From dc0b05286002963d97a957faad19fe1a5f102de2 Mon Sep 17 00:00:00 2001
From: Maxim Kolchin <kolchinmax@gmail.com>
Date: Sat, 4 Apr 2015 22:32:49 +0300
Subject: [PATCH] Extended test_Q1.10. Filed an issue #4

---
 .../parsers/publication_parser.py             |  2 +-
 .../tests/test_Q1.10/expected.output          | 35 +++++++++++++++++++
 .../CeurWsParser/tests/test_Q1.10/input.urls  | 22 ++++++++++++
 .../tests/test_Q1.10/query.sparql             | 17 ++++++++-
 4 files changed, 74 insertions(+), 2 deletions(-)
 create mode 100644 ceur-ws-crawler/CeurWsParser/tests/test_Q1.10/input.urls

diff --git a/ceur-ws-crawler/CeurWsParser/parsers/publication_parser.py b/ceur-ws-crawler/CeurWsParser/parsers/publication_parser.py
index 3b3c3da..e7a3853 100755
--- a/ceur-ws-crawler/CeurWsParser/parsers/publication_parser.py
+++ b/ceur-ws-crawler/CeurWsParser/parsers/publication_parser.py
@@ -39,7 +39,7 @@ def end_template(self):
 
     @staticmethod
     def is_invited(publication):
-        if rex.rex(publication['link'], r'.*(keynote|invite).*', re.I, default=None):
+        if rex.rex(publication['link'], r'.*(keynote|invite|talk|-inv-).*', re.I, default=None):
             return True
         else:
             return False
diff --git a/ceur-ws-crawler/CeurWsParser/tests/test_Q1.10/expected.output b/ceur-ws-crawler/CeurWsParser/tests/test_Q1.10/expected.output
index 3da89f2..e2ac5ec 100755
--- a/ceur-ws-crawler/CeurWsParser/tests/test_Q1.10/expected.output
+++ b/ceur-ws-crawler/CeurWsParser/tests/test_Q1.10/expected.output
@@ -3,8 +3,43 @@
 <http://ceur-ws.org/Vol-1081/>,"Mihai Codescu"
 <http://ceur-ws.org/Vol-1081/>,"Christoph Lange"
 <http://ceur-ws.org/Vol-1081/>,"George A. Vouros"
+<http://ceur-ws.org/Vol-1014/>,"Giuseppe De Giacomo"
+<http://ceur-ws.org/Vol-1014/>,"Michel Dumontier"
+<http://ceur-ws.org/Vol-1014/>,"Ian Pratt-Hartmann"
 <http://ceur-ws.org/Vol-1008/>,"Ethan Munson"
+<http://ceur-ws.org/Vol-994/>,"Peter Murray-Rust"
+<http://ceur-ws.org/Vol-937/>,"Hannes Mühleisen"
+<http://ceur-ws.org/Vol-937/>,"Christian Bizer"
+<http://ceur-ws.org/Vol-937/>,"Peter Mika"
+<http://ceur-ws.org/Vol-937/>,"Tim Potter"
+<http://ceur-ws.org/Vol-936/>,"Roberto Navigli"
 <http://ceur-ws.org/Vol-875/>,"Thomas Eiter"
 <http://ceur-ws.org/Vol-875/>,"Luciano Serafini"
+<http://ceur-ws.org/Vol-846/>,"Serge Abiteboul"
+<http://ceur-ws.org/Vol-846/>,"Piero Bonatti"
+<http://ceur-ws.org/Vol-846/>,"Alan Rector"
+<http://ceur-ws.org/Vol-745/>,"Marcelo Arenas"
+<http://ceur-ws.org/Vol-745/>,"Gert Smolka"
+<http://ceur-ws.org/Vol-745/>,"Heiner Stuckenschmidt"
+<http://ceur-ws.org/Vol-477/>,"Leonid Libkin"
+<http://ceur-ws.org/Vol-477/>,"Kavitha Srinivas"
+<http://ceur-ws.org/Vol-477/>,"Rajeev Goré"
+<http://ceur-ws.org/Vol-353/>,"Andrea Cali"
+<http://ceur-ws.org/Vol-353/>,"Georg Gottlob"
+<http://ceur-ws.org/Vol-353/>,"Michael Kifer"
