aboutsummaryrefslogtreecommitdiffstats
path: root/scalding/src/test
diff options
context:
space:
mode:
Diffstat (limited to 'scalding/src/test')
-rw-r--r--scalding/src/test/scala/sandcrawler/CrossrefScorableTest.scala100
-rw-r--r--scalding/src/test/scala/sandcrawler/GrobidScorableTest.scala88
-rw-r--r--scalding/src/test/scala/sandcrawler/HBaseBuilderTest.scala1
-rw-r--r--scalding/src/test/scala/sandcrawler/HBaseMimeCountTest.scala9
-rw-r--r--scalding/src/test/scala/sandcrawler/HBaseRowCountTest.scala11
-rw-r--r--scalding/src/test/scala/sandcrawler/HBaseStatusCountTest.scala13
-rw-r--r--scalding/src/test/scala/sandcrawler/ScorableFeaturesTest.scala57
-rw-r--r--scalding/src/test/scala/sandcrawler/ScorableTest.scala73
-rw-r--r--scalding/src/test/scala/sandcrawler/ScoreJobTest.scala226
-rw-r--r--scalding/src/test/scala/sandcrawler/StringUtilitiesTest.scala85
10 files changed, 651 insertions, 12 deletions
diff --git a/scalding/src/test/scala/sandcrawler/CrossrefScorableTest.scala b/scalding/src/test/scala/sandcrawler/CrossrefScorableTest.scala
new file mode 100644
index 0000000..1789d1a
--- /dev/null
+++ b/scalding/src/test/scala/sandcrawler/CrossrefScorableTest.scala
@@ -0,0 +1,100 @@
+package sandcrawler
+
+import cascading.tuple.Fields
+import cascading.tuple.Tuple
+import com.twitter.scalding.JobTest
+import com.twitter.scalding.TextLine
+import com.twitter.scalding.TupleConversions
+import com.twitter.scalding.TypedTsv
+import org.apache.hadoop.hbase.io.ImmutableBytesWritable
+import org.apache.hadoop.hbase.util.Bytes
+import org.scalatest._
+import parallelai.spyglass.hbase.HBaseConstants.SourceMode
+
+class CrossrefScorableTest extends FlatSpec with Matchers {
+ // scalastyle:off
+ 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" ]
+}
+"""
+ // scalastyle:on
+ val CrossrefStringWithTitle = CrossrefString.replace("<<TITLE>>", "Some Title")
+ val CrossrefStringWithEmptyTitle = CrossrefString.replace("<<TITLE>>", "")
+ val CrossrefStringWithoutTitle = CrossrefString.replace("title", "nottitle")
+ val MalformedCrossrefString = CrossrefString.replace("}", "")
+
+ // Unit tests
+ "CrossrefScorable.jsonToMapFeatures()" should "handle invalid JSON" in {
+ val result = CrossrefScorable.jsonToMapFeatures(MalformedCrossrefString)
+ result.slug shouldBe Scorable.NoSlug
+ }
+
+ it should "handle missing title" in {
+ val result = CrossrefScorable.jsonToMapFeatures(CrossrefStringWithoutTitle)
+ result.slug shouldBe Scorable.NoSlug
+ }
+
+ it should "handle empty title" in {
+ val result = CrossrefScorable.jsonToMapFeatures(CrossrefStringWithEmptyTitle)
+ result.slug shouldBe Scorable.NoSlug
+ }
+
+ it should "handle valid input" in {
+ val result = CrossrefScorable.jsonToMapFeatures(CrossrefStringWithTitle)
+ result.slug shouldBe "sometitle"
+ Scorable.jsonToMap(result.json) match {
+ case None => fail()
+ case Some(map) => {
+ map("title").asInstanceOf[String] shouldBe "Some Title"
+ }
+ }
+ }
+}
