pybliometrics-dev#360 ScienceDirect Object Retrieval with eid

pybliometrics/sciencedirect/__init__.py

@@ -3,4 +3,5 @@
 from pybliometrics.sciencedirect.article_retrieval import *
 from pybliometrics.sciencedirect.article_metadata import *
 from pybliometrics.sciencedirect.object_metadata import *
+from pybliometrics.sciencedirect.object_retrieval import *
 from pybliometrics.sciencedirect.sciencedirect_search import *

pybliometrics/sciencedirect/article_metadata.py

@@ -44,8 +44,10 @@ def results(self) -> Optional[List[NamedTuple]]:
             authors_list = [author.get('$') for author in chained_get(item, ['authors', 'author'], [])]
             authors_list = deduplicate(authors_list)
             authors = ';'.join(authors_list)
-            first_author = item.get('dc:creator')[0].get('$')
-            link = item.get('link')[0].get('@href')
+            first_author = item.get('dc:creator')
+            first_author = first_author[0].get('$') if first_author else None
+            link = item.get('link')
+            link = link[0].get('@href') if link else None
             doi = item.get("prism:doi") or item.get("dc:identifier")[4:] if item.get("dc:identifier") else None
             new = doc(authorKeywords=item.get('authkeywords'),
                     authors=authors,

pybliometrics/sciencedirect/object_retrieval.py

@@ -0,0 +1,56 @@
+from io import BytesIO
+from typing import Literal, Optional, Union
+
+from pybliometrics.sciencedirect import ArticleRetrieval
+from pybliometrics.superclasses import Retrieval
+from pybliometrics.utils import (
+    chained_get,
+    check_parameter_value,
+    detect_id_type,
+    VIEWS,
+)
+
+class ObjectRetrieval(Retrieval):
+    @property
+    def object(self) -> BytesIO:
+        """The object retrieved."""
+        return self._object
+
+    def __init__(self,
+                 identifier: Union[int, str],
+                 filename: str,
+                 id_type: Optional[str] = None,
+                 refresh: Union[bool, int] = False,
+                 **kwds: str
+                 ):
+        """Class to retrieve a specific object of a document.
+
+        :param identifier: The indentifier of the document.
+        :param filename: Filename of the object to be retrieved. To get a list of all available
+            objects of a document (and its corresponding filename) use the class `ObjectMetadata`.
+        :param id_type: The type of identifier supplied. Allowed values: doi, pii, scopus_id, pubmed_id, eid.
+        :param refresh: Whether to refresh the cached file if it exists. Default: False.
+        """
+        identifier = str(identifier)
+
+        if id_type is None:
+            id_type = detect_id_type(identifier)
+        else:
+            allowed_id_types = ('doi', 'pii', 'scopus_id', 'pubmed_id', 'eid')
+            check_parameter_value(id_type, allowed_id_types, "id_type")
+
+        if id_type != 'eid':
+            identifier = self._get_eid(identifier)
+        file_identifier = f'{identifier}-{filename}'
+
+        self._view = ''
+        self._refresh = refresh
+
+        super().__init__(file_identifier, 'ObjectRetrieval', 'eid', **kwds)
+
+        self._object = BytesIO(self._object)
+
+    def _get_eid(self, identifier: str) -> str:
+        """Get the EID of a document."""
+        am = ArticleRetrieval(identifier, field='eid')
+        return am.eid

pybliometrics/sciencedirect/tests/test_ObjectRetrieval.py

@@ -0,0 +1,38 @@
+"""Test class ObjectRetrieval"""
+import xml.etree.ElementTree as ET
+
+from io import BytesIO
+from PIL import Image
+
+from pybliometrics.sciencedirect import init, ObjectRetrieval
+
+init()
+
+or_1 = ObjectRetrieval('S156984322300331X',
+                       'gr10.jpg',
+                       refresh=30)
+or_2 = ObjectRetrieval('10.1016/j.rcim.2020.102086',
+                       'si92.svg',
+                       id_type='doi',
+                       refresh=30)
+
+
+def test_object():
+    """Tests whether the object is a BytesIO object."""
+    assert isinstance(or_1.object, BytesIO)
+    assert isinstance(or_2.object, BytesIO)
+
+
+def test_is_jpg():
+    """Tests whether the object is a JPEG image."""
+    obj_1 = or_1.object
+    with Image.open(obj_1) as img:
+        assert img.format.lower() == 'jpeg'
+
+
+def test_is_svg():
+    """Tests whether the object is an SVG image."""
+    obj_2 = or_2.object
+    tree = ET.parse(obj_2)
+    root = tree.getroot()
+    assert root.tag == '{http://www.w3.org/2000/svg}svg'

