def _write_is_valid_row_col(val_sheet, col_index,
index_and_range_tuple_by_header_dict):
"""
:type val_sheet: ValidationWorksheet
"""
# write is_valid_row column: e.g., =IF(AO2,TRUE,AND(AQ2:BW2))
is_absent_col_index = index_and_range_tuple_by_header_dict[
val_sheet.IS_ABSENT_HEADER][0]
is_absent_single_cell_range_str = xlsxbasics.format_range(
is_absent_col_index, "{{curr_row_index}}")
static_grid_single_row_range_str = xlsxbasics.format_range(
val_sheet.first_static_grid_col_index, "{{curr_row_index}}",
val_sheet.last_static_grid_col_index)
# When {{ and/or }} are put IN to string by a format call, as in first format call here, they aren't collapsed to
# { and }, so second (empty) format call does that.
is_valid_row_partial_formula = "IF({is_absent_1cell_range_str},TRUE,AND({static_grid_1row_range_str}))".format(
is_absent_1cell_range_str=is_absent_single_cell_range_str,
static_grid_1row_range_str=static_grid_single_row_range_str).format()
return xlsxbasics.copy_formula_throughout_range(
val_sheet.worksheet,
is_valid_row_partial_formula,
col_index,
val_sheet.first_data_row_index,
last_row_index=val_sheet.last_data_row_index)
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 _format_static_grid_cell_reference_formula_str(val_sheet, row_rank_num,
col_rank_num):
"""
:type val_sheet: xlsx_static_grid_builder.ValidationWorksheet
"""
# The contents here will be either "TRUE" or "FALSE", depending on whether the relevant cell in the metadata
# sheet is valid or invalid according to the static validation grid
#
# INDEX($AQ$2:$BW$11,MATCH(ROWS($AJ$2:AJ2),$AJ$2:$AJ$11,0),MATCH(COLUMNS($AQ$16:AQ16),$AQ$16:$BW$16,0))
static_grid_fixed_range_str = xlsxbasics.format_range(
val_sheet.first_static_grid_col_index,
val_sheet.first_data_row_index,
last_col_index=val_sheet.last_static_grid_col_index,
last_row_index=val_sheet.last_data_row_index,
first_col_fixed=True,
first_row_fixed=True,
last_col_fixed=True,
last_row_fixed=True)
static_grid_cell_contents = "INDEX({static_grid_fixed_range_str},{row_rank_num},{col_rank_num})".format(
static_grid_fixed_range_str=static_grid_fixed_range_str,
row_rank_num=row_rank_num,
col_rank_num=col_rank_num)
return static_grid_cell_contents
def _write_sample_id_col(data_sheet):
"""
:type data_sheet: xlsxbasics.MetadataWorksheet
"""
data_sheet.worksheet.set_column(data_sheet.sample_id_col_index,
data_sheet.sample_id_col_index, None, None,
data_sheet.hidden_cell_setting)
xlsxbasics.write_header(data_sheet,
"sample_id",
data_sheet.sample_id_col_index,
set_width=False)
# +1 bc range is exclusive of last number
for row_index in range(data_sheet.first_data_row_index,
data_sheet.last_allowable_row_for_sample_index + 1):
curr_cell = xlsxbasics.format_range(data_sheet.sample_id_col_index,
row_index)
id_num = row_index - data_sheet.first_data_row_index + 1
data_row_range = xlsxbasics.format_single_data_grid_row_range(
data_sheet, row_index)
completed_formula = "=IF(COUNTBLANK({data_row_range})<>COLUMNS({data_row_range}),{id_num},\"\")".format(
data_row_range=data_row_range, id_num=id_num)
data_sheet.worksheet.write_formula(curr_cell, completed_formula)
def _format_dynamic_rank_formula_str(val_sheet, curr_range_index,
index_and_range_str_tuple_by_header_dict,
for_row):
"""
:type val_sheet: xlsx_static_grid_builder.ValidationWorksheet
"""
