scrapy在python爬蟲中抓取符號出錯的解決方法

這篇文章給大家分享的是有關scrapy在python爬蟲中抓取符號出錯的解決方法的內容。小編覺得挺實用的，因此分享給大家做個參考。一起跟隨小編過來看看吧。

因為抓取到的Twitter上有表情等特殊符號，在插入數據庫時會報錯，所以，這里需要對抓取的內容信息進行清洗。

TweetScraper/utils.py 文件新增filter_emoji過濾方法

import re
 
def filter_emoji(desstr, restr=''):
    """
    filter emoji
    desstr: origin str
    restr: replace str
    """
    # filter emoji
    try:
        res = re.compile(u'[\U00010000-\U0010ffff]')
    except re.error:
        res = re.compile(u'[\uD800-\uDBFF][\uDC00-\uDFFF]')
return res.sub(restr, desstr)

在 TweetCrawler.py 文件中調用該方法：

from TweetScraper.utils import filter_emoji
 
def parse_tweet_item(self, items):
        for item in items:
            try:
                tweet = Tweet()
 
                tweet['usernameTweet'] = item.xpath('.//span[@class="username u-dir u-textTruncate"]/b/text()').extract()[0]
 
                ID = item.xpath('.//@data-tweet-id').extract()
                if not ID:
                    continue
                tweet['ID'] = ID[0]
 
                ### get text content
                tweet['text'] = ' '.join(
                    item.xpath('.//div[@class="js-tweet-text-container"]/p//text()').extract()).replace(' # ',
                                                                                                        '#').replace(
                    ' @ ', '@')
 
                ### clear data[20200416]
                # tweet['text'] = re.sub(r"[\s+\.\!\/_,$%^*(+\"\')]+|[+——?【】？~@#￥%……&*]+|\\n+|\\r+|(\\xa0)+|(\\u3000)+|\\t", "", tweet['text']);
                                
                                # 過濾掉表情符號【20200417】
                tweet['text'] = filter_emoji(tweet['text'], '')
 
                if tweet['text'] == '':
                    # If there is not text, we ignore the tweet
                    continue
 
                ### get meta data
                tweet['url'] = item.xpath('.//@data-permalink-path').extract()[0]
 
                nbr_retweet = item.css('span.ProfileTweet-action--retweet > span.ProfileTweet-actionCount').xpath(
                    '@data-tweet-stat-count').extract()
                if nbr_retweet:
                    tweet['nbr_retweet'] = int(nbr_retweet[0])
                else:
                    tweet['nbr_retweet'] = 0
 
                nbr_favorite = item.css('span.ProfileTweet-action--favorite > span.ProfileTweet-actionCount').xpath(
                    '@data-tweet-stat-count').extract()
                if nbr_favorite:
                    tweet['nbr_favorite'] = int(nbr_favorite[0])
                else:
                    tweet['nbr_favorite'] = 0
 
                nbr_reply = item.css('span.ProfileTweet-action--reply > span.ProfileTweet-actionCount').xpath(
                    '@data-tweet-stat-count').extract()
                if nbr_reply:
                    tweet['nbr_reply'] = int(nbr_reply[0])
                else:
                    tweet['nbr_reply'] = 0
 
                tweet['datetime'] = datetime.fromtimestamp(int(
                    item.xpath('.//div[@class="stream-item-header"]/small[@class="time"]/a/span/@data-time').extract()[
                        0])).strftime('%Y-%m-%d %H:%M:%S')
 
                ### get photo
                has_cards = item.xpath('.//@data-card-type').extract()
                if has_cards and has_cards[0] == 'photo':
                    tweet['has_image'] = True
                    tweet['images'] = item.xpath('.//*/div/@data-image-url').extract()
                elif has_cards:
                    logger.debug('Not handle "data-card-type":\n%s' % item.xpath('.').extract()[0])
 
                ### get animated_gif
                has_cards = item.xpath('.//@data-card2-type').extract()
                if has_cards:
                    if has_cards[0] == 'animated_gif':
                        tweet['has_video'] = True
                        tweet['videos'] = item.xpath('.//*/source/@video-src').extract()
                    elif has_cards[0] == 'player':
                        tweet['has_media'] = True
                        tweet['medias'] = item.xpath('.//*/div/@data-card-url').extract()
                    elif has_cards[0] == 'summary_large_image':
                        tweet['has_media'] = True
                        tweet['medias'] = item.xpath('.//*/div/@data-card-url').extract()
                    elif has_cards[0] == 'amplify':
                        tweet['has_media'] = True
                        tweet['medias'] = item.xpath('.//*/div/@data-card-url').extract()
                    elif has_cards[0] == 'summary':
                        tweet['has_media'] = True
                        tweet['medias'] = item.xpath('.//*/div/@data-card-url').extract()
                    elif has_cards[0] == '__entity_video':
                        pass  # TODO
                        # tweet['has_media'] = True
                        # tweet['medias'] = item.xpath('.//*/div/@data-src').extract()
                    else:  # there are many other types of card2 !!!!
                        logger.debug('Not handle "data-card2-type":\n%s' % item.xpath('.').extract()[0])
 
                is_reply = item.xpath('.//div[@class="ReplyingToContextBelowAuthor"]').extract()
                tweet['is_reply'] = is_reply != []
 
                is_retweet = item.xpath('.//span[@class="js-retweet-text"]').extract()
                tweet['is_retweet'] = is_retweet != []
 
                tweet['user_id'] = item.xpath('.//@data-user-id').extract()[0]
                yield tweet
 
                if self.crawl_user:
                    ### get user info
                    user = User()
                    user['ID'] = tweet['user_id']
                    user['name'] = item.xpath('.//@data-name').extract()[0]
                    user['screen_name'] = item.xpath('.//@data-screen-name').extract()[0]
                    user['avatar'] = \
                        item.xpath('.//div[@class="content"]/div[@class="stream-item-header"]/a/img/@src').extract()[0]
                    yield user
            except:
                logger.error("Error tweet:\n%s" % item.xpath('.').extract()[0])
                # raise

通過數據清洗，現在可以正常插入到表里了。

感謝各位的閱讀！關于scrapy在python爬蟲中抓取符號出錯的解決方法就分享到這里了，希望以上內容可以對大家有一定的幫助，讓大家可以學到更多知識。如果覺得文章不錯，可以把它分享出去讓更多的人看到吧！