Update to move df_info as attribute of the record

modules/store_tools/pluginpublicationinfo.py

@@ -10,23 +10,26 @@ PAPER_REFERENCE_KEYS = {"pagination", "title", "volume", "year"}
 class PluginPublicationInfo(object):
     """Plugin to handle publication_info in RecordPubli and RecordHepPubli
 
-    publication information are stored in DataFrame with the
-    following structure:
-
-        +------------+--------------------------------+
-        | column     |                                |
-        +============+================================+
-        | title      | abbreviation of the publisher  |
-        +------------+--------------------------------+
-        | volume     | volume                         |
-        +------------+--------------------------------+
-        | year       | year of publication            |
-        +------------+--------------------------------+
-        | pagination | page number or ranges          |
-        +------------+--------------------------------+
+    Attributes:
+        df_info (pandas.DataFrame):
+
+            publication information are stored in DataFrame with the
+            following structure:
+
+                +------------+--------------------------------+
+                | column     |                                |
+                +============+================================+
+                | title      | abbreviation of the publisher  |
+                +------------+--------------------------------+
+                | volume     | volume                         |
+                +------------+--------------------------------+
+                | year       | year of publication            |
+                +------------+--------------------------------+
+                | pagination | page number or ranges          |
+                +------------+--------------------------------+
 
     The DataFrame has to be build in the parent
-    It is retrieved via self["df_info"]
+    It is retrieved via self.df_info
 
     """
 
@@ -46,7 +49,7 @@ class PluginPublicationInfo(object):
         # * the row contains empty string when the record is not published.
         # * iloc[0] returns a serie where the index are the column's name
         #
-        columns = (self["df_info"].iloc[0]
+        columns = (self.df_info.iloc[0]
                    .replace("", np.nan)
                    .dropna()
                    .index)
@@ -60,7 +63,7 @@ class PluginPublicationInfo(object):
             bool
 
         """
-        df = self["df_info"]
+        df = self.df_info
         return len(df) > 1
 
     def paper_editor(self):
@@ -71,7 +74,7 @@ class PluginPublicationInfo(object):
                 empty string when not defined.
 
         """
-        df = self["df_info"]
+        df = self.df_info
         editor = (df.title.iloc[0] if "title" in df else "")
 
         # add space after the dot  Phys.Rev -> Phys. Rev
@@ -88,7 +91,7 @@ class PluginPublicationInfo(object):
                 * Empty string when not defined.
 
         """
-        df = self["df_info"]
+        df = self.df_info
         return (df.pagination.iloc[0] if "pagination" in df else "")
 
     def paper_reference(self):
@@ -101,7 +104,7 @@ class PluginPublicationInfo(object):
                   published in a review.
 
         """
-        df = self["df_info"]
+        df = self.df_info
 
         its = df.columns.intersection({"title",
                                        "volume",
@@ -128,7 +131,7 @@ class PluginPublicationInfo(object):
                 empty string when nothing is found.
 
         """
-        df = self["df_info"]
+        df = self.df_info
         return (df.volume.iloc[0] if "volume" in df else "")
 
     def paper_year(self):
@@ -139,5 +142,5 @@ class PluginPublicationInfo(object):
                 - Empty string if the year is not defined.
 
         """
-        df = self["df_info"]
+        df = self.df_info
         return (df.year.iloc[0] if "year" in df else "")

modules/store_tools/recordhepconf.py

@@ -11,6 +11,11 @@ from .recordheppubli import RecordHepPubli
 class RecordHepConf(RecordHepPubli):
     """Conference proceeding from inspirehep.net version 2.
 
+    Attributes:
+
+        conference (dict):
+            metadata of the conference
+            schema is https://inspirehep.net/s…records/conferences.json
 
     """
 

modules/store_tools/recordheppubli.py

@@ -129,7 +129,7 @@ class RecordHepPubli(RecordHep, PluginAuthors, PluginPublicationInfo):
                     "year",
                     "pagination"]
 
-            self["df_info"] = DataFrame([[""] * len(cols)], columns=cols)
+            self.df_info = DataFrame([[""] * len(cols)], columns=cols)
 
             return
 
@@ -149,7 +149,7 @@ class RecordHepPubli(RecordHep, PluginAuthors, PluginPublicationInfo):
             df = df.sort_values("year")
 
         # replace
-        self["df_info"] = df
+        self.df_info = df
 
     def collaboration(self):
         """The collaboration(s) signing the publication.

modules/store_tools/recordpubli.py

@@ -236,7 +236,7 @@ class RecordPubli(Record, PluginAuthors, PluginPublicationInfo):
                     "year",
                     "pagination"]
 
-            self["df_info"] = DataFrame([[""] * len(cols)], columns=cols)
+            self.df_info = DataFrame([[""] * len(cols)], columns=cols)
 
             return
 
@@ -261,7 +261,7 @@ class RecordPubli(Record, PluginAuthors, PluginPublicationInfo):
             df = df.sort_values("year")
 
         # replace
-        self["df_info"] = df
+        self.df_info = df
 
     def collaboration(self):
         """The collaboration(s) signing the publication.

tests/basis/test_04_RecordPubli.py

@@ -52,7 +52,7 @@ def test_constructor_cds_04002(record):
     assert len(authors) == record["number_of_authors"]
     assert authors.affiliation.iloc[12] == "INFN, Rome|CERN"
 
-    papers = record["df_info"]
+    papers = record.df_info
 
     assert isinstance(papers, pd.DataFrame)
     assert len(papers) == 1

tests/basis/test_05_RecordHepPubli.py

@@ -52,7 +52,7 @@ def test_constructor_ins_05002(record):
     assert len(authors) == 704
     assert authors.affiliation.iloc[12] == "INFN, Rome|CERN"
 
-    papers = record["df_info"]
+    papers = record.df_info
 
     assert isinstance(papers, pd.DataFrame)
     assert len(papers) == 1

tests/basis/test_13_CheckAndFix_article.py

@@ -66,7 +66,7 @@ def test_paper_reference_cds_13004(svc, reccds):
     # remove the publisher and volume information
     paper_ref = reccds.paper_reference()
 
-    reccds["df_info"].loc[0, ["title", "volume"]] = ["", ""]
+    reccds.df_info.loc[0, ["title", "volume"]] = ["", ""]
     svc.paper_reference(reccds)
 
     assert reccds.paper_reference() == paper_ref