def _write_sample_id_col(data_sheet):
"""
:type data_sheet: xlsxbasics.MetadataWorksheet
"""
data_sheet.worksheet.set_column(data_sheet.sample_id_col_index,
data_sheet.sample_id_col_index, None, None,
data_sheet.hidden_cell_setting)
xlsxbasics.write_header(data_sheet,
"sample_id",
data_sheet.sample_id_col_index,
set_width=False)
# +1 bc range is exclusive of last number
for row_index in range(data_sheet.first_data_row_index,
data_sheet.last_allowable_row_for_sample_index + 1):
curr_cell = xlsxbasics.format_range(data_sheet.sample_id_col_index,
row_index)
id_num = row_index - data_sheet.first_data_row_index + 1
data_row_range = xlsxbasics.format_single_data_grid_row_range(
data_sheet, row_index)
completed_formula = "=IF(COUNTBLANK({data_row_range})<>COLUMNS({data_row_range}),{id_num},\"\")".format(
data_row_range=data_row_range, id_num=id_num)
data_sheet.worksheet.write_formula(curr_cell, completed_formula)
def _write_static_helper_ranges(
val_sheet,
header_and_writer_func_tuple_list,
range_index_and_range_str_tuple_by_header_dict=None):
"""
:type val_sheet: ValidationWorksheet
"""
def get_col_index(header_index):
# NB: -2 because leaving an empty column between static validation grid and helper columns
return val_sheet.first_static_grid_col_index - header_index - 2
def get_row_index(header_index):
return val_sheet.first_helper_rows_row_index + header_index
col_index = None
write_col = False
if range_index_and_range_str_tuple_by_header_dict is None:
range_index_and_range_str_tuple_by_header_dict = {}
write_col = True
for curr_header_index, curr_tuple in enumerate(
header_and_writer_func_tuple_list):
curr_header = curr_tuple[0]
curr_write_method = curr_tuple[1]
col_index = get_col_index(
curr_header_index
) if write_col else val_sheet.helper_rows_header_col_index
row_index = None if write_col else get_row_index(curr_header_index)
xlsxbasics.write_header(val_sheet,
curr_header,
col_index,
row_index,
set_width=False)
curr_range_index = col_index if write_col else row_index
curr_range_str = curr_write_method(
val_sheet, curr_range_index,
range_index_and_range_str_tuple_by_header_dict)
range_index_and_range_str_tuple_by_header_dict[curr_header] = (
curr_range_index, curr_range_str)
if write_col and col_index is not None:
# use the col_index left over from last time through loop
val_sheet.hide_columns(val_sheet.first_static_grid_col_index - 1,
col_index)
return range_index_and_range_str_tuple_by_header_dict
def _write_static_validation_grid(val_sheet, schema_dict):
"""
:type val_sheet: ValidationWorksheet
"""
curr_grid_col_index = None
sorted_keys = xlsxbasics.sort_keys(schema_dict)
for field_index, field_name in enumerate(sorted_keys):
field_specs_dict = schema_dict[field_name]
curr_grid_col_index = val_sheet.first_static_grid_col_index + field_index
xlsxbasics.write_header(val_sheet,
field_name,
curr_grid_col_index,
set_width=False)
unformatted_formula_str = qiimp.xlsx_validation_builder.get_formula_constraint(
field_specs_dict, val_sheet.regex_handler)
if unformatted_formula_str is not None:
curr_metadata_col_index = val_sheet.first_data_col_index + field_index
# metadata_cell_range_str = xlsxbasics.format_single_col_range(val_sheet, curr_metadata_col_index,
# sheet_name=val_sheet.metadata_sheet_name)
cell_enumerator = xlsxbasics.loop_through_range(
curr_grid_col_index,
val_sheet.first_data_row_index,
last_row_index=val_sheet.last_allowable_row_for_sample_index)
for curr_col_index, curr_row_index, curr_cell_range in cell_enumerator:
metadata_col_range = xlsxbasics.format_range(
curr_metadata_col_index,
None,
sheet_name=val_sheet.metadata_sheet_name)
metadata_cell = xlsxbasics.format_range(
curr_metadata_col_index,
curr_row_index,
sheet_name=val_sheet.metadata_sheet_name)
formatted_formula_str = unformatted_formula_str.format(
cell=metadata_cell, col_range=metadata_col_range)
val_sheet.worksheet.write_formula(curr_cell_range,
formatted_formula_str)
