def get_result_from_file(filename):
from pdfminer.pdfparser import PDFParser
from pdfminer.pdfdocument import PDFDocument
from pdfminer.pdfpage import PDFPage
from pdfminer.pdfpage import PDFTextExtractionNotAllowed
from pdfminer.pdfinterp import PDFResourceManager
from pdfminer.pdfinterp import PDFPageInterpreter
from pdfminer.converter import PDFPageAggregator
from pdfminer.layout import LAParams
result = {"filename": filename, "pages": []}
fp = open(filename, "rb")
parser = PDFParser(fp)
document = PDFDocument(parser)
if not document.is_extractable:
raise PDFTextExtractionNotAllowed
rsrcmgr = PDFResourceManager()
laparams = LAParams()
laparams.char_margin = 2.0
laparams.detect_vertical = True
laparams.line_margin = 1.0
device = PDFPageAggregator(rsrcmgr, laparams=laparams)
interpreter = PDFPageInterpreter(rsrcmgr, device)
page_index = 0
for page in PDFPage.create_pages(document):
interpreter.process_page(page)
layout = device.get_result()
bounding_box = get_bounding_box(layout)
labels = get_text_labels(layout)
result["pages"].append({"index": page_index, "bounding_box": bounding_box, "labels": labels})
page_index += 1
fp.close()
return result
def extract_text_from_pdf(path_in, path_out, fichier, page_beg, page_end=0):
if (page_end == 0):
page_end = page_beg
fp = open(path_in + '/' + fichier, 'rb')
parser = PDFParser(fp)
doc = PDFDocument()
parser.set_document(doc)
doc.set_parser(parser)
doc.initialize('')
rsrcmgr = PDFResourceManager()
laparams = LAParams()
laparams.char_margin = 4.0 # 2.0 by default : two char whose distance is closer than this value are considered contiguous and get grouped into one.
laparams.word_margin = 0.3 # 0.1 by default : distance between two words is greater than this value => insert space
laparams.line_margin = 0.5 # 0.5 by default : Distance between 2 Lines under this value are grouped
device = PDFPageAggregator(rsrcmgr, laparams=laparams)
interpreter = PDFPageInterpreter(rsrcmgr, device)
extracted_text = ''
x = list(doc.get_pages())
for i in range(page_beg - 1, page_end):
page = x[i]
extracted_text += "EXTRACTION DE LA PAGE " + str(i + 1) + "\n\n"
interpreter.process_page(page)
layout = device.get_result()
for lt_obj in layout:
if isinstance(lt_obj, LTTextBox) or isinstance(lt_obj, LTTextLine):
extracted_text += lt_obj.get_text()
extracted_text += "\n"
return extracted_text
def to_txt(infile: str, outfile: str):
"""
Convert a pdf file to txt.
:param infile: pdf file path;
:param outfile: txt file path;
:return: txt file path;
"""
caching = True
rsrcmgr = PDFResourceManager(caching=caching)
codec = 'utf-8'
pagenos = set()
maxpages = 0
password = ''
laparams = LAParams()
laparams.word_margin = float(0)
laparams.line_margin = float(1)
outfp = io.open(outfile, 'wt', encoding=codec, errors='ignore')
device = TextConverter(rsrcmgr, outfp, laparams=laparams)
fp = io.open(infile, 'rb')
process_pdf(rsrcmgr,
device,
fp,
pagenos,
maxpages=maxpages,
password=password,
caching=caching,
check_extractable=True)
fp.close()
device.close()
outfp.close()
return outfile
def get_text_rows(path):
rows = defaultdict(list)
# Open a PDF file.
fp = open(path, 'rb')
# Create a PDF parser object associated with the file object.
# parser = PDFParser(fp)
# Create a PDF document object that stores the document structure.
# Password for initialization as 2nd parameter
# document = PDFDocument(parser)
# Check if the document allows text extraction. If not, abort.
# if not document.is_extractable:
# raise PDFTextExtractionNotAllowed
# Create a PDF resource manager object that stores shared resources.
rsrcmgr = PDFResourceManager()
# Create a PDF device object.
device = PDFDevice(rsrcmgr)
# BEGIN LAYOUT ANALYSIS
# Set parameters for analysis.
laparams = LAParams()
laparams.line_overlap = 0.01
laparams.line_margin = 0.01
laparams.word_margin = 0.15
# Create a PDF page aggregator object.
device = PDFPageAggregator(rsrcmgr, laparams=laparams)
# Create a PDF interpreter object.
interpreter = PDFPageInterpreter(rsrcmgr, device)
def parse_obj(lt_objs, page):
# loop over the object list
for obj in lt_objs:
# if it's a textbox, print text and location
if isinstance(obj, pdfminer.layout.LTTextBoxHorizontal):
rows[(page, -int(obj.bbox[1]))].append(
(int(obj.bbox[0]), sanitize(obj.get_text())))
# if it's a container, recurse
elif isinstance(obj, pdfminer.layout.LTFigure):
parse_obj(obj._objs, page)
# loop over all pages in the document
for page_num, page in enumerate(PDFPage.get_pages(fp)):
# read the page into a layout object
interpreter.process_page(page)
layout = device.get_result()
# extract text from this object
parse_obj(layout._objs, page_num)
for key in sorted(rows):
rows[key] = sorted(rows[key])
page, y = key
y = -y
yield (page, y, rows[key])
def parse(self, path):
out = StringIO.StringIO()
fp = None
# Directory
if os.path.isdir(path):
raise NotImplementedError()
# File
else:
fp = file(path)
rsrc = PDFResourceManager()
codec = 'utf-8'
laparams = LAParams()
laparams.char_margin = 2.0
laparams.line_margin = 2.0
laparams.word_margin = 0.0
device = TextConverter(rsrc, out, codec=codec, laparams=laparams)
doc = PDFDocument()
parser = PDFParser(fp)
parser.set_document(doc)
doc.set_parser(parser)
doc.initialize()
interpreter = PDFPageInterpreter(rsrc, device)
for page in doc.get_pages():
interpreter.process_page(page)
device.close()
sample = Sample(path, None, out.getvalue())
out.close()
return sample
def extract_layout_by_page(pdf_path, page_number):
"""
:param pdf_path: pdf file path
:param page_number: the specific page that you want to parse(start from 1)
:return: a list of pdfminer layout object
"""
fp = open(pdf_path, 'rb') # 以二进制读模式打开
# 用文件对象来创建一个pdf文档分析器
praser = PDFParser(fp) # 创建一个PDF文档
doc = PDFDocument() # 连接分析器 与文档对象
praser.set_document(doc)
doc.set_parser(praser)
doc.initialize()
# 检测文档是否提供txt转换，不提供就忽略
if not doc.is_extractable:
raise PDFTextExtractionNotAllowed
# 创建PDf 资源管理器 来管理共享资源
rsrcmgr = PDFResourceManager() # 创建一个PDF设备对象
laparams = LAParams()
laparams.line_overlap = 0.3
laparams.char_margin = 3
laparams.word_margin = 0.3
laparams.line_margin = 0.01
device = PDFPageAggregator(rsrcmgr, laparams=laparams) # 创建一个PDF解释器对象
interpreter = PDFPageInterpreter(rsrcmgr, device)
layouts = []
# 循环遍历列表，每次处理一个page的内容
pages = list(doc.get_pages())
interpreter.process_page(pages[page_number - 1])
# 接受该页面的LTPage对象
return device.get_result()
def __extract_extra__(request, item_id=None):
if not request.user.is_authenticated():
return HttpResponse('Please sign in first')
from pdfminer.layout import LAParams
from pdfminer.pdfinterp import PDFResourceManager, PDFPageInterpreter, process_pdf
from pdfminer.pdfdevice import PDFDevice, TagExtractor
from pdfminer.converter import TextConverter
from cStringIO import StringIO
laparams = LAParams()
outtype = 'text'
laparams.char_margin = 1.0
laparams.line_margin = 0.3
laparams.word_margin = 0.2
codec = 'utf-8'
caching = True
if item_id:
all_items = Item.objects.filter(id=item_id)
else:
all_items = Item.objects.all()
for item in all_items:
# Don't extract if no PDF exists; or if we already have search text
if not item.pdf_file or item.other_search_text:
continue
rsrcmgr = PDFResourceManager(caching=caching)
outfp = StringIO()
device = TextConverter(rsrcmgr, outfp, codec=codec, laparams=laparams)
fp = item.pdf_file.file
try:
process_pdf(rsrcmgr,
device,
fp,
pagenos=set(),
maxpages=0,
password='',
caching=caching,
check_extractable=True)
except AssertionError:
logger.warning('FAILED in completely PDF index "%s"' % item.title)
return HttpResponse('FAILED in completely PDF index "%s"' \
% item.title)
else:
logger.debug('Full PDF index of item "%s"' % item.title)
finally:
fp.close()
device.close()
outfp.seek(0)
page_text = outfp.read()
outfp.close()
item.other_search_text = page_text
item.save()
return HttpResponse('Full PDF indexed for item "%s"' % item.title)
def convert(fname, pages=None, M=1.0, L=0.3, W=0.2, F=0.5):
""" Converts a pdf filename into plain text.
