def check_tweet(tweet, do_format_checking=False):
# get the format of the Tweet & make sure it's probably a Tweet
original_format = is_original_format(tweet)
# make sure, to the best of our knowledge, that the Tweet is a Tweet
if "id" not in tweet:
raise (NotATweetError(
"This text has no 'id' key, it's probably not a Tweet"))
if original_format:
# check to see if it's not a Tweet at all
if "user" not in tweet:
raise (NotATweetError(
"This text has no 'user' key, it's probably not a Tweet"))
if "text" not in tweet:
raise (NotATweetError(
"This text has no 'text' key, it's probably not a Tweet"))
# check for changing keys
if do_format_checking:
check_format(get_all_keys(tweet), original_format_superset_keys,
original_format_minimum_set_keys)
else:
# check to see if it's not a Tweet at all
if "actor" not in tweet:
raise (NotATweetError(
"This text has no 'actor' key, it's probably not a Tweet"))
if "body" not in tweet:
raise (NotATweetError(
"This text has no 'body' key, it's probably not a Tweet"))
#check for changing keys
if do_format_checking:
check_format(get_all_keys(tweet), activity_streams_superset_keys,
activity_streams_minimum_set_keys)
return original_format
def _check_original_format_tweet(tweet, validation_checking=False):
for key in ["user"]:
if key not in tweet:
raise NotATweetError("This dict has no '{}' key".format(key))
if "text" not in tweet and "full_text" not in tweet:
raise NotATweetError(
"This dict has no 'text' or 'full_text' key".format())
# check for changing keys
if validation_checking:
_ = key_validation_check(get_all_keys(tweet),
original_format_superset_keys,
original_format_minimum_set_keys)
def is_original_format(tweet):
"""
Simple checker to flag the format of a tweet.
Args:
tweet (Tweet): tweet in qustion
Returns:
Bool
Example:
>>> import tweet_parser.tweet_checking as tc
>>> tweet = {"created_at": 124125125125,
... "text": "just setting up my twttr",
... "nested_field": {"nested_1": "field", "nested_2": "field2"}}
>>> tc.is_original_format(tweet)
True
"""
# deleted due to excess checking; it's a key lookup and does not need any
# operational optimization
if "created_at" in tweet:
original_format = True
elif "postedTime" in tweet:
original_format = False
else:
raise NotATweetError(
"This dict has neither 'created_at' or 'postedTime' as keys")
return original_format
def _check_activity_streams_tweet(tweet, validation_checking=False):
for key in ["actor", "body"]:
if key not in tweet:
raise NotATweetError("This dict has no '{}' key".format(key))
# check for changing keys
if validation_checking:
_ = key_validation_check(get_all_keys(tweet),
activity_streams_superset_keys,
activity_streams_minimum_set_keys)
def embedded_tweet(self):
"""
get the quote tweet or the retweet and return a tweet object of it
"""
embedded_tweet = tweet_embeds.get_embedded_tweet(self)
if embedded_tweet is not None:
try:
return Tweet(embedded_tweet)
except NotATweetError as nate:
raise (NotATweetError(
"The embedded tweet payload {} appears malformed. \nFailed with '{}'"
.format(embedded_tweet, nate)))
def retweet(self):
"""
get the retweet and return a tweet obj of the retweet
"""
retweet = tweet_embeds.get_retweet(self)
if retweet is not None:
try:
return Tweet(retweet)
except NotATweetError as nate:
raise (NotATweetError(
"The retweet payload appears malformed. Failed with '{}'".
format(nate)))
def is_original_format(tweet):
"""
helper function to categorize the format of a tweet
"""
if hasattr(tweet, "original_format"):
return tweet.original_format
else:
if "created_at" in tweet:
original_format = True
elif "postedTime" in tweet:
original_format = False
else:
raise (NotATweetError(
"This text has neither 'created_at' or 'postedTime' as keys, it's not a Tweet"
))
return original_format
def embedded_tweet(self):
"""
Get the retweeted Tweet OR the quoted Tweet and return it as a Tweet object
Returns:
Tweet (or None, if the Tweet is neither a quote tweet or a Retweet):
a Tweet representing the quote Tweet or the Retweet
(see tweet_embeds.get_embedded_tweet, this is that value as a Tweet)
Raises:
NotATweetError: if embedded tweet is malformed
"""
embedded_tweet = tweet_embeds.get_embedded_tweet(self)
if embedded_tweet is not None:
try:
return Tweet(embedded_tweet)
except NotATweetError as nate:
raise(NotATweetError("The embedded tweet payload {} appears malformed." +
" Failed with '{}'".format(embedded_tweet, nate)))
else:
return None
def check_tweet(tweet, validation_checking=False):
"""
Ensures a tweet is valid and determines the type of format for the tweet.
Args:
tweet (dict/Tweet): the tweet payload
validation_checking (bool): check for valid key structure in a tweet.
"""
if "id" not in tweet:
raise NotATweetError("This text has no 'id' key")
original_format = is_original_format(tweet)
if original_format:
_check_original_format_tweet(tweet,
validation_checking=validation_checking)
else:
_check_activity_streams_tweet(tweet,
validation_checking=validation_checking)
return original_format
def retweeted_tweet(self):
"""
The retweeted Tweet as a Tweet object
If the Tweet is not a Retweet, return None
If the Retweet payload cannot be loaded as a Tweet, this will
raise a `NotATweetError`
Returns:
Tweet: A Tweet representing the retweeted status (or None)
(see tweet_embeds.get_retweet, this is that value as a Tweet)
Raises:
NotATweetError: if retweeted tweet is malformed
"""
retweet = tweet_embeds.get_retweeted_tweet(self)
if retweet is not None:
try:
return Tweet(retweet)
except NotATweetError as nate:
raise(NotATweetError("The retweet payload appears malformed." +
" Failed with '{}'".format(nate)))
else:
return None
