load frequencies of epub

.gitignore

@@ -160,4 +160,7 @@ cython_debug/
 .idea/
 
 # Project Specific
-scripts/
+scripts/
+.vscode
+.vscode
+data/

requirements.txt

@@ -1,5 +1,10 @@
 click
 genanki
-pandas
+pandas==1.5.2
 pyyaml
-bullet
+bullet
+nltk
+EbookLib
+BeautifulSoup4
+PyMultiDictionary
+translate

setup.py

@@ -8,7 +8,7 @@ def readme():
 
 setup(
     name='ankimaker',
-    version='0.0.5',
+    version='0.0.6',
     description='Makes anki with files',
     url="https://git.lgoon.xyz/gabriel/ankimaker",
     license="BSD-3-Clause",
@@ -27,7 +27,11 @@ setup(
         "genanki",
         "pandas",
         "pyyaml",
-        "bullet"
+        "bullet",
+        "nltk",
+        "EbookLib",
+        "BeautifulSoup",
+        "wordfreq",
     ],
     long_description_content_type='text/markdown',
 )

src/ankimaker/commands/__init__.py

@@ -1,3 +1,4 @@
 from .base_click import cli
 from .from_csv import generate_anki
 from .make_config import make_csv_config
+from .from_epub import process_epub

src/ankimaker/commands/from_epub.py

@@ -0,0 +1,13 @@
+import click
+
+from ankimaker.commands import cli
+from ankimaker.tasks.epub_to_anki import process_epub
+
+
+@cli.command('epub')
+@click.option('-i', '--input', 'input_file', type=click.Path(exists=True))
+@click.option('-o', '--output', 'output_file', type=click.Path(exists=False))
+@click.option('-l', '--lang', 'language', default=None, type=click.STRING)
+@click.option('-n', '--name', 'name', required=False, type=click.STRING)
+def generate_anki(input_file, output_file, language, name):
+    process_epub(input_file=input_file, output_file=output_file, language=language, deck_name=name)

src/ankimaker/config/configuration.py

@@ -12,16 +12,20 @@ class AnkimakerConfig(yaml.YAMLObject):
     question_column = None
     answer_column = None
     separators = ','
+    input_language = None,
+    output_language = None,
     filters: List[List[FilterConfig]] = list()
 
     def __init__(
             self, separators=',', header=None, answer_column=None, question_column=None,
-            filters=tuple(), *args, **karhs
+            filters=tuple(), input_language=None, output_language=None, *args, **karhs
     ):
         self.answer_column = answer_column
         self.question_column = question_column
         self.header = header
         self.separators = separators
+        self.input_language = input_language
+        self.output_language = output_language
         self.filters = _conditionally_create_new_filters(filters)
 
     @staticmethod
@@ -34,6 +38,8 @@ class AnkimakerConfig(yaml.YAMLObject):
         AnkimakerConfig.question_column = content.question_column
         AnkimakerConfig.answer_column = content.answer_column
         AnkimakerConfig.separators = content.separators
+        AnkimakerConfig.input_language = content.input_language
+        AnkimakerConfig.output_language = content.output_language
         AnkimakerConfig.filters = _conditionally_create_new_filters(content.filters)
 
 

src/ankimaker/generator/__init__.py

@@ -3,3 +3,5 @@ from . import (
 )
 from .card import create_note
 from .model import create_model
+from .translator_generator import TranslatorGenerator
+from .question_answer_generator import QuestionAnswerGenerator

src/ankimaker/generator/question_answer_generator.py

@@ -0,0 +1,17 @@
+import genanki
+from typing import Collection, List
+
+from ankimaker import generator
+
+
+class QuestionAnswerGenerator:
+    def __init__(self):
+        self.__model = generator.create_model()
+
+    def get_cards(self, questions: Collection[str], answers: Collection[str]) -> List[genanki.Model]:
+        assert len(questions) == len(answers)
+        cards = list()
+        for content_fields in zip(questions, answers):
+            card = generator.create_note(self.__model, fields=content_fields)
+            cards.append(card)
+        return cards

src/ankimaker/generator/translator_generator.py

@@ -0,0 +1,28 @@
+import genanki
+from translate import Translator
+from typing import Collection, List
+
+from ankimaker import generator
+
+
+class TranslatorGenerator:
+    def __init__(self, original_language, destination_language):
+        """
+        :param original_language: Language of the inserted text, following https://en.wikipedia.org/wiki/ISO_639-1
+        :param destination_language: Language you want to translate to, following https://en.wikipedia.org/wiki/ISO_639-1
+        """
+        self.__translator = Translator(from_lang=original_language, to_lang=destination_language)
+        self.__model = generator.model.create_model()
+
+    def get_cards(self, content_collection: Collection[str]) -> List[genanki.Model]:
+        cards = list()
+        for content in content_collection:
+            card = self._create_card(content)
+            cards.append(card)
+        return cards
+
+    def _create_card(self, content):
+        translation = self.__translator.translate(content)
+        fields = (content, translation)
+        card = generator.create_note(self.__model, fields)
+        return card

src/ankimaker/tasks/__init__.py

@@ -1,2 +1,3 @@
 from .basic_csv_to_anki import basic_pandas_to_anki
 from .config_tasks import create_config, enhance_config
+from . import dictionary

