Add academic analytics feed

carbon/__init__.py

@@ -8,4 +8,4 @@
 
 __version__ = '0.2.0'
 
-from .app import people, person_feed
+from .app import people, person_feed, articles, article_feed

carbon/app.py

@@ -8,7 +8,7 @@
 from lxml import etree as ET
 from sqlalchemy import func, select
 
-from carbon.db import persons, orcids, dlcs, engine
+from carbon.db import persons, orcids, dlcs, engine, aa_articles
 
 AREAS = (
     'ARCHITECTURE & PLANNING AREA', 'ENGINEERING AREA',
@@ -66,6 +66,17 @@ def people():
             yield dict(zip(row.keys(), row))
 
 
+def articles():
+    """An article generator.
+
+    Returns an iterator over the AA_ARTICLE table.
+    """
+    sql = select([aa_articles])
+    with closing(engine().connect()) as conn:
+        for row in conn.execute(sql):
+            yield dict(zip(row.keys(), row))
+
+
 def initials(*args):
     """Turn `*args` into a space-separated string of initials.
 
@@ -134,6 +145,37 @@ def person_feed(out):
             yield partial(_add_person, xf)
 
 
+@contextmanager
+def article_feed(out):
+    """Generate XML feed of articles."""
+    with ET.xmlfile(out, encoding='UTF-8') as xf:
+        xf.write_declaration()
+        with xf.element('articles'):
+            yield partial(_add_article, xf)
+
+
+def _add_article(xf, article):
+    record = ET.Element('article')
+    add_child(record, 'AA_MATCH_SCORE', str(article['AA_MATCH_SCORE']))
+    add_child(record, 'ARTICLE_ID', article['ARTICLE_ID'])
+    add_child(record, 'ARTICLE_TITLE', article['ARTICLE_TITLE'])
+    add_child(record, 'ARTICLE_YEAR', article['ARTICLE_YEAR'])
+    add_child(record, 'AUTHORS', article['AUTHORS'])
+    add_child(record, 'DOI', article['DOI'])
+    add_child(record, 'ISSN_ELECTRONIC', article['ISSN_ELECTRONIC'])
+    add_child(record, 'ISSN_PRINT', article['ISSN_PRINT'])
+    add_child(record, 'IS_CONFERENCE_PROCEEDING',
+              article['IS_CONFERENCE_PROCEEDING'])
+    add_child(record, 'JOURNAL_FIRST_PAGE', article['JOURNAL_FIRST_PAGE'])
+    add_child(record, 'JOURNAL_LAST_PAGE', article['JOURNAL_LAST_PAGE'])
+    add_child(record, 'JOURNAL_ISSUE', article['JOURNAL_ISSUE'])
+    add_child(record, 'JOURNAL_VOLUME', article['JOURNAL_VOLUME'])
+    add_child(record, 'JOURNAL_NAME', article['JOURNAL_NAME'])
+    add_child(record, 'MIT_ID', article['MIT_ID'])
+    add_child(record, 'PUBLISHER', article['PUBLISHER'])
+    xf.write(record)
+
+
 def _add_person(xf, person):
     record = ET.Element('record')
     add_child(record, 'field', person['MIT_ID'], name='[Proprietary_ID]')

carbon/cli.py

@@ -3,7 +3,7 @@
 
 import click
 
-from carbon import people, person_feed
+from carbon import people, person_feed, articles, article_feed
 from carbon.db import engine
 
 
@@ -23,10 +23,14 @@ def main(db, feed_type, out):
 
     The feed will be printed to stdout if OUT is not specified.
 
-    FEED_TYPE should be 'people'.
+    FEED_TYPE should be 'people' or 'articles'.
     """
     engine.configure(db)
     if feed_type == 'people':
         with person_feed(out) as f:
             for person in people():
                 f(person)
+    elif feed_type == 'articles':
+        with article_feed(out) as f:
+            for article in articles():
+                f(article)

carbon/db.py

@@ -2,7 +2,7 @@
 from __future__ import absolute_import
 
 from sqlalchemy import (create_engine, Table, Column, String, Date, MetaData,
-                        ForeignKey)
+                        ForeignKey, Numeric, Text)
 
 
 metadata = MetaData()
@@ -36,6 +36,26 @@
                Column('ORCID', String))
 
 
+aa_articles = Table('AA_ARTICLE', metadata,
+                    Column('AA_MATCH_SCORE', Numeric(3, 1)),
+                    Column('ARTICLE_ID', String),
+                    Column('ARTICLE_TITLE', String),
+                    Column('ARTICLE_YEAR', String),
+                    Column('AUTHORS', Text),
+                    Column('DOI', String),
+                    Column('ISSN_ELECTRONIC', String),
+                    Column('ISSN_PRINT', String),
+                    Column('IS_CONFERENCE_PROCEEDING', String),
+                    Column('JOURNAL_FIRST_PAGE', String),
+                    Column('JOURNAL_LAST_PAGE', String),
+                    Column('JOURNAL_ISSUE', String),
+                    Column('JOURNAL_NAME', String),
+                    Column('JOURNAL_VOLUME', String),
+                    Column('MIT_ID', String),
+                    Column('PUBLISHER', String),
+                    )
+
+
 class Engine(object):
     """Database engine.
 

tests/conftest.py

@@ -3,11 +3,11 @@
 from contextlib import closing
 import os
 
-from lxml.builder import ElementMaker
+from lxml.builder import ElementMaker, E as B
 import pytest
 import yaml
 
