diff --git a/.gitignore b/.gitignore index e9bdb45..eda5486 100644 --- a/.gitignore +++ b/.gitignore @@ -3,7 +3,6 @@ raw_tweets logs insert_queue env -platform.ini *.pyc error_inserted_tweet.txt diff --git a/platform.ini b/platform.ini new file mode 100644 index 0000000..9b531bc --- /dev/null +++ b/platform.ini @@ -0,0 +1,69 @@ +; +; SoMe Tools platform ini/config file +; +; Step 1: In the same directory as "tweet_collector.py", create a plain text file named +; "collection.terms" and put one keyterm per line. +; +; Step 2: File >> Save as >> save this file as "platform.ini" - KEEP THE ORIGINAL AS BACKUP +; +; Step 3: Confirm "platform.ini" is in the same directory as "tweet_collector.py" +; +; Step 4: Visit: https://dev.twitter.com/apps/new +; +; Step 5: Login with your Twitter account +; +; Step 6: You'll be queries to "create an application" - you need to do this to obtain +; "oauth" credentials. The name, description, and website are filler - feel free to use +; whatever you like. Leave "callback URL" blank. Submit. +; +; Step 7: Change the values below: [collection] needs a name value, [files] require directory +; path values and file names (change these!), and [oauth] requires the values you obtained +; from the Twitter dev site + +[collection] +; name is a human readable word(s) for logging and reporting about the data being collected +name:billy +; this is the mongo database name +db_name:govdata +; this is the mongo collection name +collection_name:tweets + +; note: is there a way to take the file name from the collection name? so +; %(name)s_tweets_out.txt + +[files] +; this is where the raw tweets get stored from the collector +raw_tweets_file_path:./raw_tweets/ + +; this where the raw tweets and the processed tweets get sotred +tweet_archive_dir:./tweet_archive/ + +; this is where the processed tweets go while they wait to be inserted in mongo +tweet_insert_queue:./insert_queue/ + +; the date format is used to build the tweets out file name as well as indicate +; when the file should rool over. The fastest possible roll over (not reccomended) +; is seconds and is specified with %Y%m%d-%H%M%S. For testing use minutes (%Y%m%d-%H%M) +; or hours (%Y%m%d-%H), but for collection use hours or days (%Y%m%d) +tweets_file_date_frmt:%Y%m%d-%H +tweets_file:tweets_out.json +; the final constructed file name will be something like: ./tweets/20130822-1030-track-tweets_out.json + +terms_file:./collection.terms + +log_file:./tweet_collection.log +log_dir:./logs/ +log_config_file:./logging.conf + +[oauth-track] +consumer_key: lrZoHMNXE35WypwQbuccG6Ca5 +consumer_secret: weoM7VN7lx7IvkC8zx3hNbnc7itHpVMBuJvLccyqyJzyyAvTUM +access_token: 260546258-VJid9oU3QvxQ2UzfuGybtLELI45Pcb1D299wAooa +access_token_secret: KFKWONgBKUPBedtEnHDwd7xGQhR8zF5Y1ScZadW55tqoy + +[oauth-follow] +consumer_key: SqUhmu9GAEGrR9DGwOPL9nDjt +consumer_secret: xp2ezq2zE3djgcUbeCeipTfBcIRvQPryBLdSEig8hinOBY2cHF +access_token: 260546258-BWW0cVtX00CvYHulnM2F9peOn7HdUhsFHzUf1Rp9 +access_token_secret: UIQ9YTAxjnqoxSU5A6kjpMHbD5NZoqfuAaBGu71AQxkwj +