Added abstract retrieval

articledownloader/articledownloader.py

@@ -81,7 +81,7 @@ def get_pdf_from_doi(self, doi, writefile, mode):
     :param mode: either 'crossref' or 'elsevier', depending on how we wish to access the file 
     :type mode: str 
 
-    :returns: True on succesful write, False otherwise
+    :returns: True on successful write, False otherwise
     :rtype: bool
     '''
 
@@ -105,9 +105,6 @@ def get_pdf_from_doi(self, doi, writefile, mode):
     if mode == 'elsevier':
       if self.check_els_entitlement(doi):
         try:
-          name = re.sub('[\(\)]', '', doi)
-          name = re.sub('\s+', '', name)
-
           pdf_url='http://api.elsevier.com/content/article/doi:' + doi + '?view=FULL'
           self.headers['Accept'] = 'application/pdf'
 
@@ -122,6 +119,36 @@ def get_pdf_from_doi(self, doi, writefile, mode):
 
         return False
 
+  def get_abstract_from_doi(self, doi, mode):
+    '''
+    Returns abstract as a unicode string given a DOI
+
+    :param doi: DOI string for the article we want to grab metadata for 
+    :type doi: str 
+
+    :param mode: Only supports 'elsevier' for now
+    :type mode: str 
+
+    :returns: An abstract (or None on failure)
+    :rtype: unicode
+    '''
+
+    if mode == 'elsevier':
+      if self.check_els_entitlement(doi):
+        try:
+          url='http://api.elsevier.com/content/article/doi:' + doi + '?view=FULL'
+          self.headers['Accept'] = 'application/json'
+
+          r = requests.get(url, headers=self.headers)
+          if r.status_code == 200:
+            abstract = unicode(json.loads(r.text)['full-text-retrieval-response']['coredata']['dc:description'])
+            return abstract
+        except requests.exceptions.ConnectionError:
+          # API download limit exceeded
+          return None
+
+        return None
+
   def load_queries_from_csv(self, csvf):
     '''
     Loads a list of queries from a CSV file 

articledownloader/tests/test.py

@@ -5,7 +5,7 @@
 
 class Tester(TestCase):
   def setUp(self):
-    self.downloader = ArticleDownloader('NO_API_KEY')
+    self.downloader = ArticleDownloader(environ.get('ELS_API_KEY'))
     self.doi = '10.1016/j.nantod.2008.10.014'
     self.pdf_file = TemporaryFile(mode='wb')
 
@@ -21,9 +21,12 @@ def test_download(self):
     self.downloader.get_pdf_from_doi(self.doi, self.pdf_file, 'elsevier')
     self.downloader.get_pdf_from_doi(self.doi, self.pdf_file, 'crossref')
 
+  def test_abstract_download(self):
+    self.downloader.get_abstract_from_doi(self.doi, 'elsevier')
+
   def test_entitlement(self):
     #Test entitlement
-    self.assertFalse(self.downloader.check_els_entitlement(self.doi))
+    self.assertTrue(self.downloader.check_els_entitlement(self.doi))
 
   def test_search(self):
     #Search test

setup.py

@@ -2,11 +2,11 @@
 setup(
 name = 'articledownloader',
 packages = ['articledownloader'], # this must be the same as the name above
-version = '2.2',
+version = '2.3',
 description = 'A class for downloading scientific journal articles',
 author = 'Edward Kim',
 author_email = 'eddotman@gmail.com',
 url = 'https://github.com/eddotman/article-downloader', # use the URL to the github repo
-download_url = 'https://www.github.com/eddotman/article-downloader/tarball/2.2',
+download_url = 'https://www.github.com/eddotman/article-downloader/tarball/2.3',
 keywords = ['journal', 'paper', 'article', 'downloader'], # arbitrary keywords
 )