Add method 'parse' to class 'Invoices'

* Add method 'parse' to class 'Invoices'
* Add tax rate detection when parsing PDF invoices
* Wire PDF parsing with class `Volksbank`
* Move helper method 'convert_tax_rates`

Also, this fixes #16

knv_cli/command.py

@@ -52,8 +52,8 @@ def convert_date(self, string: str) -> str:
 
 
     def convert_number(self, string) -> str:
-        # Convert integers & floats
-        string = str(string)
+        # Convert to string & clear whitespaces
+        string = str(string).strip()
 
         # Take care of thousands separator, as in '1.234,56'
         if '.' in string and ',' in string:
@@ -63,7 +63,3 @@ def convert_number(self, string) -> str:
         integer = f'{string:.2f}'
 
         return str(integer)
-
-
-    def convert_tax_rate(self, string: str) -> str:
-        return str(string).replace(',00', '') + '%'

knv_cli/gateways/volksbank.py

@@ -157,18 +157,37 @@ def match_payments(self, data: list, invoice_handler: Invoices = None) -> None:
                 payment['Vorgang'] = matching_invoices
 
                 # There are two ways extracting information about invoices ..
-                if matching_orders:
-                    # .. via fetching order data
-                    taxes = self.match_invoices(matching_invoices, matching_orders)
+                if invoice_handler:
+                    # .. via parsing invoice files
+                    matching_invoices = self.parse_invoices(matching_invoices, invoice_handler)
+
+                    payment['Vorgang'] = [invoice['Rechnungsnummer'] for invoice in matching_invoices]
+                    payment['Rechnungssumme'] = '0.00'
+                    payment['Bestellung'] = []
+
+                    taxes = {}
+
+                    for invoice in matching_invoices:
+                        payment['Rechnungssumme'] = self.convert_number(float(payment['Rechnungssumme']) + float(invoice['Gesamtbetrag']))
+                        payment['Bestellung'].append(invoice['Steuern'])
+
+                        for  tax_rate, tax_amount in invoice['Steuern'].items():
+                            if not tax_rate in taxes:
+                                taxes[tax_rate] = tax_amount
+
+                            else:
+                                taxes[tax_rate] = self.convert_number(float(taxes[tax_rate]) + float(tax_amount))
 
-                    if taxes:
-                        payment['Bestellung'] = taxes
+                    for tax_rate, tax_amount in taxes.items():
+                        payment[tax_rate + ' MwSt'] = tax_amount
 
                 else:
-                    # .. via parsing invoice files
-                    pass
-                    # if invoice_handler:
-                    #     taxes = self.parse_invoices(matching_invoices, invoice_handler)
+                    if matching_orders:
+                        # .. via fetching order data
+                        taxes = self.match_invoices(matching_invoices, matching_orders)
+
+                        if taxes:
+                            payment['Bestellung'] = taxes
 
                 # Reverse-lookup orders if no matching order number(s) yet
                 # if not matching_orders:
@@ -218,4 +237,10 @@ def match_invoices(self, invoices: list, orders: list) -> dict:
 
 
     def parse_invoices(self, invoices: list, handler: Invoices) -> list:
-        pass
+        results = []
+
+        for invoice in invoices:
+            if handler.has(invoice):
+                results.append(handler.parse(invoice))
+
+        return results

knv_cli/knv/invoices.py

@@ -5,8 +5,12 @@
 from datetime import datetime
 from os.path import basename
 
+import PyPDF2
 
-class Invoices:
+from ..command import Command
+
+
+class Invoices():
     # PROPS
 
     regex = '*_Invoices_TimeFrom*_TimeTo*.zip'
@@ -39,25 +43,203 @@ def remove(self, invoice_number: str) -> None:
         del self.invoices[invoice_number]
 
