diff --git a/pdf_parser.py b/pdf_parser.py index 08c5b1d2929488b6d20eb9138c0215234d4bf718..f157cf40325523caabfd0e5f91d9bad2639110f5 100644 --- a/pdf_parser.py +++ b/pdf_parser.py @@ -11,7 +11,7 @@ from pdfminer.converter import PDFPageAggregator from pdfminer.layout import LTTextBoxHorizontal -def get_text_elements(p_path): +def get_textelements_from_pdf(p_path): """ Gets all text-elements of the given document as list. """ @@ -114,10 +114,11 @@ def is_point_in_box(p_point, p_box, p_padding=5): return x0 <= x and x <= x1 and y0 <= y and y <= y1 -def read_template(p_template, p_text_element_list): +def get_contents_from_textelements(p_template, p_text_element_list): """ Analysis all text elements according to the given template. """ + text_values = [] for key, pos in p_template.items(): key_box = None for element in p_text_element_list: @@ -138,6 +139,8 @@ def read_template(p_template, p_text_element_list): for i in range(max_padding): if is_point_in_box(estimated_position, element.bbox, i): text = element.get_text().replace("\n", " ").split(" ") + text_values.append({key: text[offset]}) print("Found value %s: '%s' with offset %d" % (key, text[offset], i)) break #print("Found no value for '%s' with max padding %d" % (key, max_padding)) + return text_values diff --git a/template_manager.py b/template_manager.py index 8e18e49b25141171fde87578dcb61aff91f93b9b..59d7b34bee43589dc6e83428cee417eab03911db 100644 --- a/template_manager.py +++ b/template_manager.py @@ -1,4 +1,4 @@ -from pdf_parser import get_text_elements, generate_template, read_template +from pdf_parser import get_textelements_from_pdf, generate_template, get_contents_from_textelements import json @@ -7,7 +7,7 @@ def create_template(p_name, p_path, p_examples): Generates a template from example and pdf file and stores it as json """ - elements = get_text_elements(p_path) + elements = get_textelements_from_pdf(p_path) template = generate_template(p_examples, elements) data = [p_examples, template] @@ -20,12 +20,12 @@ def evaluate_pdf(p_template_name, p_path): Analysis a pdf according to the given template """ + print('\nEvaluating:', p_path) with open(p_template_name + '.template') as f: - data = json.load(f) - - template = data[1] - elements = get_text_elements(p_path) - content = read_template(template, elements) + data = json.load(f) # [{'Rechnung': 'EB589544', 'Rechnungsdatum': '01.09.2016', 'Mandatsreferenz': '8ADB836C18A8491092EA62F7F35F8A28'}, {'Rechnung': [[125.29500000000002, 29.985000000000014], 2], 'Rechnungsdatum': [[125.29500000000002, 29.985000000000014], 0], 'Mandatsreferenz': [[125.90500000000006, 6.095000000000027], 1]}] + template = data[1] # {'Rechnung': [[125.29500000000002, 29.985000000000014], 2], 'Rechnungsdatum': [[125.29500000000002, 29.985000000000014], 0], 'Mandatsreferenz': [[125.90500000000006, 6.095000000000027], 1]} + textelements = get_textelements_from_pdf(p_path) # [<LTTextBoxHorizontal(0) 60.000,712.080,231.864,721.376 'sipgate GmbH - Gladbacher Str. 74 - 40219 Düsseldorf\n'>, <LTTextBoxHorizontal(1) 65.000,663.580,163.690,699.200 'Warpzone e.V.