Created
October 15, 2011 06:41
-
-
Save sweemeng/1289157 to your computer and use it in GitHub Desktop.
My Own python pdf toolkit
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| from pdfminer.pdfparser import PDFParser,PDFDocument | |
| from pdfminer.pdfinterp import PDFResourceManager,PDFPageInterpreter | |
| from pdfminer.pdfdevice import PDFDevice | |
| from pdfminer.layout import LAParams | |
| from pdfminer.converter import PDFPageAggregator | |
| from pdfminer.layout import LTTextBox, LTTextLine, LTFigure, LTImage, LTChar,LTTextBoxHorizontal | |
| class Document: | |
| def __init__(self,filename,password=''): | |
| self.extracted_text = '' | |
| self.xml_text = '' | |
| self.fp = open(filename,'rb') | |
| self.parser = PDFParser(self.fp) | |
| self.doc = PDFDocument() | |
| self.parser.set_document(self.doc) | |
| self.doc.set_parser(self.parser) | |
| self.password = password | |
| self.doc.initialize(self.password) | |
| self.rsrcmgr = PDFResourceManager() | |
| self.laparams = LAParams() | |
| self.device = PDFPageAggregator(self.rsrcmgr,laparams=self.laparams) | |
| self.interpreter = PDFPageInterpreter(self.rsrcmgr,self.device) | |
| def get_layout(self): | |
| for page in self.doc.get_pages(): | |
| self.interpreter.process_page(page) | |
| layout = self.device.get_result() | |
| yield layout | |
| def get_text(self): | |
| for i in self.get_layout(): | |
| for j in i: | |
| if isinstance(j,LTTextBox) or \ | |
| isinstance(j,LTTextBoxHorizontal) or \ | |
| isinstance(j,LTTextLine): | |
| if j.get_text(): | |
| self.extracted_text = self.extracted_text + j.get_text() | |
| return self.extracted_text | |
| def to_xml(self): | |
| self.temp_xml = '<?xml version="1.0" encoding="UTF-8" ?>\n' | |
| for i in self.get_layout(): | |
| for j in i: | |
| if isinstance(j,LTTextBox): | |
| if j.get_text(): | |
| temp = '<LTTextBox>' + j.get_text() + '</LTTextBox>\n' | |
| self.temp_xml = self.temp_xml + temp | |
| if isinstance(j,LTTextBoxHorizontal): | |
| if j.get_text(): | |
| temp = '<LTTextBoxHorizontal>' + j.get_text() + '</LTTextBoxHorizontal>\n' | |
| self.temp_xml = self.temp_xml + temp | |
| if isinstance(j,LTTextLine): | |
| if j.get_text(): | |
| temp = '<LTTextLine>' + j.get_text() + '</LTTextLine>\n' | |
| self.temp_xml = self.temp_xml + temp | |
| return self.temp_xml |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment