def make_users_events_table(event_col_names, all_cols):
"""event_col_names is a list of all the column names that represent events
(these can be alpha or numeric). This function assumes that all_cols has
already been sorted, merged, had voided records excluded, etc."""
event_cols = all_cols.select(*event_col_names)
rows = iterate_user_events(all_cols.user_id, event_cols)
return ColumnGroup.from_rows(["user_id", "event_id", "value"], rows)
def make_users_events_table(event_col_names, all_cols):
"""event_col_names is a list of all the column names that represent events
(these can be alpha or numeric). This function assumes that all_cols has
already been sorted, merged, had voided records excluded, etc."""
event_cols = all_cols.select(*event_col_names)
rows = iterate_user_events(all_cols.user_id, event_cols)
return ColumnGroup.from_rows(["user_id", "event_id", "value"], rows)
def make_events_table(event_headers):
"""The column names are of the format "Friday Dinner #123123". We want to
take all the column names and make a table of event ids and descriptions"""
events = [
parse_event_header(header) for header in event_headers
if is_event_header(header)
]
return ColumnGroup.from_rows(["event_id", "name"], sorted(events))
def non_harris_event_cols(class_year, num_rows):
"""Add columns for non-Harris events. These can be auto-populated if Harris
package events include them (look in config.py). They are initialized to all
blank."""
blank_col = DataColumn.init_with(num_rows, '')
return ColumnGroup([
(format_as_event_header(event_id, "NH"), blank_col)
for event_id in config.non_harris_events_for_year(class_year)
])
def parse_doc(infile):
return ColumnGroup.from_csv(infile,
delimiter="\t",
force_unique_col_names=True,
encoding="latin-1")
def make_events_table(event_headers):
"""The column names are of the format "Friday Dinner #123123". We want to
take all the column names and make a table of event ids and descriptions"""
events = [parse_event_header(header)
for header in event_headers if is_event_header(header)]
return ColumnGroup.from_rows(["event_id", "name"], sorted(events))
def parse_doc(infile):
return ColumnGroup.from_csv(infile,
delimiter="\t",
force_unique_col_names=True,
encoding="latin-1")
