From a7fc7a956ca0d3c30a15b824bed1ca80022119d4 Mon Sep 17 00:00:00 2001
From: leo <Leo.Strohmidel@gmx.net>
Date: Wed, 27 Jun 2018 20:20:56 +0200
Subject: [PATCH] +comments
---
pdf_parser.py | 22 ++++++++++++++++++++++
template_manager.py | 7 +++++++
2 files changed, 29 insertions(+)
diff --git a/pdf_parser.py b/pdf_parser.py
index d1727ba..0d5dd66 100644
--- a/pdf_parser.py
+++ b/pdf_parser.py
@@ -12,6 +12,9 @@ from pdfminer.layout import LTTextBoxHorizontal
def get_text_elements(p_path):
+ """
+ Gets all text-elements of the given document as list.
+ """
fp = open(p_path, 'rb')
parser = PDFParser(fp)
document = PDFDocument(parser)
@@ -37,6 +40,11 @@ def get_text_elements(p_path):
def find_relative_position(p_key_box, p_value_box):
+ """
+ Calculates the distance from key_box to value_box in order to provide a
+ position relative to key_box. Example:
+ key_box conatains "Total" and the value_box "34.42€"
+ """
kx0, ky0, kx1, ky1 = p_key_box
vx0, vy0, vx1, vy1 = p_value_box
center_value_x = (abs(vx0 - vx1) / 2.0) + vx0
@@ -45,6 +53,9 @@ def find_relative_position(p_key_box, p_value_box):
def get_word_offset(p_sentence, p_word, p_seperator=" "):
+ """
+ Gets the count (index) of a word in a sentence/string
+ """
# das wievielte wort eines satzes
p_sentence = str(p_sentence)
p_sentence = p_sentence.replace("\n", " ")
@@ -58,6 +69,11 @@ def get_word_offset(p_sentence, p_word, p_seperator=" "):
def generate_template(p_example_dict, p_text_element_list):
+ """
+ Generates a template from examples.
+ The exmample is a dict containing key/value pair that are found in the
+ element list.
+ """
template = {}
for key, value in p_example_dict.items():
key_box = None
@@ -86,6 +102,9 @@ def generate_template(p_example_dict, p_text_element_list):
def is_point_in_box(p_point, p_box, p_padding=5):
+ """
+ Checks wether a point is in a specific box. The point can be padding of.
+ """
x0, y0, x1, y1 = p_box
x, y = p_point
x0 -= p_padding
@@ -96,6 +115,9 @@ def is_point_in_box(p_point, p_box, p_padding=5):
def read_template(p_template, p_text_element_list):
+ """
+ Analysis all text elements according to the given template.
+ """
for key, pos in p_template.iteritems():
key_box = None
for element in p_text_element_list:
diff --git a/template_manager.py b/template_manager.py
index 574ea0d..8e18e49 100644
--- a/template_manager.py
+++ b/template_manager.py
@@ -3,6 +3,10 @@ import json
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)
template = generate_template(p_examples, elements)
data = [p_examples, template]
@@ -12,6 +16,9 @@ def create_template(p_name, p_path, p_examples):
def evaluate_pdf(p_template_name, p_path):
+ """
+ Analysis a pdf according to the given template
+ """
with open(p_template_name + '.template') as f:
data = json.load(f)
--
GitLab