diff --git a/scalding/src/test/scala/sandcrawler/GrobidScorableTest.scala b/scalding/src/test/scala/sandcrawler/GrobidScorableTest.scala
new file mode 100644
index 0000000..661824b
--- /dev/null
+++ b/scalding/src/test/scala/sandcrawler/GrobidScorableTest.scala
@@ -0,0 +1,88 @@
+package sandcrawler
+
+import cascading.tuple.Fields
+import cascading.tuple.Tuple
+import com.twitter.scalding.JobTest
+import com.twitter.scalding.TextLine
+import com.twitter.scalding.TupleConversions
+import com.twitter.scalding.TypedTsv
+import org.apache.hadoop.hbase.io.ImmutableBytesWritable
+import org.apache.hadoop.hbase.util.Bytes
+import org.scalatest._
+import parallelai.spyglass.hbase.HBaseConstants.SourceMode
+
+class GrobidScorableTest 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 Key = "Dummy Key"
+
+ // Unit tests
+
+ "GrobidScorable.jsonToMapFeatures()" should "handle invalid JSON" in {
+ val result = GrobidScorable.jsonToMapFeatures(Key, MalformedGrobidString)
+ result.slug shouldBe Scorable.NoSlug
+ }
+
+ it should "handle missing title" in {
+ val result = GrobidScorable.jsonToMapFeatures(Key, GrobidStringWithoutTitle)
+ result.slug shouldBe Scorable.NoSlug
+ }
+
+ it should "handle valid input" in {
+ val result = GrobidScorable.jsonToMapFeatures(Key, GrobidStringWithTitle)
+ result.slug shouldBe "dummyexamplefile"
+ Scorable.jsonToMap(result.json) match {
+ case None => fail()
+ case Some(map) => {
+ map should contain key "title"
+ map("title").asInstanceOf[String] shouldBe "Dummy Example File"
+ }
+ }
+ }
+}
diff --git a/scalding/src/test/scala/sandcrawler/HBaseBuilderTest.scala b/scalding/src/test/scala/sandcrawler/HBaseBuilderTest.scala
index 603a4c7..c61cb22 100644
--- a/scalding/src/test/scala/sandcrawler/HBaseBuilderTest.scala
+++ b/scalding/src/test/scala/sandcrawler/HBaseBuilderTest.scala
@@ -22,6 +22,7 @@ class HBaseBuilderTest extends FlatSpec with Matchers {
fields should have length 0
}
+ //scalastyle:off no.whitespace.before.left.bracket
it should "throw IllegalArgumentException on malformed input" in {
a [IllegalArgumentException] should be thrownBy {
HBaseBuilder.parseColSpecs(List("file_size"))
diff --git a/scalding/src/test/scala/sandcrawler/HBaseMimeCountTest.scala b/scalding/src/test/scala/sandcrawler/HBaseMimeCountTest.scala
index fde2290..d6d283f 100644
--- a/scalding/src/test/scala/sandcrawler/HBaseMimeCountTest.scala
+++ b/scalding/src/test/scala/sandcrawler/HBaseMimeCountTest.scala
@@ -1,15 +1,18 @@
package sandcrawler
-import cascading.tuple.{Tuple, Fields}
-import com.twitter.scalding.{JobTest, Tsv, TupleConversions}
+import cascading.tuple.Fields
+import cascading.tuple.Tuple
+import com.twitter.scalding.JobTest
+import com.twitter.scalding.Tsv
+import com.twitter.scalding.TupleConversions
import org.apache.hadoop.hbase.io.ImmutableBytesWritable
import org.apache.hadoop.hbase.util.Bytes
import org.junit.runner.RunWith
import org.scalatest.FunSpec
import org.scalatest.junit.JUnitRunner
import org.slf4j.LoggerFactory
-import parallelai.spyglass.hbase.HBaseSource
