at = computed.setdefault(key, {}).setdefault(sensitive, {})
for book in books:
print book, name, canonical, position
at.setdefault(book, set()).add(position)
for (canonical, name), sensitivity in computed.items():
sensitive = sensitivity.get(True, {})
insensitive = sensitivity.get(False, {})
for book, positions in insensitive.items():
positions.difference_update(sensitive.get(book, set()))
def citations():
for name in NAMES:
key = (name["Canonical"], name["Name"].lower())
at = computed.get(key, {})
sensitive, insensitive = (
", ".join(
"%s#%d" % (book, len(positions))
for book, positions in sorted(
at.get(sensitivity, {}).items(),
key = lambda (book, positions): BOOKS.book_index(book)
)
if len(positions)
)
for sensitivity in (True, False)
)
yield name["Canonical"], name["Name"], sensitive, insensitive
codecs.open("names-citations.tsv", "w", "utf-8").write(tsv.format(citations()))
def cross_reference_txt():
from tsv import format
open('tolkiengateway.tsv', 'w').write(format(cross_reference()))
def dump():
import tsv
import codecs
data = tsv.format(sorted(analyze()))
codecs.open("tuckborough.tsv", "w", "utf-8").write(data)