# xlsxbasics.format_and_write_array_formula(val_sheet, curr_grid_col_index, unformatted_formula_str,
# write_col=True, cell_range_str=metadata_cell_range_str)
# hide all the columns in the static grid. Use value of curr_grid_col_index left over from last time thru loop.
if curr_grid_col_index:
val_sheet.hide_columns(val_sheet.first_static_grid_col_index,
curr_grid_col_index)
def _write_dynamic_header_cell(val_sheet, curr_col_index, col_rank):
# e.g., =IF(B2=" ", " ", INDEX($AS$1:$BY$1, 1, MATCH(COLUMNS($AS$1005:AS1005),$AS$1005:$BY$1005,0)))
first_data_cell_in_col = xlsxbasics.format_range(
curr_col_index, val_sheet.first_data_row_index)
static_grid_header_row = xlsxbasics.format_single_static_grid_row_range(
val_sheet,
val_sheet.name_row_index,
first_col_fixed=True,
first_row_fixed=True,
last_col_fixed=True,
last_row_fixed=True)
# if ALL samples are valid in this column, then all the data cells for this column will have a space
# in them. Conversely, if ANY sample is invalid in this column, the data cells will either be *empty* (have
# not even a space) if a particular sample is valid in that column or will have the word "Fix". I check just
# the FIRST data cell in this column to see if it holds a space; if it does, the whole column is valid and
# should be hidden, so the "header" will also just be a space.
header_formula = "=IF({first_data_cell_in_col}=\" \", \" \", INDEX({static_grid_header_row}, 1, {col_rank}))" \
.format(first_data_cell_in_col=first_data_cell_in_col, static_grid_header_row=static_grid_header_row,
col_rank=col_rank)
xlsxbasics.write_header(val_sheet, header_formula, curr_col_index)
def write_workbook(study_name, schema_dict, form_dict, metadata_wizard_settings):
num_allowable_samples = 1000
# TODO: someday: either expand code to use num_samples and add real code to get in from interface, or take out unused hook
num_samples = 0
num_columns = len(schema_dict.keys())
a_regex_handler = metadata_wizard_settings.regex_handler
# create workbook
file_base_name = slugify(study_name)
file_name = '{0}_{1}.xlsx'.format(file_base_name, randrange(1000, 9999))
output_path = metadata_wizard_settings.get_output_path(file_name)
workbook = xlsxwriter.Workbook(output_path, {'strings_to_numbers': False,
'strings_to_formulas': True,
'strings_to_urls': True})
# write metadata worksheet
phi_renamed_schema_dict = qiimp.schema_builder.rewrite_field_names_with_phi_if_relevant(schema_dict)
metadata_worksheet = xlsxbasics.MetadataWorksheet(workbook, num_columns, num_samples, a_regex_handler,
num_allowable_samples=num_allowable_samples)
qiimp.xlsx_metadata_grid_builder.write_metadata_grid(metadata_worksheet, phi_renamed_schema_dict,
DescriptionWorksheet.get_sheet_name())
# write validation worksheet
validation_worksheet = qiimp.xlsx_static_grid_builder.ValidationWorksheet(workbook, num_columns,
