Merge pull request #1441 from henrykironde/testcite

Add citation lookup in Python interface

retriever/__main__.py

@@ -7,12 +7,12 @@
 
 from retriever.engines import engine_list, choose_engine
 from retriever.lib.datasets import datasets, dataset_names, license
-from retriever.lib.defaults import sample_script, CITATION, SCRIPT_SEARCH_PATHS, LICENSE
+from retriever.lib.defaults import sample_script, CITATION, SCRIPT_SEARCH_PATHS, LICENSE, VERSION
 from retriever.lib.engine_tools import reset_retriever
 from retriever.lib.get_opts import parser
 from retriever.lib.install import _install
 from retriever.lib.repository import check_for_updates
-from retriever.lib.scripts import SCRIPT_LIST, reload_scripts, get_script, name_matches
+from retriever.lib.scripts import SCRIPT_LIST, reload_scripts, get_script, name_matches, get_script_citation
 from retriever.lib.create_scripts import create_package
 from retriever.lib.provenance import commit, commit_log
 
@@ -66,12 +66,9 @@ def main():
                 print("\nCitation for retriever:\n")
                 print(CITATION)
             else:
-                scripts = name_matches(script_list, args.dataset)
-                for dataset in scripts:
-                    print("\nDataset:  {}".format(dataset.name))
-                    print("Citation:   {}".format(dataset.citation))
-                    print("Description:   {}\n".format(dataset.description))
-
+                citations = get_script_citation(args.dataset)
+                for citation in citations:
+                    print("Citation:   {}".format(citation))
             return
 
         if args.command == 'license':

retriever/lib/__init__.py

@@ -18,11 +18,12 @@
 from .scripts import reload_scripts
 from .scripts import get_script_upstream
 from .scripts import get_dataset_names_upstream
+from .scripts import get_script_citation
 
 __all__ = [
     'check_for_updates', 'commit', 'commit_log', 'create_package', 'datasets',
     'dataset_names', 'download', 'reload_scripts', 'reset_retriever', 'install_csv',
     'install_mysql', 'install_postgres', 'install_sqlite', 'install_msaccess',
     'install_json', 'install_xml', 'fetch', 'get_script_upstream',
-    'get_dataset_names_upstream'
+    'get_dataset_names_upstream', 'get_script_citation'
 ]

retriever/lib/scripts.py

@@ -279,6 +279,19 @@ def get_script_version_upstream(dataset, repo=REPOSITORY):
     return version
 
 
+def get_script_citation(dataset=None):
+    """Get the citation list for a script"""
+    if dataset is None:
+        return [VERSION]
+    scripts = name_matches(reload_scripts(), dataset)
+    if scripts:
+        citations = []
+        for script in scripts:
+            citations.append(script.citation)
+        return citations
+    return None
+
+
 def get_dataset_names_upstream(keywords=None, licenses=None, repo=REPOSITORY):
     """Search all datasets upstream by keywords and licenses. If the keywords
     or licenses argument is passed, Github's search API is used for looking

test/test_retriever.py

@@ -491,6 +491,12 @@ def test_format_insert_value_string_complex():
     assert test_engine.format_insert_value(test_str, 'char') == test_str
 
 
+def test_get_script_citation():
+    """Test get citation of a script"""
+    cite = rt.get_script_citation("iris")
+    expected_cite = "R. A. Fisher. 1936."
+    assert expected_cite.lower() in cite[0].lower()
+
 def test_getmd5_lines():
     """Test md5 sum calculation given a line."""
     lines = ['a,b,c', '1,2,3', '4,5,6']