求助：用python程序直接读取pdf内容并解析提前里面的资料读取pdf里的内容信息，并将pdf里的楼宇名称和中英文
求助：用python程序直接读取pdf内容并解析提前里面的资料
读取pdf里的内容信息，并将pdf里的楼宇名称和中英文地址提取出来，保存到csv file里，
pdf内容说明：前面的一些序言是不要的，每条完整的记录是用一行空白行来隔开的，这也就说明有些楼宇名称及地址是分布在几行中
例如：
A. Sun Plaza 3 San Hop Lane, 屯門新合里 3 號 1997
*愛善中心 Tuen Mun
注意此空白行为一条记录的分割
A.D. & F.D. of Pok Oi Hospital 111 So Kwun Wat Rd., 屯門 -
Mrs. Cheng Yam On Tuen Mun 掃管笏路 111 號
Millennium School
博愛醫院歷屆總理聯誼會
鄭任安夫人千禧小學
提取出来的资料为：
楼宇名称 地址 建筑年份
英文： A. Sun Plaza San Hop Lane,Tuen Mun 1997
中文： *愛善中心 屯門新合里 3 號 1997
A.D. & F.D. of Pok Oi Hospital Mrs. Cheng Yam On Millennium School 111 So Kwun Wat Rd.,Tuen Mun -
博愛醫院歷屆總理聯誼會鄭任安夫人千禧小學 屯門掃管笏路 111 號 -
首选python,实在不行，Java也行
[解决办法]
试试 PDFTron
[解决办法]
- Python code
#---------------------------------------------------# Copyright (c) 2001-2012 by PDFTron Systems Inc. All Rights Reserved.# Consult legal.txt regarding legal and license information.#---------------------------------------------------import sitesite.addsitedir("../../../Lib")import sysif sys.version_info.major < 3: from PDFNetPython2 import *else: from PDFNetPython3 import *def printStyle (style): print(" style=\"font-family:" + style.GetFontName() + "; font-size:" + str(style.GetFontSize()) + "; sans-serif: " + str(style.IsSerif()) + "; color:" + str(style.GetColor())+ "\"")def dumpAllText (reader): element = reader.Next() while element != None: type = element.GetType() if type == Element.e_text_begin: print("Text Block Begin") elif type == Element.e_text_end: print("Text Block End") elif type == Element.e_text: bbox = element.GetBBox() print("BBox: " + str(bbox.GetX1()) + ", " + str(bbox.GetY1()) + ", " + str(bbox.GetX2()) + ", " + str(bbox.GetY2())) print(element.GetTextString()) elif type == Element.e_text_new_line: print("New Line") elif type == Element.e_form: reader.FormBegin() dumpAllText(reader) reader.End() element = reader.Next()# A utility method used to extract all text content from# a given selection rectangle. The recnagle coordinates are# expressed in PDF user/page coordinate system.def ReadTextFromRect (page, pos, reader): reader.Begin(page) srch_str = RectTextSearch(reader, pos) reader.End() return srch_str#A helper method for ReadTextFromRectdef RectTextSearch (reader, pos): element = reader.Next() srch_str2 = "" while element != None: type = element.GetType() if type == Element.e_text: bbox = element.GetBBox() if (bbox.IntersectRect(bbox, pos)): arr = element.GetTextString() srch_str2 += arr srch_str2 += "\n" elif type == Element.e_text_new_line: None elif type == Element.e_form: reader.FormBegin() srch_str2 += RectTextSearch(reader, pos) print(srch_str2) reader.End() element = reader.Next() return srch_str2 def main(): PDFNet.Initialize() # Relative path to the folder containing test files. input_path = "../../TestFiles/newsletter.pdf" example1_basic = True example2_xml = True example3_wordlist = True example4_advanced = True example5_low_level = True # Sample code showing how to use high-level text extraction APIs. doc = PDFDoc(input_path) doc.InitSecurityHandler() page = doc.GetPage(1) if page == None: print("page no found") txt = TextExtractor() txt.Begin(page) # Read the page # Example 1. Get all text on the page in a single string. # Words will be separated witht space or new line characters. if example1_basic: print("Word count: " + str(txt.GetWordCount())) print("- GetAsText --------------------------" + txt.GetAsText()) print("-----------------------") # Example 