def export_sample_messages_csv(messages, consent_withdrawn_field, analysis_configurations, f,
filter_code_ids=None, limit_per_code=sys.maxsize):
"""
Computes the sample messages and exports them to a CSV.
:param messages: Objects to sample the messages from.
:type messages: iterable of core_data_modules.traced_data.TracedData
:param consent_withdrawn_field: Field in each messages object which records if consent is withdrawn.
:type consent_withdrawn_field: str
:param analysis_configurations: Configurations for the datasets to include in the sample_messages.
:type analysis_configurations: iterable of core_data_modules.analysis.AnalysisConfiguration
:param f: File to write the sample_messages CSV to.
:type f: file-like
:param filter_code_ids: The code ids to sample messages for, or None. If None, exports sample messages for all
code ids.
:type filter_code_ids: list of str | None
:param limit_per_code: The maximum number of sample messages to export per code.
Defaults to sys.maxsize, i.e. to no practical limit.
:type limit_per_code: int
"""
analysis_utils.write_csv(
compute_sample_messages(messages, consent_withdrawn_field, analysis_configurations,
filter_code_ids, limit_per_code),
sample_messages_keys,
f
)
def export_theme_distributions_csv(individuals, consent_withdrawn_field,
theme_configurations, breakdown_configurations, f):
"""
Computes the theme_distributions and exports them to a CSV.
The CSV will contain the headers:
- Dataset, set to the dataset_name in each of the `theme_configurations`, de-duplicated for clarity).
- Theme, set to {dataset_name}_{code.string_value}, for each code in each theme_configuration.
- Total Participants
- Total Participants %
and raw total and % headers for each scheme and code in the `breakdown_configurations`.
:param individuals: Individuals to compute the theme_distributions for.
:type individuals: iterable of core_data_modules.traced_data.TracedData
:param consent_withdrawn_field: Field in each individuals object which records if consent is withdrawn.
:type consent_withdrawn_field: str
:param theme_configurations: Configuration for the themes. Each configuration should contain:
- dataset_name. This will be used to index the returned theme_distributions dict
(the "theme dataset_name" above).
- code_scheme. This will be used to index each themes dictionary in the returned
theme_distributions dict (the "theme" above). Themes are formatted as
{theme_configuration.dataset_name}_{code.string_value} for each code in the
code_scheme.
:type theme_configurations: iterable of core_data_modules.analysis.AnalysisConfiguration
:param breakdown_configurations: Configuration for the breakdowns dict.
For details, see `theme_distributions._make_breakdowns_dict`.
:type breakdown_configurations: iterable of core_data_modules.analysis.AnalysisConfiguration
:param f: File to write the theme_distributions CSV to.
:type f: file-like
"""
theme_distributions = compute_theme_distributions(individuals, consent_withdrawn_field,
theme_configurations, breakdown_configurations)
# Denormalize the theme_distributions into a flat array of dicts that can be written to disk.
csv_data = []
last_dataset_name = None
for dataset_name, themes in theme_distributions.items():
for theme, breakdowns in themes.items():
row = {
"Dataset": dataset_name if dataset_name != last_dataset_name else "",
"Theme": theme
}
row.update(breakdowns)
csv_data.append(row)
last_dataset_name = dataset_name
headers = ["Dataset", "Theme"]
headers.extend(_make_breakdowns_dict(breakdown_configurations).keys())
analysis_utils.write_csv(
csv_data,
headers,
f
)
def export_repeat_participations_csv(individuals, consent_withdrawn_field,
analysis_configurations, f):
"""
Computes the repeat_participations and exports them to a CSV.
:param individuals: Individuals to compute the repeat participations for.
:type individuals: iterable of core_data_modules.traced_data.TracedData
:param consent_withdrawn_field: Field in each individuals object which records if consent is withdrawn.
:type consent_withdrawn_field: str
:param analysis_configurations: Configurations for the datasets/coded fields to include in the engagement_counts.
:type analysis_configurations: iterable of core_data_modules.analysis.AnalysisConfiguration
:param f: File to write the engagement_counts CSV to.
:type f: file-like
"""
analysis_utils.write_csv(
compute_repeat_participations(individuals, consent_withdrawn_field,
analysis_configurations).values(),
repeat_participations_keys, f)
def export_cross_tabs_csv(individuals, consent_withdrawn_field,
analysis_configuration_1, analysis_configuration_2,
f):
"""
Computes cross-tabs and exports them to a file.
:param individuals: Individuals to compute the cross-tabs for.
:type individuals: iterable of core_data_modules.traced_data.TracedData
:param consent_withdrawn_field: Field in each individuals object which records if consent is withdrawn.
:type consent_withdrawn_field: str
:param analysis_configuration_1: Configuration for the first cross-tab variable.
:type analysis_configuration_1: core_data_modules.analysis.AnalysisConfiguration
:param analysis_configuration_2: Configuration for the second cross-tab variable.
:type analysis_configuration_2: core_data_modules.analysis.AnalysisConfiguration
:param f: File to write the cross_tabs CSV to.
:type f: file-like
"""
cross_tabs = compute_cross_tabs(individuals, consent_withdrawn_field,
analysis_configuration_1,
analysis_configuration_2)
analysis_utils.write_csv(cross_tabs, list(cross_tabs[0].keys()), f)
def export_engagement_counts_csv(messages, individuals,
consent_withdrawn_field,
analysis_configurations, f):
"""
Computes the engagement_counts and exports them to a CSV.
:param messages: Messages to compute the engagement_counts for.
:type messages: iterable of core_data_modules.traced_data.TracedData
:param individuals: Individuals to compute the engagement_counts for.
:type individuals: iterable of core_data_modules.traced_data.TracedData
:param consent_withdrawn_field: Field in each messages/individuals object which records if consent is withdrawn.
:type consent_withdrawn_field: str
:param analysis_configurations: Configurations for the datasets/coded fields to include in the engagement_counts.
:type analysis_configurations: iterable of core_data_modules.analysis.AnalysisConfiguration
:return: Dictionary of dataset_name | "Total" -> dict with keys `engagement_counts_keys`.
:rtype: OrderedDict of str -> dict
:param f: File to write the engagement_counts CSV to.
:type f: file-like
"""
analysis_utils.write_csv(
compute_engagement_counts(messages, individuals,
consent_withdrawn_field,
analysis_configurations).values(),
engagement_counts_keys, f)
def export_traffic_analysis_csv(messages, consent_withdrawn_field,
analysis_configurations, time_field,
traffic_labels, f):
"""
Computes the traffic_analysis and exports the results to a CSV.
:param messages: Messages to compute the engagement_counts for.
:type messages: iterable of core_data_modules.traced_data.TracedData
:param consent_withdrawn_field: Field in each messages object which records if consent is withdrawn.
:type consent_withdrawn_field: str
:param analysis_configurations: Configurations for the datasets/coded fields to check for opt-ins/relevance in
the time ranges.
:type analysis_configurations: iterable of core_data_modules.analysis.AnalysisConfiguration
:param time_field: Field in each messages object which records when the message was sent.
:type time_field: str
:param traffic_labels: Time ranges to search and labels to apply.
:type traffic_labels: iterable of TrafficLabel
:param f: File to write the engagement_counts CSV to.
:type f: file-like
"""
analysis_utils.write_csv(
compute_traffic_analysis(messages, consent_withdrawn_field,
analysis_configurations, time_field,
traffic_labels), traffic_analysis_keys, f)