Each value is specified not as an actual length, but as a proportion of the length
to the size of each character in question.
Parameters define layout analysis. In a PDF text is in several chunks of various types.
Text extraction needs to recover text chunks which ar regarded as continuous if
elements distance is closer than the char_margin (identified as M) and thus are
grouped into one block. Two lines are part of the same text if they are closer than
the line_margin (L). If the distance between two words is greater than the word_margin (W),
blank characters (spaces) shall be inserted as necessary to keep format.
Boxes flow (F) specifies how much a horizontal and vertical position of a text matters
when determining text flow order. The value should be within the range from -1.0
(only horizontal position matters) to +1.0 (only vertical position matters).
Keyword arguments:
fname -- PDF file name (string)
pages -- Set of pages to extract (set)
M -- char_margin (float)
L -- line_margin (float)
W -- word_margin (float)
F -- boxes_flow (float)
Return:
text: pdf contents as plain text
"""
if not pages:
pagenums = set()
else:
pagenums = set(pages)
output = BytesIO()
codec = "utf-8"
manager = PDFResourceManager()
laparams = LAParams()
laparams.all_texts = True
laparams.detect_vertical = False
laparams.char_margin = M
laparams.line_margin = L
laparams.word_margin = W
laparams.boxes_flow = F
converter = TextConverter(manager, output, codec=codec, laparams=laparams)
interpreter = PDFPageInterpreter(manager, converter)
infile = open(fname, 'rb')
for page in PDFPage.get_pages(infile, pagenums):
interpreter.process_page(page)
infile.close()
converter.close()
text = output.getvalue()
output.close
return text
def readText(self,path, outtype='text', opts={}):
outfile = path[:-3] + outtype
outdir = '/'.join(path.split('/')[:-1])
# debug option
pagenos = set()
maxpages = 0
# output option
# ?outfile = None
# ?outtype = None
outdir = None
#layoutmode = 'normal'
codec = 'utf-8'
pageno = 1
scale = 1
showpageno = True
laparams = LAParams()
for (k, v) in opts:
if k == '-d': debug += 1
elif k == '-p': pagenos.update( int(x)-1 for x in v.split(',') )
elif k == '-m': maxpages = int(v)
elif k == '-P': password = v
elif k == '-o': outfile = v
elif k == '-n': laparams = None
elif k == '-A': laparams.all_texts = True
elif k == '-V': laparams.detect_vertical = True
elif k == '-M': laparams.char_margin = float(v)
elif k == '-L': laparams.line_margin = float(v)
elif k == '-W': laparams.word_margin = float(v)
elif k == '-F': laparams.boxes_flow = float(v)
elif k == '-Y': layoutmode = v
elif k == '-O': outdir = v
elif k == '-t': outtype = v
elif k == '-c': codec = v
elif k == '-s': scale = float(v)
print laparams
#
#PDFDocument.debug = debug
#PDFParser.debug = debug
CMapDB.debug = self.debug
PDFResourceManager.debug = self.debug
PDFPageInterpreter.debug = self.debug
PDFDevice.debug = self.debug
#
rsrcmgr = PDFResourceManager()
#outtype = 'text'
outfp = StringIO()
device = HTMLConverter(rsrcmgr, outfp, codec=codec, laparams=laparams)
fp = file(path, 'rb')
process_pdf(rsrcmgr, device, fp, pagenos, maxpages=maxpages, password=password, check_extractable=True)
fp.close()
device.close()
print outfp.getvalue()
outfp.close()
return
def convert_to_text_file(filename_in, filename_out, rewrite=False):
"""
Parse file according to BORME PDF format
filename:
filenameOut:
"""
if os.path.isdir(filename_out):
filename_out = os.path.join(filename_out, os.path.basename(filename_in))
if os.path.exists(filename_out) and not rewrite:
logging.info('Skipping file %s already exists and rewriting is disabled!' % filename_out)
return False
# conf
codec = 'utf-8'
laparams = LAParams()
imagewriter = None
pagenos = set()
maxpages = 0
password = ''
rotation = 0
# <LAParams: char_margin=2.0, line_margin=0.5, word_margin=0.1 all_texts=False>
laparams.detect_vertical = True
laparams.all_texts = False
laparams.char_margin = 2.0
laparams.line_margin = 0.5
laparams.word_margin = 0.1
caching = True
rsrcmgr = PDFResourceManager(caching=caching)
outfp = open(filename_out, 'w')
device = TextConverter(rsrcmgr, outfp, codec=codec, laparams=laparams, imagewriter=imagewriter)
fp = open(filename_in, 'rb')
interpreter = PDFPageInterpreter(rsrcmgr, device)
# https://github.com/euske/pdfminer/issues/72
#page = PDFPage()
#PDFPage.cropbox =
# y esto?
for page in PDFPage.get_pages(fp, pagenos,
maxpages=maxpages, password=password,
caching=caching, check_extractable=True):
page.rotate = (page.rotate + rotation) % 360
interpreter.process_page(page)
fp.close()
device.close()
outfp.close()
return True
def extractrefs(infile, outfile):
pagenos = set()
caching = True
infp = open(infile, 'rb')
outfp = open(outfile, 'w')
rsrcmgr = PDFResourceManager()
laparams = LAParams()
laparams.line_margin = 1.4
device = RefsExtractor(rsrcmgr, outfp, laparams=laparams)
interpreter = PDFPageInterpreter(rsrcmgr, device)
for page in PDFPage.get_pages(infp, pagenos,
caching=caching,
check_extractable=True):
interpreter.process_page(page)
infp.close()
outfp.close()
def output_pdf_to_table(path):
fp = open(path, "rb")
rsrcmgr = PDFResourceManager()
laparams = LAParams()
laparams.line_margin = line_margin_threshold
codec = 'utf-8'
device = PDFPageAggregator(rsrcmgr, laparams=laparams)
interpreter = PDFPageInterpreter(rsrcmgr, device)
password=""
maxpages=pages_to_view
caching=True
pagenos=set()
for page in PDFPage.get_pages(fp, pagenos, maxpages=maxpages,
password=password,caching=caching, check_extractable=True):
interpreter.process_page(page)
layout = device.get_result()
getRows(layout)
def extract_text(self):
resource_manager = PDFResourceManager()
fake_file_handle = io.StringIO()
laparams = LAParams()
laparams.char_margin = 1000
laparams.word_margin = 0.01
laparams.line_margin = 0.01
converter = TextConverter(resource_manager,
fake_file_handle,
laparams=laparams)
page_interpreter = PDFPageInterpreter(resource_manager, converter)
page_interpreter.device.handle_undefined_char = lambda f, c: chr(c)
with open(self.path, 'rb') as fh:
for page in PDFPage.get_pages(fh,
caching=True,
check_extractable=True):
page_interpreter.process_page(page)
self._text = fake_file_handle.getvalue()
self.rows = self._text.split('\n')
converter.close()
fake_file_handle.close()
def read_pdf_data(filename):
fp = open(filename, 'rb')
rsrcmgr = PDFResourceManager()
laparams = LAParams()
laparams.line_margin = 0.1
device = My(rsrcmgr, sys.stdout, laparams=laparams)
device.reset()
interpreter = PDFPageInterpreter(rsrcmgr, device)
result_data = []
count = 0
for page in PDFPage.get_pages(fp, set()):
interpreter.process_page(page)
result_data.append(device.group)
device.word = ""
device.group = []
device.word_pos_info = {}
count += 1
# break
fp.close()
return result_data
def readpdf(pdfFile):
fp = open(pdfFile, 'rb')
parser = PDFParser(fp)
doc = PDFDocument(parser)
#doc.initialize('password') # leave empty for no password
rsrcmgr = PDFResourceManager()
laparams = LAParams()
laparams.char_margin = float('1.1') #too small and it splits the description, too big and Quantity-Unit-Part number are not separated: 1.1 seems to work
laparams.line_margin = float('0.8')
device = PDFPageDetailedAggregator(rsrcmgr, laparams=laparams)
interpreter = PDFPageInterpreter(rsrcmgr, device)
for page in PDFPage.create_pages(doc):
interpreter.process_page(page)
# receive the LTPage object for this page
device.get_result()
#print(device.rows)
df = pd.DataFrame(device.rows, columns=['Page', 'x', 'y', 'c1','c2','String'])
return df
def _get_content(fname):
rsrcmgr = PDFResourceManager(caching=caching)
laparams = LAParams()
laparams.line_margin = 1.0
laparams.boxes_flow = 1.0
imagewriter = None
with io.BytesIO() as outfp:
device = TextConverter(
rsrcmgr, outfp, laparams=laparams, imagewriter=imagewriter
)
interpreter = PDFPageInterpreter(rsrcmgr, device)
with open(fname, "rb") as f:
for page in PDFPage.get_pages(
f,
pagenos,
maxpages=maxpages,
password=password,
caching=caching,
check_extractable=True,
):
interpreter.process_page(page)
return outfp.getvalue().decode("utf-8")
def pdf_to_string(pdf_file):
fp = open(pdf_file, 'rb')
parser = PDFParser(fp)
doc = PDFDocument()
parser.set_document(doc)
doc.set_parser(parser)
doc.initialize('')
rsrcmgr = PDFResourceManager()
# Configuração das margens
laparams = LAParams()
laparams.line_margin = 0.3
laparams.word_margin = 0.3
device = PDFPageAggregator(rsrcmgr, laparams=laparams)
