"""
Joins multiple single-file invoices in the StormWare Pohoda XML format to a
single XML file.

https://www.stormware.cz/xml/schema/version_2/invoice.xsd
https://www.stormware.cz/schema/version_2/data.xsd

Usage:
$ python join_invoices.py 'single_invoices/*.xml' merged_invoice.xml

File encoding: windows-1250, CRLF.
"""

import argparse
import glob
from tqdm import tqdm
import xmltodict

def merge_invoices(input_path, output_path):
    merged_doc = None
    for path in tqdm(glob.glob(input_path)):
        # use the first document as the template for the top-level structure
        if merged_doc is None:
            merged_doc = load_root_doc(path)
        doc = load_xml(path)
        item = extract_item_with_id(doc)
        merged_doc['dat:dataPack']['dat:dataPackItem'].append(item)
    save_xml(merged_doc, output_path)

def load_root_doc(path):
    root_doc = load_xml(path)
    del root_doc['dat:dataPack']['dat:dataPackItem']
    # envelope id must not be empty, so let's leave there some dummy value
    root_doc['dat:dataPack']['@id'] = "00001"
    root_doc['dat:dataPack']['dat:dataPackItem'] = []
    return root_doc

def extract_item_with_id(doc):
    item = doc['dat:dataPack']['dat:dataPackItem']
    item['@id'] = doc['dat:dataPack']['@id']
    return item

def load_xml(path):
    with open(path, 'rb') as f:
        return xmltodict.parse(f)

def save_xml(doc, path):
    with open(path, 'wb') as f:
        xmltodict.unparse(doc, output=f, pretty=True, encoding='Windows-1250', newl='\r\n')

def parse_args():
    parser = argparse.ArgumentParser(description='Join multiple Pohoda invoices')
    parser.add_argument('input_path', metavar='INPUT_PATH')
    parser.add_argument('output_path', metavar='OUTPUT_PATH')
    return parser.parse_args()

if __name__ == '__main__':
    args = parse_args()
    merge_invoices(args.input_path, args.output_path)