auf python 3 konvertiert, schönheitsop

pdf_parser.py

@@ -10,29 +10,31 @@ from pdfminer.converter import PDFPageAggregator
 
 from pdfminer.layout import LTTextBoxHorizontal
 
+
 def get_text_elements(p_path):
-        fp = open(p_path, 'rb')
-        parser = PDFParser(fp)
-        document = PDFDocument(parser)
-        if not document.is_extractable:
-            raise PDFTextExtractionNotAllowed
-        rsrcmgr = PDFResourceManager()
-        laparams = LAParams(line_margin=0.1)
-        device = PDFPageAggregator(rsrcmgr, laparams=laparams)
-        interpreter = PDFPageInterpreter(rsrcmgr, device)
-
-        pages = []
-        for page in PDFPage.create_pages(document):
-            interpreter.process_page(page)
-            layout = device.get_result()
-
-            elements_on_page = []
-            for element in layout:
-                if type(element) is LTTextBoxHorizontal:
-                    pages.append(element)
-            #pages.append(elements_on_page)
-
-        return pages
+    fp = open(p_path, 'rb')
+    parser = PDFParser(fp)
+    document = PDFDocument(parser)
+    if not document.is_extractable:
+        raise PDFTextExtractionNotAllowed
+    rsrcmgr = PDFResourceManager()
+    laparams = LAParams(line_margin=0.1)
+    device = PDFPageAggregator(rsrcmgr, laparams=laparams)
+    interpreter = PDFPageInterpreter(rsrcmgr, device)
+
+    pages = []
+    for page in PDFPage.create_pages(document):
+        interpreter.process_page(page)
+        layout = device.get_result()
+
+        elements_on_page = []
+        for element in layout:
+            if type(element) is LTTextBoxHorizontal:
+                pages.append(element)
+        #pages.append(elements_on_page)
+
+    return pages
+
 
 def find_relative_position(p_key_box, p_value_box):
     kx0, ky0, kx1, ky1 = p_key_box
@@ -41,9 +43,12 @@ def find_relative_position(p_key_box, p_value_box):
     center_value_y = (abs(vy0 - vy1) / 2.0) + vy0
     return (center_value_x - kx0, center_value_y - ky0)
 
+
 def get_word_offset(p_sentence, p_word, p_seperator=" "):
-    p_sentence = p_sentence.encode()
+    # das wievielte wort eines satzes
+    p_sentence = str(p_sentence)
     p_sentence = p_sentence.replace("\n", " ")
+    # print('get_word_offset p_sentence(' + str(p_sentence) + '), p_word(' + str(p_word) + ')')
     p_sentence = p_sentence.split(p_seperator)
     for i in range(len(p_sentence)):
         current_word = p_sentence[i]
@@ -51,9 +56,10 @@ def get_word_offset(p_sentence, p_word, p_seperator=" "):
             return i
     raise RuntimeError("Sentence does not contains word: %s -> %s" % (p_word, p_sentence))
 
+
 def generate_template(p_example_dict, p_text_element_list):
     template = {}
-    for key, value in p_example_dict.iteritems():
+    for key, value in p_example_dict.items():
         key_box = None
         value_box = None
 
@@ -78,6 +84,7 @@ def generate_template(p_example_dict, p_text_element_list):
 
     return template
 
+
 def is_point_in_box(p_point, p_box, p_padding=5):
     x0, y0, x1, y1 = p_box
     x, y = p_point
@@ -87,6 +94,7 @@ def is_point_in_box(p_point, p_box, p_padding=5):
     y1 += p_padding
     return x0 <= x and x <= x1 and y0 <= y and y <= y1
 
+
 def read_template(p_template, p_text_element_list):
     for key, pos in p_template.iteritems():
         key_box = None

pdf_parser_test.py

-#from pdf_parser import get_text_elements, generate_template, read_template
+# from pdf_parser import get_text_elements, generate_template, read_template
 from template_manager import create_template
 
-#path = "/home/reverend/Programmierung/git_projects/warpzone-rechnungen/2017/Eingangsrechnungen/2017-08-12 rabe getraenke 33971.pdf"
+# path = "/home/reverend/Programmierung/git_projects/warpzone-rechnungen/2017/Eingangsrechnungen/2017-08-12 rabe getraenke 33971.pdf"
 path = "/home/reverend/Programmierung/git_projects/warpzone-rechnungen/2016/Eingangsrechnungen/2016-09-01 sipgate EB589544.pdf"
-#e = get_text_elements(path)
+path = "pdf/2016-09-01 sipgate EB589544.pdf"
+# e = get_text_elements(path)
 
-#examples = {"Rechnungsbetrag": "183,43", "Datum": "12.8.2017", "RECHNUNGSNr": "33971"}
+# examples = {"Rechnungsbetrag": "183,43", "Datum": "12.8.2017", "RECHNUNGSNr": "33971"}
 examples = {"Rechnung": "EB589544", "Rechnungsdatum": "01.09.2016", "Mandatsreferenz": "8ADB836C18A8491092EA62F7F35F8A28"}
 create_template("sipgate", path, examples)
 
-#g = generate_template(examples, e[0])
+# g = generate_template(examples, e[0])
 
-#path = "/home/reverend/Programmierung/git_projects/warpzone-rechnungen/2016/Eingangsrechnungen/2016-07-27 sipgate EB540908.pdf"
-#e = get_text_elements(path)
-#read_template(g, e[0])
+# path = "/home/reverend/Programmierung/git_projects/warpzone-rechnungen/2016/Eingangsrechnungen/2016-07-27 sipgate EB540908.pdf"
+# e = get_text_elements(path)
+# read_template(g, e[0])

template_manager.py

 from pdf_parser import get_text_elements, generate_template, read_template
 import json
 
+
 def create_template(p_name, p_path, p_examples):
     elements = get_text_elements(p_path)
     template = generate_template(p_examples, elements)
@@ -9,6 +10,7 @@ def create_template(p_name, p_path, p_examples):
     with open(p_name + '.template', 'w') as outfile:
         json.dump(data, outfile)
 
+
 def evaluate_pdf(p_template_name, p_path):
 
     with open(p_template_name + '.template') as f: