_taxamatch_hash 'Xantherii'
|
|
94
|
+
g2 = make_taxamatch_hash 'Xantherrr'
|
|
95
|
+
@tm.match_genera(g1, g2).should == {:phonetic_match=>false, :match=>true, :edit_distance=>2}
|
|
96
|
+
#First letter is the same and distance is 3 should match, no phonetic match
|
|
97
|
+
g1 = make_taxamatch_hash 'Xantheriii'
|
|
98
|
+
g2 = make_taxamatch_hash 'Xantherrrr'
|
|
99
|
+
@tm.match_genera(g1, g2).should == {:phonetic_match=>false, :match=>true, :edit_distance=>3}
|
|
100
|
+
#Should not match if one of words is shorter than 2x edit distance and distance is 2 or 3
|
|
101
|
+
g1 = make_taxamatch_hash 'Xant'
|
|
102
|
+
g2 = make_taxamatch_hash 'Xanthe'
|
|
103
|
+
@tm.match_genera(g1, g2).should == {:phonetic_match=>false, :match=>false, :edit_distance=>2}
|
|
104
|
+
#Should not match if edit distance > 3 and no phonetic match
|
|
105
|
+
g1 = make_taxamatch_hash 'Xantheriiii'
|
|
106
|
+
g2 = make_taxamatch_hash 'Xantherrrrr'
|
|
107
|
+
@tm.match_genera(g1, g2).should == {:phonetic_match=>false, :match=>false, :edit_distance=>4}
|
|
108
|
+
end
|
|
109
|
+
|
|
110
|
+
it 'should compare species' do
|
|
111
|
+
#Exact match
|
|
112
|
+
s1 = make_taxamatch_hash 'major'
|
|
113
|
+
s2 = make_taxamatch_hash 'major'
|
|
114
|
+
@tm.match_species(s1, s2).should == {:phonetic_match=>true, :match=>true, :edit_distance=>0}
|
|
115
|
+
#Phonetic match always works
|
|
116
|
+
s1 = make_taxamatch_hash 'xanteriiiiiiii'
|
|
117
|
+
s2 = make_taxamatch_hash 'zantereeeeeeee'
|
|
118
|
+
@tm.match_species(s1, s2).should == {:phonetic_match=>true, :match=>true, :edit_distance=>5}
|
|
119
|
+
#Phonetic match works with different endings
|
|
120
|
+
s1 = make_taxamatch_hash 'majorum'
|
|
121
|
+
s2 = make_taxamatch_hash 'majoris'
|
|
122
|
+
@tm.match_species(s1, s2).should == {:phonetic_match=>true, :match=>true, :edit_distance=>2}
|
|
123
|
+
#Distance 4 matches if first 3 chars are the same
|
|
124
|
+
s1 = make_taxamatch_hash 'majorrrrr'
|
|
125
|
+
s2 = make_taxamatch_hash 'majoraaaa'
|
|
126
|
+
@tm.match_species(s1, s2).should == {:phonetic_match=>false, :match=>true, :edit_distance=>4}
|
|
127
|
+
#Should not match if Distance 4 matches and first 3 chars are not the same
|
|
128
|
+
s1 = make_taxamatch_hash 'majorrrrr'
|
|
129
|
+
s2 = make_taxamatch_hash 'marorraaa'
|
|
130
|
+
@tm.match_species(s1, s2).should == {:phonetic_match=>false, :match=>false, :edit_distance=>4}
|
|
131
|
+
#Distance 2 or 3 matches if first 1 char is the same
|
|
132
|
+
s1 = make_taxamatch_hash 'morrrr'
|
|
133
|
+
s2 = make_taxamatch_hash 'moraaa'
|
|
134
|
+
@tm.match_species(s1, s2).should == {:phonetic_match=>false, :match=>true, :edit_distance=>3}
|
|
135
|
+
#Should not match if Distance 2 or 3 and first 1 char is not the same
|
|
136
|
+
s1 = make_taxamatch_hash 'morrrr'
|
|
137
|
+
s2 = make_taxamatch_hash 'torraa'
|
|
138
|
+
@tm.match_species(s1, s2).should == {:phonetic_match=>false, :match=>false, :edit_distance=>3}
|
|
139
|
+
#Distance 1 will match anywhere
|
|
140
|
+
s1 = make_taxamatch_hash 'major'
|
|
141
|
+
s2 = make_taxamatch_hash 'rajor'
|
|
142
|
+
@tm.match_species(s1, s2).should == {:phonetic_match=>false, :match=>true, :edit_distance=>1}
|
|
143
|
+
#Will not match if distance 3 and length is less then twice of the edit distance
|
|
144
|
+
s1 = make_taxamatch_hash 'marrr'
|
|
145
|
+
s2 = make_taxamatch_hash 'maaaa'
|
|
146
|
+