pybliometrics/superclasses/base.py

@@ -53,13 +53,17 @@ def __init__(self,
         search_request = "query" in params
         # Check if ref retrieval for abstract
         ab_ref_retrieval = (api == 'AbstractRetrieval') and (params['view'] == 'REF')
+        # Check if object retrieval
+        obj_retrieval = (api == 'ObjectRetrieval')
 
         if fname.exists() and not self._refresh:
             self._mdate = mod_ts
             if search_request:
                 self._json = [loads(line) for line in
                               fname.read_text().split("\n") if line]
                 self._n = len(self._json)
+            elif obj_retrieval:
+                self._object = fname.read_bytes()
             else:
                 self._json = loads(fname.read_text())
         else:
@@ -111,6 +115,9 @@ def __init__(self,
                     self._json = data
                 else:
                     data = None
+            elif obj_retrieval:
+                self._object = resp.content
+                data = []
             else:
                 data = loads(resp.text)
                 self._json = data
@@ -120,8 +127,11 @@ def __init__(self,
             self._header = header
             # Finally write data unless download=False
             if download:
-                text = [dumps(item, separators=(',', ':')) for item in data]
-                fname.write_text("\n".join(text))
+                if obj_retrieval:
+                    fname.write_bytes(self._object)
+                else:
+                    text = [dumps(item, separators=(',', ':')) for item in data]
+                    fname.write_text("\n".join(text))
 
     def get_cache_file_age(self) -> int:
         """Return the age of the cached file in days."""

pybliometrics/superclasses/retrieval.py

@@ -33,7 +33,7 @@ def __init__(self,
         """
         # Construct URL and cache file name
         url = URLS[api]
-        if api in ("AbstractRetrieval", "PlumXMetrics", "ArticleRetrieval", "ObjectMetadata"):
+        if api in ("AbstractRetrieval", "PlumXMetrics", "ArticleRetrieval", "ObjectMetadata", "ObjectRetrieval"):
             url += id_type + "/"
         if api == 'CitationOverview':
             stem = identifier.replace("/", "")

pybliometrics/utils/constants.py

@@ -27,6 +27,7 @@
     'ArticleMetadata': BASE_PATH_SCIENCEDIRECT/'article_metadata/',
     'ArticleRetrieval': BASE_PATH_SCIENCEDIRECT/'article_retrieval',
     'ObjectMetadata': BASE_PATH_SCIENCEDIRECT/'object_metadata',
+    'ObjectRetrieval': BASE_PATH_SCIENCEDIRECT/'object_retrieval',
     'ScienceDirectSearch': BASE_PATH_SCIENCEDIRECT/'science_direct_search'
 }
 
@@ -59,6 +60,7 @@
     'ArticleMetadata': RETRIEVAL_BASE + 'metadata/article/',
     'ArticleRetrieval': RETRIEVAL_BASE + 'article/',
     'ObjectMetadata': RETRIEVAL_BASE + 'object/',
+    'ObjectRetrieval': RETRIEVAL_BASE + 'object/',
     'ScienceDirectSearch': SEARCH_BASE + 'sciencedirect/'
 }
 
@@ -78,6 +80,7 @@
     "ArticleRetrieval": ["META", "META_ABS", "META_ABS_REF", "FULL", "ENTITLED"],
     "ArticleMetadata": ["STANDARD", "COMPLETE"],
     "ObjectMetadata": ["META"],
+    "ObjectRetrieval": [""],
     "ScienceDirectSearch": ["STANDARD"]
 }
 
@@ -97,6 +100,7 @@
     'ArticleMetadata': 6,
     'ArticleRetrieval': 10,
     'ObjectMetadata': 0,
+    'ObjectRetrieval': 0,
     'ScienceDirectSearch': 2
 }