def get_timing_for_funnel(eventsfull: DataFrame, funnel: list, useResolvedUrls: bool) -> list:
"""Get a list of funnel step times (amounts of time users spend before navigating to next step) for a funnel
:param eventsfull: full events DataFrame (that includes non-navigate events)
:param funnel: funnel of interest
:param useResolvedUrls: indicates whether original or resolved URLs should be used
:return: list of funnel step times for each step
"""
funneltimes = []
for i in range(len(funnel)):
funneltimes.append([])
events = analyze_clicks.remove_non_navigation(eventsfull)
if useResolvedUrls:
columnToUse = analyze_traffic.RESOLVEDURL
else:
columnToUse = analyze_traffic.PAGEURL
if useResolvedUrls:
url_regex_resolver.resolve_urls(events, manage_resolutions.get_regex_dict(), analyze_traffic.PAGEURL, analyze_traffic.RESOLVEDURL)
si = analyze_traffic.build_session_index(events, columnToUse)
sessFound = analyze_traffic.get_unordered_sessions_for_funnel(si, funnel)
sessOrdered = analyze_traffic.get_sessions_with_ordered(events, sessFound, funnel, columnToUse, strict=True)
for sid in sessOrdered:
sess_df = events.loc[sid]
indices = analyze_traffic.get_sublist_indices(funnel, sess_df[columnToUse].tolist(), True)
for index in indices:
timestamps = []
timespent = []
for i in range(len(funnel)):
timestamps.append(sess_df.iloc[index + i].loc[EVENTSTART])
if i > 0:
delta = timestamps[i] - timestamps[i - 1]
funneltimes[i-1].append(delta.total_seconds())
return funneltimes
def get_funnel_stats(events: DataFrame,
funnel: list,
useResolvedUrls: bool,
limit_rows: int = 0) -> list:
"""Get conversion statistics for a funnel
:param events: events DataFrame
:param funnel: funnel of interest
:param useResolvedUrls: indicates whether original or resolved URLs should be used
:param limit_rows: number of rows of events DataFrame to use (use all rows if 0)
:return: sorted list of funnel conversions by step
"""
if useResolvedUrls:
columnToUse = analyze_traffic.RESOLVEDURL
else:
columnToUse = analyze_traffic.PAGEURL
if limit_rows != 0:
events = events.head(limit_rows)
if useResolvedUrls:
url_regex_resolver.resolve_urls(events,
manage_resolutions.get_regex_dict(),
analyze_traffic.PAGEURL,
analyze_traffic.RESOLVEDURL)
si = analyze_traffic.build_session_index(events, columnToUse)
funnelCounts = analyze_traffic.get_funnel_conversion_stats(
events, si, funnel, columnToUse)
funnelCounts = list(funnelCounts)
return funnelCounts
def get_sessions_for_funnel(
events: pd.DataFrame,
funnel: list,
useResolvedUrls: bool,
OrgId: str = None,
is_staging: bool = False,
strict: bool = True,
numSessions: int = 0,
) -> list:
"""Get a list of sessions where each session contains the specified funnel
:param events: events DataFrame
:param funnel: funnel of interest
:param useResolvedUrls: indicates whether original or resolved URLs should be used
:param OrgId: FullStory OrgId for the organization
:param is_staging: set to True if FullStory staging environment should be used (for debugging purposes)
:param strict: If `True`, the session has to follow the funnel steps in exact order (with no diversions between the steps). The `False` option is currently not supported.
:param numSessions: number of sessions to return (if 0, return all available)
:return: list of session URLs
"""
if useResolvedUrls:
columnToUse = RESOLVEDURL
else:
columnToUse = PAGEURL
if useResolvedUrls:
url_regex_resolver.resolve_urls(events,
manage_resolutions.get_regex_dict(),
PAGEURL, RESOLVEDURL)
sids = build_and_get_sids_for_funnel(events, funnel, columnToUse, strict)
if numSessions != 0:
sids = sids[:numSessions]
sessions = list(map(lambda p: get_session_link(p, OrgId, is_staging),
sids))
return sessions
def get_in_outs(events: DataFrame, funnel: list, useResolvedUrls: bool, limit_rows: int = 0) -> (dict, dict):
"""Get information about inflows and outflows for a funnel
:param events: events DataFrame
:param funnel: funnel of interest
:param useResolvedUrls: indicates whether original or resolved URLs should be used
:param limit_rows: number of rows of events DataFrame to use (use all rows if 0)
:return: a pair of dictionaries, with inflow and outflow URL frequency counts
"""
if useResolvedUrls:
columnToUse = analyze_traffic.RESOLVEDURL
else:
columnToUse = analyze_traffic.PAGEURL
if limit_rows != 0:
events = events.head(limit_rows)
if useResolvedUrls:
url_regex_resolver.resolve_urls(events, manage_resolutions.get_regex_dict(), analyze_traffic.PAGEURL, analyze_traffic.RESOLVEDURL)
si = analyze_traffic.build_session_index(events, columnToUse)
ingressCounts, egressCounts = analyze_traffic.get_funnel_in_outs(events, si, funnel, columnToUse, analyze_traffic.REFERAL)
return ingressCounts, egressCounts
def get_top_funnels_df(funurl: str, funlen: int, useResolvedUrls: bool, events: DataFrame, limit_rows: int = 0) -> dict:
"""Get top funnels of specified length which contain the specified URL
:param funurl: URL that should be contained in the funnel
:param funlen: funnel length
:param useResolvedUrls: indicates whether original or resolved URLs should be used
:param events: events DataFrame
:param limit_rows: number of rows of events DataFrame to use (use all rows if 0)
:return: dictionary of funnels and their frequencies
"""
if useResolvedUrls:
columnToUse = analyze_traffic.RESOLVEDURL
else:
columnToUse = analyze_traffic.PAGEURL
if limit_rows != 0:
events = events.head(limit_rows)
if useResolvedUrls:
url_regex_resolver.resolve_urls(events, manage_resolutions.get_regex_dict(), analyze_traffic.PAGEURL, analyze_traffic.RESOLVEDURL)
si = analyze_traffic.build_session_index(events, columnToUse)
funnelCounts = get_funnel_lists(events, si, funurl, funlen, columnToUse)
return funnelCounts
def get_popular(events: pd.DataFrame,
useResolvedUrls: bool,
limit_rows: int = 0) -> dict:
"""Returns a dictionary of visited URLs and visit counts for each URL
:param events: events DataFrame
:param useResolvedUrls: boolean indicating whether original or resolved URLs should be used
:param limit_rows: number of rows from the original DataFrame to use (if 0, then use entire DataFrame)
:return:
"""
if useResolvedUrls:
columnToUse = analyze_traffic.RESOLVEDURL
else:
columnToUse = analyze_traffic.PAGEURL
if limit_rows != 0:
events = events.head(limit_rows)
if useResolvedUrls:
url_regex_resolver.resolve_urls(events,
manage_resolutions.get_regex_dict(),
analyze_traffic.PAGEURL,
analyze_traffic.RESOLVEDURL)
si = analyze_traffic.build_session_index(events, columnToUse)
urlCounts = analyze_traffic.get_counts_for_url(si)
return urlCounts