1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
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
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
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
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
|
package sandcrawler
import cascading.tuple.Fields
import cascading.tuple.Tuple
import com.twitter.scalding.{JobTest, TextLine, TypedTsv, TupleConversions}
import org.apache.hadoop.hbase.io.ImmutableBytesWritable
import org.apache.hadoop.hbase.util.Bytes
import org.scalatest._
import parallelai.spyglass.hbase.HBaseConstants.SourceMode
class HBaseCrossrefScoreTest extends FlatSpec with Matchers {
val GrobidString = """
{
"title": "<<TITLE>>",
"authors": [
{"name": "Brewster Kahle"},
{"name": "J Doe"}
],
"journal": {
"name": "Dummy Example File. Journal of Fake News. pp. 1-2. ISSN 1234-5678",
"eissn": null,
"issn": null,
"issue": null,
"publisher": null,
"volume": null
},
"date": "2000",
"doi": null,
"citations": [
{ "authors": [{"name": "A Seaperson"}],
"date": "2001",
"id": "b0",
"index": 0,
"issue": null,
"journal": "Letters in the Alphabet",
"publisher": null,
"title": "Everything is Wonderful",
"url": null,
"volume": "20"},
{ "authors": [],
"date": "2011-03-28",
"id": "b1",
"index": 1,
"issue": null,
"journal": "The Dictionary",
"publisher": null,
"title": "All about Facts",
"url": null,
"volume": "14"}
],
"abstract": "Everything you ever wanted to know about nothing",
"body": "Introduction \nEverything starts somewhere, as somebody [1] once said. \n\n In Depth \n Meat \nYou know, for kids. \n Potatos \nQED.",
"acknowledgement": null,
"annex": null
}
"""
val GrobidStringWithTitle = GrobidString.replace("<<TITLE>>", "Dummy Example File")
val GrobidStringWithoutTitle = GrobidString.replace("title", "nottitle")
val MalformedGrobidString = GrobidString.replace("}", "")
val CrossrefString =
"""
{ "_id" : { "$oid" : "5a553d5988a035a45bf50ed3" },
"indexed" : { "date-parts" : [ [ 2017, 10, 23 ] ],
"date-time" : "2017-10-23T17:19:16Z",
"timestamp" : { "$numberLong" : "1508779156477" } },
"reference-count" : 0,
"publisher" : "Elsevier BV",
"issue" : "3",
"license" : [ { "URL" : "http://www.elsevier.com/tdm/userlicense/1.0/",
"start" : { "date-parts" : [ [ 1996, 1, 1 ] ],
"date-time" : "1996-01-01T00:00:00Z",
"timestamp" : { "$numberLong" : "820454400000" } },
"delay-in-days" : 0, "content-version" : "tdm" }],
"content-domain" : { "domain" : [], "crossmark-restriction" : false },
"published-print" : { "date-parts" : [ [ 1996 ] ] },
"DOI" : "<<DOI>>",
"type" : "journal-article",
"created" : { "date-parts" : [ [ 2002, 7, 25 ] ],
"date-time" : "2002-07-25T15:09:41Z",
"timestamp" : { "$numberLong" : "1027609781000" } },
"page" : "186-187",
"source" : "Crossref",
"is-referenced-by-count" : 0,
"title" : [ "<<TITLE>>" ],
"prefix" : "10.1016",
"volume" : "9",
"author" : [ { "given" : "W", "family" : "Gaier", "affiliation" : [] } ],
"member" : "78",
"container-title" : [ "Journal de Pédiatrie et de Puériculture" ],
"link" : [ { "URL" : "http://api.elsevier.com/content/article/PII:0987-7983(96)87729-2?httpAccept=text/xml",
"content-type" : "text/xml",
"content-version" : "vor",
"intended-application" : "text-mining" },
{ "URL" :
"http://api.elsevier.com/content/article/PII:0987-7983(96)87729-2?httpAccept=text/plain",
"content-type" : "text/plain",
"content-version" : "vor",
"intended-application" : "text-mining" } ],
"deposited" : { "date-parts" : [ [ 2015, 9, 3 ] ],
"date-time" : "2015-09-03T10:03:43Z",
"timestamp" : { "$numberLong" : "1441274623000" } },
"score" : 1,
"issued" : { "date-parts" : [ [ 1996 ] ] },
"references-count" : 0,
"alternative-id" : [ "0987-7983(96)87729-2" ],
"URL" : "http://dx.doi.org/10.1016/0987-7983(96)87729-2",
"ISSN" : [ "0987-7983" ],