num_samples, a_regex_handler)
index_and_range_str_tuple_by_header_dict = qiimp.xlsx_static_grid_builder.write_static_validation_grid_and_helpers(
validation_worksheet, phi_renamed_schema_dict)
qiimp.xlsx_dynamic_grid_builder.write_dynamic_validation_grid(
validation_worksheet, index_and_range_str_tuple_by_header_dict)
# write descriptions worksheet
descriptions_worksheet = DescriptionWorksheet(workbook, num_columns, num_samples, a_regex_handler)
xlsxbasics.write_header(descriptions_worksheet, "field name", 0)
xlsxbasics.write_header(descriptions_worksheet, "field description", 1)
sorted_keys = xlsxbasics.sort_keys(phi_renamed_schema_dict)
for field_index, field_name in enumerate(sorted_keys):
row_num = field_index + 1 + 1 # plus 1 to move past name row, and plus 1 again because row nums are 1-based
field_specs_dict = phi_renamed_schema_dict[field_name]
message = qiimp.xlsx_validation_builder.get_field_constraint_description(field_specs_dict, a_regex_handler)
descriptions_worksheet.worksheet.write("A{0}".format(row_num), field_name, metadata_worksheet.header_format)
descriptions_worksheet.worksheet.write("B{0}".format(row_num), message)
# write schema worksheet--note, don't use the phi_renamed_schema_dict but the original schema_dict
schema_worksheet = xlsxbasics.create_worksheet(workbook, xlsxbasics.SheetNames.schema.value)
schema_worksheet.write_string("A1", yaml.dump(schema_dict, default_flow_style=False))
schema_worksheet.hide()
# write form worksheet
form_worksheet = xlsxbasics.create_worksheet(workbook, xlsxbasics.SheetNames.form.value)
form_worksheet.write_string("A1", yaml.dump(form_dict, default_flow_style=False))
form_worksheet.hide()
# write readme worksheet
readme_format = workbook.add_format({'align': 'left', 'valign': 'top'})
readme_format.set_text_wrap()
readme_worksheet = xlsxbasics.create_worksheet(workbook, xlsxbasics.SheetNames.readme.value)
readme_worksheet.set_column(0, 0, 100) # Width of column A set to 100.
tutorial_hyperlink_str = "=HYPERLINK(\"{0}\", \"Click here for instructions on using this spreadsheet.\")".format(
metadata_wizard_settings.TUTORIAL_LINK)
readme_worksheet.write_formula(0,0, tutorial_hyperlink_str,
# TODO: someday: centralize link format definitions.
# this link format is a copy-paste of the one in
# xlsx_dynamic_grid_builder._write_dynamic_name_link_col ; assuming we want
# all the link formats in the spreadsheet to look the same, they should be defined
# in one central place.
xlsxbasics.make_format(workbook, {'font_color': 'blue', 'underline': 1}))
readme_worksheet.write_string('A3', metadata_wizard_settings.make_readme_text(), readme_format)
# close workbook
workbook.close()
return file_name
def _write_dynamic_name_link_col(val_sheet,
index_and_range_str_tuple_by_header_dict):
"""
:type val_sheet: xlsx_static_grid_builder.ValidationWorksheet
"""