 
-    # HELPER methods
+    # PARSING methods
 
-    @staticmethod
-    def invoice2date(string: str) -> str:
-        # Distinguish between delimiters ..
-        # (1) .. hyphen ('Shopkonfigurator')
-        delimiter = '-'
+    def parse(self, invoice_file) -> list:
+        # Make sure given invoice is available for parsing
+        if self.invoice2number(invoice_file) not in self.invoices:
+            raise Exception
 
-        # (2) .. underscore ('Barsortiment')
-        if delimiter not in string:
-            delimiter = '_'
+        # Normalize input
+        invoice_file = self.invoices[self.invoice2number(invoice_file)]
+
+        # Extract general information from file name
+        invoice_date = self.invoice2date(invoice_file)
+        invoice_number = self.invoice2number(invoice_file)
+
+        # Prepare data storage
+        invoice = {
+            'Rechnungsnummer': invoice_number,
+            'Datum': invoice_date,
+            'Versandkosten': '0.00',
+            'Gesamtbetrag': 'keine Angabe',
+            'Steuern': {},
+            'Gutscheine': 'keine Angabe',
+        }
+
+        content = []
+
+        # Fetch content from invoice file
+        with open(invoice_file, 'rb') as file:
+            pdf = PyPDF2.PdfFileReader(file)
+
+            for page in pdf.pages:
+                content += [text.strip() for text in page.extractText().splitlines() if text]
+
+        # Determine invoice kind, as those starting with 'R' are formatted quite differently
+        if invoice_number[:1] == 'R':
+            # Parse content, looking for information about ..
+            # (1) .. general information
+            for line in content:
+                if 'Rechnungsbetrag gesamt brutto' in line:
+                    invoice['Gesamtbetrag'] = self.convert_number(content[content.index(line) + 1])
+
+                if 'Versandpauschale' in line or 'Versandkosten' in line:
+                    invoice['Versandkosten'] = self.convert_number(content[content.index(line) + 2])
+
+                # Edge case with two lines preceeding shipping cost
+                if 'versandt an' in line:
+                    invoice['Versandkosten'] = self.convert_number(content[content.index(line) + 2])
+
+            # (2) .. taxes
+            for tax_rate in ['5', '7', '16', '19']:
+                tax_string = 'MwSt. ' + tax_rate + ',00 %'
+
+                if tax_string in content:
+                    invoice['Steuern'][tax_rate + '%'] = self.convert_number(content[content.index(tax_string) + 2])
+
+            # (3) .. coupons
+            if 'Gutschein' in content:
+                coupons = []
+
+                for index in self.build_indices(content, 'Gutschein'):
+                    coupons.append({
+                        'Anzahl': int(content[index - 1]),
+                        'Wert': self.convert_number(content[index + 2]),
+                    })
+
+                invoice['Gutscheine'] = coupons
+
+        else:
+            # Gather general information
+            for index, line in enumerate(content):
+                # TODO: Get values via regexes
+                if 'Versandkosten:' in line:
+                    invoice['Versandkosten'] = self.convert_number(line.replace('Versandkosten:', ''))
+
+                if 'Gesamtbetrag' in line:
+                    invoice['Gesamtbetrag'] = self.convert_number(line.replace('Gesamtbetrag', ''))
+
+            # Fetch first occurence of ..
+            # (1) .. 'Nettobetrag' (= starting point)
+            starting_point = self.get_index(content, 'Nettobetrag')
+
+            # (2) .. 'Gesamtbetrag' (= terminal point)
+            terminal_point = self.get_index(content, 'Gesamtbetrag')
+
+            # Try different setup, since some invoices are the other way around
+            reverse_order = starting_point > terminal_point
+
+            if reverse_order:
+                # In this case, fetch last occurence of 'EUR' (= terminal point)
+                terminal_point = self.get_index(content, 'EUR', True)
+
+            costs = content[starting_point:terminal_point + 1]
+
+            # Determine tax rates:
+            # reduced = 5% or 7%
+            # full = 16% or 19%
+            tax_rates = [self.format_tax_rate(tax_rate) for tax_rate in costs[:2]]
+
+            reduced_tax = 0
+            full_tax = 0
+
+            if len(costs) < 8:
+                costs_list = costs[4].replace('MwSt. gesamt:', '').split()
+
+                reduced_tax = costs_list[0]
+                full_tax = costs_list[1]
+
+                if len(costs_list) < 3:
+                    full_tax = costs[5]
+
+            elif len(costs) == 9:
+                reduced_tax = costs[4].split(':')[-1]
+                full_tax = costs[5]
 
