Attachment 184438 Details for Bug 152804 – The Python script used to generate the .odt file

The Python script used to generate the .odt file

make-invoice.py (text/x-python), 8.19 KB, created by Mitch on 2023-01-02 04:14:00 UTC

(hide)

Description:

Filename:

MIME Type:

Creator: Mitch

Created: 2023-01-02 04:14:00 UTC

Size: 8.19 KB

patch

obsolete

>import argparse
>import csv
>import dateutil.parser as dparser
>from datetime import datetime
>import fileinput
>from pathlib import Path
>import xml.etree.ElementTree as ElementTree
>import tempfile
>import zipfile
>
>def read_hour_rows(csv_file_path):
>    with open(csv_file_path, 'r') as csvfile:
>        reader = csv.reader(csvfile, quotechar='"')
>        data = []
>        row_index = 0
>        for row in reader:
>            column_count = len(row)
>            if (column_count != 3):
>                raise ValueError("In csv file " + csv_file_path + ", row at index " + str(row_index)
>                    + " has " + str(column_count) + " columns instead of 3: \n" + str(row))
>
>#            print(str(len(row)) + "_" + str(row) + "_")
>
>            data += [row]
>            row_index += 1
>    return data
>
># Based on https://crcok.wordpress.com/2014/10/25/unzipping-editing-and-zipping-odt-documents-in-python/.
># Extracts content.xml from odt file, modifies it, writes it back and re-zips it into a new odt.
>
>arg_parser = argparse.ArgumentParser()
>arg_parser.add_argument('invoice_number')
>args = arg_parser.parse_args()
>
># A temporary directory to store the files we will extract from the odt.
>tmp_dir_url = tempfile.gettempdir() + "/invoice-template"
># The directory that contains this script
>this_dir_path = Path(__file__).parent.resolve()
>
># This is one of the files that will be extracted when we unzip the odt.
>xml_file_url = tmp_dir_url + "/content.xml"
>
>template_odt_path = str(this_dir_path / 'invoice-template.odt')
>invoice_odt_path = str(this_dir_path / 'invoice-') + str(args.invoice_number) + ".odt"
>
>#
># Unzip ODT
>#
>print(" -- Extracting ---------------------")
>print("%s -> %s" % (template_odt_path, tmp_dir_url))
>
># Load the template into memory.
>zipdata = zipfile.ZipFile(template_odt_path)
># Extract its contents into the file in the temporary directory.
>zipdata.extractall(tmp_dir_url)
>
>
>#
># Find and replace tokens
>#
>print(" -- Replacing -------------")
>print(xml_file_url)
>
>xml_namespaces = { 'table': 'urn:oasis:names:tc:opendocument:xmlns:table:1.0' }
>xml_tree = ElementTree.parse(xml_file_url)
>xml_tree_root = xml_tree.getroot()
>
>
># Set the invoice number and date in InvoiceNumberTable.
>invoice_number_tables = list(xml_tree_root.iterfind(".//*[@table:name='InvoiceNumberTable']", xml_namespaces))
>if (len(list(invoice_number_tables)) == 0):
>    raise ValueError("Can't find InvoiceNumberTable")
>
>invoice_number_table = invoice_number_tables[0]
>if (len(invoice_number_table) != 4):
>    raise ValueError("Expected 4 rows in InvoiceNumberTable, but got " + str(len(invoice_number_table)))
>
>
>invoice_number_row = invoice_number_table[2]
>if ("table-row" not in invoice_number_row.tag):
>    raise ValueError("Expected row at index 2 to be table-row " + invoice_number_row.tag)
>
>invoice_number_cell = invoice_number_row[1]
>if ("table-cell" not in invoice_number_cell.tag):
>    raise ValueError("Expected table-cell for row at index 2 but got " + invoice_number_cell.tag)
>
>invoice_number_p = invoice_number_cell[0]
>if ("p" not in invoice_number_p.tag):
>    raise ValueError("Expected p for row at index 2 but got " + invoice_number_p.tag)
>
>invoice_number_p.text = args.invoice_number
>
>
>invoice_data_row = invoice_number_table[3]
>if ("table-row" not in invoice_data_row.tag):
>    raise ValueError("Expected row at index 3 to be table-row " + invoice_data_row.tag)
>
>invoice_data_cell = invoice_data_row[1]
>if ("table-cell" not in invoice_data_cell.tag):
>    raise ValueError("Expected table-cell for row at index 3 but got " + invoice_data_cell.tag)
>
>invoice_data_p = invoice_data_cell[0]
>if ("p" not in invoice_data_p.tag):
>    raise ValueError("Expected p for row at index 3 but got " + invoice_data_p.tag)
>
>invoice_data_p.text = datetime.today().strftime('%d/%m/%Y')
>
>
># Look for HoursTable.
># Convert it into a list because otherwise there is a warning about generators not being subscriptable.
>hours_tables = list(xml_tree_root.iterfind(".//*[@table:name='HoursTable']", xml_namespaces))
>if (len(list(hours_tables)) == 0):