# e.g., =IF(B2=" "," ",HYPERLINK(CONCATENATE("#metadata!", ADDRESS(INDEX($AL$2:$AL$11,MATCH(ROWS($AJ$2:AJ2),
# $AJ$2:$AJ$11,0),0),$AQ$14)),INDEX($AO$2:$AO$11,MATCH(ROWS($AJ$2:AJ2),$AJ$2:$AJ$11,0),0)))
# Create the standard blue, underlined url link format.
url_format = xlsxbasics.make_format(val_sheet.workbook, {
'font_color': 'blue',
'underline': 1
})
xlsxbasics.write_header(val_sheet, val_sheet.SAMPLE_NAME_HEADER,
val_sheet.name_link_col_index)
# NB: grid goes as far as last allowable row for samples, not just to number of expected samples, in case user
# adds some extra ones :)
for curr_row_index in range(
val_sheet.first_data_row_index,
val_sheet.last_allowable_row_for_sample_index + 1):
row_rank_num = _format_dynamic_rank_formula_str(
val_sheet,
curr_row_index,
index_and_range_str_tuple_by_header_dict,
for_row=True)
row_in_metadata_fixed_range_str = index_and_range_str_tuple_by_header_dict[
val_sheet.ROW_IN_METADATA_HEADER][1]
metadata_row_index_str = "INDEX({row_in_metadata_fixed_range_str},{row_rank_num},0)".format(
row_in_metadata_fixed_range_str=row_in_metadata_fixed_range_str,
row_rank_num=row_rank_num)
link_address = "CONCATENATE(\"#metadata!\",ADDRESS({metadata_row_index},{metadata_name_col_index}))".format(
metadata_row_index=metadata_row_index_str,
metadata_name_col_index=val_sheet.name_col_index)
helper_name_fixed_range_str = index_and_range_str_tuple_by_header_dict[
val_sheet.SAMPLE_NAME_HEADER][1]
# this index formula will get the value of the name for this sample from the helper col next to the static grid
helper_name_val = "INDEX({conditional_name_fixed_range_str},{row_num},0)".format(
conditional_name_fixed_range_str=helper_name_fixed_range_str,
row_num=row_rank_num)
curr_cell = xlsxbasics.format_range(val_sheet.name_link_col_index,
curr_row_index)
# If this sample is entirely valid as shown by the fact that the first data cell in the dynamic grid for this
# row is just an empty string, write a space into the dynamic name cell. Otherwise, write a link to the
# name column for this sample in the metadata sheet. NB: it does NOT work to look at
# the value in the is_valid helper column for this sample (either True or False) because the samples
# change order based on validation status ...
first_data_cell_in_first_data_col = xlsxbasics.format_range(
val_sheet.name_link_col_index + 1, curr_row_index)
full_formula = "=IF({first_validation_cell}=\" \",\" \",HYPERLINK({link_address},{helper_name_val}))".format(
first_validation_cell=first_data_cell_in_first_data_col,
link_address=link_address,
helper_name_val=helper_name_val)
val_sheet.worksheet.write_formula(curr_cell, full_formula, url_format)
def write_metadata_grid(data_worksheet, schema_dict, field_descs_sheet_name):
"""
:type data_worksheet: xlsxbasics.MetadataWorksheet
"""
_write_sample_id_col(data_worksheet)
unlocked = xlsxbasics.make_format(data_worksheet.workbook, is_locked=False)
# format as text to prevent autoformatting!
unlocked_text = xlsxbasics.make_format(data_worksheet.workbook,
{'num_format': '@'},
is_locked=False)
sorted_keys = xlsxbasics.sort_keys(schema_dict)
for field_index, field_name in enumerate(sorted_keys):
field_specs_dict = schema_dict[field_name]
curr_col_index = field_index + 1 # add one bc sample id is in first col
xlsxbasics.write_header(data_worksheet, field_name, field_index + 1)
curr_format = unlocked_text if _determine_if_format_should_be_text(
field_specs_dict) else unlocked
data_worksheet.worksheet.set_column(curr_col_index, curr_col_index,
None, curr_format)
col_range = xlsxbasics.format_range(curr_col_index, None)
starting_cell_name = xlsxbasics.format_range(
curr_col_index, data_worksheet.first_data_row_index)
whole_col_range = xlsxbasics.format_range(
curr_col_index,
data_worksheet.first_data_row_index,
last_row_index=data_worksheet.last_allowable_row_for_sample_index)
validation_dict = _get_validation_dict(field_name, field_specs_dict,
data_worksheet.regex_handler,
field_descs_sheet_name)
value_key = "value"
if validation_dict is not None:
if value_key in validation_dict:
unformatted_validation_formula = validation_dict[value_key]
formatted_validation_formula = unformatted_validation_formula.format(
cell=starting_cell_name, col_range=col_range)
validation_dict[value_key] = formatted_validation_formula
validation_return_code = data_worksheet.worksheet.data_validation(
whole_col_range, validation_dict)
# NB: xlsxwriter's data_validation docstring *claims* it returns 0 if it succeeds, but in fact if it
# succeeds it doesn't return an error code at all, hence the then None check ...
if validation_return_code is not None and validation_return_code < 0:
raise ValueError(
"Worksheet validation failed with return code '{0}'; check user warnings."