@tm.match_species(s1, s2).should == {:phonetic_match=>false, :match=>false, :edit_distance=>3}
|
|
147
|
+
end
|
|
148
|
+
|
|
149
|
+
it 'should match mathes' do
|
|
150
|
+
#No trobule case
|
|
151
|
+
gmatch = {:match => true, :phonetic_match => true, :edit_distance => 1}
|
|
152
|
+
smatch = {:match => true, :phonetic_match => true, :edit_distance => 1}
|
|
153
|
+
@tm.match_matches(gmatch, smatch).should == {:phonetic_match=>true, :edit_distance=>2, :match=>true}
|
|
154
|
+
#Will not match if either genus or sp. epithet dont match
|
|
155
|
+
gmatch = {:match => false, :phonetic_match => false, :edit_distance => 1}
|
|
156
|
+
smatch = {:match => true, :phonetic_match => true, :edit_distance => 1}
|
|
157
|
+
@tm.match_matches(gmatch, smatch).should == {:phonetic_match=>false, :edit_distance=>2, :match=>false}
|
|
158
|
+
gmatch = {:match => true, :phonetic_match => true, :edit_distance => 1}
|
|
159
|
+
smatch = {:match => false, :phonetic_match => false, :edit_distance => 1}
|
|
160
|
+
@tm.match_matches(gmatch, smatch).should == {:phonetic_match=>false, :edit_distance=>2, :match=>false}
|
|
161
|
+
#Should not match if binomial edit distance > 4 NOTE: EVEN with full phonetic match
|
|
162
|
+
gmatch = {:match => true, :phonetic_match => true, :edit_distance => 3}
|
|
163
|
+
smatch = {:match => true, :phonetic_match => true, :edit_distance => 2}
|
|
164
|
+
@tm.match_matches(gmatch, smatch).should == {:phonetic_match=>true, :edit_distance=>5, :match=>false}
|
|
165
|
+
#Should not have phonetic match if one of the components does not match phonetically
|
|
166
|
+
gmatch = {:match => true, :phonetic_match => false, :edit_distance => 1}
|
|
167
|
+
smatch = {:match => true, :phonetic_match => true, :edit_distance => 1}
|
|
168
|
+
@tm.match_matches(gmatch, smatch).should == {:phonetic_match=>false, :edit_distance=>2, :match=>true}
|
|
169
|
+
gmatch = {:match => true, :phonetic_match => true, :edit_distance => 1}
|
|
170
|
+
smatch = {:match => true, :phonetic_match => false, :edit_distance => 1}
|
|
171
|
+
@tm.match_matches(gmatch, smatch).should == {:phonetic_match=>false, :edit_distance=>2, :match=>true}
|
|
172
|
+
#edit distance should be equal the sum of of edit distances
|
|
173
|
+
gmatch = {:match => true, :phonetic_match => true, :edit_distance => 2}
|
|
174
|
+
smatch = {:match => true, :phonetic_match => true, :edit_distance => 2}
|
|
175
|
+
@tm.match_matches(gmatch, smatch).should == {:phonetic_match=>true, :edit_distance=>4, :match=>true}
|
|
176
|
+
end
|
|
177
|
+
end
|
|
178
|
+
|
|
179
|
+
|
|
@@ -0,0 +1,19 @@
|
|
|
1
|
+
###
|
|
2
|
+
#
|
|
3
|
+
# Tests for string comparison by taxamatch algorithm
|
|
4
|
+
#
|
|
5
|
+
##
|
|
6
|
+
|
|
7
|
+
# additional authorship should match
|
|
8
|
+
Puma concolor|Puma concolor L.|true
|
|
9
|
+
|
|
10
|
+
# one-letter misspeling in species epithet should match
|
|
11
|
+
Puma concolor|Puma cancolor|true
|
|
12
|
+
|
|
13
|
+
Pomatomus saltatrix|Pomatomus saltratix|true
|
|
14
|
+
Pomatomus saltator|Pomatomus saltatrix|true
|
|
15
|
+
|
|
16
|
+
Loligo pealeii|Loligo plei|false
|
|
17
|
+
|
|
18
|
+
# different authors should not match
|
|
19
|
+
#Puma concolor Linnaeus|Puma concolor Kurtz|false
|
data/taxamatch_rb.gemspec
CHANGED
|
@@ -2,7 +2,7 @@
|
|
|
2
2
|
|
|
3
3
|
Gem::Specification.new do |s|
|
|
4
4
|
s.name = %q{taxamatch_rb}
|
|
5
|
-
s.version = "0.1.
