litcorpt 0.0.2

API to access Portuguese Literary Corpus

litcorpt

LITerary CORpus in PorTuguese is a API to access a literary corpus in portuguese language.

The API provides access to the corpus without all the fuzz to download and write a loader for different types of data sources. It is exposed as a simple document database.

How to install.

Simply:

pip install litcorpt

Getting started

After installation in you Python just

import litcorpt
corpus_db = litcorpt.load_corpus(verbose=True)
print(f'There are {len(corpus_db)} documents in corpus')

It will load the whole corpus. When running by the first time, it will download from internet, process and build the whole dataset.

The download size is around 115MB and is automaticly handled by the library. It is downloaded just at first time you load it. After the first time it will load from local disk. The time to load data locally takes around 34 ms. This value was measured in my own computer (your mileage may vary).

Basic Usage

Most of time you just want to retrieve the whole corpus as a list of documents. You can do that with this one liner.

corpus = litcorpt.corpus(corpus_db)

This operation just append to a list all contents for all documents. Since a document may have more than one content.

Advanced usage

Besides the fetchall usage, many custom queries can be done. Is possible to search by matches, regexes, fields.

All book titles of an author (Eça de Queirós)

We are ignoring documents where Queirós is an editor.

As a regular for loop

q = litcorpt.Query()
search = corpus_db.search(q.authors.any((q.lastname == 'Queirós') &
                                     (q.firstname == 'Eça de')))

titles = []
for document in search:
  titles.append(document['title'])

As a list comprehension shorter but harder to read.

q = litcorpt.Query()
titles = [ document['title'] for document in corpus_db.search(q.authors.any((q.lastname == 'Queirós') & (q.firstname == 'Eça de')))]

Building a corpus with Eça de Queirós

q = litcorpt.Query()
search = corpus_db.search(q.authors.any((q.lastname == 'Queirós') &
                                     (q.firstname == 'Eça de')))

queiros_corpus = []
for documents in search:
    for document in documents['contents']:
        queiros_corpus.append(document)

Building a bibliography

Here we handle the case where there is no author.

bibliography = []
for document in corpus_db.all():
    authors = []
    for author in document.get('authors', [{'lastname': 'Anonymous'}]):
        authors.append(', '.join(filter(None, list(author.values())[:2])))
    bibliography.append(f'{" and ".join(authors)}. {document["title"]}.')

Count documents by Author Surname

Here we use Python's Counter to count the surnames and using a dict comprehension to filter the authors that occurs more than 5 times. You still can access the whole counting the lastnames variable

As a list comprehension

q = litcorpt.Query()
from collections import Counter
lastnames = Counter([ author['lastname'] for document in corpus_db.search(q.authors.exists()) for author in document['authors'] ])
most_common_surnames = {lastname: count for lastname, count in lastnames.items() if count >= 5}

Unrolling the comprehension

q = litcorpt.Query()
from collections import Counter

lastnames = []

for document in corpus_db.search(q.authors.exists()):
  for author in document['authors']:
    lastnames.append(author['lastname'])

lastnames = Counter(lastnames)

most_common_surnames = {}
for lastname, count in lastnames.items():
  if count >= 5:
    most_common_surnames[lastname] = count

Extra: Sorting by decreasing frequency, then alphabeticaly.

sorted(most_common_surnames.items(), key=lambda item: (-item[1], item[0]))

Display all Subjects

First we group all subjects

q = litcorpt.Query()
subjects = []
for document in corpus_db.search(q.subject.exists()):
  subjects.extend(document['subject'])

Then we can count, and sort by descending frequency (Python 3.6> dicts are ordered by default).

from collections import Counter
subject_frequency = Counter(subjects)
subject_frequency = dict(sorted(subject_frequency.items(), key=lambda item: -item[1]))

And also group the unique items for reference.

subject_list = list(subject_frequency.keys())

Building a corpus given a list of Subjects

First we pick a list of subjects (this is just an example with a few valid entries, and some not valid).

subjects = [ 'portuguese drama',
             'france',
             'drama',
             'women',
             '<INVALID SUBJECT>' ]

Then we proceed with search and corpus building

q = litcorpt.Query()
search = corpus_db.search(q.subject.any(subjects))
drama_corpus = [ document for documents in search for document in documents['contents'] ]

If we want we can easily list the titles in our new drama_corpus

titles = [ document['title'] for document in search ]

Of course we can do the same by any of the fields in document.

The structure of a document.

The corpus database is a list of documents. A document is often related with a literary document (book, text, play, etc) and contains the following fields:

'title': (str) The document title

'authors': (list) A list of authors. Each author contains:

'lastname' (str) The author last name,

'firstname'(str) The author first name,

'life': (str) The author life.

'editors': (list) A list of editors. Same fields as authors.

'language': (list) A list of languages in document (often portuguese)

'subject': (list) A list of topics covered in the document

'category': (list) A list of genres, often just a single entry.

'contents': (list) The document contents. Often just one entry.

'note': (str) Notes and observations about the document.

There are also some less used keys as

'datasource': (str) The data source

'sourcekey': (str) The data source key index.

'loc': (str) The Library of Congress catalog string

'href': (str) The url path inside the data source (used to fetch the metadata)

'files': (list) A list of files related to document in data source.

These keys can be changed, added or removed in future. Everything is mutable.

Customizing

By default, the corpus is stored at

${HOME}/litcorpt_data

If you wish to put in a different place, just set the LITCORPUSPT_DATAPATH environment variable in your system configuration. For example for bash, add this to your ~/.bashrc

export LITCORPUSPT_DATAPATH="/whatever/place/you/want"

Then call your programs using litcorpt or your ipython session

TODO

• Maybe build some custom functions to handle the most common filter use cases.
• Make it easier to retrieve the corpus given a filter.
• Crawling paralellization
• Use of logging API
• Better define the book "object"
• Refactor =/ the API