interpreter = PDFPageInterpreter(rsrcmgr, device)
for page in doc.get_pages():
interpreter.process_page(page)
layout = device.get_result()
for lt_obj in layout:
print(lt_obj)
def from_pdf(pdfFile):
try:
pagenos = set()
strfp = StringIO()
codec = 'utf-8'
laparams = LAParams()
#laparams.char_margin = 10
laparams.line_margin = 20
#laparams.word_margin = 10
laparams.boxes_flow = -1
rsrcmgr = PDFResourceManager()
device = TextConverter(rsrcmgr, strfp, codec=codec, laparams=laparams)
fp = file(pdfFile, 'rb')
interpreter = PDFPageInterpreter(rsrcmgr, device)
for page in PDFPage.get_pages(fp, pagenos, check_extractable=True):
interpreter.process_page(page)
except Exception, e:
print e
traceback.print_exc()
pass
def get_result_from_file(filename):
from pdfminer.pdfparser import PDFParser
from pdfminer.pdfdocument import PDFDocument
from pdfminer.pdfpage import PDFPage
from pdfminer.pdfpage import PDFTextExtractionNotAllowed
from pdfminer.pdfinterp import PDFResourceManager
from pdfminer.pdfinterp import PDFPageInterpreter
from pdfminer.converter import PDFPageAggregator
from pdfminer.layout import LAParams
result = {"filename": filename, "pages": []}
fp = open(filename, 'rb')
parser = PDFParser(fp)
document = PDFDocument(parser)
if not document.is_extractable:
raise PDFTextExtractionNotAllowed
rsrcmgr = PDFResourceManager()
laparams = LAParams()
laparams.char_margin = 2.0
laparams.detect_vertical = True
laparams.line_margin = 1.0
device = PDFPageAggregator(rsrcmgr, laparams=laparams)
interpreter = PDFPageInterpreter(rsrcmgr, device)
page_index = 0
for page in PDFPage.create_pages(document):
interpreter.process_page(page)
layout = device.get_result()
bounding_box = get_bounding_box(layout)
labels = get_text_labels(layout)
result["pages"].append({
"index": page_index,
"bounding_box": bounding_box,
"labels": labels
})
page_index += 1
fp.close()
return result
def pdf_to_string(pdf_file):
fp = open(pdf_file, 'rb')
parser = PDFParser(fp)
doc = PDFDocument()
parser.set_document(doc)
doc.set_parser(parser)
doc.initialize('')
rsrcmgr = PDFResourceManager()
laparams = LAParams()
laparams.line_margin = 0.3
laparams.word_margin = 0.3
device = PDFPageAggregator(rsrcmgr, laparams=laparams)
interpreter = PDFPageInterpreter(rsrcmgr, device)
extracted_text = ''
for page in doc.get_pages():
interpreter.process_page(page)
layout = device.get_result()
for lt_obj in layout:
if isinstance(lt_obj, (LTTextBox, LTTextLine)):
extracted_text += lt_obj.get_text()
return extracted_text
def main(argv):
def usage():
print(('usage: %s [-d] [-p pagenos] [-m maxpages] [-P password] [-o output] [-C] '
'[-n] [-A] [-V] [-M char_margin] [-L line_margin] [-W word_margin] [-F boxes_flow] '
'[-Y layout_mode] [-O output_dir] [-t text|html|xml|tag] [-c codec] [-s scale] file ...' % argv[0]))
return 100
try:
(opts, args) = getopt.getopt(argv[1:], 'dp:m:P:o:CnAVM:L:W:F:Y:O:t:c:s:')
except getopt.GetoptError:
return usage()
if not args: return usage()
debug = False
# input option
password = ''
pagenos = set()
maxpages = 0
# output option
outfile = None
outtype = None
outdir = None
layoutmode = 'normal'
codec = 'utf-8'
pageno = 1
scale = 1
caching = True
showpageno = True
laparams = LAParams()
for (k, v) in opts:
if k == '-d': debug = True
elif k == '-p': pagenos.update( int(x)-1 for x in v.split(',') )
elif k == '-m': maxpages = int(v)
elif k == '-P': password = v
elif k == '-o': outfile = v
elif k == '-C': caching = False
elif k == '-n': laparams = None
elif k == '-A': laparams.all_texts = True
elif k == '-V': laparams.detect_vertical = True
elif k == '-M': laparams.char_margin = float(v)
elif k == '-L': laparams.line_margin = float(v)
elif k == '-W': laparams.word_margin = float(v)
elif k == '-F': laparams.boxes_flow = float(v)
elif k == '-Y': layoutmode = v
elif k == '-O': outdir = v
elif k == '-t': outtype = v
elif k == '-c': codec = v
elif k == '-s': scale = float(v)
if debug:
set_debug_logging()
rsrcmgr = PDFResourceManager(caching=caching)
if not outtype:
outtype = 'text'
if outfile:
if outfile.endswith('.htm') or outfile.endswith('.html'):
outtype = 'html'
elif outfile.endswith('.xml'):
outtype = 'xml'
elif outfile.endswith('.tag'):
outtype = 'tag'
if outfile:
outfp = io.open(outfile, 'wt', encoding=codec, errors='ignore')
close_outfp = True
else:
outfp = sys.stdout
close_outfp = False
if outtype == 'text':
device = TextConverter(rsrcmgr, outfp, laparams=laparams)
elif outtype == 'xml':
device = XMLConverter(rsrcmgr, outfp, laparams=laparams, outdir=outdir)
elif outtype == 'html':
device = HTMLConverter(rsrcmgr, outfp, scale=scale, layoutmode=layoutmode,
laparams=laparams, outdir=outdir, debug=debug)
elif outtype == 'tag':
device = TagExtractor(rsrcmgr, outfp)
else:
return usage()
for fname in args:
fp = io.open(fname, 'rb')
process_pdf(rsrcmgr, device, fp, pagenos, maxpages=maxpages, password=password,
caching=caching, check_extractable=True)
fp.close()
device.close()
if close_outfp:
outfp.close()
def ConvertPdf(pdfpath, outfp, opts={}):
import sys
from pdfminer.pdfdocument import PDFDocument
from pdfminer.pdfparser import PDFParser
from pdfminer.pdfinterp import PDFResourceManager, PDFPageInterpreter
from pdfminer.pdfdevice import PDFDevice, TagExtractor
from pdfminer.pdfpage import PDFPage
from pdfminer.converter import XMLConverter, HTMLConverter, TextConverter
from pdfminer.cmapdb import CMapDB
from pdfminer.layout import LAParams
from pdfminer.image import ImageWriter
debug = 0
# input option
password = ''
pagenos = set()
maxpages = 0
# output option
outfile = None
outtype = None
imagewriter = None
rotation = 0
layoutmode = 'normal'
codec = 'utf-8'
pageno = 1
scale = 1
caching = True
showpageno = True
laparams = LAParams()
for (k, v) in opts:
if k == '-d': debug += 1
elif k == '-p': pagenos.update(int(x) - 1 for x in v.split(','))
elif k == '-m': maxpages = int(v)
elif k == '-P': password = v
elif k == '-o': outfile = v
elif k == '-C': caching = False
elif k == '-n': laparams = None
elif k == '-A': laparams.all_texts = True
elif k == '-V': laparams.detect_vertical = True
elif k == '-M': laparams.char_margin = float(v)
elif k == '-L': laparams.line_margin = float(v)
elif k == '-W': laparams.word_margin = float(v)
elif k == '-F': laparams.boxes_flow = float(v)
elif k == '-Y': layoutmode = v
elif k == '-O': imagewriter = ImageWriter(v)
elif k == '-R': rotation = int(v)
elif k == '-t': outtype = v
elif k == '-c': codec = v
elif k == '-s': scale = float(v)
#
CMapDB.debug = debug
PDFResourceManager.debug = debug
PDFDocument.debug = debug
PDFParser.debug = debug
PDFPageInterpreter.debug = debug
PDFDevice.debug = debug
#
rsrcmgr = PDFResourceManager()
if not outtype:
outtype = 'txt'
if outtype == 'txt':
device = TextConverter(rsrcmgr,
outfp,
codec=codec,
laparams=laparams,
imagewriter=imagewriter)
elif outtype == 'xml':
device = XMLConverter(rsrcmgr,
outfp,
codec=codec,
laparams=laparams,
imagewriter=imagewriter)
elif outtype == 'html':
device = HTMLConverter(rsrcmgr,
outfp,
codec=codec,
scale=scale,
layoutmode=layoutmode,
laparams=laparams,
imagewriter=imagewriter)
elif outtype == 'tag':
device = TagExtractor(rsrcmgr, outfp, codec=codec)
fp = file(pdfpath, 'rb')
interpreter = PDFPageInterpreter(rsrcmgr, device)
for page in PDFPage.get_pages(fp,
pagenos,
maxpages=maxpages,
password=password,
caching=caching,
check_extractable=True):
page.rotate = (page.rotate + rotation) % 360
interpreter.process_page(page)
fp.close()
device.close()
return True
def main(argv):
# debug option
debug = 0
# input option
password = ''
pagenos = set()
maxpages = 0
# output option
outfile = None
outtype = None
imagewriter = None
rotation = 0
stripcontrol = False
layoutmode = 'normal'
codec = 'utf-8'
pageno = 1
scale = 1
caching = True
showpageno = False
laparams = LAParams()
using_optparse = False
parser = ArgumentParser(prog='pdf2txt.py',
description='Convert pdf to txt',
formatter_class=ArgumentDefaultsHelpFormatter)
if using_optparse:
DEBUG(3, 'using optparse')
parser.add_argument = parser.add_option
parser.parse_known_args = parser.parse_args
parser.disable_interspersed_args()
parser.add_argument('-d',
dest='debuglevel',
action='count',
default=0,
help='Debug (repeat for more verbose debugging)')
parser.add_argument(
'-p',
'--pages',
dest='pagenos',
action='store',
type=str,
default='',
help=
'Specifies the comma-separated list of the page numbers to be extracted. Page numbers start at one. By default, it extracts text from all the pages.'
