Skip to content

wind2sing/aCrawler

Folders and files

NameName
Last commit message
Last commit date

Latest commit

Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 

Repository files navigation

aCrawler

PyPI Documentation Status

πŸ” A powerful web-crawling framework, based on aiohttp.

Feature

  • Write your crawler in one Python script with asyncio
  • Schedule task with priority, fingerprint, exetime, recrawl...
  • Middleware: add handlers before or after task's execution
  • Simple shortcuts to speed up scripting
  • Parse html conveniently with Parsel
  • Parse with rules and chained processors
  • Support JavaScript/browser-automation with pyppeteer
  • Stop and Resume: crawl periodically and persistently
  • Distributed work support with Redis

Installation

To install, simply use pip:

$ pip install acrawler

(Optional)
$ pip install uvloop      #(only Linux/macOS, for faster asyncio event loop)
$ pip install aioredis    #(if you need Redis support)
$ pip install motor       #(if you need MongoDB support)
$ pip install aiofiles    #(if you need FileRequest)

Documentation

Documentation and tutorial are available online at https://acrawler.readthedocs.io/ and in the docs directory.

Sample Code

Scrape imdb.com

from acrawler import Crawler, Request, ParselItem, Handler, register, get_logger


class MovieItem(ParselItem):
   log = True
   css = {
      # just some normal css rules
      # see Parsel for detailed information
      "date": ".subtext a[href*=releaseinfo]::text",
      "time": ".subtext time::text",
      "rating": "span[itemprop=ratingValue]::text",
      "rating_count": "span[itemprop=ratingCount]::text",
      "metascore": ".metacriticScore span::text",

      # if you provide a list with additional functions,
      # they are considered as field processor function
      "title": ["h1::text", str.strip],

      # the following four fules is for getting all matching values
      # the rule starts with [ and ends with ] comparing to normal rules
      "genres": "[.subtext a[href*=genres]::text]",
      "director": "[h4:contains(Director) ~ a[href*=name]::text]",
      "writers": "[h4:contains(Writer) ~ a[href*=name]::text]",
      "stars": "[h4:contains(Star) ~ a[href*=name]::text]",
   }


class IMDBCrawler(Crawler):
   config = {"MAX_REQUESTS": 4, "DOWNLOAD_DELAY": 1}

   async def start_requests(self):
      yield Request("https://www.imdb.com/chart/moviemeter", callback=self.parse)

   def parse(self, response):
      yield from response.follow(
            ".lister-list tr .titleColumn a::attr(href)", callback=self.parse_movie
      )

   def parse_movie(self, response):
      url = response.url_str
      yield MovieItem(response.sel, extra={"url": url.split("?")[0]})


@register()
class HorrorHandler(Handler):
   family = "MovieItem"
   logger = get_logger("horrorlog")

   async def handle_after(self, item):
      if item["genres"] and "Horror" in item["genres"]:
            self.logger.warning(f"({item['title']}) is a horror movie!!!!")


@MovieItem.bind()
def process_time(value):
   # a self-defined field processing function
   # process time to minutes
   # '3h 1min' -> 181
   if value:
      res = 0
      segs = value.split(" ")
      for seg in segs:
            if seg.endswith("min"):
               res += int(seg.replace("min", ""))
            elif seg.endswith("h"):
               res += 60 * int(seg.replace("h", ""))
      return res
   return value


if __name__ == "__main__":
   IMDBCrawler().run()

Scrape quotes.toscrape.com

# Scrape quotes from http://quotes.toscrape.com/
from acrawler import Parser, Crawler, ParselItem, Request


logger = get_logger("quotes")


class QuoteItem(ParselItem):
   log = True
   default = {"type": "quote"}
   css = {"author": "small.author::text"}
   xpath = {"text": ['.//span[@class="text"]/text()', lambda s: s.strip("β€œ")[:20]]}


class AuthorItem(ParselItem):
   log = True
   default = {"type": "author"}
   css = {"name": "h3.author-title::text", "born": "span.author-born-date::text"}

class QuoteCrawler(Crawler):

   main_page = r"quotes.toscrape.com/page/\d+"
   author_page = r"quotes.toscrape.com/author/.*"
   parsers = [
      Parser(
            in_pattern=main_page,
            follow_patterns=[main_page, author_page],
            item_type=QuoteItem,
            css_divider=".quote",
      ),
      Parser(in_pattern=author_page, item_type=AuthorItem),
   ]

   async def start_requests(self):
      yield Request(url="http://quotes.toscrape.com/page/1/")


if __name__ == "__main__":
   QuoteCrawler().run()

See examples.

Todo

  • Replace parsel with parselx
  • clean redundant handlers
  • Cralwer's name for distinguishing
  • Use dynaconf as configuration manager
  • Add delta_key support for request
  • Monitor all crawlers in web
  • Write detailed Documentation
  • Testing

About

πŸ” A powerful web-crawling framework, based on aiohttp.

Topics

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Languages