|
|
5
|
+
s.version = "0.1.2"
|
|
6
6
|
|
|
7
7
|
s.required_rubygems_version = Gem::Requirement.new(">= 0") if s.respond_to? :required_rubygems_version=
|
|
8
8
|
s.authors = ["Dmitry Mozzherin"]
|
|
@@ -26,6 +26,7 @@ Gem::Specification.new do |s|
|
|
|
26
26
|
"features/support/matchers.rb",
|
|
27
27
|
"features/taxamatch_rb.feature",
|
|
28
28
|
"lib/taxamatch_rb.rb",
|
|
29
|
+
"lib/taxamatch_rb/authormatch.rb",
|
|
29
30
|
"lib/taxamatch_rb/damerau_levenshtein_mod.rb",
|
|
30
31
|
"lib/taxamatch_rb/normalizer.rb",
|
|
31
32
|
"lib/taxamatch_rb/parser.rb",
|
|
@@ -34,6 +35,7 @@ Gem::Specification.new do |s|
|
|
|
34
35
|
"spec/spec.opts",
|
|
35
36
|
"spec/spec_helper.rb",
|
|
36
37
|
"spec/taxamatch_rb_spec.rb",
|
|
38
|
+
"spec/taxamatch_test.txt",
|
|
37
39
|
"taxamatch_rb.gemspec"
|
|
38
40
|
]
|
|
39
41
|
s.has_rdoc = true
|
metadata
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
|
2
2
|
name: dimus-taxamatch_rb
|
|
3
3
|
version: !ruby/object:Gem::Version
|
|
4
|
-
version: 0.1.
|
|
4
|
+
version: 0.1.2
|
|
5
5
|
platform: ruby
|
|
6
6
|
authors:
|
|
7
7
|
- Dmitry Mozzherin
|
|
@@ -55,6 +55,7 @@ files:
|
|
|
55
55
|
- features/support/matchers.rb
|
|
56
56
|
- features/taxamatch_rb.feature
|
|
57
57
|
- lib/taxamatch_rb.rb
|
|
58
|
+
- lib/taxamatch_rb/authormatch.rb
|
|
58
59
|
- lib/taxamatch_rb/damerau_levenshtein_mod.rb
|
|
59
60
|
- lib/taxamatch_rb/normalizer.rb
|
|
60
61
|
- lib/taxamatch_rb/parser.rb
|
|
@@ -63,6 +64,7 @@ files:
|
|
|
63
64
|
- spec/spec.opts
|
|
64
65
|
- spec/spec_helper.rb
|
|
65
66
|
- spec/taxamatch_rb_spec.rb
|
|
67
|
+
- spec/taxamatch_test.txt
|
|
66
68
|
- taxamatch_rb.gemspec
|
|
67
69
|
has_rdoc: true
|
|
68
70
|
homepage: http://github.com/dimus/taxamatch_rb
|