)
parser.add_argument('-c',
'--codec',
dest='codec',
action='store',
type=str,
default='utf-8',
help='Specifies the output codec.')
parser.add_argument(
'-t',
'--type',
dest='outtype',
action='store',
type=str,
default='shape',
choices=['text', 'html', 'xml', 'tag', 'shape'],
help='Specifies the output format, one of: shape, text, html, xml, tag'
)
parser.add_argument(
'-m',
dest='maxpages',
action='store',
type=int,
default=0,
help=
'Specifies the maximum number of pages to extract. By default (0), it extracts all the pages in a document.'
)
parser.add_argument(
'-P',
'--password',
dest='password',
action='store',
type=str,
default='',
help='Provides the user password to access PDF contents.')
parser.add_argument(
'-o',
'--output',
dest='outfile',
action='store',
type=str,
default=None,
help=
'Specifies the output file name. By default, it prints the extracted contents to stdout in text format.'
)
parser.add_argument(
'-C',
'--no-caching',
dest='caching',
action='store_false',
default=True,
help=
'Suppress object caching. This will reduce the memory consumption but also slows down the process.'
)
parser.add_argument('-n',
'--no-layout',
dest='layout',
action='store_false',
default=True,
help='Suppress layout analysis.')
parser.add_argument('--show-pageno',
dest='show_pageno',
action='store_true',
default=False,
help='Show page numbers.')
parser.add_argument(
'-A',
'--analyze-all',
dest='all_texts',
action='store_true',
default=False,
help=
'Forces to perform layout analysis for all the text strings, including text contained in figures.'
)
parser.add_argument('-V',
'--detect-vertical',
dest='detect_vertical',
action='store_true',
default=False,
help='Allows vertical writing detection.')
parser.add_argument(
'-M',
dest='char_margin',
action='store',
type=float,
default=2.0,
help=
'Two text chunks whose distance is closer than the char_margin (shown as M) is considered continuous and get grouped into one.'
)
parser.add_argument(
'-L',
dest='line_margin',
action='store',
type=float,
default=0.5,
help=
'Two lines whose distance is closer than the line_margin (L) is grouped as a text box, which is a rectangular area that contains a "cluster" of text portions.'
)
parser.add_argument(
'-W',
dest='word_margin',
action='store',
type=float,
default=0.1,
help=
'It may be required to insert blank characters (spaces) as necessary if the distance between two words is greater than the word_margin (W), as a blank between words might not be represented as a space, but indicated by the positioning of each word.'
)
parser.add_argument(
'-F',
dest='boxes_flow',
action='store',
type=float,
default=0.5,
help=
'Specifies how much a horizontal and vertical position of a text matters when determining a text order. The value should be within the range of -1.0 (only horizontal position matters) to +1.0 (only vertical position matters).'
)
parser.add_argument(
'-Y',
'--layout-mode',
dest='layoutmode',
action='store',
type=str,
default='normal',
choices=['exact', 'normal', 'loose'],
help=
'Specifies how the page layout should be preserved. (Currently only applies to HTML format.) One of: exact, normal, loose.'
)
parser.add_argument('-O',
'--image-writer',
dest='imagewriter',
action='store',
type=str,
default=None,
help='imagewriter')
parser.add_argument('-R',
'--rotation',
dest='rotation',
action='store',
type=int,
default=0,
help='rotation')
parser.add_argument('-S',
'--strip-control',
dest='stripcontrol',
action='store_true',
default=False,
help='stripcontrol')
parser.add_argument(
'-s',
dest='scale',
action='store',
type=float,
default=1,
help='Specifies the output scale. Can be used in HTML format only.')
parser.add_argument(
'--draw-lines',
dest='draw_lines',
action='store_true',
help=
"Draw crude page representation, coloured TextLines (= short pieces of text). Valid only for the `shape' output."
)
parser.add_argument(
'--draw-boxes',
dest='draw_boxes',
action='store_true',
help=
"Draw crude page representation, coloured TextBoxes (= grouped text lines). Valid only for the `shape' output."
)
parser.add_argument(
'--draw-blocks',
dest='draw_blocks',
action='store_true',
help=
"Draw crude page representation, coloured TextBlocks (= grouped TextBoxes). Valid only for the `shape' output."
)
parser.add_argument(
'--shear-limit',
dest='shear_limit',
action='store',
default=0.1,
type=float,
help=
"If the text is sheared above this limit, reject it. Valid only for the `shape' output."
)
parser.add_argument(
'--rotation-limit',
dest='rotation_limit',
action='store',
default=2,
type=float,
help=
"If the text is rotated above this angle (in degrees), reject it. Valid only for the `shape' output."
)
parser.add_argument(
'--line-height-diff',
dest='line_height_diff',
action='store',
type=float,
default=0.1,
help=
'Two lines whose vertical sizes differ more than this ratio are not to be considered of the same paragraph (but e.g. one of them is a heading).'
)
parser.add_argument('--heading-before',
dest='heading_before',
action='store',
type=str,
default='',
help='String to put before each heading, e.g. <h1>')
parser.add_argument('--heading-after',
dest='heading_after',
action='store',
type=str,
default='',
help='String to put after each heading, e.g. </h1>')
parser.add_argument(
'--box-separator',
dest='box_separator',
action='store',
type=str,
default=r'\n\n',
help=
r'Separate boxes with this string. Use \n for new line, \t for TAB, other escape sequences are not recognized.'
)
parser.add_argument(
'--block-separator',
dest='block_separator',
action='store',
type=str,
default=r'\n\n',
help=
r'Separate blocks with this string. Use \n for new line, \t for TAB, other escape sequences are not recognized.'
)
parser.add_argument(
'--indent-separator',
dest='indent_separator',
action='store',
type=str,
default=r'\n\n',
help=
r'Separate indented lines with this string. Use \n for new line, \t for TAB, other escape sequences are not recognized.'
)
parser.add_argument(
'--indent-string',
dest='indent_string',
action='store',
type=str,
default=r'\t',
help=
r'Put this string in front of indented lines. Use \n for new line, \t for TAB, other escape sequences are not recognized.'
)
parser.add_argument(
'--indent-limit',
dest='indent_limit',
action='store',
type=float,
default=3,
help=
'If the line is indented more then this (approximately characters), it will separated by --indent-separator from the previous one.'
