def test_get_xml_from_textmarc_success(self):
textmarc = """100__ $$aDoe, J.$$uCERN
245__ $$aPion production by 24 GeV/c protons in hydrogen
260__ $$c1961
300__ $$a15"""
output = get_xml_from_textmarc(1, textmarc)
self.assertEqual(output['resultMsg'], 'textmarc_parsing_success')
xml_expected_output = """<record>
<controlfield tag="001">1</controlfield>
<datafield tag="100" ind1=" " ind2=" ">
<subfield code="a">Doe, J.</subfield>
<subfield code="u">CERN</subfield>
</datafield>
<datafield tag="245" ind1=" " ind2=" ">
<subfield code="a">Pion production by 24 GeV/c protons in hydrogen</subfield>
</datafield>
<datafield tag="260" ind1=" " ind2=" ">
<subfield code="c">1961</subfield>
</datafield>
<datafield tag="300" ind1=" " ind2=" ">
<subfield code="a">15</subfield>
</datafield>
</record>
"""
self.assertEqual(re.sub("\s+", " ", output['resultXML'].strip()),
re.sub("\s+", " ", xml_expected_output.strip()))
def test_get_xml_from_textmarc_wrong_content(self):
textmarc = """100__ $$a
245__ $$aPion production by 24 GeV/c protons in hydrogen
260__ $$c1961
300__ $$a15"""
output = get_xml_from_textmarc(1, textmarc)
self.assertEqual(output['resultMsg'], 'textmarc_parsing_error')
def test_get_xml_from_textmarc_success(self):
textmarc = """100__ $$aDoe, J.$$uCERN
245__ $$aPion production by 24 GeV/c protons in hydrogen
260__ $$c1961
300__ $$a15"""
output = get_xml_from_textmarc(1, textmarc)
self.assertEqual(output['resultMsg'], 'textmarc_parsing_success')
xml_expected_output = """<record>
<controlfield tag="001">1</controlfield>
<datafield tag="100" ind1=" " ind2=" ">
<subfield code="a">Doe, J.</subfield>
<subfield code="u">CERN</subfield>
</datafield>
<datafield tag="245" ind1=" " ind2=" ">
<subfield code="a">Pion production by 24 GeV/c protons in hydrogen</subfield>
</datafield>
<datafield tag="260" ind1=" " ind2=" ">
<subfield code="c">1961</subfield>
</datafield>
<datafield tag="300" ind1=" " ind2=" ">
<subfield code="a">15</subfield>
</datafield>
</record>
"""
self.assertEqual(re.sub("\s+", " ", output['resultXML'].strip()),
re.sub("\s+", " ", xml_expected_output.strip()))
def test_get_xml_from_textmarc_wrong_content(self):
textmarc = """100__ $$a
245__ $$aPion production by 24 GeV/c protons in hydrogen
260__ $$c1961
300__ $$a15"""
output = get_xml_from_textmarc(1, textmarc)
self.assertEqual(output['resultMsg'], 'textmarc_parsing_error')
def test_get_xml_from_textmarc_wrong_field(self):
textmarc = """1sasd00__ $$aDoe, J.$$uCERN
245__ $$aPion production by 24 GeV/c protons in hydrogen
260__ $$c1961
300__ $$a15"""
output = get_xml_from_textmarc(1, textmarc)
self.assertEqual(output['resultMsg'], 'textmarc_parsing_error')
expected_error = [1, "1sasd00__ $$aDoe, J.$$uCERN"]
self.assertEqual(output['parse_error'][:-1], expected_error)
def test_get_xml_from_textmarc_wrong_field(self):
textmarc = """1sasd00__ $$aDoe, J.$$uCERN
245__ $$aPion production by 24 GeV/c protons in hydrogen
260__ $$c1961
300__ $$a15"""
output = get_xml_from_textmarc(1, textmarc)
self.assertEqual(output['resultMsg'], 'textmarc_parsing_error')
expected_error = [1, "1sasd00__ $$aDoe, J.$$uCERN"]