.format(validation_return_code))
_add_default_if_any(data_worksheet, field_specs_dict, curr_col_index)
max_samples_msg = "No more than {0} samples can be entered in this worksheet. If you need to submit metadata" \
" for >{0} samples, please contact CMI directly.".format(data_worksheet.num_allowable_samples)
xlsxbasics.write_header(
data_worksheet, max_samples_msg,
data_worksheet.first_data_col_index,
data_worksheet.last_allowable_row_for_sample_index + 1)
def write_metadata_grid(data_worksheet, schema_dict, field_descs_sheet_name):
"""
:type data_worksheet: xlsxbasics.MetadataWorksheet
"""
_write_sample_id_col(data_worksheet)
unlocked = xlsxbasics.make_format(data_worksheet.workbook, is_locked=False)
# format as text to prevent autoformatting!
unlocked_text = xlsxbasics.make_format(data_worksheet.workbook,
{'num_format': '@'},
is_locked=False)
sorted_keys = xlsxbasics.sort_keys(schema_dict)
for field_index, field_name in enumerate(sorted_keys):
field_specs_dict = schema_dict[field_name]
curr_col_index = field_index + 1 # add one bc sample id is in first col
xlsxbasics.write_header(data_worksheet, field_name, field_index + 1)
curr_format = unlocked_text if _determine_if_format_should_be_text(
field_specs_dict) else unlocked
# Note: although the xlsxwriter docs say
# "If you wish to set the format without changing the width you can
# pass None as the width parameter" (https://xlsxwriter.readthedocs.io/
# worksheet.html), it appears that if I call set_column with None for
# the width *after* calling write_header (which sets the width
# explicitly), the width of the column is reset to the default, thus
# requiring me to explicitly set the min column width here too :(
data_worksheet.worksheet.set_column(curr_col_index, curr_col_index,
xlsxbasics.get_min_col_width(),
curr_format)
col_range = xlsxbasics.format_range(curr_col_index, None)
starting_cell_name = xlsxbasics.format_range(
curr_col_index, data_worksheet.first_data_row_index)
whole_col_range = xlsxbasics.format_range(
curr_col_index,
data_worksheet.first_data_row_index,
last_row_index=data_worksheet.last_allowable_row_for_sample_index)
validation_dict = _get_validation_dict(field_name, field_specs_dict,
data_worksheet.regex_handler,
field_descs_sheet_name)
value_key = "value"
if validation_dict is not None:
if value_key in validation_dict:
unformatted_validation_formula = validation_dict[value_key]
formatted_validation_formula = unformatted_validation_formula.format(
cell=starting_cell_name, col_range=col_range)
validation_dict[value_key] = formatted_validation_formula
validation_return_code = data_worksheet.worksheet.data_validation(
whole_col_range, validation_dict)
# NB: xlsxwriter's data_validation docstring *claims* it returns 0 if it succeeds, but in fact if it
# succeeds it doesn't return an error code at all, hence the then None check ...
if validation_return_code is not None and validation_return_code < 0:
raise ValueError(
"Worksheet validation failed with return code '{0}'; check user warnings."
.format(validation_return_code))
_add_default_if_any(data_worksheet, field_specs_dict, curr_col_index)
max_samples_msg = "No more than {0} samples can be entered in this worksheet. If you need to submit metadata" \
" for >{0} samples, please contact CMI directly.".format(data_worksheet.num_allowable_samples)
xlsxbasics.write_header(
data_worksheet,
max_samples_msg,
data_worksheet.first_data_col_index,
data_worksheet.last_allowable_row_for_sample_index + 1,
set_width=False)