)
parser.add_argument(
'--page-separator',
dest='page_separator',
action='store',
type=str,
default=r'\n\n',
help=
r'Separate pages with this string. Use \n for new line, \t for TAB, other escape sequences are not recognized.'
)
parser.add_argument(
'--norm-whitespace',
dest='norm_whitespace',
action='store_true',
default=False,
help=
'Normalize whitespace (remove duplicate spaces, replace end of lines with spaces).'
)
parser.add_argument(
'--print-stats',
dest='print_stats',
action='store_true',
default=False,
help=
'Instead of the text, output some simple statistics about the file.')
parser.add_argument(
'--max-blocks',
dest='max_blocks',
action='store',
default=0,
type=int,
help=
'If there is more than this blocks per page, do not return any text. Use to discriminate abnormal files (run --print-stats first to find out the number of boxes per "normal" file). 0 means no limit. 50 is maybe a good value.'
)
parser.add_argument(
'--max-textlines',
dest='max_textlines',
action='store',
default=0,
type=int,
help=
'If there is more than this textlines per any block, do not return any text. Use to discriminate abnormal files (run --print-stats first to find out the number of boxes per "normal" page). 0 means no limit. 18 is maybe a good value.'
)
parser.add_argument(
'--line-height-method',
dest='line_height_method',
action='store',
type=str,
default='bbox',
choices=['bbox', 'mean', 'median'],
help=
'Method to calculate height of line (relevant if there are characters with uneven height). bbox takes the bounding box (rectangle encompassing the line), mean the arithmetic mean of the height of all the characters, median is the median of the height of all the characters. Use mean or median if there are outlier characters, e.g. one big character at the beginning of line.'
)
parser.add_argument(dest='pdffile',
help='List of PDF files to go through',
default=None,
nargs='+')
args, rest = parser.parse_known_args()
global debuglevel
debuglevel = debug = args.debuglevel
DEBUG(3, 'args:', str(args))
DEBUG(3, 'rest:', str(rest))
DEBUG(3, 'optparse:', using_optparse)
if args.pagenos:
pagenos.update(int(x) - 1 for x in args.pagenos.split(','))
maxpages = args.maxpages
outfile = args.outfile
password = args.password
caching = args.caching
showpageno = args.show_pageno
if not args.layout:
laparams = None
if laparams and args.all_texts:
laparams.all_texts = True
if laparams and args.detect_vertical:
laparams.detect_vertical = True
if laparams:
laparams.char_margin = args.char_margin
laparams.line_margin = args.line_margin
laparams.word_margin = args.word_margin
laparams.boxes_flow = args.boxes_flow
layoutmode = args.layoutmode
if args.imagewriter:
imagewriter = ImageWriter(args.imagewriter)
rotation = args.rotation
stripcontrol = args.stripcontrol
outtype = args.outtype
codec = args.codec
scale = args.scale
args.box_separator = unescape_string(args.box_separator)
args.block_separator = unescape_string(args.block_separator)
args.indent_separator = unescape_string(args.indent_separator)
args.indent_string = unescape_string(args.indent_string)
args.page_separator = unescape_string(args.page_separator)
global options
options = args
PDFDocument.debug = debug
PDFParser.debug = debug
CMapDB.debug = debug
PDFPageInterpreter.debug = debug
rsrcmgr = PDFResourceManager(caching=caching)
if not outtype:
outtype = 'text'
if outfile:
if outfile.endswith('.htm') or outfile.endswith('.html'):
outtype = 'html'
elif outfile.endswith('.xml'):
outtype = 'xml'
elif outfile.endswith('.tag'):
outtype = 'tag'
if outfile:
outfp = file(outfile, 'w')
DEBUG(2, 'output goes to', outfile)
else:
outfp = sys.stdout
DEBUG(2, 'output goes to stdout')
if outtype == 'shape':
device = ShapeTextConverter(rsrcmgr,
outfp,
codec=codec,
laparams=laparams,
showpageno=showpageno,
imagewriter=imagewriter)
elif outtype == 'text':
device = TextConverter(rsrcmgr,
outfp,
codec=codec,
laparams=laparams,
imagewriter=imagewriter)
elif outtype == 'xml':
device = XMLConverter(rsrcmgr,
outfp,
codec=codec,
laparams=laparams,
imagewriter=imagewriter,
stripcontrol=stripcontrol)
elif outtype == 'html':
device = HTMLConverter(rsrcmgr,
outfp,
codec=codec,
scale=scale,
layoutmode=layoutmode,
laparams=laparams,
imagewriter=imagewriter,
debug=debug)
elif outtype == 'tag':
device = TagExtractor(rsrcmgr, outfp, codec=codec)
else:
return usage()
for fname in options.pdffile:
DEBUG(2, 'processing', fname)
fp = file(fname, 'rb')
interpreter = PDFPageInterpreter(rsrcmgr, device)
for page in PDFPage.get_pages(fp,
pagenos,
maxpages=maxpages,
password=password,
caching=caching,
check_extractable=True):
page.rotate = (page.rotate + rotation) % 360
interpreter.process_page(page)
fp.close()
device.close()
outfp.close()
DEBUG(2, 'finished.')
return
def main(argv):
import getopt
def usage():
print(
f'usage: {argv[0]} [-P password] [-o output] [-t text|html|xml|tag]'
' [-O output_dir] [-c encoding] [-s scale] [-R rotation]'
' [-Y normal|loose|exact] [-p pagenos] [-m maxpages]'
' [-S] [-C] [-n] [-A] [-V] [-M char_margin] [-L line_margin]'
' [-W word_margin] [-F boxes_flow] [-d] input.pdf ...')
return 100
try:
(opts, args) = getopt.getopt(argv[1:],
'dP:o:t:O:c:s:R:Y:p:m:SCnAVM:W:L:F:')
except getopt.GetoptError:
return usage()
if not args:
return usage()
# debug option
debug = 0
# input option
password = b''
pagenos = set()
maxpages = 0
# output option
outfile = None
outtype = None
imagewriter = None
rotation = 0
stripcontrol = False
layoutmode = 'normal'
encoding = 'utf-8'
# pageno = 1
scale = 1
caching = True
# showpageno = True
laparams = LAParams()
for (k, v) in opts:
if k == '-d':
debug += 1
elif k == '-P':
password = v.encode('ascii')
elif k == '-o':
outfile = v
elif k == '-t':
outtype = v
elif k == '-O':
imagewriter = ImageWriter(v)
elif k == '-c':
encoding = v
elif k == '-s':
scale = float(v)
elif k == '-R':
rotation = int(v)
elif k == '-Y':
layoutmode = v
elif k == '-p':
pagenos.update(int(x) - 1 for x in v.split(','))
elif k == '-m':
maxpages = int(v)
elif k == '-S':
stripcontrol = True
elif k == '-C':
caching = False
elif k == '-n':
laparams = None
elif k == '-A':
laparams.all_texts = True
elif k == '-V':
laparams.detect_vertical = True
elif k == '-M':
laparams.char_margin = float(v)
elif k == '-W':
laparams.word_margin = float(v)
elif k == '-L':
laparams.line_margin = float(v)
elif k == '-F':
laparams.boxes_flow = float(v)
#
PDFDocument.debug = debug
PDFParser.debug = debug
CMapDB.debug = debug
PDFPageInterpreter.debug = debug
#
rsrcmgr = PDFResourceManager(caching=caching)
if not outtype:
outtype = 'text'
if outfile:
if outfile.endswith('.htm') or outfile.endswith('.html'):
outtype = 'html'
elif outfile.endswith('.xml'):
outtype = 'xml'
elif outfile.endswith('.tag'):
outtype = 'tag'
if outfile:
outfp = open(outfile, 'w', encoding=encoding)
else:
outfp = sys.stdout
if outtype == 'text':
device = TextConverter(rsrcmgr,
outfp,
laparams=laparams,
imagewriter=imagewriter)
elif outtype == 'xml':
device = XMLConverter(rsrcmgr,
outfp,
laparams=laparams,
imagewriter=imagewriter,
stripcontrol=stripcontrol)
elif outtype == 'html':
device = HTMLConverter(rsrcmgr,
outfp,
scale=scale,
layoutmode=layoutmode,
laparams=laparams,
imagewriter=imagewriter,
debug=debug)
elif outtype == 'tag':
device = TagExtractor(rsrcmgr, outfp)
else:
return usage()
for fname in args:
with open(fname, 'rb') as fp:
interpreter = PDFPageInterpreter(rsrcmgr, device)
for page in PDFPage.get_pages(fp,
pagenos,
maxpages=maxpages,
password=password,
caching=caching,
check_extractable=True):
page.rotate = (page.rotate + rotation) % 360
interpreter.process_page(page)
device.close()
outfp.close()
bad_words = [
'Personal', 'Information', 'Projects', 'Internship', 'Technologies'
]
with open('cv.txt') as oldfile, open('cv_new.txt', 'w') as newfile:
for line in oldfile:
if not any(bad_word in line for bad_word in bad_words):
newfile.write(line)
file = open("cv_new.txt", "r")
s = file.read()
s = s.split('\n')
while ("" in s):
s.remove("")
while (" " in s):
s.remove(" ")
while ("\x0c" in s):
s.remove("\x0c")
details = []
i = 0
while (i < len(s)):
s1 = s[i].split(': ')
if (len(s1) > 1):
details.append(s1[1])
i += 1
sql = "INSERT INTO entries (name, post, exp) VALUES (%s, %s, %s)"
val = (details[0], details[1], details[2])
mycursor.execute(sql, val)
mydb.commit()
print(mycursor.rowcount, "record inserted.")
