edc-pdutils 0.3.21

Use pandas with clinicedc/edc projects

edc-pdutils

Use pandas with the Edc

Using the management command to export to CSV and STATA

To export as CSV where the delimiter is |

python manage.py export_models -a ambition_subject -p /ambition/export

To export as STATA dta:

python manage.py export_models -a ambition_subject -f stata -p /ambition/export

To export encrypted fields as well:

python manage.py export_models -a ambition_subject -f stata -p /ambition/export  --decrypt

To export using a simpler filename that drops the tablename app_label prefix and does not include a datestamp suffix:

python manage.py export_models -a ambition_subject -f stata -p /ambition/export  --use_simple_filename

Export manually

To export Crf data, for example:

from edc_pdutils.df_exporters import CsvCrfTablesExporter
from edc_pdutils.df_handlers import CrfDfHandler

app_label = 'ambition_subject'
csv_path = '/Users/erikvw/Documents/ambition/export/'
date_format = '%Y-%m-%d'
sep = '|'
exclude_history_tables = True

class MyDfHandler(CrfDfHandler):
    visit_tbl = f'{app_label}_subjectvisit'
    exclude_columns = ['form_as_json', 'survival_status','last_alive_date',
                       'screening_age_in_years', 'registration_datetime',
                       'subject_type']

class MyCsvCrfTablesExporter(CsvCrfTablesExporter):
    visit_column = 'subject_visit_id'
    datetime_fields = ['randomization_datetime']
    df_handler_cls = MyDfHandler
    app_label = app_label
    export_folder = csv_path

sys.stdout.write('\n')
exporter = MyCsvCrfTablesExporter(
    export_folder=csv_path,
    exclude_history_tables=exclude_history_tables
)
exporter.to_csv(date_format=date_format, delimiter=sep)

To export INLINE data for any CRF configured with an inline, for example:

class MyDfHandler(CrfDfHandler):
    visit_tbl = 'ambition_subject_subjectvisit'
    exclude_columns = ['form_as_json', 'survival_status','last_alive_date',
                       'screening_age_in_years', 'registration_datetime',
                       'subject_type']


class MyCsvCrfInlineTablesExporter(CsvCrfInlineTablesExporter):
    visit_columns = ['subject_visit_id']
    df_handler_cls = MyDfHandler
    app_label = 'ambition_subject'
    export_folder = csv_path
    exclude_inline_tables = [
        'ambition_subject_radiology_abnormal_results_reason',
        'ambition_subject_radiology_cxr_type']
sys.stdout.write('\n')
exporter = MyCsvCrfInlineTablesExporter()
exporter.to_csv(date_format=date_format, delimiter=sep)

Using model_to_dataframe

from edc_pdutils.model_to_dataframe import ModelToDataframe
from edc_pdutils.utils import get_model_names
from edc_pdutils.df_exporters.csv_exporter import CsvExporter

app_label = 'ambition_subject'
csv_path = '/Users/erikvw/Documents/ambition/export/'
date_format = '%Y-%m-%d'
sep = '|'

for model_name in get_model_names(
        app_label=app_label,
        # with_columns=with_columns,
        # without_columns=without_columns,
    ):
    m = ModelToDataframe(model=model_name)
    exporter = CsvExporter(
        data_label=model_name,
        date_format=date_format,
        delimiter=sep,
        export_folder=csv_path,
    )
    exported = exporter.to_csv(dataframe=m.dataframe)

Settings

EXPORT_FILENAME_TIMESTAMP_FORMAT: True/False (Default: False)

By default a timestamp of the current date is added as a suffix to CSV export filenames.

By default a timestamp of format %Y%m%d%H%M%S is added.

EXPORT_FILENAME_TIMESTAMP_FORMAT may be set to an empty string or a valid format for strftime.

If EXPORT_FILENAME_TIMESTAMP_FORMAT is set to an empty string, “”, a suffix is not added.

For example:

# default
registered_subject_20190203112555.csv

# EXPORT_FILENAME_TIMESTAMP_FORMAT = "%Y%m%d"
registered_subject_20190203.csv

# EXPORT_FILENAME_TIMESTAMP_FORMAT = ""
registered_subject.csv