def get(
self,
sort: str = "date",
dates: bool = True,
output: Union[str, None] = None,
) -> str:
"""
Exposed function for getting data on our Facebook friends.
@param sort: the column we want to sort by.
Can be either of {date|name}. Default is `dates`.
@param dates: boolean flag on do we want the dates column.
Default is `True`.
@param output: where do we want to write the return value,
can be any of: {csv|json|/some/path.{json|csv}}.
@return: either the data formatted as csv or json,
or a success message about where was the data saved.
"""
data = self.data
if sort == "name":
data = data.sort_values(by="name")
if not dates:
data.reset_index(drop=True, inplace=True)
return utils.df_to_file(output, data)
def get(
self,
kind: str = "private",
channels: Union[str, List[str]] = None,
cols: List[str] = None,
output: str = "csv",
):
"""
Conversations.
@param kind: one of private or group,
depending on which messaging do you want to analyze.
@param channels: channel names you want to filter for.
@param cols: column names you want to include in the output.
@param output: where do we want to write the return value,
can be any of: {csv|json|/some/path.{json|csv}}.
@return: either the data formatted as csv or json,
or a success message about where was the data saved.
"""
data = getattr(self, kind)
filtered_channels = (data.keys() if channels is None else
list(set(data.keys()) & set(channels)))
if not filtered_channels:
return (f"Could not filter for these channels: {channels}. "
f"Please double-check the channel names.")
res = utils.stack_dfs(
*[data.get(channel).data for channel in filtered_channels])
filtered_cols = (list(res.columns) if cols is None else
[col for col in cols if col in res.columns])
if not filtered_cols:
return (f"Could not filter for these columns: {cols}. "
f"Please double-check the column names. Hint: if cols "
f"is None, all columns will be returned. ")
return utils.df_to_file(output, res[filtered_cols])
def test_df_to_file(self, sample_df):
path = f"{os.path.dirname(os.path.realpath(__file__))}/test.csv"
res = utils.df_to_file(path, sample_df)
assert res == f"Data was written to {path}"
os.unlink(path)
path = f"{os.path.dirname(os.path.realpath(__file__))}/test.json"
res = utils.df_to_file(path, sample_df)
assert res == f"Data was written to {path}"
os.unlink(path)
path = "csv"
res = utils.df_to_file(path, sample_df)
assert isinstance(res, str)
path = "json"
res = utils.df_to_file(path, sample_df)
assert isinstance(res, str)
assert isinstance(json.loads(res), dict)
res = utils.df_to_file("/home/whatever/a/gibberish", sample_df)
assert res == "Directory does not exist: `/home/whatever/a`"
def get(self, output: Union[str, None] = None) -> str:
"""
Exposed function for getting data on people.
@param output: where do we want to write the return value,
an be any of: {csv|json|/some/path.{json|csv}}.
@return: either the data formatted as csv or json,
or a success message about where was the data saved.
"""
name = [person.name for person in self.data.values()]
friend = [person.friend for person in self.data.values()]
message_dir = [person.thread_path for person in self.data.values()]
media_dir = [person.media_dir for person in self.data.values()]
df = pd.DataFrame({
"name": name,
"friend": friend,
"message_dir": message_dir,
"media_dir": media_dir,
})
return utils.df_to_file(output, df)
def wrapper(*args, **kwargs: Any):
res = func(*args, **kwargs)
if not len(res):
return
return utils.df_to_file(kwargs.get("output"), res)