import parallelai.spyglass.hbase.HBaseConstants.SourceMode
+import parallelai.spyglass.hbase.HBaseSource
import scala._
@RunWith(classOf[JUnitRunner])
diff --git a/scalding/src/test/scala/sandcrawler/HBaseRowCountTest.scala b/scalding/src/test/scala/sandcrawler/HBaseRowCountTest.scala
index 3424a36..c4ca5aa 100644
--- a/scalding/src/test/scala/sandcrawler/HBaseRowCountTest.scala
+++ b/scalding/src/test/scala/sandcrawler/HBaseRowCountTest.scala
@@ -1,15 +1,18 @@
package sandcrawler
-import cascading.tuple.{Tuple, Fields}
-import com.twitter.scalding.{JobTest, Tsv, TupleConversions}
+import cascading.tuple.Fields
+import cascading.tuple.Tuple
+import com.twitter.scalding.JobTest
+import com.twitter.scalding.Tsv
+import com.twitter.scalding.TupleConversions
import org.apache.hadoop.hbase.io.ImmutableBytesWritable
import org.apache.hadoop.hbase.util.Bytes
import org.junit.runner.RunWith
import org.scalatest.FunSpec
import org.scalatest.junit.JUnitRunner
import org.slf4j.LoggerFactory
-import parallelai.spyglass.hbase.HBaseSource
import parallelai.spyglass.hbase.HBaseConstants.SourceMode
+import parallelai.spyglass.hbase.HBaseSource
import scala._
/**
@@ -47,12 +50,10 @@ class HBaseRowCountTest extends FunSpec with TupleConversions {
outputBuffer =>
it("should return the test data provided.") {
- println("outputBuffer.size => " + outputBuffer.size)
assert(outputBuffer.size === 1)
}
it("should return the correct count") {
- println("raw output => " + outputBuffer)
assert(outputBuffer(0).getObject(0) === 8)
}
}
diff --git a/scalding/src/test/scala/sandcrawler/HBaseStatusCountTest.scala b/scalding/src/test/scala/sandcrawler/HBaseStatusCountTest.scala
index d7689cd..0da0b9c 100644
--- a/scalding/src/test/scala/sandcrawler/HBaseStatusCountTest.scala
+++ b/scalding/src/test/scala/sandcrawler/HBaseStatusCountTest.scala
@@ -1,15 +1,19 @@
package sandcrawler
-import cascading.tuple.{Tuple, Fields}
-import com.twitter.scalding.{JobTest, Tsv, TypedTsv, TupleConversions}
+import cascading.tuple.Fields
+import cascading.tuple.Tuple
+import com.twitter.scalding.JobTest
+import com.twitter.scalding.Tsv
+import com.twitter.scalding.TupleConversions
+import com.twitter.scalding.TypedTsv
import org.apache.hadoop.hbase.io.ImmutableBytesWritable
import org.apache.hadoop.hbase.util.Bytes
import org.junit.runner.RunWith
import org.scalatest.FunSpec
import org.scalatest.junit.JUnitRunner
import org.slf4j.LoggerFactory
-import parallelai.spyglass.hbase.HBaseSource
import parallelai.spyglass.hbase.HBaseConstants.SourceMode
+import parallelai.spyglass.hbase.HBaseSource
import scala._
@RunWith(classOf[JUnitRunner])
@@ -25,7 +29,8 @@ class HBaseStatusCountTest extends FunSpec with TupleConversions {
val statusType1Bytes = Bytes.toBytes(statusType1)
val statusType2Bytes = Bytes.toBytes(statusType2)
- val sampleData = List(
+ val sampleData : List[List[Array[Byte]]] = List(
+ // TODO(bnewbold): now to express a null (empty value) in this list?
