12409

pending completion

Build # 12409

Build Type

Pull #3436

travis-ci

Committed by

web-flow

Commit Message

utils: make is_pdf_link accept long lines

The previous (i.e. third) iteration of `is_pdf_link` didn't work when
all the content was on a single line, as it was then reading too much
resulting in the connection being killed.
This is avoided by reading a fixed number of bytes, instead of a fixed
number of lines.

Signed-off-by: Micha Moskovic <michamos@gmail.com>

Pull Request Pull Request #3436: utils: make is_pdf_link accept long lines

Run Details

2 of 2 new or added lines in 1 file covered. (100.0%)

7216 of 9239 relevant lines covered (78.1%)

2.42 hits per line

Source File
Press 'n' to go to next uncovered line, 'b' for previous

94.32

/inspirehep/utils/export.py

# -*- coding: utf-8 -*-
#
# This file is part of INSPIRE.
# Copyright (C) 2014-2017 CERN.
#
# INSPIRE is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# INSPIRE is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with INSPIRE. If not, see <http://www.gnu.org/licenses/>.
#
# In applying this license, CERN does not waive the privileges and immunities
# granted to it by virtue of its status as an Intergovernmental Organization
# or submit itself to any jurisdiction.

from __future__ import absolute_import, division, print_function

import time

from inspirehep.utils.record_getter import get_es_record


class MissingRequiredFieldError(LookupError):

    """Base class for exceptions in this module.
    The exception should be raised when the specific,
    required field doesn't exist in the record.
    """

    def __init__(self, field):
        self.field = field

    def __str__(self):
        return "Missing field: " + self.field


class Export(object):
    """Base class used for export formats."""

    def __init__(self, record, *args, **kwargs):
        self.record = record

    def _get_citation_key(self):
        """Returns citation keys."""
        result = []
        citation_key = ''
        if 'external_system_numbers' in self.record:
            for field in self.record['external_system_numbers']:
                if 'institute' in field and \
                    (field['institute'] == 'INSPIRETeX' or
                        field['institute'] == 'SPIRESTeX'):
                    result.append(field)
            for key in result:
                if key['institute'] in ('INSPIRETeX', 'SPIRESTeX'):
                    if 'value' in key:
                        citation_key = key['value']
        if isinstance(citation_key, list):
            return citation_key[0].replace(' ', '')
        else:
            return citation_key.replace(' ', '')

    def _get_doi(self):
        """Return doi"""
        if 'dois' in self.record:
            doi_list = []
            for doi in self.record['dois']:
                doi_list.append(doi['value'])
            return ', '.join(doi for doi in list(set(doi_list)))
        else:
            return ''

    @property
    def arxiv_field(self):
        """Return arXiv field if exists"""
        if 'arxiv_eprints' in self.record:
            for field in self.record['arxiv_eprints']:
                return field

    def _get_arxiv(self):
        """Return arXiv and arXiv category"""
        arxiv = ''
        if self.arxiv_field:
            if 'value' in self.arxiv_field:
                arxiv = self.arxiv_field['value']
                if self.arxiv_field.get('categories', []):
                    arxiv += ' ['
                    arxiv += ",".join(self.arxiv_field.get('categories', []))
                    arxiv += ']'
        return arxiv

    def _get_report_number(self):
        """Return report number separated by commas"""
        report_number = []
        if 'report_numbers' in self.record:
            for field in self.record['report_numbers']:
                if 'value' in field:
                    report_number.append(field['value'])
            return ', '.join(str(p) for p in report_number)
        else:
            return report_number

    def _get_slac_citation(self):
        """Return SLACcitation"""
        cite_line = ''
        cite_element = ''
        if self.arxiv_field:
            if 'value' in self.arxiv_field:
                cite_element = self.arxiv_field['value'].upper()
                cite_line = '%%CITATION = ' + \
                            cite_element + ';%%'
        elif self._get_pubnote():
            cite_element = self._get_pubnote()
            cite_line = '%%CITATION = ' + cite_element + ';%%'
        elif 'report_numbers' in self.record:
            for field in self.record.get('arxiv_eprints', []):
                if 'categories' in field:
                    cite_element = field['value'].upper()
                    cite_line = '%%CITATION = ' + cite_element + ';%%'
            if not cite_element and self.record['report_numbers']:
                cite_element = self.record[
                    'report_numbers'][0]['value'].upper()
                cite_line = '%%CITATION = ' + cite_element + ';%%'
        else:
            cite_element = str(self.record['control_number'])
            cite_line = '%%CITATION = ' + 'INSPIRE-' + \
                cite_element + ';%%'
        return cite_line

    def _get_citation_number(self):
        """Returns how many times record was cited. If 0, returns nothing"""
        today = time.strftime("%d %b %Y")
        record = get_es_record('lit', self.record['control_number'])
        citations = ''
        try:
            times_cited = record['citation_count']
            if times_cited != 0:
                if times_cited > 1:
                    citations = '%d citations counted in INSPIRE as of %s' \
                                % (times_cited, today)
                else:
                    citations = '%d citation counted in INSPIRE as of %s'\
                                % (times_cited, today)
        except KeyError:
            pass
        return citations