return
def convert_pdf_to_text(self):
"""
PDFファイルをテキストに変換
PDFは2段に段組みされたものも含む
"""
laparams = LAParams() # パラメータインスタンス
laparams.boxes_flow = None # -1.0（水平位置のみが重要）から+1.0（垂直位置のみが重要）default 0.5
laparams.word_margin = 0.2 # default 0.1
laparams.char_margin = 2.0 # default 2.0
laparams.line_margin = 0 # default 0.5
# 出力ファイルのオープン ファイルがある時は上書きされる
with open(self.output_path, "w", encoding="utf-8") as f:
# 初期化
self.text_l = "" # 左側の文字列
self.text_r = "" # 右側の文字列
print("Analyzing from {} page to {} page(0:to last)".format(
self.start_page, self.last_page))
# 対象ページを読み、テキスト抽出する。（maxpages：0は全ページ）
for page_layout in extract_pages(
self.input_path, maxpages=0,
laparams=laparams): # ファイルにwithしている
# 抽出するページの選別。extract_pagesの引数では、開始ページだけの指定に対応できないため
if page_layout.pageid < self.start_page:
continue # 指定開始ページより前は飛ばす
if self.last_page and self.last_page < page_layout.pageid:
break # 指定終了ページ以降は中断
# ページの幅から段組みの境界を計算(用紙幅の半分とする)
if self.border == 0:
self.border = int(page_layout.width / 2)
if page_layout.pageid == self.start_page:
print("Check on page #{}".format(page_layout.pageid))
print("Page Info width:{}, heght:{}".format(
page_layout.width, page_layout.height))
print("Calc result border: {}, footer: {}".format(
self.border, self.footer))
# 要素の出現順の確認(debug)
# for element in self.flatten_lttext(page_layout, LTTextBox):
# print("bbox{} {}".format(element.bbox, element.get_text()[:20]))
# 要素のイテレータをたどり入れ子の要素を1次元に取り出す。戻るイテレータはLTTextBox型のみ
# 要素の行の上側y1で降順、行の左側x0で昇順にソートする。
for element in sorted(self.flatten_lttext(
page_layout, LTTextBox),
key=lambda x: (-x.y1, x.x0)):
# for element in self.flatten_lttext(page_layout, LTTextBox):
if element.y1 < self.footer: continue # フッター位置の文字は抽出しない
if element.y0 > self.header: continue # ヘッダー位置の文字は抽出しない
_text = element.get_text()
# debug
# print("y1:{}, y0:{}■{}".format(element.y1, element.y0, _text))
if element.x1 < self.border:
# 文字列全体が左側
self.text_l += _text
else:
if element.x0 >= self.border:
# 文字列全体が右側
self.text_r += _text
else:
# 文字列が境界をまたいでいる場合
# 右側に既に文章があれば先に出力する
if self.text_r:
self.write2text(f)
self.text_l += _text
# 1ページ分処理したら書き込む
self.write2text(f)
def main(argv):
import getopt
def usage():
print ('usage: %s [-d] [-p pagenos] [-m maxpages] [-P password] [-o output] '
'[-n] [-A] [-D writing_mode] [-M char_margin] [-L line_margin] [-W word_margin] '
'[-O output_dir] [-t text|html|xml|tag] [-c codec] [-s scale] file ...' % argv[0])
return 100
try:
(opts, args) = getopt.getopt(argv[1:], 'dp:m:P:o:nAD:M:L:W:O:t:c:s:')
except getopt.GetoptError:
return usage()
if not args: return usage()
# debug option
debug = 0
# input option
password = ''
pagenos = set()
maxpages = 0
# output option
outfile = None
outtype = None
outdir = None
codec = 'utf-8'
pageno = 1
scale = 1
showpageno = True
laparams = LAParams()
for (k, v) in opts:
if k == '-d': debug += 1
elif k == '-p': pagenos.update( int(x)-1 for x in v.split(',') )
elif k == '-m': maxpages = int(v)
elif k == '-P': password = v
elif k == '-o': outfile = v
elif k == '-n': laparams = None
elif k == '-A': laparams.all_texts = True
elif k == '-D': laparams.writing_mode = v
elif k == '-M': laparams.char_margin = float(v)
elif k == '-L': laparams.line_margin = float(v)
elif k == '-W': laparams.word_margin = float(v)
elif k == '-O': outdir = v
elif k == '-t': outtype = v
elif k == '-c': codec = v
elif k == '-s': scale = float(v)
#
CMapDB.debug = debug
PDFResourceManager.debug = debug
PDFDocument.debug = debug
PDFParser.debug = debug
PDFPageInterpreter.debug = debug
PDFDevice.debug = debug
#
rsrc = PDFResourceManager()
if not outtype:
outtype = 'text'
if outfile:
if outfile.endswith('.htm') or outfile.endswith('.html'):
outtype = 'html'
elif outfile.endswith('.xml'):
outtype = 'xml'
elif outfile.endswith('.tag'):
outtype = 'tag'
if outfile:
outfp = file(outfile, 'w')
else:
outfp = sys.stdout
if outtype == 'text':
device = TextConverter(rsrc, outfp, codec=codec, laparams=laparams)
elif outtype == 'xml':
device = XMLConverter(rsrc, outfp, codec=codec, laparams=laparams, outdir=outdir)
elif outtype == 'html':
device = HTMLConverter(rsrc, outfp, codec=codec, scale=scale, laparams=laparams, outdir=outdir)
elif outtype == 'tag':
device = TagExtractor(rsrc, outfp, codec=codec)
else:
return usage()
for fname in args:
fp = file(fname, 'rb')
process_pdf(rsrc, device, fp, pagenos, maxpages=maxpages, password=password)
fp.close()
device.close()
outfp.close()
return
if __name__ == "__main__":
doc = open_pdf(sys.argv[1])
Point = Route = False
pages = page_count(doc)
if pages == 68:
Point = True
elif pages == 1143:
Route = True
else:
sys.stderr.write("PDF file not of recognised (NRG) format\n")
sys.exit(1)
rsrcmgr = PDFResourceManager()
laparams = LAParams()
laparams.line_margin = 0 # Forces every line to be absolutely positioned
laparams.word_margin = 0.2 # Prevents space before narrow characters
device = PDFPageAggregator(rsrcmgr, laparams=laparams)
interpreter = PDFPageInterpreter(rsrcmgr, device)
writer = BufferedWriter(sys.stdout)
for (pageno, page) in enumerate(doc.get_pages()):
interpreter.process_page(page)
layout = device.get_result() # returns LTPage
(text, other) = fsplit(lambda obj: isinstance(obj, LTText), layout)
header_y = 0
if Point:
# Locates bottom of header separator (lowest non-text < 10px height)
def pdfminerr(argv):
global pdfminerr, install
import getopt
def usage():
print ("usage: just put the path to the pdf file in pdf.txt, and make sure you create a seprate folder and put nothing there except for this repository.")