List(Bytes.toBytes("sha1:K2DKSSVTXWPRMFDTWSTCQW3RVWRIOV3Q"), statusType1Bytes),
List(Bytes.toBytes("sha1:C3YNNEGH5WAG5ZAAXWAEBNXJWT6CZ3WU"), statusType1Bytes),
List(Bytes.toBytes("sha1:SDKUVHC3YNNEGH5WAG5ZAAXWAEBNX4WT"), statusType2Bytes),
diff --git a/scalding/src/test/scala/sandcrawler/ScorableFeaturesTest.scala b/scalding/src/test/scala/sandcrawler/ScorableFeaturesTest.scala
new file mode 100644
index 0000000..80d92aa
--- /dev/null
+++ b/scalding/src/test/scala/sandcrawler/ScorableFeaturesTest.scala
@@ -0,0 +1,57 @@
+package sandcrawler
+
+import org.scalatest._
+
+// scalastyle:off null
+class ScorableFeaturesTest extends FlatSpec with Matchers {
+ private def titleToSlug(s : String) : String = {
+ new ScorableFeatures(title = s).toSlug
+ }
+
+ "toMapFeatures()" should "work with gnarly inputs" in {
+ new ScorableFeatures(title = null).toMapFeatures
+ new ScorableFeatures(title = "something", doi = null, sha1 = null, year = 123).toMapFeatures
+ }
+
+ "mapToSlug()" should "extract the parts of titles before a colon" in {
+ titleToSlug("HELLO:there") shouldBe "hellothere"
+ }
+
+ it should "extract an entire colon-less string" in {
+ titleToSlug("hello THERE") shouldBe "hellothere"
+ }
+
+ it should "return Scorable.NoSlug if given empty string" in {
+ titleToSlug("") shouldBe Scorable.NoSlug
+ }
+
+ it should "return Scorable.NoSlug if given null" in {
+ titleToSlug(null) shouldBe Scorable.NoSlug
+ }
+
+ it should "strip punctuation" in {
+ titleToSlug("HELLO!:the:re") shouldBe "hellothere"
+ titleToSlug("a:b:c") shouldBe "abc"
+ titleToSlug(
+ "If you're happy and you know it, clap your hands!") shouldBe "ifyourehappyandyouknowitclapyourhands"
+ titleToSlug(":;\"\'") shouldBe Scorable.NoSlug
+ }
+
+ it should "filter stub titles" in {
+ titleToSlug("abstract") shouldBe Scorable.NoSlug
+ titleToSlug("title!") shouldBe Scorable.NoSlug
+ titleToSlug("a real title which is not on blacklist") shouldBe "arealtitlewhichisnotonblacklist"
+ }
+
+ it should "strip special characters" in {
+ titleToSlug(":;!',|\"\'`.#?!-@*/\\=+~%$^{}()[]<>-_") shouldBe Scorable.NoSlug
+ // TODO: titleToSlug("©™₨№…") shouldBe Scorable.NoSlug
+ // TODO: titleToSlug("πµΣσ") shouldBe Scorable.NoSlug
+ }
+
+ it should "remove whitespace" in {
+ titleToSlug("foo bar : baz ::") shouldBe "foobarbaz"
+ titleToSlug("\na\t:b:c") shouldBe "abc"
+ titleToSlug("\n \t \r ") shouldBe Scorable.NoSlug
+ }
+}
diff --git a/scalding/src/test/scala/sandcrawler/ScorableTest.scala b/scalding/src/test/scala/sandcrawler/ScorableTest.scala
new file mode 100644
index 0000000..f63bef8
--- /dev/null
+++ b/scalding/src/test/scala/sandcrawler/ScorableTest.scala
@@ -0,0 +1,73 @@
+package sandcrawler
+
+import cascading.tuple.Fields
+import cascading.tuple.Tuple
+import com.twitter.scalding.JobTest
+import com.twitter.scalding.TextLine
+import com.twitter.scalding.TupleConversions
+import com.twitter.scalding.TypedTsv
+import org.apache.hadoop.hbase.io.ImmutableBytesWritable
+import org.apache.hadoop.hbase.util.Bytes
+import org.scalatest._
+import parallelai.spyglass.hbase.HBaseConstants.SourceMode
+
+class ScorableTest extends FlatSpec with Matchers {
