def nadia_process_xls(filename, data_headers_location, data_values_location,
y_axis_location, global_data_vector, global_const_vector, outfile):
"""Given an XLS file and all the settings -- such as where are headers,
where is data, where is Y axis, etc. -- this function translates such
XLS file in a sequence of JSON that match very closely the format of
a CKAN package."""
sheet = open_sheet(filename)
data = data_section(sheet, data_headers_location, data_values_location)
y_axis(sheet, y_axis_location, data)
data = remove_empty_lines(data)
for name, location in global_data_vector:
global_data(data, sheet, name, location)
for name, value in global_const_vector:
global_const(data, name, value)
jsonize(data, outfile)
# Ensure that the package name is not too long or the server
# will have a boo.
#
name = slugify(package.author.lower())
if not package.name.startswith(name):
package.name = name + "_" + package.name
package.name = package.name[:100]
#
# AFAIK vars() here will work as long as all the variables of
# `package` have been initialized using __init__(). This is
# what the code above already does. Nonetheless I whish to add
# this comment for future robusteness of the code.
#
octets = json.dumps(vars(package), indent=indent)
fp.write(octets)
fp.write("\n")
if __name__ == "__main__":
sheet = open_sheet("test/sample1.xls")
data = data_section(sheet, "C3:R6", "C7:R24")
y_axis(sheet, "B7:B24", data)
data = remove_empty_lines(data)
global_data(data, sheet, "author", "A10")
global_data(data, sheet, "url", "A11")
global_data(data, sheet, "mission", "A12")
global_const(data, "category", "geodati")
jsonize(data, sys.stdout, 4)