1	# -- coding: utf-8 --
2	#
3	# This file is part of INSPIRE.
4	# Copyright (C) 2014-2017 CERN.
5	#
6	# INSPIRE is free software: you can redistribute it and/or modify
7	# it under the terms of the GNU General Public License as published by
8	# the Free Software Foundation, either version 3 of the License, or
9	# (at your option) any later version.
10	#
11	# INSPIRE is distributed in the hope that it will be useful,
12	# but WITHOUT ANY WARRANTY; without even the implied warranty of
13	# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
14	# GNU General Public License for more details.
15	#
16	# You should have received a copy of the GNU General Public License
17	# along with INSPIRE. If not, see <http://www.gnu.org/licenses/>.
18	#
19	# In applying this license, CERN does not waive the privileges and immunities
20	# granted to it by virtue of its status as an Intergovernmental Organization
21	# or submit itself to any jurisdiction.
22
23	from __future__ import absolute_import, division, print_function	5✔
24
25	import time	5✔
26
27	from inspirehep.utils.record_getter import get_es_record	5✔
28
29
30	class MissingRequiredFieldError(LookupError):	5✔
31
32	"""Base class for exceptions in this module.
33	The exception should be raised when the specific,
34	required field doesn't exist in the record.
35	"""
36
37	def __init__(self, field):	5✔
38	self.field = field	×
39
40	def __str__(self):	5✔
41	return "Missing field: " + self.field	×
42
43
44	class Export(object):	5✔
45	"""Base class used for export formats."""
46
47	def __init__(self, record, args, *kwargs):	5✔
48	self.record = record	2✔
49
50	def _get_citation_key(self):	5✔
51	"""Returns citation keys."""
52	result = []	2✔
53	citation_key = ''	2✔
54	if 'external_system_numbers' in self.record:	2✔
55	for field in self.record['external_system_numbers']:	1✔
56	if 'institute' in field and \	1✔
57	(field['institute'] == 'INSPIRETeX' or
58	field['institute'] == 'SPIRESTeX'):
59	result.append(field)	1✔
60	for key in result:	1✔
61	if key['institute'] in ('INSPIRETeX', 'SPIRESTeX'):	1✔
62	if 'value' in key:	1✔
63	citation_key = key['value']	1✔
64	if isinstance(citation_key, list):	2✔
65	return citation_key[0].replace(' ', '')	1✔
66	else:
67	return citation_key.replace(' ', '')	2✔
68
69	def _get_doi(self):	5✔
70	"""Return doi"""
71	if 'dois' in self.record:	2✔
72	doi_list = []	2✔
73	for doi in self.record['dois']:	2✔
74	doi_list.append(doi['value'])	2✔
75	return ', '.join(doi for doi in list(set(doi_list)))	2✔
76	else:
77	return ''	1✔
78
79	@property	5✔
80	def arxiv_field(self):
81	"""Return arXiv field if exists"""
82	if 'arxiv_eprints' in self.record:	2✔
83	for field in self.record['arxiv_eprints']:	1✔
84	return field	1✔
85
86	def _get_arxiv(self):	5✔
87	"""Return arXiv and arXiv category"""
88	arxiv = ''	2✔
89	if self.arxiv_field:	2✔
90	if 'value' in self.arxiv_field:	1✔
91	arxiv = self.arxiv_field['value']	1✔
92	if self.arxiv_field.get('categories', []):	1✔
93	arxiv += ' ['	1✔
94	arxiv += ",".join(self.arxiv_field.get('categories', []))	1✔
95	arxiv += ']'	1✔
96	return arxiv	2✔
97
98	def _get_report_number(self):	5✔
99	"""Return report number separated by commas"""
100	report_number = []	1✔
101	if 'report_numbers' in self.record:	1✔
102	for field in self.record['report_numbers']:	1✔
103	if 'value' in field:	1✔
104	report_number.append(field['value'])	1✔
105	return ', '.join(str(p) for p in report_number)	1✔
106	else:
107	return report_number	1✔
108
109	def _get_slac_citation(self):	5✔
110	"""Return SLACcitation"""
111	cite_line = ''	2✔
112	cite_element = ''	2✔
113	if self.arxiv_field:	2✔
114	if 'value' in self.arxiv_field:	1✔
115	cite_element = self.arxiv_field['value'].upper()	1✔
116	cite_line = '%%CITATION = ' + \	1✔
117	cite_element + ';%%'
118	elif self._get_pubnote():	2✔
119	cite_element = self._get_pubnote()	1✔
120	cite_line = '%%CITATION = ' + cite_element + ';%%'	1✔
121	elif 'report_numbers' in self.record:	2✔
122	for field in self.record.get('arxiv_eprints', []):	1✔
123	if 'categories' in field:	×
124	cite_element = field['value'].upper()	×
125	cite_line = '%%CITATION = ' + cite_element + ';%%'	×
126	if not cite_element and self.record['report_numbers']:	1✔
127	cite_element = self.record[	1✔
128	'report_numbers'][0]['value'].upper()
129	cite_line = '%%CITATION = ' + cite_element + ';%%'	1✔
130	else:
131	cite_element = str(self.record['control_number'])	2✔
132	cite_line = '%%CITATION = ' + 'INSPIRE-' + \	2✔
133	cite_element + ';%%'
134	return cite_line	2✔
135
136	def _get_citation_number(self):	5✔
137	"""Returns how many times record was cited. If 0, returns nothing"""
138	today = time.strftime("%d %b %Y")	2✔
139	record = get_es_record('lit', self.record['control_number'])	2✔
140	citations = ''	2✔
141	try:	2✔
142	times_cited = record['citation_count']	2✔
143	if times_cited != 0:	1✔
144	if times_cited > 1:	1✔
145	citations = '%d citations counted in INSPIRE as of %s' \	1✔
146	% (times_cited, today)
147	else:
148	citations = '%d citation counted in INSPIRE as of %s'\	1✔
149	% (times_cited, today)
150	except KeyError:	2✔
151	pass	2✔
152	return citations	2✔

inspirehep / inspire-next / 12409

Source File Press 'n' to go to next uncovered line, 'b' for previous

Source File
Press 'n' to go to next uncovered line, 'b' for previous