# #1: Get the row_rank for the metadata row that should be shown in this validation row
# (e.g., if we're in validation row 5, we should be showing the metadata row with row_rank 5).
# MATCH(ROWS($AJ$2:AJ2),$AJ$2:$AJ$11,0)
# or
# # 2: Get the col_rank for the metadata column that should be shown in this validation column
# (e.g., if we're in validation column 6, we should be showing the metadata column with col_rank 6).
# MATCH(COLUMNS($AQ$16:AQ16),$AQ$16:$BW$16,0)
rank_header = val_sheet.ROW_RANK_HEADER if for_row else val_sheet.COL_RANK_HEADER
rank_index_and_range_tuple = index_and_range_str_tuple_by_header_dict[
rank_header]
rank_range_index = rank_index_and_range_tuple[0]
rank_fixed_range_str = rank_index_and_range_tuple[1]
if for_row:
excel_func_name = "ROWS"
first_col_index = rank_range_index
first_row_index = val_sheet.first_data_row_index
last_col_index = rank_range_index
last_row_index = curr_range_index
else:
excel_func_name = "COLUMNS"
first_col_index = val_sheet.first_static_grid_col_index
first_row_index = rank_range_index
last_col_index = curr_range_index
last_row_index = rank_range_index
rank_to_curr_point_range = xlsxbasics.format_range(
first_col_index,
first_row_index,
last_col_index=last_col_index,
last_row_index=last_row_index,
first_col_fixed=True,
first_row_fixed=True)
rank_num = "MATCH({excel_func_name}({rank_to_curr_point_range}),{rank_fixed_range_str},0)".format(
excel_func_name=excel_func_name,
rank_to_curr_point_range=rank_to_curr_point_range,
rank_fixed_range_str=rank_fixed_range_str)
return rank_num
def write_dynamic_validation_grid(val_sheet,
index_and_range_str_tuple_by_header_dict):
"""
:type val_sheet: xlsx_static_grid_builder.ValidationWorksheet
"""
_write_dynamic_name_link_col(val_sheet,
index_and_range_str_tuple_by_header_dict)
# apparently can't add alignment to a conditional format :(
centered_format = xlsxbasics.make_format(val_sheet.workbook,
{'align': 'center'})
# at outer level, move across columns
for curr_col_index in range(val_sheet.first_data_col_index,
val_sheet.last_data_col_index + 1):
val_sheet.worksheet.set_column(curr_col_index, curr_col_index, None,
centered_format)
curr_static_grid_col_index = val_sheet.first_static_grid_col_index + curr_col_index - 1
col_rank = _format_dynamic_rank_formula_str(
val_sheet,
curr_static_grid_col_index,
index_and_range_str_tuple_by_header_dict,
for_row=False)
col_already_valid_condition = _format_range_already_valid_formula_str(
val_sheet,
col_rank,
index_and_range_str_tuple_by_header_dict,
for_row=False)
_write_dynamic_header_cell(val_sheet, curr_col_index, col_rank)
# at inner level, move down rows
for curr_row_index in range(val_sheet.first_data_row_index,
val_sheet.last_data_row_index + 1):
cell_formula = _generate_dynamic_grid_cell_formula_str(
val_sheet, col_rank, col_already_valid_condition,
curr_row_index, index_and_range_str_tuple_by_header_dict)
curr_cell = xlsxbasics.format_range(curr_col_index, curr_row_index)
val_sheet.worksheet.write_formula(curr_cell, cell_formula)
_write_dynamic_grid_conditional_formatting(val_sheet)
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_dynamic_grid_conditional_formatting(val_sheet):
"""
:type val_sheet: xlsx_static_grid_builder.ValidationWorksheet
"""
# Light red fill with dark red text
red_format = xlsxbasics.make_format(val_sheet.workbook, {
'bg_color': '#FFC7CE',
'font_color': '#9C0006',
'underline': 1
})
# Green fill with (same) green text
green_format = xlsxbasics.make_format(val_sheet.workbook, {
'bg_color': '#C6EFCE',
'font_color': '#C6EFCE'
})
# 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 :)
dynamic_grid_range = xlsxbasics.format_range(
val_sheet.first_data_col_index, val_sheet.first_data_row_index,
val_sheet.last_data_col_index,
val_sheet.last_allowable_row_for_sample_index)
val_sheet.worksheet.conditional_format(dynamic_grid_range, {
'type': 'cell',
'criteria': '==',
'value': "\"\"",
'format': green_format
})
val_sheet.worksheet.conditional_format(dynamic_grid_range, {
'type': 'cell',
'criteria': '==',
'value': "\"Fix\"",
'format': red_format
})
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)