def line_per_theory(theory, terms):
'''(str, List[str]) -> str
Generate a row of data for the theory. Return a string containing that data.
'''
bibs = []
for term in terms:
try:
bibs.append(
bib_utils.get_bib(
os.sep.join(['ALL', '{},{}.bib'.format(theory, term)])))
except FileNotFoundError:
pass
bib = bib_utils.merge_bibs(bibs)
occurrences = len(bib.entries_dict)
bibs = []
for term in terms:
try:
bibs.append(
bib_utils.get_bib(
os.sep.join(['CSE', '{},{}.bib'.format(theory, term)])))
except FileNotFoundError:
pass
bib = bib_utils.merge_bibs(bibs)
cs_occurrences = len(bib.entries_dict)
venue_counts = bib_utils.get_venue_counts(bib)
print(theory, venue_counts)
venue_counts = venue_counts[:min(3, len(venue_counts))]
top_venues = '; '.join(
['{} ({})'.format(*venue) for venue in venue_counts if venue[1] > 3])
return '{} & {} & {} & {} \\\\'.format(theory, occurrences, cs_occurrences,
top_venues)
def gen_crosstab(theory_terms_d, typ):
'''(Dict[str: str], str) -> str
Return a latex table that represents the crosstab of theories.
'''
theory_bibs_d = OrderedDict()
for (theory, terms) in theory_terms_d.items():
bibs = []
for term in terms:
try:
bibs.append(bib_utils.get_bib(os.sep.join([typ, '{},{}.bib'.format(theory, term)])))
except FileNotFoundError:
pass
bib = bib_utils.merge_bibs(bibs)
venues = bib_utils.get_venues(bib) # Just in case we want venues later
papers = bib_utils.extract_paper_list(bib)
theory_bibs_d[theory] = (papers, venues)
body = []
counter = 1
to_remove = []
for (theory, data) in theory_bibs_d.items():
row = [theory]
papers = data[0]
any_data = False
for (theory_compare, comp_data) in theory_bibs_d.items():
if theory != theory_compare:
intersection = len(papers.intersection(comp_data[0]))
if intersection > 0:
any_data = True
row.append(str(intersection))
else:
row.append('')
else:
row.append('-')
if not any_data:
to_remove.append(counter)
counter += 1
body.append(row)
# Removing any rows with no data
for index in to_remove[::-1]:
for row in body:
row.pop(index)
body.pop(index - 1)
format_str = 'l' * (len(body) + 1)
header = 'Theory (id) & {}'.format('&'.join([str(i + 1) for i in range(len(body))]))
body = ['{} ({}) & {}\\\\'.format(row[0], count + 1, '&'.join(row[1:])) for (count, row) in enumerate(body)]
return table_str.format(format_str, header, '\n'.join(body))
total_occurrences += len(series_d[series])
venue_list.append('& {} & {} &\\\\'.format(series,
len(series_d[series])))
# Fenceposting -- but inserting at front to allow for rowspan
series = related_series[0]
total_occurrences += len(series_d[series])
venue_list.insert(0, '\\multirow{{{}}}{{*}}{{{}}} & {} & {} & \\multirow{{{}}}{{*}}{{{}}}\\\\'\
.format(len(related_series), venue, series, len(series_d[series]), len(related_series), total_occurrences))
body_list.extend(venue_list)
return table_str.format(format_str, header_str, '\n'.join(body_list),
label)
if __name__ == '__main__':
from docopt import docopt
arguments = docopt(__doc__)
analysis_name = arguments.get('<analysis_name>')
bibfiles = arguments.get('<bibfile>')
bibs = [bib_utils.get_bib(bibfile)
for bibfile in bibfiles] # Burning space!
merged_bib = bib_utils.merge_bibs(bibs)
output_fname = os.sep.join([OUTPUT_FOLDER, '{}.tex'.format(analysis_name)])
open(output_fname, 'w').write(
gen_venue_occurrence_table(merged_bib,
analysis_name.replace('_', '\\_')))