-from carbon.db import engine, metadata, persons, orcids, dlcs
+from carbon.db import engine, metadata, persons, orcids, dlcs, aa_articles
 
 
 @pytest.fixture(scope="session", autouse=True)
@@ -26,21 +26,33 @@ def records():
     return r
 
 
+@pytest.fixture(scope="session")
+def aa_data():
+    current_dir = os.path.dirname(os.path.realpath(__file__))
+    data = os.path.join(current_dir, 'fixtures/articles.yml')
+    with open(data) as fp:
+        r = list(yaml.load_all(fp))
+    return r
+
+
 @pytest.yield_fixture
-def load_data(records):
+def load_data(records, aa_data):
     with closing(engine().connect()) as conn:
         conn.execute(persons.delete())
         conn.execute(orcids.delete())
         conn.execute(dlcs.delete())
+        conn.execute(aa_articles.delete())
         for r in records:
             conn.execute(persons.insert(), r['person'])
             conn.execute(orcids.insert(), r['orcid'])
             conn.execute(dlcs.insert(), r['dlc'])
+        conn.execute(aa_articles.insert(), aa_data)
     yield
     with closing(engine().connect()) as conn:
         conn.execute(persons.delete())
         conn.execute(orcids.delete())
         conn.execute(dlcs.delete())
+        conn.execute(aa_articles.delete())
 
 
 @pytest.fixture
@@ -90,3 +102,27 @@ def xml_data(E, xml_records):
 def E():
     return ElementMaker(namespace='http://www.symplectic.co.uk/hrimporter',
                         nsmap={None: 'http://www.symplectic.co.uk/hrimporter'})
+
+
+@pytest.fixture
+def articles_data(aa_data):
+    return B.articles(
+        B.article(
+            B.AA_MATCH_SCORE('0.9'),
+            B.ARTICLE_ID('1234567'),
+            B.ARTICLE_TITLE('Interaction between hatsopoulos microfluids and the Yawning Abyss of Chaos.'),
+            B.ARTICLE_YEAR('1999'),
+            B.AUTHORS('McRandallson, Randall M.|Lord, Dark'),
+            B.DOI('10.0000/1234LETTERS56'),
+            B.ISSN_ELECTRONIC('0987654'),
+            B.ISSN_PRINT('01234567'),
+            B.IS_CONFERENCE_PROCEEDING('0'),
+            B.JOURNAL_FIRST_PAGE('666'),
+            B.JOURNAL_LAST_PAGE('666'),
+            B.JOURNAL_ISSUE('10'),
+            B.JOURNAL_VOLUME('1'),
+            B.JOURNAL_NAME('Bunnies'),
+            B.MIT_ID('123456789'),
+            B.PUBLISHER('MIT Press')
+            )
+        )

tests/fixtures/articles.yml

@@ -0,0 +1,17 @@
+---
+AA_MATCH_SCORE: 0.9
+ARTICLE_ID: '1234567'
+ARTICLE_TITLE: Interaction between hatsopoulos microfluids and the Yawning Abyss of Chaos.
+ARTICLE_YEAR: '1999'
+AUTHORS: McRandallson, Randall M.|Lord, Dark
+DOI: 10.0000/1234LETTERS56
+ISSN_ELECTRONIC: '0987654'
+ISSN_PRINT: '01234567'
+IS_CONFERENCE_PROCEEDING: '0'
+JOURNAL_FIRST_PAGE: '666'
+JOURNAL_LAST_PAGE: '666'
+JOURNAL_ISSUE: '10'
+JOURNAL_VOLUME: '1'
+JOURNAL_NAME: Bunnies
+MIT_ID: '123456789'
+PUBLISHER: MIT Press

tests/test_app.py

@@ -5,8 +5,9 @@
 from lxml import etree as ET
 import pytest
 
-from carbon import people
-from carbon.app import person_feed, ns, NSMAP, add_child, initials
+from carbon import people, articles
+from carbon.app import (person_feed, ns, NSMAP, add_child, initials,
+                        article_feed,)
 
 
 pytestmark = pytest.mark.usefixtures('load_data')
@@ -101,3 +102,16 @@ def test_person_feed_uses_utf8_encoding(records, xml_records, E):
         f(r)
     assert b.getvalue() == ET.tostring(xml, encoding="UTF-8",
                                        xml_declaration=True)
+
+
+def test_articles_generates_articles():
+    arts = list(articles())
+    assert 'Yawning Abyss of Chaos' in arts[0]['ARTICLE_TITLE']
+
+
+def test_article_feed_adds_article(aa_data, articles_data):
+    b = BytesIO()
+    with article_feed(b) as f:
+        f(aa_data[0])
+    assert b.getvalue() == ET.tostring(articles_data, encoding='UTF-8',
+                                       xml_declaration=True)

tests/test_cli.py

@@ -16,8 +16,15 @@ def runner():
     return CliRunner()
 
 
-def test_hr_returns_people(runner, E, xml_data):
+def test_people_returns_people(runner, xml_data):
     res = runner.invoke(main, ['sqlite://', 'people'])
     assert res.exit_code == 0
     assert res.output_bytes == \
         ET.tostring(xml_data, encoding="UTF-8", xml_declaration=True)
+
+
+def test_articles_returns_articles(runner, articles_data):
+    res = runner.invoke(main, ['sqlite://', 'articles'])
+    assert res.exit_code == 0
+    assert res.output_bytes == \
+        ET.tostring(articles_data, encoding='UTF-8', xml_declaration=True)