-        date_string = basename(string).split(delimiter)[1].replace('.pdf', '')
+                if 'MwSt. gesamt' in costs[5]:
+                    costs_list = costs[5].split(':')[-1].split()
+
+                    reduced_tax = costs_list[0]
+                    full_tax = costs_list[1]
+
+                if 'MwSt. gesamt' in costs[6]:
+                    reduced_tax = costs[6].split(':')[-1]
+                    full_tax = costs[7]
+
+
+            elif len(costs) in [10, 11]:
+                index = 6 if 'MwSt.' in costs[6] else 5
+
+                reduced_tax = costs[index].split(':')[-1].split()[0]
+                full_tax = costs[index + 1].split()[0]
+
+            else:
+                reduced_tax = costs[5].split()[0]
+                full_tax = costs[2].split()[2]
+
+                if reduced_tax == 'MwSt.':
+                    reduced_tax = costs[5].split(':')[-1]
+                    full_tax = costs[6]
+
+            invoice['Steuern'][tax_rates[0]] = self.convert_number(reduced_tax)
+            invoice['Steuern'][tax_rates[1]] = self.convert_number(full_tax)
+
+        return invoice
+
+
+    # PARSING HELPER methods
+
+    def format_tax_rate(self, string: str) -> str:
+        return (string[:-1].replace('Nettobetrag', '')).strip()
+
+
+    def get_index(self, haystack: list, needle: str, last: bool = False) -> int:
+        position = 0
+
+        if last:
+            position = -1
+
+        return [i for i, string in enumerate(haystack) if needle in string][position]
+
+
+    def build_indices(self, haystack: list, needle: str) -> list:
+        return [count for count, line in enumerate(haystack) if line == needle]
+
+
+    def convert_number(self, string) -> str:
+        # Clear whitespaces & convert to string (suck it, `int` + `float`)
+        string = str(string).replace('EUR', '').strip()
+
+        # Take care of thousands separator, as in '1.234,56'
+        if '.' in string and ',' in string:
+            string = string.replace('.', '')
+
+        string = float(string.replace(',', '.'))
+        integer = f'{string:.2f}'
+
+        return str(integer)
+
+
+    # HELPER methods
+
+    def invoice2date(self, string: str) -> str:
+        date_string = self.split_string(string)[1].replace('.pdf', '')
 
         return datetime.strptime(date_string, '%Y%m%d').strftime('%Y-%m-%d')
 
 
-    @staticmethod
-    def invoice2number(string: str) -> str:
+    def invoice2number(self, string: str) -> str:
+        string_list = self.split_string(string)
+
+        # Sort out invoice numbers
+        if len(string_list) == 1:
+            return string
+
+        return string_list[-1].replace('.pdf', '')
+
+
+    def split_string(self, string: str) -> str:
         # Strip path information
         string = basename(string)
 
@@ -69,8 +251,4 @@ def invoice2number(string: str) -> str:
         if delimiter not in string:
             delimiter = '_'
 
-        # (3) .. as well as invoice numbers
-        if delimiter not in string:
-            return string
-
-        return string.split(delimiter)[-1].replace('.pdf', '')
+        return string.split(delimiter)

knv_cli/knv/orders.py

@@ -76,3 +76,9 @@ def process_data(self, order_data: list) -> dict:
     def orders(self):
         # Sort orders by date & order number, output as list
         return sorted(list(self.data.values()), key=itemgetter('Datum', 'ID'))
+
+
+    # DATA HELPER methods
+
+    def convert_tax_rate(self, string: str) -> str:
+        return str(string).replace(',00', '') + '%'