+<http://ceur-ws.org/Vol-250/>,"Renee Miller"
+<http://ceur-ws.org/Vol-250/>,"Hector Levesque"
+<http://ceur-ws.org/Vol-250/>,"Alex Borgida"
+<http://ceur-ws.org/Vol-81/>,"John Mylopoulos"
+<http://ceur-ws.org/Vol-81/>,"Richard Hull"
+<http://ceur-ws.org/Vol-53/>,"Patrick Blackburn"
+<http://ceur-ws.org/Vol-53/>,"Andrei Voronkov"
+<http://ceur-ws.org/Vol-53/>,"Maurizio Lenzerini"
+<http://ceur-ws.org/Vol-33/>,"P. Devanbu"
+<http://ceur-ws.org/Vol-33/>,"C. Goble"
+<http://ceur-ws.org/Vol-33/>,"D. Harel"
+<http://ceur-ws.org/Vol-22/>,"Natasha Alechina"
+<http://ceur-ws.org/Vol-22/>,"Robert MacGregor"
+<http://ceur-ws.org/Vol-22/>,"Deborah McGuinness"
 <http://ceur-ws.org/Vol-1/>,"M. Lenzerini"
 <http://ceur-ws.org/Vol-1/>,"M. Scholl"
diff --git a/ceur-ws-crawler/CeurWsParser/tests/test_Q1.10/input.urls b/ceur-ws-crawler/CeurWsParser/tests/test_Q1.10/input.urls
new file mode 100644
index 0000000..0f18b3f
--- /dev/null
+++ b/ceur-ws-crawler/CeurWsParser/tests/test_Q1.10/input.urls
@@ -0,0 +1,22 @@
+input_urls = ["http://ceur-ws.org/",
+"http://ceur-ws.org/Vol-1085/",
+"http://ceur-ws.org/Vol-1081/",
+"http://ceur-ws.org/Vol-1014/",
+"http://ceur-ws.org/Vol-1008/",
+"http://ceur-ws.org/Vol-994/",
+"http://ceur-ws.org/Vol-937/",
+"http://ceur-ws.org/Vol-936/",
+"http://ceur-ws.org/Vol-875/",
+"http://ceur-ws.org/Vol-859/",
+"http://ceur-ws.org/Vol-846/",
+"http://ceur-ws.org/Vol-745/",
+"http://ceur-ws.org/Vol-573/",
+"http://ceur-ws.org/Vol-477/",
+"http://ceur-ws.org/Vol-353/",
+"http://ceur-ws.org/Vol-250/",
+"http://ceur-ws.org/Vol-81/",
+"http://ceur-ws.org/Vol-53/",
+"http://ceur-ws.org/Vol-33/",
+"http://ceur-ws.org/Vol-22/",
+"http://ceur-ws.org/Vol-1/"
+]
diff --git a/ceur-ws-crawler/CeurWsParser/tests/test_Q1.10/query.sparql b/ceur-ws-crawler/CeurWsParser/tests/test_Q1.10/query.sparql
index 1144808..3979fbd 100755
--- a/ceur-ws-crawler/CeurWsParser/tests/test_Q1.10/query.sparql
+++ b/ceur-ws-crawler/CeurWsParser/tests/test_Q1.10/query.sparql
@@ -2,12 +2,27 @@ SELECT ?proc ?author_name {
   VALUES ?proc {
     <http://ceur-ws.org/Vol-1085/>
     <http://ceur-ws.org/Vol-1081/>
+    <http://ceur-ws.org/Vol-1014/>
     <http://ceur-ws.org/Vol-1008/>
+    <http://ceur-ws.org/Vol-994/>
+    <http://ceur-ws.org/Vol-937/>
+    <http://ceur-ws.org/Vol-936/>
     <http://ceur-ws.org/Vol-875/>
+    <http://ceur-ws.org/Vol-859/>
+    <http://ceur-ws.org/Vol-846/>
+    <http://ceur-ws.org/Vol-745/>
+    <http://ceur-ws.org/Vol-573/>
+    <http://ceur-ws.org/Vol-477/>
+    <http://ceur-ws.org/Vol-353/>
+    <http://ceur-ws.org/Vol-250/>
+    <http://ceur-ws.org/Vol-81/>
+    <http://ceur-ws.org/Vol-53/>
+    <http://ceur-ws.org/Vol-33/>
+    <http://ceur-ws.org/Vol-22/>
     <http://ceur-ws.org/Vol-1/>
   }
   ?proc dcterms:hasPart ?pub .
   ?pub a swc:InvitedPaper .
-  ?author dc:creator ?pub ;
+  ?author foaf:made ?pub ;
           foaf:name ?author_name .
 }