return 100
try:
(opts, args) = getopt.getopt(argv[1:], 'dp:m:P:o:CnAVM:L:W:F:Y:O:t:c:s:')
except getopt.GetoptError:
return usage()
if not args: return usage()
# debug option
debug = 0
# input option
password = ''
pagenos = set()
maxpages = 0
# output option
outfile = None
outtype = None
imagewriter = None
layoutmode = 'normal'
codec = 'utf-8'
pageno = 1
scale = 1
caching = True
showpageno = True
laparams = LAParams()
for (k, v) in opts:
if k == '-d': debug += 1
elif k == '-p': pagenos.update( int(x)-1 for x in v.split(',') )
elif k == '-m': maxpages = int(v)
elif k == '-P': password = v
elif k == '-o': outfile = v
elif k == '-C': caching = False
elif k == '-n': laparams = None
elif k == '-A': laparams.all_texts = True
elif k == '-V': laparams.detect_vertical = True
elif k == '-M': laparams.char_margin = float(v)
elif k == '-L': laparams.line_margin = float(v)
elif k == '-W': laparams.word_margin = float(v)
elif k == '-F': laparams.boxes_flow = float(v)
elif k == '-Y': layoutmode = v
elif k == '-O': imagewriter = ImageWriter(v)
elif k == '-t': outtype = v
elif k == '-c': codec = v
elif k == '-s': scale = float(v)
PDFDocument.debug = debug
PDFParser.debug = debug
CMapDB.debug = debug
PDFResourceManager.debug = debug
PDFPageInterpreter.debug = debug
PDFDevice.debug = debug
rsrcmgr = PDFResourceManager(caching=caching)
if not outtype:
outtype = 'text'
if outfile:
if outfile.endswith('.htm') or outfile.endswith('.html'):
outtype = 'html'
elif outfile.endswith('.xml'):
outtype = 'xml'
elif outfile.endswith('.tag'):
outtype = 'tag'
if outfile:
outfp = file(outfile, 'w')
else:
outfp = sys.stdout
if outtype == 'text':
device = TextConverter(rsrcmgr, outfp, codec=codec, laparams=laparams,
imagewriter=imagewriter)
elif outtype == 'xml':
device = XMLConverter(rsrcmgr, outfp, codec=codec, laparams=laparams,
imagewriter=imagewriter)
elif outtype == 'html':
device = HTMLConverter(rsrcmgr, outfp, codec=codec, scale=scale,
layoutmode=layoutmode, laparams=laparams,
imagewriter=imagewriter)
elif outtype == 'tag':
device = TagExtractor(rsrcmgr, outfp, codec=codec)
else:
return usage()
for fname in args:
fp = file(fname, 'rb')
process_pdf(rsrcmgr, device, fp, pagenos, maxpages=maxpages, password=password,
caching=caching, check_extractable=True)
fp.close()
device.close()
outfp.close()
return
def convert_pdf(path, outtype='txt', opts={}):
outfile = path[:-3] + outtype
outdir = '/'.join(path.split('/')[:-1])
debug = 0
# input option
password = ''
pagenos = set()
maxpages = 0
# output option
codec = 'utf-8'
pageno = 1
scale = 1
showpageno = True
laparams = LAParams()
for (k, v) in opts:
if k == '-d': debug += 1
elif k == '-p': pagenos.update( int(x)-1 for x in v.split(',') )
elif k == '-m': maxpages = int(v)
elif k == '-P': password = v
elif k == '-o': outfile = v
elif k == '-n': laparams = None
elif k == '-A': laparams.all_texts = True
elif k == '-D': laparams.writing_mode = v
elif k == '-M': laparams.char_margin = float(v)
elif k == '-L': laparams.line_margin = float(v)
elif k == '-W': laparams.word_margin = float(v)
elif k == '-O': outdir = v
elif k == '-t': outtype = v
elif k == '-c': codec = v
elif k == '-s': scale = float(v)
#
CMapDB.debug = debug
PDFResourceManager.debug = debug
PDFDocument.debug = debug
PDFParser.debug = debug
PDFPageInterpreter.debug = debug
PDFDevice.debug = debug
#
rsrcmgr = PDFResourceManager()
if not outtype:
outtype = 'txt'
if outfile:
if outfile.endswith('.htm') or outfile.endswith('.html'):
outtype = 'html'
elif outfile.endswith('.xml'):
outtype = 'xml'
elif outfile.endswith('.tag'):
outtype = 'tag'
if outfile:
outfp = file(outfile, 'w')
else:
outfp = sys.stdout
if outtype == 'txt':
device = TextConverter(rsrcmgr, outfp, codec=codec, laparams=laparams)
elif outtype == 'xml':
device = XMLConverter(rsrcmgr, outfp, codec=codec, laparams=laparams, outdir=outdir)
elif outtype == 'html':
device = HTMLConverter(rsrcmgr, outfp, codec=codec, scale=scale, laparams=laparams, outdir=outdir)
elif outtype == 'tag':
device = TagExtractor(rsrcmgr, outfp, codec=codec)
else:
return usage()
fp = file(path, 'rb')
process_pdf(rsrcmgr, device, fp, pagenos, maxpages=maxpages, password=password)
fp.close()
device.close()
outfp.close()
return
def pdf2txt(argv):
import getopt
(opts, args) = getopt.getopt(argv[0:], 'dp:m:P:o:CnAVM:L:W:F:Y:O:t:c:s:')
# debug option
debug = 0
# input option
password = ''
pagenos = set()
maxpages = 0
# output option
outfile = None
outtype = None
outdir = None
layoutmode = 'normal'
codec = 'utf-8'
pageno = 1
scale = 1
caching = True
showpageno = True
laparams = LAParams()
for (k, v) in opts:
if k == '-d': debug += 1
elif k == '-p': pagenos.update(int(x) - 1 for x in v.split(','))
elif k == '-m': maxpages = int(v)
elif k == '-P': password = v
elif k == '-o': outfile = v
elif k == '-C': caching = False
elif k == '-n': laparams = None
elif k == '-A': laparams.all_texts = True
elif k == '-V': laparams.detect_vertical = True
elif k == '-M': laparams.char_margin = float(v)
elif k == '-L': laparams.line_margin = float(v)
elif k == '-W': laparams.word_margin = float(v)
elif k == '-F': laparams.boxes_flow = float(v)
elif k == '-Y': layoutmode = v
elif k == '-O': outdir = v
elif k == '-t': outtype = v
elif k == '-c': codec = v
elif k == '-s': scale = float(v)
#
PDFDocument.debug = debug
PDFParser.debug = debug
CMapDB.debug = debug
PDFResourceManager.debug = debug
PDFPageInterpreter.debug = debug
PDFDevice.debug = debug
#
rsrcmgr = PDFResourceManager(caching=caching)
if not outtype:
outtype = 'text'
if outfile:
if outfile.endswith('.htm') or outfile.endswith('.html'):
outtype = 'html'
elif outfile.endswith('.xml'):
outtype = 'xml'
elif outfile.endswith('.tag'):
outtype = 'tag'
if outfile:
outfp = file(outfile, 'w')
else:
outfp = sys.stdout
if outtype == 'text':
device = TextConverter(rsrcmgr, outfp, codec=codec, laparams=laparams)
elif outtype == 'xml':
device = XMLConverter(rsrcmgr,
outfp,
codec=codec,
laparams=laparams,
outdir=outdir)
elif outtype == 'html':
device = HTMLConverter(rsrcmgr,
outfp,
codec=codec,
scale=scale,
layoutmode=layoutmode,
laparams=laparams,
outdir=outdir)
elif outtype == 'tag':
device = TagExtractor(rsrcmgr, outfp, codec=codec)
else:
return usage()
for fname in args:
fp = file(fname, 'rb')
process_pdf(rsrcmgr,
device,
fp,
pagenos,
maxpages=maxpages,
password=password,
caching=caching,
check_extractable=True)
fp.close()
device.close()
outfp.close()
return
import string
from string import punctuation
from io import BytesIO
from openpyxl import load_workbook
from bs4 import BeautifulSoup
# pdf path
document = open('/Users/anan/Desktop/pdf/Part1_test.pdf', 'rb')
# crate pdf manager
rsrcmgr = PDFResourceManager()
retster = BytesIO()
