Text coordinates in markup.

Author: cann3v

auxil.py

@@ -2,13 +2,15 @@
 import consts
 
 import re
-from loguru import logger
 from sys import stdout, platform
 from random import randint, choice
 import os
 import subprocess as sb
 from argparse import ArgumentTypeError
 
+from PyPDF2 import PdfReader
+from loguru import logger
+
 def logger_config(v):
 	logger.remove()
 	if int(v) == 0:
@@ -243,3 +245,199 @@ def calculate_seal_coords(sign_coords, new_page=False):
 	seal_coords = [[x1-pd, y1-pd], [x2+pd, y2+pd]]
 
 	return seal_coords
+
+# coords: list - список из пар координат (x, y)
+def calculate_borders(original_coords, creator_and_date=False, task=False):
+
+	def calculate(coords):
+		if len(coords) == 1:
+			x1 = min(coords)
+			y1 = coords[0][1]
+			y2 = coords[0][1]
+		
+		elif len(coords) > 1:
+			x1 = min(coords)
+			y1 = 10000
+			y2 = 0
+
+			for pair in coords:
+				if pair[1] < y1:
+					y1 = pair[1]
+
+				if pair[1] > y2:
+					y2 = pair[1]
+
+		else:
+			return []
+
+		x1 = PDFunits_to_px(x1[0])
+		y1 = PDFunits_to_px(y1)
+		y2 = PDFunits_to_px(y2)
+
+		x_offset = consts.text_borders[0]
+		y_offset = consts.text_borders[1]
+
+		if task:
+			return [[x1 - x_offset, y1 - y_offset * 2], [2385, y2 + y_offset / 2]]
+		elif  creator_and_date:
+			return [[x1 - x_offset, y1 - y_offset * 0.75], [2385, y2 + y_offset / 4]]
+		else:
+			return [[x1 - x_offset, y1 - y_offset * 1.6], [2385, y2 + y_offset / 2]]
+
+	if original_coords == ["page_break"]:
+		return original_coords
+	elif "page_break" in original_coords:
+		splitted_coords = []
+		result = []
+		for pair in original_coords:
+			if pair != "page_break":
+				splitted_coords.append(pair)
+			else:
+				result.append(calculate(splitted_coords))
+				result.append("page_break") 
+				splitted_coords = []
+	else:
+		return calculate(original_coords)
+
+	return result
+
+
+# pdf_path: str - путь к pdf файлу
+# data: tuple - кортеж данных для генерации
+def calculate_text_coords(pdf_path, data):
+	header, header_coords = data[0], []
+	name, name_coords = data[1], []
+	intro, intro_coords = data[2], []
+	instruction, instruction_coords = data[3], [[] * i for i in range(len(data[3]))]
+	responsible, responsible_coords = data[4], []
+	creator, creator_coords = data[5], []
+	date, date_coords = data[6], []
+
+	reader = PdfReader(pdf_path)
+	for page in reader.pages:
+	# page = reader.pages[0]
+
+		raw_data = []
+		def visitor_t(text, cm, tm, fontDict, fontSize):
+			raw_data.append([text, tm[4], tm[5]]) # [text, x1, y1]
+
+		page.extract_text(visitor_text=visitor_t) # Посимвольное извлечение координат текста
+
+		for i in range(len(raw_data)):
+			raw_data[i][1] = int(raw_data[i][1])
+			raw_data[i][2] = int(raw_data[i][2])
+
+		# Объединение символов с одинаковыми координатами в строки
+		# text - список со строками
+		# coords - список с соответствующими строкам координатами
+		text, coords = [], []
+		for i in range(len(raw_data)):
+			if [raw_data[i][1], raw_data[i][2]] not in coords:
+				coords.append([raw_data[i][1], raw_data[i][2]])
+				text.append(raw_data[i][0])
+			else:
+				text[-1] += raw_data[i][0]
+
+		if (len(text) != len(coords)):
+			logger.error("[text] != [coords]")
+			raise SystemExit
+
+		formatted_markup = {}
+		for i in range(len(text)):
+			if text[i] == '':
+				continue
+
+			formatted_markup[text[i]] = coords[i]
+
+		# Если ключ словаря (строка в документе) входит в секцию данных для генерации,
+		# то добавить координаты строки в соответствующий список
+		for k in list(formatted_markup):
+
+			if ((k.replace('\n', '') in header) and (not name_coords)):
+				val = formatted_markup.pop(k)
+				header_coords.append(val)
+
+			if ((k.replace('\n', '') in name) and (not intro_coords)):
+				try:
+					val = formatted_markup.pop(k)
+					name_coords.append(val)
+				except KeyError:
+					pass
+
+			if (k.replace('\n', '') in intro):
+				try:
+					val = formatted_markup.pop(k)
+					intro_coords.append(val)
+				except KeyError:
+					pass
+
+			for i in range(len(instruction)):
+				task = instruction[i]["task_text"]
+				if (k.replace('\n', '') in task):
+					try:
+						val = formatted_markup.pop(k)
+						instruction_coords[i].append(val)
+					except KeyError:
+						pass
+
+			if (k.replace('\n', '') in responsible):
+				try:
+					val = formatted_markup.pop(k)
+					responsible_coords.append(val)
+				except KeyError:
+					pass
+
+			if (k.replace('\n', '') in creator):
+				try:
+					val = formatted_markup.pop(k)
+					creator_coords.append(val)
+				except KeyError:
+					pass
+
+			if (k.replace('\n', '') in '.' + date):
+				try:
+					val = formatted_markup.pop(k)
+					date_coords.append(val)
+				except KeyError:
+					pass
+
+		header_coords.append("page_break")
+		name_coords.append("page_break")
+		intro_coords.append("page_break")
+		instruction_coords.append(["page_break"])
+		responsible_coords.append("page_break")
+		creator_coords.append("page_break")
+		date_coords.append("page_break")
+
+
+	header_coords = calculate_borders(header_coords)
+	
+	name_coords = calculate_borders(name_coords)
+	
+	intro_coords = calculate_borders(intro_coords)
+	
+	task_coords = []
+	for task in instruction_coords:
+		task_coords.append(calculate_borders(task, task=True))
+
+	responsible_coords = calculate_borders(responsible_coords)
+
+	creator_coords = calculate_borders(creator_coords, creator_and_date=True)
+
+	date_coords = calculate_borders(date_coords, creator_and_date=True)
+
+	# try:
+	# 	intro_coords[0][1] -= 55
+	# except TypeError:
+	# 	pass
+
+	# for i in range(len(task_coords)):
+	# 	try:
+	# 		task_coords[i][0][1] -= 80
+	# 	except TypeError:
+	# 		continue
+
+	result = [header_coords, name_coords, intro_coords, task_coords, responsible_coords,
+		creator_coords, date_coords]
+
+	return result