+ val JsonString = """
+{
+ "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
+}
+"""
+ "jsonToMap()" should "return a map, given a legal JSON string" in {
+ Scorable.jsonToMap(JsonString) should not be (None)
+ }
+
+ it should "return None, given illegal JSON" in {
+ Scorable.jsonToMap("illegal{,json{{") should be (None)
+ }
+
+ "computeOutput()" should "return Scorable.MaxScore if given identical ReduceFeatures" in {
+ val score = Scorable.computeSimilarity(
+ new ReduceFeatures(JsonString), new ReduceFeatures(JsonString))
+ score shouldBe Scorable.MaxScore
+ }
+}
diff --git a/scalding/src/test/scala/sandcrawler/ScoreJobTest.scala b/scalding/src/test/scala/sandcrawler/ScoreJobTest.scala
new file mode 100644
index 0000000..f92ba31
--- /dev/null
+++ b/scalding/src/test/scala/sandcrawler/ScoreJobTest.scala
@@ -0,0 +1,226 @@
+package sandcrawler
+
+import cascading.tuple.Fields
+import cascading.tuple.Tuple
+import com.twitter.scalding.JobTest
+import com.twitter.scalding.TextLine
+import com.twitter.scalding.TupleConversions
+import com.twitter.scalding.TypedTsv
+import org.apache.hadoop.hbase.io.ImmutableBytesWritable
+import org.apache.hadoop.hbase.util.Bytes
+import org.scalatest._
+import parallelai.spyglass.hbase.HBaseConstants.SourceMode
+
+class ScoreJobTest extends FlatSpec with Matchers {
+ //scalastyle:off
+ val JsonString = """
+{
+ "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
+}
+"""
+ // scalastyle:on
+ val JsonStringWithTitle = JsonString.replace("<<TITLE>>", "Dummy Example File")
+ val JsonStringWithoutTitle = JsonString.replace("title", "nottitle")
+ val MalformedJsonString = JsonString.replace("}", "")
+
+ // scalastyle:off
+ 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" ]
+}
+"""
+ // scalastyle:on
+ val CrossrefStringWithTitle = CrossrefString.replace("<<TITLE>>", "SomeTitle")
+ val CrossrefStringWithoutTitle = CrossrefString.replace("title", "nottitle")
+ val MalformedCrossrefString = CrossrefString.replace("}", "")
+ val CrossrefStrings = List(
+ CrossrefString.replace("<<TITLE>>", "Title 2: TNG").replace("<<DOI>>", "DOI-0"),
+ CrossrefString.replace("<<TITLE>>", "Title 1: TNG 2A").replace("<<DOI>>", "DOI-0.5"),
+ CrossrefString.replace("<<TITLE>>", "Title 1: TNG 3").replace("<<DOI>>", "DOI-0.75"),
+ CrossrefString.replace("<<TITLE>>", "Title 2: Rebooted").replace("<<DOI>>", "DOI-1"))
+
+ // Pipeline tests
+ val output = "/tmp/testOutput"
+ val input = "/tmp/testInput"
+ val (testTable, testHost) = ("test-table", "dummy-host:2181")
+
+ val Sha1Strings : List[String] = List(
+ "sha1:K2DKSSVTXWPRMFDTWSTCQW3RVWRIOV3Q",
+ "sha1:C3YNNEGH5WAG5ZAAXWAEBNXJWT6CZ3WU",
+ "sha1:SDKUVHC3YNNEGH5WAG5ZAAXWAEBNX4WT",
+ "sha1:35985C3YNNEGH5WAG5ZAAXWAEBNXJW56",
+ "sha1:93187A85273589347598473894839443",
+ "sha1:024937534094897039547e9824382943")
+
+ val JsonStrings : List[String] = List(
+ JsonString.replace("<<TITLE>>", "Title 1"),
+ JsonString.replace("<<TITLE>>", "Title 2: TNG"),
+ JsonString.replace("<<TITLE>>", "Title 3: The Sequel"),
+ // This will have bad status.
+ JsonString.replace("<<TITLE>>", "Title 1"),
+ MalformedJsonString,
+ // This will have bad status.