self.assertEqual(output['parse_error'][:-1], expected_error)
def test_accept_fft_tags_in_textmarc(self):
textmarc = """100__ $$aDoe, J.$$uCERN
FFT__ $$ahttp://scd-theses.u-strasbg.fr/1818/01/RICAUD_Helene_2008.pdf$$dFulltext"""
output = get_xml_from_textmarc(1, textmarc)
xml_expected_output = """<record>
<controlfield tag="001">1</controlfield>
<datafield tag="100" ind1=" " ind2=" ">
<subfield code="a">Doe, J.</subfield>
<subfield code="u">CERN</subfield>
</datafield>
<datafield tag="FFT" ind1=" " ind2=" ">
<subfield code="a">http://scd-theses.u-strasbg.fr/1818/01/RICAUD_Helene_2008.pdf</subfield>
<subfield code="d">Fulltext</subfield>
</datafield>
</record>"""
self.assertEqual(re.sub("\s+", " ", output['resultXML'].strip()),
re.sub("\s+", " ", xml_expected_output.strip()))
def test_accept_fft_tags_in_textmarc(self):
textmarc = """100__ $$aDoe, J.$$uCERN
FFT__ $$ahttp://scd-theses.u-strasbg.fr/1818/01/RICAUD_Helene_2008.pdf$$dFulltext"""
output = get_xml_from_textmarc(1, textmarc)
xml_expected_output = """<record>
<controlfield tag="001">1</controlfield>
<datafield tag="100" ind1=" " ind2=" ">
<subfield code="a">Doe, J.</subfield>
<subfield code="u">CERN</subfield>
</datafield>
<datafield tag="FFT" ind1=" " ind2=" ">
<subfield code="a">http://scd-theses.u-strasbg.fr/1818/01/RICAUD_Helene_2008.pdf</subfield>
<subfield code="d">Fulltext</subfield>
</datafield>
</record>"""
self.assertEqual(re.sub("\s+", " ", output['resultXML'].strip()),
re.sub("\s+", " ", xml_expected_output.strip()))
def test_get_xml_from_textmarc_wrong_content(self):
textmarc = """100__ $$a
245__ $$aPion production by 24 GeV/c protons in hydrogen
260__ $$c1961
300__ $$a15"""
output = get_xml_from_textmarc(1, textmarc)
# We expect success now, ignoring empty field, see ticket:1269.
# <http://invenio-software.org/ticket/1269>
self.assertEqual(output['resultMsg'], 'textmarc_parsing_success')
self.assertEqual(output['resultXML'], """<record>
<controlfield tag="001">1</controlfield>
<datafield tag="245" ind1=" " ind2=" ">
<subfield code="a">Pion production by 24 GeV/c protons in hydrogen</subfield>
</datafield>
<datafield tag="260" ind1=" " ind2=" ">
<subfield code="c">1961</subfield>
</datafield>
<datafield tag="300" ind1=" " ind2=" ">
<subfield code="a">15</subfield>
</datafield>
</record>
""")
def test_get_xml_from_textmarc_wrong_content(self):
textmarc = """100__ $$a
245__ $$aPion production by 24 GeV/c protons in hydrogen
260__ $$c1961
300__ $$a15"""
output = get_xml_from_textmarc(1, textmarc)
# We expect success now, ignoring empty field, see ticket:1269.
# <http://invenio-software.org/ticket/1269>
self.assertEqual(output['resultMsg'], 'textmarc_parsing_success')
self.assertEqual(output['resultXML'], """<record>
<controlfield tag="001">1</controlfield>
<datafield tag="245" ind1=" " ind2=" ">
<subfield code="a">Pion production by 24 GeV/c protons in hydrogen</subfield>
</datafield>
<datafield tag="260" ind1=" " ind2=" ">
<subfield code="c">1961</subfield>
</datafield>
<datafield tag="300" ind1=" " ind2=" ">
<subfield code="a">15</subfield>
</datafield>
</record>
""")
