dict_dl/dict_dl.py

import json
import os
import random
import re
import string
import time
from datetime import datetime
from pathlib import Path
from urllib.parse import unquote
from xml.etree import ElementTree as ET

import requests
from bs4 import BeautifulSoup
from requests.exceptions import ConnectionError

letters = string.ascii_lowercase
unusual = lambda prefix: not all( [c in letters for c in prefix.lower()])
# def uq(s):
#     return unquote(s).split("?")[0]
uq = unquote


def uqall(data):
    if isinstance(data, list):
        iterator = enumerate(data)
    elif isinstance(data, dict):
        for k in list(data.keys()):
            if "%" in k:
                data[uq(k)] = data.pop(k)
        iterator = data.items()
    elif isinstance(data, str):
        data = [data]
        iterator = enumerate(data)
    else:
        raise TypeError("can only traverse list or dict")
    for i, value in iterator:
        if isinstance(value, (list, dict)):
            uqall(value)
        elif isinstance(value, str):
            if "%" in value:
                data[i] = uq(value)
    return data


def clear_whitespace(data):
    if isinstance(data, list):
        iterator = enumerate(data)
    elif isinstance(data, dict):
        iterator = data.items()
    elif isinstance(data, str):
        data = [data]
        iterator = enumerate(data)
    else:
        raise TypeError("can only traverse list or dict")
    for i, value in iterator:
        if isinstance(value, (list, dict)):
            clear_whitespace(value)
        elif isinstance(value, str):
            data[i] = re.sub(r"[\n\t\s]+", " ", value).strip()
    return data


def randtime(a, b, k=0):
    if k:
        return [random.uniform(a, b) for _ in range(k)]
    else:
        return random.uniform(a, b)


def remove_between(string, a, b):
    otag_pos = 0
    ctag_pos = 0
    for i in range(len(string)):
        if string[i : i + len(a)] == a:
            otag_pos = i
        elif string[i : i + len(b)] == b:
            ctag_pos = i + len(b)
        if otag_pos and ctag_pos:
            return remove_between(string[:otag_pos] + string[ctag_pos:], a, b)
    return string.strip()


def remove_tag(string, tag="script"):
    otag = f"<{tag}"
    ctag = f"</{tag}>"
    otag_pos = 0
    ctag_pos = 0
    for i in range(len(string)):
        if string[i : i + len(otag)] == otag:
            otag_pos = i
        elif string[i : i + len(ctag)] == ctag:
            ctag_pos = i + len(ctag)
        if otag_pos and ctag_pos:
            return remove_tag(string[:otag_pos] + string[ctag_pos:], tag)
    return string


def all_text(e):
    return clear_whitespace(" ".join(e.itertext()))


def only_first_text(e):
    return clear_whitespace([next(e.itertext())])[0]


def only_text(e):
    return " ".join(all_text(e))


def url2str(url: str) -> str:
    headers = {
        "user-agent": "Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/102.0.5005.115 Safari/537.36"
    }
    # bad_html = requests.get(url, headers=headers)
    bad_html = requests.get(url)
    tree = BeautifulSoup(bad_html.text, features="lxml")
    xml_str = str(tree)
    xml_str = remove_tag(xml_str, "head")
    xml_str = remove_tag(xml_str)
    # with open("test.html", "w") as f:
    #     f.write(xml_str)
    return xml_str


# aliases
rb = remove_between
cw = clear_whitespace
ot = only_text
at = all_text
oft = only_first_text


class WordParser:
    """WordParser needs additional methods to work with Queue:
    - self.neighbours = words found on the site
    - self.todict() = returning a dict with the parsed info"""

    def __init__(self, word, url_prefix):
        self.time = datetime.now().strftime("%Y%m%d-%H%M%S")
        self.word = word
        self.url = f"{url_prefix}{word}"
        self.xml_string = url2str(self.url)
        self.root = ET.fromstring(self.xml_string)


class FileSet(set):
    def __init__(self, file):
        self.file = file
        super().__init__({line.strip() for line in open(self.file, "r")})

    def load(self):
        self.update({line.strip() for line in open(self.file, "r")})

    def save(self):
        if self:
            with open(self.file, "w") as f:
                f.write("\n".join([w for w in self if w]))

    def append(self):
        if self:
            self |= {line.strip() for line in open(self.file, "r")}
            self.save()


class DictFile(dict):
    def __init__(self, file):
        self.file = file
        if os.path.isfile(self.file):
            with open(self.file, "r") as f:
                super().__init__(json.load(f))
        else:
            super()

    def save(self):
        with open(self.file, "w") as f:
            json.dump(self, f, separators=(",", ":"), indent=2, sort_keys=False)


class FullDictionary(dict):
    def __init__(self, dir_prefix, suffix):
        self.__dict__.update(locals())
        full_dict = {}
        start = time.time()
        for db_file in Path(self.dir_prefix).glob(f"*{self.suffix}"):
            with open(db_file, "r") as f:
                full_dict |= json.load(f)
        self.readtime = time.time() - start
        super().__init__(full_dict)


class Queue:
    def __init__(
        self,
        Parser,
        dir_prefix,
        suffix,
        time_base=1.01,
        time_exponent=10,
        prefix_length=3,
    ):
        self.__dict__.update(locals())
        self.words = set()
        self.queue = FileSet(f"{dir_prefix}queue")
        self.snafus = FileSet(f"{dir_prefix}snafus")
        self.redo = FileSet(f"{dir_prefix}redo")

    def wait(self):
        a = self.time_base**self.time_exponent
        b = self.time_base ** (self.time_exponent * 3)
        time.sleep(randtime(a, b))

    def loadDB(self):
        d = FullDictionary(self.dir_prefix, self.suffix)
        self.words |= set(d.keys())
        print(d.readtime)

    def add_word(self):
        self.redo.load()
        self.queue -= self.words
        self.queue -= self.snafus
        self.queue |= self.redo
        len_queue = len(self.queue)  # actual queue
        p = random.choice(list(self.queue))
        try:
            start_parsing = time.time()
            w = self.Parser(p)  # fetch new word
            word_dict = w.todict()

            print(
                f"{p}  |  "
                f"{len(self.words)} words collected, "
                f"{len_queue} words waiting in queue"
                # f", {start_db_stuff-start_parsing:.06f}s"
                # f"/{time.time() - start_db_stuff:.06f}s"
            )

            start_db_stuff = time.time()

            prefix = p[: self.prefix_length].lower()
            if unusual(prefix):
                prefix = "_" * self.prefix_length

            dict_part = DictFile(f"{self.dir_prefix}{prefix}{self.suffix}")
            dict_part |= word_dict
            dict_part.save()
            del dict_part

            self.words |= set(word_dict.keys())

            self.queue |= set(w.neighbours)
            self.queue -= {p}
            self.redo -= {p}
            self.redo.save()
            self.wait()
        except (
            AssertionError,
            ET.ParseError,
        ):
            self.queue.save()
            print("snafu... ", p)
            self.redo -= {p}
            self.redo.save()
            self.snafus |= {p}
            self.snafus.append()
            self.wait()
        except ConnectionError:
            self.queue.save()
            self.time_exponent += 1
            self.wait()


if __name__ == "__main__":
    f = FileSet("en_merriam_webster/queue")
    # d = DictFile("en_merriam_webster/abc_mw.json")
    # d.save()
    print(d)