+ JsonString.replace("<<TITLE>>", "Title 2")
+ )
+
+ // bnewbold: status codes aren't strings, they are uint64
+ val Ok : Long = 200
+ val Bad : Long = 400
+ val StatusCodes = List(Ok, Ok, Ok, Bad, Ok, Bad)
+
+ val SampleDataHead : List[Tuple] = (Sha1Strings, JsonStrings, StatusCodes)
+ .zipped
+ .toList
+ .map { case (sha, json, status) => List(Bytes.toBytes(sha), Bytes.toBytes(json), Bytes.toBytes(status)) }
+ .map { l => new Tuple(l.map(s => {new ImmutableBytesWritable(s)}):_*) }
+
+ // Add example of lines without GROBID data
+ val SampleData = SampleDataHead :+ new Tuple(
+ new ImmutableBytesWritable(Bytes.toBytes("sha1:35985C3YNNEGH5WAG5ZAA88888888888")), null, null)
+
+ JobTest("sandcrawler.ScoreJob")
+ .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](GrobidScorable.getHBaseSource(testTable, testHost), SampleData)
+ .source(TextLine(input), List(
+ 0 -> CrossrefStrings(0),
+ 1 -> CrossrefStrings(1),
+ 2 -> CrossrefStrings(2),
+ 3 -> CrossrefStrings(3)))
+ .sink[(String, Int, String, String)](TypedTsv[(String, Int, String, String)](output)) {
+ // Grobid titles and slugs (in parentheses):
+ // Title 1 (title1)
+ // Title 2: TNG (title2tng)
+ // Title 3: The Sequel (title3thesequel)
+ // crossref titles and slugs (in parentheses):
+ // Title 2: TNG (title2tng)
+ // Title 1: TNG 2A (title1tng2a)
+ // Title 1: TNG 3 (title1tng3)
+ // Title 2: Rebooted (title2rebooted)
+ // XXX: Join should have 3 "title1" slugs and 1 "title2tng" slug
+ outputBuffer =>
+ "The pipeline" should "return a 1-element list" in {
+ outputBuffer should have length 1
+ }
+
+ it should "has right # of entries with each slug" in {
+ val slugs = outputBuffer.map(_._1)
+ val countMap : Map[String, Int] = slugs.groupBy(identity).mapValues(_.size)
+ // XXX: countMap("title1") shouldBe 3
+ countMap("title2tng") shouldBe 1
+ }
+
+ def bundle(slug : String, grobidIndex : Int, crossrefIndex : Int) : (String, Int, String, String) = {
+ val mf1 : MapFeatures = GrobidScorable.jsonToMapFeatures(
+ Sha1Strings(grobidIndex),
+ JsonStrings(grobidIndex))
+ val mf2 : MapFeatures = CrossrefScorable.jsonToMapFeatures(
+ CrossrefStrings(crossrefIndex))
+ val score = Scorable.computeSimilarity(
+ ReduceFeatures(mf1.json),
+ ReduceFeatures(mf2.json))
+ (slug, score, mf1.json, mf2.json)
+ }
+
+ it should "have right output values" in {
+ //outputBuffer.exists(_ == bundle("title1", 0, 0))
+ //outputBuffer.exists(_ == bundle("title1", 0, 2))
+ //outputBuffer.exists(_ == bundle("title1", 0, 1))
+ outputBuffer.exists(_ == bundle("title2tng", 1, 3))
+ }
+ }
+ .run
+ .finish
+}
diff --git a/scalding/src/test/scala/sandcrawler/StringUtilitiesTest.scala b/scalding/src/test/scala/sandcrawler/StringUtilitiesTest.scala
new file mode 100644
index 0000000..410819b
--- /dev/null
+++ b/scalding/src/test/scala/sandcrawler/StringUtilitiesTest.scala
@@ -0,0 +1,85 @@
+package sandcrawler
+
+import org.scalatest._
+
+class StringUtilitiesTest extends FlatSpec with Matchers {
+ "removeAccents()" should "handle the empty string" in {
+ StringUtilities.removeAccents("") shouldBe ""
+ }
+
