我正在使用Tweepy库(Python)和Kafka从Twitter检索推文.该文本以UTF-8编码,如下所示：

self.producer.send('my-topic', data.encode('UTF-8'))

其中“数据”是字符串.然后,此数据以键值格式存储到Oracle NoSQL数据库中.因此,推文本身已被编码.我使用Java执行此操作：

Value myValue = Value.createValue(msg.value().getBytes("UTF-8"));

最后,这些推文由用Java开发的Formatter检索.为了将其存储在关系模式中,我必须解析该tweet,以便将其检索为String.

String data = new String(value.toByteArray(),StandardCharsets.UTF_8);

如您所见,我在执行的所有步骤中都维护UTF-8编码.但是,当我在数据库中看到推文时,它总是被剪切.例如：

RT @briIIohead: the hardest pill i had to swallow this year was learning that no matter how good you could be to somebody, no matter how mu?

注意它的结尾是“？”符号,并且已被明确切割.好吧,每条漫长的推文都会发生这种情况.我的意思是,如果文本长约30个字符,则可以正常显示,但是任何超过100个左右的内容都会被剪切掉.

起初,我以为可能是我的表定义,但是字段“ Text”被声明为VARCHAR2(400 CHAR),这是一条推文在社交网络中可以包含的最大字符数.

