Merge branch '84-fix-insert-marcxml' into 'master'

Resolve "Fail to insert MarcXML"

Closes #84

See merge request !85

controllers/harvest.py

@@ -322,6 +322,7 @@ def insert_marcxml():
         if not tool:
             return inline_alert(T("Error"), T("Select an harvester."))
 
+        tool.harvester.host = selector.host
         tool.process_xml(selector.xml)
 
     except ToolException as e:

modules/harvest_tools/checkandfix.py

@@ -5,7 +5,7 @@
 import re
 import regex
 
-from base import search_synonym, ToolException
+from base import format_author_fr, search_synonym, ToolException
 from exception import CheckException
 from gluon import current
 from invenio_tools import (DECODE_REF,
@@ -18,6 +18,7 @@ from invenio_tools import (DECODE_REF,
                            REG_OAI,
                            REG_YEAR)
 from itertools import imap
+from pandas import DataFrame
 from plugin_dbui import CLEAN_SPACES, get_id
 
 
@@ -679,7 +680,11 @@ class CheckAndFix(object):
         if not isinstance(record, RecordThesis):
             raise CheckException(MSG_NO_THESIS)
 
-    def my_affiliation(self, record, id_project, id_team):
+    def my_affiliation(self,
+                       record,
+                       id_project,
+                       id_team,
+                       func=format_author_fr):
         """Check that authors of my institute are signatories.
 
         Launch a recovery procedure when affiliations are not defined.
@@ -689,6 +694,7 @@ class CheckAndFix(object):
             record (RecordPubli): record describing a publication.
             id_project (int): identifier of the project in the database
             id_team (int): identifier of the team in the database
+            func (reference): function used to format the author names.
 
         Raises:
             CheckException: when there is no authors from my institute.
@@ -704,16 +710,21 @@ class CheckAndFix(object):
         if not rescue_list:
             raise CheckException(MSG_NO_MY_AUTHOR)
 
+        # format the author in the same way as the rescue list
         # compute the intersection between the authors and the rescue list
-        set_1 = set(record.authors_as_list())
-        set_2 = set(rescue_list)
+        df = (DataFrame(record.authors_as_list(), columns=["raw_author"])
+              .assign(format_author=lambda x:
+                      x.raw_author.apply(lambda y: func(y)))
+              .set_index("format_author"))
+
+        rescue_list = [el.decode("utf-8") for el in rescue_list]
+        intersection = df.index & rescue_list
 
-        li = list(set_1.intersection(set_2))
-        if not li:
+        if intersection.size == 0:
             raise CheckException(MSG_NO_MY_AUTHOR)
 
         # cache the result for a latter use
-        self.__my_authors[record.id()] = li
+        self.__my_authors[record.id()] = intersection.values.tolist()
 
     def paper_reference(self, record):
         """Check that editor, page, volume and paper year are defined

modules/model_selector.py

@@ -363,12 +363,14 @@ class Selector(object):
             "marc12_selector",
             Field("id_teams", "reference teams", label="Team"),
             Field("id_projects", "reference projects", label="Project"),
+            Field("host", "string", default=STORES[0], label="Store"),
             Field("xml", "text", comment=tp_xml, label="MARCXML"),
             Field("controller", "string", label="Automaton"),
             Field("id_categories", "reference categories", label="Category"),
             Field("mode", "string", default=mode_dry_run))
 
         table.controller.requires = IS_IN_SET(CONTROLLERS)
+        table.host.requires = IS_IN_SET(STORES)
         table.id_categories.requires = IS_IN_DB(db, "categories.code")
         table.id_projects.requires = IS_IN_DB(db, "projects.project")
         table.id_teams.requires = IS_IN_DB(db, "teams.team")