clic.db.corpora: Resolve corpora lists

corpora_to_book_ids

There are various ways we can specify which books to search in the corpora parameter. They are specified below.

Make a very cut-down library of books, with the word “the” so we can search for all of them:

>>> from clic.db.corpus import put_corpus
>>> db_cur = test_database(
... alice='''
... Alice’s Adventures in Wonderland
... Lewis Carroll
...
... the
... '''.strip(),
...
... willows='''
... The Wind in the Willows
... Kenneth Grahame
...
... the
... '''.strip(),
...
... sense='''
... Sense and Sensibility
... Jane Austen
...
... the
... '''.strip(),
...
... northanger='''
... Northanger Abbey
... Jane Austen
...
... the
... '''.strip(),
...
... gulliver='''
... Gulliver's Travels
... Jonathan Swift
...
... the
... '''.strip())
>>> put_corpus(db_cur, dict(name='ChiLit', contents=['alice', 'willows'], title='', ordering=1))
>>> put_corpus(db_cur, dict(name='ArTs', contents=['sense', 'northanger', 'gulliver'], title='', ordering=2))

We will use concordance for these examples, but the choice shouldn’t matter:

>>> from ..concordance import concordance

book name

We can use book names directly:

>>> just_metadata(concordance(db_cur, ['alice', 'willows'], q=["the"], metadata=['book_titles']))
{'book_titles':
  {'alice': ['Alice’s Adventures in Wonderland', 'Lewis Carroll'],
   'willows': ['The Wind in the Willows', 'Kenneth Grahame']}}

author

We can use the author name to get any books by a given author:

>>> just_metadata(concordance(db_cur, ['author:Jane Austen'], q=["the"], metadata=['book_titles']))
{'book_titles':
  {'northanger': ['Northanger Abbey', 'Jane Austen'],
   'sense': ['Sense and Sensibility', 'Jane Austen']}}

corpus name

We can use the corpus name:

>>> just_metadata(concordance(db_cur, ['corpus:ChiLit'], q=["the"], metadata=['book_titles']))
{'book_titles':
  {'alice': ['Alice’s Adventures in Wonderland', 'Lewis Carroll'],
   'willows': ['The Wind in the Willows', 'Kenneth Grahame']}}

old names

There are some old names that whilst are deprecated, should still work:

>>> just_metadata(concordance(db_cur, ['Other'], q=["the"], metadata=['book_titles']))
{'book_titles':
  {'gulliver': ["Gulliver's Travels", 'Jonathan Swift'],
   'northanger': ['Northanger Abbey', 'Jane Austen'],
   'sense': ['Sense and Sensibility', 'Jane Austen']}}

clic.db.corpora.corpora_to_book_ids(cur, corpora)

Resolve list of corpora into set of book IDs. Where corpora could be: - (book_name): Fetch relevant book ID - “corpra:DNov”: All books in that corpora - “author:Thomas Hardy” All books by that author