\nJan-Marten Brüggemann\nAm Hawerkamp 31\n'>, <LTTextBoxHorizontal(2) 65.000,639.470,129.350,651.660 '48155 Münster\n'>, <LTTextBoxHorizontal(3) 375.000,663.580,451.560,723.200 'Rechnungsdatum\nLeistungsdatum\nRechnungsnummer\nBezahlung per\nKundennummer\n'>, <LTTextBoxHorizontal(4) 465.000,663.470,535.590,723.660 '01.09.2016\n01.09.2016\nEB589544\nSEPA-Lastschrift\n1967957\n'>, <LTTextBoxHorizontal(5) 375.000,617.500,395.280,629.120 'Seite\n'>, <LTTextBoxHorizontal(6) 465.000,617.390,485.230,629.580 '1 / 1\n'>, <LTTextBoxHorizontal(7) 60.000,555.824,200.976,575.328 'Rechnung EB589544\n'>, <LTTextBoxHorizontal(8) 90.000,537.390,184.160,549.580 'Art.-Nr. Bezeichnung\n'>, <LTTextBoxHorizontal(9) 60.000,525.500,78.050,549.580 'Pos.\n \n'>, <LTTextBoxHorizontal(10) 280.000,537.390,307.500,549.580 'Menge\n'>, <LTTextBoxHorizontal(11) 318.000,525.500,364.840,549.580 'Einzelpreis\nnetto\n'>, <LTTextBoxHorizontal(12) 378.000,525.500,424.840,549.580 'Einzelpreis\nbrutto\n'>, <LTTextBoxHorizontal(13) 438.000,525.500,452.930,549.580 'USt\n \n'>, <LTTextBoxHorizontal(14) 468.000,525.500,520.580,549.580 'Gesamtpreis\nnetto\n'>, <LTTextBoxHorizontal(15) 60.000,489.500,67.330,501.120 ' 1\n'>, <LTTextBoxHorizontal(16) 90.000,489.500,95.000,501.120 '1\n'>, <LTTextBoxHorizontal(17) 130.000,489.500,252.050,501.120 'sipgate.de, Telefonieguthaben\n'>, <LTTextBoxHorizontal(18) 302.440,489.500,307.440,501.120 '1\n'>, <LTTextBoxHorizontal(19) 348.360,489.500,365.250,501.120 '8,40\n'>, <LTTextBoxHorizontal(20) 402.800,489.500,424.650,501.120 '10,00\n'>, <LTTextBoxHorizontal(21) 438.000,489.500,455.590,501.120 '19%\n'>, <LTTextBoxHorizontal(22) 498.950,489.500,534.210,501.120 '8,40 EUR\n'>, <LTTextBoxHorizontal(23) 310.000,447.500,408.610,459.120 'Summe Positionen netto\n'>, <LTTextBoxHorizontal(24) 310.000,411.500,419.970,423.120 '19% USt. auf EUR 8,40 (DE)\n'>, <LTTextBoxHorizontal(25) 492.740,447.390,531.790,459.580 '8,40 EUR\n'>, <LTTextBoxHorizontal(26) 492.780,411.390,531.830,423.580 '1,60 EUR\n'>, <LTTextBoxHorizontal(27) 310.000,369.390,383.390,381.580 'Rechnungsbetrag\n'>, <LTTextBoxHorizontal(28) 487.780,369.390,532.390,381.580 '10,00 EUR\n'>, <LTTextBoxHorizontal(29) 60.000,307.390,414.900,331.580 'Der Betrag in Höhe von 10,00 EUR wird am 06.09.2016 von Ihrem Konto abgebucht.\nInhaber: WARPZONE E.V., IBAN: DE70XXXXXXXXXXXXXXXXX3738, BIC: WELADED1MST\n'>, <LTTextBoxHorizontal(30) 60.000,283.390,311.810,295.580 'Mandatsreferenz: 8ADB836C18A8491092EA62F7F35F8A28\n'>, <LTTextBoxHorizontal(31) 60.000,223.500,485.630,271.120 'Die Umsatzsteuer wird in Höhe des in Ihrem Land geltenden Umsatzsteuersatzes in Rechnung gestellt. Der\nUmsatzsteuersatz bestimmt sich nach dem Ort der Leistung, d.h. nach dem Sitz Ihres Unternehmens bzw.\nnach Ihrem gewöhnlichen Aufenthaltsort. Sollten Ihre Rechnung (Anschrift bzw. Umsatzsteuersatz) nicht\nkorrekt sein, informieren Sie bitte unsere Kundenbetreuung unter basic@sipgate.de.\n'>, <LTTextBoxHorizontal(32) 50.000,58.000,377.192,77.296 'sipgate GmbH, Gladbacher Str. 74, 40219 Düsseldorf, HRB 39841 Düsseldorf, GF: Tim Mois, Thilo Salmon\nUSt-ID: DE219349391, Finanzamt Düsseldorf, Steuer-Nr.: 106/5724/7147, Support: basic@sipgate.de\n'>, <LTTextBoxHorizontal(33) 50.000,28.000,334.008,47.296 'Bank: Commerzbank Düsseldorf, IBAN: DE10 3004 0000 0181 1488 06, BIC: COBADEFFXXX\nGläubiger-ID: DE73ZZZ00000359204\n'>] + contents = get_contents_from_textelements(template, textelements) # None with open("output.shmebulok", "w") as out: - json.dump(content, out) + json.dump(contents, out)