>    raise ValueError("Can't find HoursTable")
>
>hour_table = hours_tables[0]
>
># Add the data we read from hours.csv into each cell.
>hours_csv_file_path = this_dir_path / 'hours.csv'
>hour_rows = read_hour_rows(hours_csv_file_path)
>hour_row_count = len(hour_rows)
>xml_table_row_count = len(list(hour_table))
>if (hour_row_count > xml_table_row_count):
>    raise ValueError("Too many hours (" + str(len(hour_rows)) + ")"
>        + " to fit into invoice template table (" + str(xml_table_row_count) + ")")
>
># Gather all of the rows so that row_index starts from the first row we're interested in.
>rows = []
>for row in hour_table:
>    if ("table-row" in row.tag):
>        rows.append(row)
>
># Add data to cells.
>row_index = 0
>for row in rows:
>#    print("looping on index " + str(row_index))
>
>    if (row_index >= hour_row_count):
>        # We reached the last hour row.
>        break
>
>#    print("- looking at row index " + str(row_index) + ", tag is: " + row.tag)
>
>    if ("table-row" in row.tag):
>        if (len(list(row)) != 3):
>            raise ValueError("Expected row to have three children")
>
>        cell_index = 0
>        for cell in row:
>#            print("  - looking at cell index " + str(cell_index))
>            if ("table-cell" not in cell.tag):
>                raise ValueError("Expected table-cell to be a child of table-row")
>
>            p = cell[0]
>
>            if ("p" not in p.tag):
>                raise ValueError("Expected p to be a child of table-cell")
>
>            p.text = hour_rows[row_index][cell_index]
>#            print("  - setting text of row " + str(row_index) + " cell " + str(cell_index) + " to " + str(p.text))
>
>            cell_index += 1
>    row_index += 1
>
>
>
># Set the total hours/amount formula since us unzipping it seems to screw it up...
>summary_tables = list(xml_tree_root.iterfind(".//*[@table:name='SummaryTable']", xml_namespaces))
>
>if (len(list(summary_tables)) == 0):
>    raise ValueError("Can't find SummaryTable")
>
>summary_number_table = summary_tables[0]
># Bit nasty hard-coding these, but since the layout won't change: first three items are table-column, next three are table-row.
>if (len(summary_number_table) != 6):
>    raise ValueError("Expected 6 rows in SummaryTable, but got " + str(len(summary_number_table)))
>
># We're interested in the rows. The first one contains the cell that stores the number of total hours.
>total_hours_row = summary_number_table[3]
>if ("table-row" not in total_hours_row.tag):
>    raise ValueError("Expected row at index 3 to be table-row " + total_hours_row.tag)
>
># The 3rd column is the right-most one that stores the number of total hours.
>total_hours_cell = total_hours_row[2]
>if ("table-cell" not in total_hours_cell.tag):
>    raise ValueError("Expected table-cell but got " + total_hours_cell.tag)
>
>for key in total_hours_cell.attrib:
>    if ("formula" in key):
>        # Apparently the preceding '=' is added automatically.
>        total_hours_cell.attrib[key] = "sum<HoursTable.C2:C24>"
>
># The last row contains the cell that stores the total amount due.
>total_amount_due_row = summary_number_table[5]
>if ("table-row" not in total_amount_due_row.tag):
>    raise ValueError("Expected row at index 5 to be table-row " + total_amount_due_row.tag)
>
># The 3rd column is the right-most one that stores the total amount due.
>total_amount_due_cell = total_amount_due_row[2]
>if ("table-cell" not in total_amount_due_cell.tag):
>    raise ValueError("Expected table-cell but got " + total_amount_due_cell.tag)
>
>for key in total_amount_due_cell.attrib:
>    if ("formula" in key):
>        total_amount_due_cell.attrib[key] = "<SummaryTable.C1>*<SummaryTable.C2>"
>
>
># Write the data back in.
>xml_tree.write(str(xml_file_url))
>
>
># Zip contents of the temporary directory to ODT
># Use file list from the original archive
># This preserves the file structure in the new Zip file
># The most important is that the "mimetype" is the first file in archive
>
>print(" -- Compressing --------------------")
>print("%s -> %s" % (tmp_dir_url , invoice_odt_path))
>
>with zipfile.ZipFile(invoice_odt_path, 'w') as outzip:
>    zipinfos = zipdata.infolist()
>    for zipinfo in zipinfos:
>        file_name = zipinfo.filename # The name and path as stored in the archive
>        file_url = tmp_dir_url + "/" + file_name # The actual name and path
>        outzip.write(file_url, file_name)

Actions: View

Attachments on bug 152804: 184435 | 184436 | 184437 | 184438