+ it should "not change a string with unaccented characters" in {
+ StringUtilities.removeAccents("abc123") shouldBe "abc123"
+ }
+
+ it should "remove accents from Ls" in {
+ StringUtilities.removeAccents("E\u0141\u0142en") shouldBe "ELlen"
+ }
+
+ it should "remove accents from Es without changing case" in {
+ val result = StringUtilities.removeAccents("\u00e9")
+ result should have length 1
+ result shouldBe "e"
+ }
+
+ it should "convert the ø in Soren" in {
+ StringUtilities.removeAccents("Søren") shouldBe "Soren"
+ StringUtilities.removeAccents("SØREN") shouldBe "SOREN"
+ }
+
+ "removePunctuation" should "work on the empty string" in {
+ StringUtilities.removePunctuation("") shouldBe ""
+ }
+
+ it should "work on non-empty text strings" in {
+ StringUtilities.removePunctuation("Hello, world!") shouldBe "Hello world"
+ StringUtilities.removePunctuation(":-)") shouldBe ""
+ StringUtilities.removePunctuation("<<---a---b--->") shouldBe "ab"
+ }
+
+ // Tests adapted from https://oldfashionedsoftware.com/2009/11/19/string-distance-and-refactoring-in-scala/
+ "stringDistance" should "work on empty strings" in {
+ StringUtilities.stringDistance("", "") shouldBe 0
+ StringUtilities.stringDistance("a", "") shouldBe 1
+ StringUtilities.stringDistance("", "a") shouldBe 1
+ StringUtilities.stringDistance("abc", "") shouldBe 3
+ StringUtilities.stringDistance("", "abc") shouldBe 3
+ }
+
+ it should "work on equal strings" in {
+ StringUtilities.stringDistance("", "") shouldBe 0
+ StringUtilities.stringDistance("a", "a") shouldBe 0
+ StringUtilities.stringDistance("abc", "abc") shouldBe 0
+ }
+
+ it should "work where only inserts are needed" in {
+ StringUtilities.stringDistance("", "a") shouldBe 1
+ StringUtilities.stringDistance("a", "ab") shouldBe 1
+ StringUtilities.stringDistance("b", "ab") shouldBe 1
+ StringUtilities.stringDistance("ac", "abc") shouldBe 1
+ StringUtilities.stringDistance("abcdefg", "xabxcdxxefxgx") shouldBe 6
+ }
+
+ it should "work where only deletes are needed" in {
+ StringUtilities.stringDistance( "a", "") shouldBe 1
+ StringUtilities.stringDistance( "ab", "a") shouldBe 1
+ StringUtilities.stringDistance( "ab", "b") shouldBe 1
+ StringUtilities.stringDistance("abc", "ac") shouldBe 1
+ StringUtilities.stringDistance("xabxcdxxefxgx", "abcdefg") shouldBe 6
+ }
+
+ it should "work where only substitutions are needed" in {
+ StringUtilities.stringDistance( "a", "b") shouldBe 1
+ StringUtilities.stringDistance( "ab", "ac") shouldBe 1
+ StringUtilities.stringDistance( "ac", "bc") shouldBe 1
+ StringUtilities.stringDistance("abc", "axc") shouldBe 1
+ StringUtilities.stringDistance("xabxcdxxefxgx", "1ab2cd34ef5g6") shouldBe 6
+ }
+
+ it should "work where many operations are needed" in {
+ StringUtilities.stringDistance("example", "samples") shouldBe 3
+ StringUtilities.stringDistance("sturgeon", "urgently") shouldBe 6
+ StringUtilities.stringDistance("levenshtein", "frankenstein") shouldBe 6
+ StringUtilities.stringDistance("distance", "difference") shouldBe 5
+ StringUtilities.stringDistance("java was neat", "scala is great") shouldBe 7
+ }
+}