"issn-type" : [ { "value" : "0987-7983", "type" : "print" } ],
"subject" : [ "Pediatrics, Perinatology, and Child Health" ]
}
"""
val CrossrefStringWithTitle = CrossrefString.replace("<<TITLE>>", "SomeTitle")
val CrossrefStringWithoutTitle = CrossrefString.replace("title", "nottitle")
val MalformedCrossrefString = CrossrefString.replace("}", "")
// Unit tests
"titleToSlug()" should "extract the parts of titles before a colon" in {
val slug = HBaseCrossrefScore.titleToSlug("HELLO:there")
slug should contain ("hello")
}
it should "extract an entire colon-less string" in {
val slug = HBaseCrossrefScore.titleToSlug("hello THERE")
slug should contain ("hello there")
}
"grobidToSlug()" should "get the right slug for a grobid json string" in {
val slug = HBaseCrossrefScore.grobidToSlug(GrobidStringWithTitle)
slug should contain ("dummy example file")
}
it should "return None if given json string without title" in {
val slug = HBaseCrossrefScore.grobidToSlug(GrobidStringWithoutTitle)
slug shouldBe None
}
it should "return None if given a malformed json string" in {
val slug = HBaseCrossrefScore.grobidToSlug(MalformedGrobidString)
slug shouldBe None
}
"crossrefToSlug()" should "get the right slug for a crossref json string" in {
val slug = HBaseCrossrefScore.crossrefToSlug(CrossrefStringWithTitle)
slug should contain ("sometitle")
}
it should "return None if given json string without title" in {
val slug = HBaseCrossrefScore.grobidToSlug(CrossrefStringWithoutTitle)
slug shouldBe None
}
it should "return None if given a malformed json string" in {
val slug = HBaseCrossrefScore.grobidToSlug(MalformedCrossrefString)
slug shouldBe None
}
// Pipeline tests
val output = "/tmp/testOutput"
val input = "/tmp/testInput"
val (testTable, testHost) = ("test-table", "dummy-host:2181")
val grobidSampleData = List(
List(Bytes.toBytes("sha1:K2DKSSVTXWPRMFDTWSTCQW3RVWRIOV3Q"),
Bytes.toBytes(GrobidString.replace("<<TITLE>>", "Title 1"))),
List(Bytes.toBytes("sha1:C3YNNEGH5WAG5ZAAXWAEBNXJWT6CZ3WU"),
Bytes.toBytes(GrobidString.replace("<<TITLE>>", "Title 2: TNG"))),
List(Bytes.toBytes("sha1:SDKUVHC3YNNEGH5WAG5ZAAXWAEBNX4WT"),
Bytes.toBytes(GrobidString.replace("<<TITLE>>", "Title 3: The Sequel"))),
List(Bytes.toBytes("sha1:35985C3YNNEGH5WAG5ZAAXWAEBNXJW56"),
Bytes.toBytes(MalformedGrobidString)))
JobTest("sandcrawler.HBaseCrossrefScoreJob")
.arg("test", "")
.arg("app.conf.path", "app.conf")
.arg("output", output)
.arg("hbase-table", testTable)
.arg("zookeeper-hosts", testHost)
.arg("crossref-input", input)
.arg("debug", "true")
.source[Tuple](HBaseCrossrefScore.getHBaseSource(testTable, testHost),
grobidSampleData.map(l => new Tuple(l.map(s => {new ImmutableBytesWritable(s)}):_*)))
.source(TextLine(input), List(
0 -> CrossrefString.replace("<<TITLE>>", "Title 1: TNG").replace("<<DOI>>", "DOI-0"),
1 -> CrossrefString.replace("<<TITLE>>", "Title 1: TNG 2").replace("<<DOI>>", "DOI-0.5"),
2 -> CrossrefString.replace("<<TITLE>>", "Title 1: TNG 3").replace("<<DOI>>", "DOI-0.75"),
3 -> CrossrefString.replace("<<TITLE>>", "Title 2: Rebooted").replace("<<DOI>>", "DOI-1")))
.sink[(Int, String, String, String, String)](TypedTsv[(Int,
String, String, String, String)](output)) {
// Grobid titles:
// "Title 1", "Title 2: TNG", "Title 3: The Sequel"
// crossref slugs:
// "Title 1: TNG", "Title 1: TNG 2", "Title 1: TNG 3", "Title 2 Rebooted"
// Join should have 3 "Title 1" slugs and 1 "Title 2" slug
outputBuffer =>
it should "return a 4-element list" in {
outputBuffer should have length 4
}
/*
it should "return the right first entry" in {
val (slug, slug0, slug1, sha1, grobidJson, crossrefJson) = outputBuffer(0)
slug shouldBe "title 1"
slug shouldBe slug0
slug shouldBe slug1
sha1 shouldBe new String(grobidSampleData(0)(0), "UTF-8")
grobidJson shouldBe new String(grobidSampleData(0)(1), "UTF-8")
}
*/
}
.run
.finish
}
|
