diff options
| -rw-r--r-- | ankivoc.py | 57 |
1 files changed, 40 insertions, 17 deletions
@@ -2,18 +2,24 @@ from bs4 import BeautifulSoup, Comment import requests -import sys import re +import csv + def lookup(word): r = requests.get("https://www.oxforddictionaries.com/definition/english/" + - word) + word.replace(" ", "-")) soup = BeautifulSoup(r.text, "html.parser") + entry_count = 0 entry = "<meta charset=\"utf-8\">" for res in soup.find_all("section", class_="senseGroup"): - entry += res.find_all("span", class_="partOfSpeech")[0].get_text() + parts = res.find_all("span", class_="partOfSpeech") + if not len(parts): + continue + entry += parts[0].get_text() + entry_count += 1 # remove comments for el in res(text=lambda text: isinstance(text, Comment)): @@ -31,31 +37,27 @@ def lookup(word): for syno in el.find_all("div"): syno.name = "li" - #el.insert_before(soup.new_tag("br")) - # simplify structure for el in sense.find_all("span") + \ sense.find_all("div"): - # skip "more" tags - if "onclick" not in el.attrs: - el.unwrap() + el.unwrap() # remove unused classes for el in sense.find_all(): del el["class"] + del el["href"] del sense["class"] # hide examples - for el in sense.find_all("li"): - el["style"] = "font-size: 75%;" + for el in sense.find_all("ul"): + el["style"] = "font-size: 75%; list-style-type: none;" - entry += str(sense) #.prettify(formatter="html") - #entry += "<br/>" + entry += str(sense) entry += "<br/>" # ugly hack. Why is this necessary: - #- for el in sense.find_all("a", class_="moreInformationExemples"): - #- AttributeError: 'NoneType' object has no attribute 'next_element' + # - for el in sense.find_all("a", class_="moreInformationExemples"): + # - AttributeError: 'NoneType' object has no attribute 'next_element' soup = BeautifulSoup(entry, "html.parser") for el in soup.find_all("li"): if "Get more examples" == el.get_text() or \ @@ -66,7 +68,28 @@ def lookup(word): text = str(soup) # another hack, I'm out of time text = re.sub(r"(\d)([A-Za-z])", r"\1 \2", text) - return text + return text, entry_count + + +def main(): + with open("words.txt", "r") as f: + with open("errors.txt", "a+") as err: + issues = [] + words = f.read().split("\n") + words = list(set(words)) # remove dupes + w = csv.writer(open("cards.csv", "a+")) + for word in words: + if len(word) <= 1: + continue + entry, number = lookup(word) + if entry and number > 0: + print("found " + word) + w.writerow([word, entry]) + else: + print("skipping " + word) + issues.append(word) + + err.write("\n".join(issues)) -#lookup("clerk") -print(lookup("smudge")) +if __name__ == "__main__": + main() |