Python替换/删除字符串中的所有网址

Question

I am trying to replace all URLs with an empty string in a string. Following JSON is a string . it's not an object. But I have difficulty capturing various permutations.

This is my python script. However if you look at https://regex101.com/r/r6tQ3B/2/ you notice that regex also removes ending " and also doesn't really capture shorthand "t.co" or the urls in the middle.

for filename in dataFiles:
    path = 'data/' + filename
    with open(path) as r:
        text = re.sub(r'https?:\/\/\S*', '"', text, flags=re.MULTILINE)
    with open(path, "w") as w:
        w.write(text)

Test: https://regex101.com/r/r6tQ3B/1/

{
   "created_at":"Fri Aug 12 10:04:00 +0000 2016",
   "id":764039724818272256,
   "text":"@theblaze https://t.com/TY9DlZ584c @realDonaldTrump https://t.com/TY9DlZ584c",
   "in_reply_to_screen_name":"theblaze",
   "source":"<a href=\"http://twitter.com/download/iphone\" rel=\"nofollow\">Twitter for iPhone</a>",
   "user":{
      "id":366636488,
      "id_str":"366636488",
      "name":"GIL DUPUY",
      "screen_name":"DUPUY77",
      "location":"Miami",
      "url":"http://ggm-dupuy.com",
      "description":"Fashion photographer, love action and adventure, care for the less fortunate, don't tolerate any kind of racism regardless of race or religion",
      "verified":false,
      "followers_count":186,
      "friends_count":446,
      "utc_offset":null,
      "time_zone":null,
      "lang":"en",
      "default_profile_image":false,
      "following":null,
      "notifications":null
   },
   "geo":null,
   "coordinates":null,
   "place":{
      "name":"Frontenac",
      "full_name":"Frontenac, MO",
      "country_code":"US",
      "country":"United States",
      "attributes":{
         
      }
   },
   "retweet_count":0,
   "favorite_count":0,
   "extended_entities":{
      "media":[
         {
            "id":764039718237409281,
            "id_str":"764039718237409281",
            "indices":[
               27,
               50
            ],
            "media_url":"http://pbs.twimg.com/media/CppqE1_UkAE2qFj.jpg",
            "media_url_https":"https://pbs.twimg.com/media/CppqE1_UkAE2qFj.jpg",
            "url":"https://t.com/TY9DlZ584c",
            "display_url":"pic.twitter.com/TY9DlZ584c",
            "expanded_url":"http://twitter.com/DUPUY77/status/764039724818272256/photo/1",
            "type":"photo",
            "sizes":{
               "medium":{
                  "w":640,
                  "h":1136,
                  "resize":"fit"
               },
               "large":{
                  "w":640,
                  "h":1136,
                  "resize":"fit"
               },
               "thumb":{
                  "w":150,
                  "h":150,
                  "resize":"crop"
               },
               "small":{
                  "w":383,
                  "h":680,
                  "resize":"fit"
               }
            }
         }
      ]
   },
   "favorited":false,
   "retweeted":false,
   "possibly_sensitive":false,
   "lang":"und"
}

Answer 1

Try this pattern \\s?(https?:\\/\\/[^\\\\\\s"]*)

Not very clean, but work for your example.

Answer 2

Removes all urls in "", all urls without "", and urls starting with pic.twitter (these seem the only ones without http(s)).
Assumes, there is no whitespace or " in the url: r"(?:https?:\\/\\/|pic\\.twitter)[^\\s\\"\\\\]*"