working. Now starting with user loop

extract-twitter-accounts.R

@@ -18,18 +18,6 @@ acc_url <- "http://www.bundestwitter.de/api/politiker"
 #acc_json <- readLines("politiker.txt")
 acc_df <- fromJSON(acc_url)
 
-# --------------
-
-
-
-
-url   <- "https://api.twitter.com/1.1/statuses/user_timeline.json";
-query <- c(screen_name="mxmehl", include_rts=1, count="200", exclude_replies="true", trim_user="true", include_entities="false");
-
-as.character(as.numeric(result$id_str[nrow(result)]) - 1)
-str_detect(result$created_at[nrow(result)], "2013$")
-
-result <- fromJSON(correctJSON(twitter_api_call(url, query, params)))
 
 # ---------------
 # http://www.joyofdata.de/blog/twitters-rest-api-v1-1-with-r-for-linux-and-windows/
@@ -53,6 +41,7 @@ max_id <- "999999999999999999"
 loop <- 1
 keep <- c("created_at", "id_str", "text", "retweet_count")
 rm(tweets_full)
+last_id <- NULL
 repeat {
   # Define specific search query
   query <- c(include_rts=1, exclude_replies="true", trim_user="true", include_entities="false", 
@@ -75,6 +64,10 @@ repeat {
     tweets_full <- tweets_full[keep]
   }
   
+  # Now sleep 3 second to dodge 300queries/15min limit
+  cat("User:",user,"in loop:",loop,"- now waiting 3 secs...\n")
+  Sys.sleep(3)
+  
   # Is the last tweet in tweets_full from 2013?
   status <- str_detect(tweets_full$created_at[nrow(tweets_full)], "2013$")
   # Last loop is reached. Now clear the data frame
@@ -100,8 +93,9 @@ repeat {
   
   # The last tweet is still from 2014, so we need another loop
   else {
-    max_id <- as.character(as.numeric(tweets_full$id_str[nrow(tweets_full)]) - 2)
-    loop <- loop + 1
+    # Setting max_id to gather next 200 tweets
+    max_id <- tweets_full$id_str[nrow(tweets_full)]
+    loop <- loop + 1  # just for stats
   }
 }