2. Get XML logical structure for the page. if example2_xml: text = txt.GetAsXML(TextExtractor.e_words_as_elements | TextExtractor.e_output_bbox | TextExtractor.e_output_style_info) print("- GetAsXML --------------------------" + text) print("-----------------------") # Example 3. Extract words one by one. if example3_wordlist: word = Word() line = txt.GetFirstLine() while line.IsValid(): word = line.GetFirstWord() while word.IsValid(): print(word.GetString()) word = word.GetNextWord() line = line.GetNextLine() print("-----------------------") print("Example 4") # Example 4. A more advanced text extraction example. # The output is XML structure containing paragraphs, lines, words, # as well as style and positioning information. if example4_advanced: bbox = Rect(); cur_flow_id = -1 cur_para_id = -1 # For each line on the page... line = txt.GetFirstLine() while line.IsValid(): if line.GetNumWords() == 0: continue word = line.GetFirstWord() if cur_flow_id != line.GetFlowID(): if cur_flow_id != -1: if cur_para_id != -1: cur_para_id = -1; print("</Para>") print("</Flow>") cur_flow_id = line.GetFlowID() print("<Flow id=\"" + str(cur_flow_id) +"\">") if cur_para_id != line.GetParagraphID(): if cur_para_id != -1: print("</Para>") cur_para_id= line.GetParagraphID() print("<Para id=\"" +str(cur_para_id)+ "\">") bbox = line.GetBBox() line_style = line.GetStyle() sys.stdout.write("<Line box=\"" + str(bbox.GetX1()) + ", " + str(bbox.GetY1()) + ", " + str(bbox.GetX2()) + ", " + str(bbox.GetY2()) + "\"") printStyle (line_style) sys.stdout.write(">") # For each word in the line... word = line.GetFirstWord() while word.IsValid(): # Output the bounding box for the word bbox = word.GetBBox() sys.stdout.write("<Word box=\"" + str(bbox.GetX1()) + ", " + str(bbox.GetY1()) + ", " + str(bbox.GetX2()) + ", " + str(bbox.GetY2()) + "\"") sz = word.GetStringLen() if sz == 0: continue # If the word style is different from the parent style, output the new style. s = word.GetStyle() if s != line_style: printStyle (s); sys.stdout.write(">" + word.GetString() + "</Word>\n") word = word.GetNextWord() line = line.GetNextLine() if cur_flow_id != -1: if cur_para_id != -1: cur_para_id = -1 sys.stdout.write("</Para>\n") sys.stdout.write("</Flow>\n") txt.Destroy() doc.Close() print("Done.") # Sample code showing how to use low-level text extraction APIs. if example5_low_level: doc = PDFDoc(input_path) doc.InitSecurityHandler() # Example 1. Extract all text content from the document reader = ElementReader() itr = doc.GetPageIterator() while itr.HasNext(): reader.Begin(itr.Current()) dumpAllText(reader) reader.End() itr.Next() # Example 2. Extract text content based on the # selection rectangle. print("----------------------------------------------------") print("Extract text based on the selection rectangle.") print("----------------------------------------------------") itr = doc.GetPageIterator() first_page = itr.Current() s1 = ReadTextFromRect(first_page, Rect(27, 392, 563, 534), reader) print("Field 1: " + s1) s1 = ReadTextFromRect(first_page, Rect(28, 551, 106, 623), reader); print("Field 2: " + s1) s1 = ReadTextFromRect(first_page, Rect(208, 550, 387, 621), reader); print("Field 3: " + s1) doc.Close() print("Done.") if __name__ == '__main__': main()
[解决办法]
能解析pdf文档的函数库不多，但处理pdf文档的工具还是挺多的。所以分两步：
1. 提取文本：pdftotext（Linux），在线转换（自己搜），windows下肯定也有相应的工具
2. 处理文本：各种脚本语言