preprints.py 4.32 KB
Newer Older
1 2 3 4 5 6 7
# -*- coding: utf-8 -*-
""" harvest_tools.preprints

"""
import traceback


8
from automaton import Automaton
9
from base import family_name_fr, MSG_CRASH, MSG_LOAD
10
from invenio_tools import CheckException, RecordConf, RecordThesis
11 12 13
from plugin_dbui import UNDEF_ID


14 15 16 17
MSG_PREPRINT_IS_PAPER = "Reject preprint is a published paper"
MSG_PREPRINT_IS_CONFERENCE = "Reject preprint is a conference"
MSG_PREPRINT_IS_THESIS = "Reject preprint is a thesis"
MSG_PREPRINT_NO_NUMBER = "Reject no preprint number"
18 19


20
class Preprints(Automaton):
21
    """Automaton for preprints.
22 23

    """
24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75
    def check_record(self, record):
        """Check the content of the preprint in order to fix non conformities.

        @type record: L{Record}
        @param record:

        @rtype: bool
        @return: C{False} when a non conformity is found and can not be
        corrected.

        """
        if not Automaton.check_record(self, record):
            return False

        if self.dbg:
            print "select preprint record"

        if record.is_published():
            self.logs[-1].reject(MSG_PREPRINT_IS_PAPER, record.year())
            return False

        if isinstance(record, RecordConf):
            self.logs[-1].reject(MSG_PREPRINT_IS_CONFERENCE, record.year())
            return False

        if isinstance(record, RecordThesis):
            self.logs[-1].reject(MSG_PREPRINT_IS_THESIS, record.year())
            return False

        if not record.preprint_number():
            self.logs[-1].reject(MSG_PREPRINT_NO_NUMBER, record.year())
            return False

        try:
            self.check.my_authors(record,
                                  reference=self._my_author_list(record),
                                  cmpFct=family_name_fr)

            self.check.oai(record)
            self.check.submitted(record)
            self.check.year(record)

        except CheckException as e:
            self.logs[-1].reject(e, record.year())
            return False

        except Exception as e:
            self.logs[-1].reject(MSG_CRASH % e, record.year(), translate=False)
            print traceback.format_exc()
            return False

        return True
76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132

    def load_db(self, record):
        """Load a preprint in the database.

        @type record: L{Record}
        @param record:

        @rtype: int
        @return: one when the record is inserted / updated in the database
        zero otherwise.

        """
        db = self.db

        # alias
        first_author = record.first_author()
        oai_url = record.oai_url()
        preprint = record.preprint_number()
        title = record.title()
        submitted = record.submitted()[0]
        year = record.year()

        # check the collaboration
        id_collaboration = self.check_collaboration(record.collaboration())

        # check against preprint or article already published
        rec_id, status = self.check_by_fields(first_author=first_author,
                                              id_projects=self.id_project,
                                              id_teams=self.id_team,
                                              oai_url=oai_url,
                                              preprint=preprint,
                                              submitted=submitted,
                                              title=title,
                                              year=year)
        if rec_id:
            return status

        # eventually insert a new preprint
        if not self.dry_run:

            db.publications.insert(authors=record.authors(),
                                   authors_institute=record.my_authors,
                                   first_author=first_author,
                                   id_categories=self.id_category,
                                   id_collaborations=id_collaboration,
                                   id_projects=self.id_project,
                                   id_status=UNDEF_ID,
                                   id_teams=self.id_team,
                                   origin=oai_url,
                                   preprint=preprint,
                                   publication_url=record.paper_url(),
                                   submitted=submitted,
                                   title=title,
                                   year=year)

        self.logs[-1].load(MSG_LOAD, year)
        return 1