关于如何发现剪切文本和插入“？”的任何想法符号结尾？

“数据”的外观如下：

{"created_at":"Tue May 28 09:23:36 +0000 2019","id":1133302792129351681,"id_str":"1133302792129351681","text":"RT @AppleEDU: Learn, create, and do more with iPad in your classroom. Get the free Everyone Can Create curriculum and bring projects to lif\u2026","source":"Twitter for iPhone<\/a>","truncated":false,"in_reply_to_status_id":null,"in_reply_to_status_id_str":null,"in_reply_to_user_id":null,"in_reply_to_user_id_str":null,"in_reply_to_screen_name":null,"user":{"id":1060510851889750022,"id_str":"1060510851889750022","name":"Rem.0112","screen_name":"0112Rem","location":"Mawson Lakes, Adelaide","url":null,"description":null,"translator_type":"none","protected":false,"verified":false,"followers_count":739,"friends_count":1853,"listed_count":10,"favourites_count":33406,"statuses_count":36936,"created_at":"Thu Nov 08 12:34:25 +0000 2018","utc_offset":null,"time_zone":null,"geo_enabled":true,"lang":null,"contributors_enabled":false,"is_translator":false,"profile_background_color":"F5F8FA","profile_background_image_url":"","profile_background_image_url_https":"","profile_background_tile":false,"profile_link_color":"1DA1F2","profile_sidebar_border_color":"C0DEED","profile_sidebar_fill_color":"DDEEF6","profile_text_color":"333333","profile_use_background_image":true,"profile_image_url":"http:\/\/pbs.twimg.com\/profile_images\/1093157842163355649\/6oAdJTCs_normal.jpg","profile_image_url_https":"https:\/\/pbs.twimg.com\/profile_images\/1093157842163355649\/6oAdJTCs_normal.jpg","profile_banner_url":"https:\/\/pbs.twimg.com\/profile_banners\/1060510851889750022\/1546155144","default_profile":true,"default_profile_image":false,"following":null,"follow_request_sent":null,"notifications":null},"geo":null,"coordinates":null,"place":null,"contributors":null,"retweeted_status":{"created_at":"Thu May 23 15:15:16 +0000 2019","id":1131579354964725760,"id_str":"1131579354964725760","text":"Learn, create, and do more with iPad in your classroom. Get the free Everyone Can Create curriculum and bring proje\u2026 https:\/\/t.co\/aeeSPTXtFx","source":"Twitter Ads Composer<\/a>","truncated":true,"in_reply_to_status_id":null,"in_reply_to_status_id_str":null,"in_reply_to_user_id":null,"in_reply_to_user_id_str":null,"in_reply_to_screen_name":null,"user":{"id":468741166,"id_str":"468741166","name":"Apple Education","screen_name":"AppleEDU","location":"Cupertino, CA","url":null,"description":"Spark new ideas, create more aha moments, and teach in ways you\u2019ve always imagined. Follow @AppleEDU for tips, updates, and inspiration.","translator_type":"none","protected":false,"verified":true,"followers_count":728781,"friends_count":273,"listed_count":2594,"favourites_count":13189,"statuses_count":2766,"created_at":"Thu Jan 19 21:26:14 +0000 2012","utc_offset":null,"time_zone":null,"geo_enabled":false,"lang":null,"contributors_enabled":false,"is_translator":false,"profile_background_color":"F0F0F0","profile_background_image_url":"http:\/\/abs.twimg.com\/images\/themes\/theme1\/bg.png","profile_background_image_url_https":"https:\/\/abs.twimg.com\/images\/themes\/theme1\/bg.png","profile_background_tile":false,"profile_link_color":"0088CC","profile_sidebar_border_color":"C0DEED","profile_sidebar_fill_color":"DDEEF6","profile_text_color":"333333","profile_use_background_image":false,"profile_image_url":"http:\/\/pbs.twimg.com\/profile_images\/892429342046691328\/2SOlm_09_normal.jpg","profile_image_url_https":"https:\/\/pbs.twimg.com\/profile_images\/892429342046691328\/2SOlm_09_normal.jpg","profile_banner_url":"https:\/\/pbs.twimg.com\/profile_banners\/468741166\/1530123538","default_profile":false,"default_profile_image":false,"following":null,"follow_request_sent":null,"notifications":null},"geo":null,"coordinates":null,"place":null,"contributors":null,"is_quote_status":false,"extended_tweet":{"full_text":"Learn, create, and do more with iPad in your classroom. Get the free Everyone Can Create curriculum and bring projects to life through music, drawing, video and photography.","display_text_range":[0,173],"entities":{"hashtags":[],"urls":[],"user_mentions":[],"symbols":[]}},"quote_count":0,"reply_count":3,"retweet_count":3,"favorite_count":58,"entities":{"hashtags":[],"urls":[{"url":"https:\/\/t.co\/aeeSPTXtFx","expanded_url":"https:\/\/twitter.com\/i\/web\/status\/1131579354964725760","display_url":"twitter.com\/i\/web\/status\/1\u2026","indices":[117,140]}],"user_mentions":[],"symbols":[]},"favorited":false,"retweeted":false,"scopes":{"followers":false},"filter_level":"low","lang":"en"},"is_quote_status":false,"quote_count":0,"reply_count":0,"retweet_count":0,"favorite_count":0,"entities":{"hashtags":[],"urls":[],"user_mentions":[{"screen_name":"AppleEDU","name":"Apple Education","id":468741166,"id_str":"468741166","indices":[3,12]}],"symbols":[]},"favorited":false,"retweeted":false,"filter_level":"low","lang":"en","timestamp_ms":"1559035416048"}

我还必须提到,这整个块都是经过编码的.然后解码,最后解析为要引入数据库.除剪切的“文本”外,所有字段均已正确解码和解析

解决方法:

根据official文档,一条推文最多包含“ 140”个字符(即广泛定义)；但最近他们将其更改为280.

该文件说：

Twitter counts the length of a Tweet using the Normalization Form C (NFC) version of the text.

因此,他们首先对文本进行了规范化(我将让您找出Java是如何实现的).后来他们说：

Twitter also counts the number of codepoints in the text rather than UTF-8 bytes.

从而：

String test = "RT @briIIohead: the hardest pill i had to swallow this year was learning that no matter how good you could be to somebody, no matter how mu";
System.out.println(test.codePoints().count()); // 139

最初的推文似乎是280个“字符”,而您使用的库对此并不了解,因此仅使用了前140个.由于这样做会进行一些分块,因此分块似乎也是错误的,它也会在末尾删除一些“部分”字节.当您尝试打印这些字符时-java不知道那些字节(末尾)的实际含义(由于某些错误的分块),只是简单地说？ (这是在根本不了解某些内容时显示的默认策略).