consts.py

@@ -42,3 +42,5 @@
 
 page_w = 612 # PDFUnits
 page_h = 792 # PDFUnits
+
+text_borders = (20, 60) # text offset in px

gen.py

@@ -115,8 +115,14 @@ def generate(data, out, formats, size, samples_dir, is_image):
 
 		if 'p' in formats:
 			pdf_path = write.write_pdf_linux(docx_path, out, count)
+
+			generation_data = (header, name, intro, instruction,
+				responsible, creator, date[0])
+			
 			if is_image:
-				write.write_coords(json_path, pdf_path)
+				write.write_coords(json_path, pdf_path, generation_data, is_image=True)
+			else:
+				write.write_coords(json_path, pdf_path, generation_data)
 
 		if 'j' in formats:
 			write.write_jpg(out, count)

write.py

@@ -197,39 +197,52 @@ def visitor_sign(text, cm, tm, fontDict, fontSize):
 
 	return (tmx, tmy, len(page.images))
 
-def write_coords(json_path, pdf_path):	
-	# Координаты логотипа
-	logo_coords = auxil.calculate_logo_coords()
+def write_coords(json_path, pdf_path, data, is_image=False):
 
-	# Координаты подписи
-	(tmx, tmy, im_count) = extract_tm(pdf_path, -1)
+	with open(json_path, "r") as json_file:
+		json_dict = json.load(json_file)
 
-	if im_count >= 2: # Если на последней странице есть подпись и печать
+	if is_image:
 
-		if (tmx == 0) or (tmy == 0): # Если на странице только подпись и печать
-			sign_coords = auxil.calculate_sign_coords(tmx, tmy, new_page=True)
-			
-		else:
-			sign_coords = auxil.calculate_sign_coords(tmx, tmy)
+		# Координаты логотипа
+		logo_coords = auxil.calculate_logo_coords()
 
-	else: # Если на последней странице только печать
-		(tmx, tmy, _) = extract_tm(pdf_path, -2)
-		sign_coords = auxil.calculate_sign_coords(tmx, tmy)
+		# Координаты подписи
+		(tmx, tmy, im_count) = extract_tm(pdf_path, -1)
 
-	# Координаты печати
-	if im_count >= 2:
-		seal_coords = auxil.calculate_seal_coords(sign_coords)
+		if im_count >= 2: # Если на последней странице есть подпись и печать
 
-	else:
-		seal_coords = auxil.calculate_seal_coords([], new_page=True)
+			if (tmx == 0) or (tmy == 0): # Если на странице только подпись и печать
+				sign_coords = auxil.calculate_sign_coords(tmx, tmy, new_page=True)
+				
+			else:
+				sign_coords = auxil.calculate_sign_coords(tmx, tmy)
 
-	with open(json_path, "r") as json_file:
-		json_dict = json.load(json_file)
+		else: # Если на последней странице только печать
+			(tmx, tmy, _) = extract_tm(pdf_path, -2)
+			sign_coords = auxil.calculate_sign_coords(tmx, tmy)
+
+		# Координаты печати
+		if im_count >= 2:
+			seal_coords = auxil.calculate_seal_coords(sign_coords)
+
+		else:
+			seal_coords = auxil.calculate_seal_coords([], new_page=True)
 
-	json_dict["Images"] = {}
-	json_dict["Images"]["logo_coordinates"] = logo_coords
-	json_dict["Images"]["signature_coordinates"] = sign_coords
-	json_dict["Images"]["seal_coordinates"] = seal_coords
+		json_dict["Images"] = {}
+		json_dict["Images"]["logo_coordinates"] = logo_coords
+		json_dict["Images"]["signature_coordinates"] = sign_coords
+		json_dict["Images"]["seal_coordinates"] = seal_coords
+
+	text_coords = auxil.calculate_text_coords(pdf_path, data)
+	json_dict["Text"] = {}
+	json_dict["Text"]["header"] = text_coords[0]
+	json_dict["Text"]["name"] = text_coords[1]
+	json_dict["Text"]["intro"] = text_coords[2]
+	json_dict["Text"]["tasks"] = text_coords[3]
+	json_dict["Text"]["responsible"] = text_coords[4]
+	json_dict["Text"]["creator"] = text_coords[5]
+	json_dict["Text"]["date"] = text_coords[6]
 
 	with open(json_path, "w") as jsonf:
 		json.dump(json_dict, jsonf, ensure_ascii=False, indent=4)