src/ankimaker/tasks/basic_csv_to_anki.py

@@ -18,14 +18,19 @@ def load_csv(path: str) -> pd.DataFrame:
 
 
 def add_df_to_deck(df: pd.DataFrame, deck: genanki.Deck) -> genanki.Deck:
-    model = generator.create_model()
-
-    for entry in df.to_dict('records'):
-        question = entry[Config.question_column]
-        answer = entry[Config.answer_column]
-        content_fields = (question, answer)
-        note = generator.create_note(model, fields=content_fields)
-        deck.add_note(note)
+    questions = df[Config.question_column].to_list()
+    if Config.answer_column is None:
+        generator_engine = generator.TranslatorGenerator(
+            original_language=Config.input_language,
+            destination_language=Config.output_language,
+        )
+        cards = generator_engine.get_cards(questions)
+    else:
+        answers = df[Config.answer_column]
+        generator_engine = generator.QuestionAnswerGenerator()
+        cards = generator_engine.get_cards(questions, answers)
+    for card in cards:
+        deck.add_note(card)
     return deck
 
 

src/ankimaker/tasks/dictionary.py

@@ -0,0 +1,24 @@
+from multiprocessing import Pool
+from itertools import repeat
+from typing import Iterable, Optional
+from http.client import RemoteDisconnected as HttpClientRemoteDisconnected
+
+from PyMultiDictionary import MultiDictionary
+
+
+def get_and_process_word_definition(language: str, word: str) -> Optional[str]:
+    try:
+        dictionary = MultiDictionary()
+        definition = dictionary.meaning(lang=language, word=word)
+        if len(definition[1]) <= 1:
+            return None
+        definition = definition[1].split('.')[0]
+    except HttpClientRemoteDisconnected:
+        return None
+    return definition
+
+
+def get_word_definitions_from_dictionary(language: str, word_collection: Iterable[str]) -> Iterable[str]:
+    with Pool(7) as p:
+        definitions = p.starmap(get_and_process_word_definition, zip(repeat(language), word_collection))
+    return definitions

src/ankimaker/tasks/epub/__init__.py

@@ -0,0 +1 @@
+from . load_epub import generate_corpus_from_epub_file

src/ankimaker/tasks/epub/load_epub.py

@@ -0,0 +1,75 @@
+import nltk
+import ebooklib
+import pandas as pd
+from ebooklib import epub
+from bs4 import BeautifulSoup
+
+nltk.download('stopwords')
+nltk.download('punkt')
+
+blacklist = (
+    '[document]',
+    'noscript',
+    'header',
+    'html',
+    'meta',
+    'head',
+    'input',
+    'script',
+    'style',
+    # there may be more elements you don't want, such as "style", etc.
+)
+
+
+def make_word_frequency_series(corpus):
+    nltk_occurrences = nltk.FreqDist(corpus)
+    occurrences: pd.Series = pd.Series(dict(nltk_occurrences))
+    frequencies = occurrences / sum(occurrences)
+    frequencies = frequencies.sort_values(ascending=False)
+    return frequencies
+
+
+def epub2thtml(epub_path):
+    book = epub.read_epub(epub_path)
+    chapters = []
+    for item in book.get_items():
+        if item.get_type() == ebooklib.ITEM_DOCUMENT:
+            chapters.append(item.get_content())
+    return chapters
+
+
+def chap2text(chap):
+    output = ''
+    soup = BeautifulSoup(chap, 'html.parser')
+    text = soup.find_all(text=True)
+    for t in text:
+        if t.parent.name not in blacklist:
+            output += '{} '.format(t)
+    return output
+
+
+def thtml2ttext(thtml):
+    Output = []
+    for html in thtml:
+        text = chap2text(html)
+        Output.append(text)
+    return Output
+
+
+def epub2text(epub_path):
+    chapters = epub2thtml(epub_path)
+    ttext = thtml2ttext(chapters)
+    return ttext
+
+
+def generate_corpus_from_epub_file(input_path, language):
+    epub_doc = epub2text(input_path)
+    corpus = []
+    sw = nltk.corpus.stopwords.words(language)
+    for content in epub_doc:
+        for w in nltk.word_tokenize(content, language=language):
+            w = w.lower()
+            if w not in sw and len(w) > 1:
+                corpus.append(w)
+    return corpus
+

src/ankimaker/tasks/epub_to_anki.py

@@ -0,0 +1,24 @@
+from ankimaker import generator
+
+from ankimaker.tasks import epub
+from ankimaker.tasks import dictionary
+
+
+def create_collection_and_filter_out_on_empty_definitions(words_from_epub, definitions):
+    collection = [(words, defi) for words, defi in zip(words_from_epub, definitions) if defi is not None]
+    return collection
+
+
+def process_epub(input_file, output_file, language, deck_name):
+    words_from_epub = epub.generate_corpus_from_epub_file(input_file)
+    definitions = dictionary.get_word_definitions_from_dictionary(language, words_from_epub)
+    collection = create_collection_and_filter_out_on_empty_definitions(words_from_epub, definitions)
+    generator_engine = generator.QuestionAnswerGenerator()
+
+    deck = generator.deck.create_deck(deck_name)
+
+    words_from_epub, definitions = map(list, zip(*collection))
+    cards = generator_engine.get_cards(words_from_epub, definitions)
+    for card in cards:
+        deck.add_note(card)
+    generator.deck.save_deck(deck, output_file)