# Set parameters for analysis.
laparams = LAParams()
laparams.char_margin = 4 # 5 4best
laparams.line_margin = 5 # test:3 part:5
laparams.word_margin = 4
laparams.boxes_flow = 0.5
# Create a PDF page TMLConverter object
device = HTMLConverter(rsrcmgr, retster, codec='utf-8', laparams=laparams)
# create pdf interpreter
interpreter = PDFPageInterpreter(rsrcmgr, device)
for page in PDFPage.get_pages(document):
interpreter.process_page(page)
content = retster.getvalue().decode()
# 以 Beautiful Soup 解析 HTML 程式碼
soup = BeautifulSoup(content, 'html.parser')
def readPDF2HTML(pdfFile, opts={}):
# open a PDF file
fp = StringIO(pdfFile.read())
retstr = StringIO()
# create a PDF parser object associated with the file object
parser = PDFParser(fp)
# create a PDF document allows text extraction
document = PDFDocument(parser) # password if needed
# check if document allows text extraction without password
if not document.is_extractable:
raise PDFTextExtractionNotAllowed
# create a PDF resource manager object that sotres shared resources
rsrcmgr = PDFResourceManager()
# create a PDF device object
laparams = LAParams()
for (k, v) in opts:
if k == '-d':
debug += 1
elif k == '-p':
pagenos.update(int(x) - 1 for x in v.split(','))
elif k == '-m':
maxpages = int(v)
elif k == '-P':
password = v
elif k == '-o':
outfile = v
elif k == '-n':
laparams = None
elif k == '-A':
laparams.all_texts = True
elif k == '-V':
laparams.detect_vertical = True
elif k == '-M':
laparams.char_margin = float(v)
elif k == '-L':
laparams.line_margin = float(v)
elif k == '-W':
laparams.word_margin = float(v)
elif k == '-F':
laparams.boxes_flow = float(v)
elif k == '-Y':
layoutmode = v
elif k == '-O':
outdir = v
elif k == '-t':
outtype = v
elif k == '-c':
codec = v
elif k == '-s':
scale = float(v)
codec = 'utf-8'
device = HTMLConverter(rsrcmgr, retstr, codec=codec, laparams=laparams)
# create a PDF interpreter object
interpreter = PDFPageInterpreter(rsrcmgr, device)
pagenos = set()
# process each page contained in the document
for page in PDFPage.get_pages(fp, pagenos):
interpreter.process_page(page)
# close streams and return text content
fp.close()
content = retstr.getvalue()
device.close()
retstr.close()
return content
def main(argv):
import getopt
def usage():
print(
"usage: %s [-d] [-p pagenos] [-m maxpages] [-P password] [-o output]"
" [-C] [-n] [-A] [-V] [-M char_margin] [-L line_margin] [-W word_margin]"
" [-F boxes_flow] [-Y layout_mode] [-O output_dir] [-R rotation] [-S]"
" [-t text|html|xml|tag] [-c codec] [-s scale]"
" file ..." % argv[0]
)
return 100
try:
(opts, args) = getopt.getopt(argv[1:], "dp:m:P:o:CnAVM:L:W:F:Y:O:R:St:c:s:")
except getopt.GetoptError:
return usage()
if not args:
return usage()
# input option
password = b""
pagenos = set()
maxpages = 0
# output option
outfile = None
outtype = None
imagewriter = None
rotation = 0
stripcontrol = False
layoutmode = "normal"
codec = "utf-8"
pageno = 1
scale = 1
caching = True
showpageno = True
laparams = LAParams()
for (k, v) in opts:
if k == "-d":
logging.getLogger().setLevel(logging.DEBUG)
elif k == "-p":
pagenos.update(int(x) - 1 for x in v.split(","))
elif k == "-m":
maxpages = int(v)
elif k == "-P":
password = v
elif k == "-o":
outfile = v
elif k == "-C":
caching = False
elif k == "-n":
laparams = None
elif k == "-A":
laparams.all_texts = True
elif k == "-V":
laparams.detect_vertical = True
elif k == "-M":
laparams.char_margin = float(v)
elif k == "-L":
laparams.line_margin = float(v)
elif k == "-W":
laparams.word_margin = float(v)
elif k == "-F":
laparams.boxes_flow = float(v)
elif k == "-Y":
layoutmode = v
elif k == "-O":
imagewriter = ImageWriter(v)
elif k == "-R":
rotation = int(v)
elif k == "-S":
stripcontrol = True
elif k == "-t":
outtype = v
elif k == "-c":
codec = v
elif k == "-s":
scale = float(v)
#
rsrcmgr = PDFResourceManager(caching=caching)
if not outtype:
outtype = "text"
if outfile:
if outfile.endswith(".htm") or outfile.endswith(".html"):
outtype = "html"
elif outfile.endswith(".xml"):
outtype = "xml"
elif outfile.endswith(".tag"):
outtype = "tag"
if outfile:
outfp = open(outfile, "wb")
else:
outfp = sys.stdout
if outfp.encoding is not None:
codec = None
if outtype == "text":
device = TextConverter(rsrcmgr, outfp, codec=codec, laparams=laparams, imagewriter=imagewriter)
elif outtype == "xml":
device = XMLConverter(
rsrcmgr, outfp, codec=codec, laparams=laparams, imagewriter=imagewriter, stripcontrol=stripcontrol
)
elif outtype == "html":
device = HTMLConverter(
rsrcmgr, outfp, codec=codec, scale=scale, layoutmode=layoutmode, laparams=laparams, imagewriter=imagewriter
)
elif outtype == "tag":
device = TagExtractor(rsrcmgr, outfp, codec=codec)
else:
return usage()
for fname in args:
fp = open(fname, "rb")
interpreter = PDFPageInterpreter(rsrcmgr, device)
for page in PDFPage.get_pages(
fp, pagenos, maxpages=maxpages, password=password, caching=caching, check_extractable=True
):
page.rotate = (page.rotate + rotation) % 360
interpreter.process_page(page)
fp.close()
device.close()
outfp.close()
return
##ATTEMPT 1
# Create a PDF parser object associated with the file object.
#parser = PDFParser(open_file)
# Create a PDF document object that stores the document structure.
#doc = PDFDocument(parser)
# Connect the parser and document objects.
#print parser.nextline()
#print parser.nextline()
#print parser.nextline()
##ATTEMPT 2
#Code from pdf2txt.py
laparams = LAParams()
laparams.char_margin = 2.0
laparams.line_margin=0.5
laparams.word_margin=0.1
laparams.all_texts=False
rsrcmgr = PDFResourceManager()
device = TextConverter(rsrcmgr, fp_out, codec='utf-8', laparams=laparams)
interpreter = PDFPageInterpreter(rsrcmgr, device)
pdf_pages = PDFPage.get_pages(fp_in, set())
pagenum = 0
pagelim = 3
for page in pdf_pages:
pagenum += 1
if pagenum > pagelim:
continue
print "Transcribing page " + str(pagenum) + " from PDF to text"
interpreter.process_page(page)
def main(fname, k, v):
# debug option
debug = 0
# input option
password = ''
pagenos = set()
maxpages = 0
# output option
outfile = None
outtype = None
imagewriter = None
rotation = 0
stripcontrol = False
layoutmode = 'normal'
codec = 'utf-8'
pageno = 1
scale = 1
caching = True
showpageno = True
laparams = LAParams()
if k == '-d':
debug += 1
elif k == '-p':
pagenos.update( int(x)-1 for x in v.split(',') )
elif k == '-m':
maxpages = int(v)
elif k == '-P':
password = v
elif k == '-o':
outfile = v
elif k == '-C':
caching = False
elif k == '-n':
laparams = None
elif k == '-A':
laparams.all_texts = True
elif k == '-V':
laparams.detect_vertical = True
elif k == '-M':
laparams.char_margin = float(v)
elif k == '-L':
laparams.line_margin = float(v)
elif k == '-W':
laparams.word_margin = float(v)
elif k == '-F':
laparams.boxes_flow = float(v)
elif k == '-Y':
layoutmode = v
elif k == '-O':
imagewriter = ImageWriter(v)
elif k == '-R':
rotation = int(v)
elif k == '-S':
stripcontrol = True
elif k == '-t':
outtype = v
elif k == '-c':
codec = v
elif k == '-s':
scale = float(v)
#
PDFDocument.debug = debug
PDFParser.debug = debug
CMapDB.debug = debug
PDFPageInterpreter.debug = debug
#
rsrcmgr = PDFResourceManager(caching=caching)
if not outtype:
outtype = 'text'
if outfile:
if outfile.endswith('.htm') or outfile.endswith('.html'):
outtype = 'html'
elif outfile.endswith('.xml'):
outtype = 'xml'
elif outfile.endswith('.tag'):
outtype = 'tag'
if outfile:
outfp = file(outfile, 'w')
else:
outfp = sys.stdout
if outtype == 'text':
device = TextConverter(rsrcmgr, outfp, codec=codec, laparams=laparams,
imagewriter=imagewriter)
elif outtype == 'xml':
device = XMLConverter(rsrcmgr, outfp, codec=codec, laparams=laparams,
imagewriter=imagewriter,
stripcontrol=stripcontrol)
elif outtype == 'html':
device = HTMLConverter(rsrcmgr, outfp, codec=codec, scale=scale,
layoutmode=layoutmode, laparams=laparams,
imagewriter=imagewriter, debug=debug)
elif outtype == 'tag':
device = TagExtractor(rsrcmgr, outfp, codec=codec)
else:
return usage()
fp = file(fname, 'rb')
interpreter = PDFPageInterpreter(rsrcmgr, device)
for page in PDFPage.get_pages(fp, pagenos,
maxpages=maxpages, password=password,
caching=caching, check_extractable=True):
page.rotate = (page.rotate+rotation) % 360
interpreter.process_page(page)
fp.close()
device.close()
outfp.close()
return
def main(argv):
import getopt
def usage():
print(
'usage: %s [-d] [-p pagenos] [-m maxpages] [-P password] [-o output] '
'[-n] [-A] [-D writing_mode] [-M char_margin] [-L line_margin] [-W word_margin] '
'[-O output_dir] [-t text|html|xml|tag] [-c codec] [-s scale] file ...'
% argv[0])
return 100
try:
(opts, args) = getopt.getopt(argv[1:], 'dp:m:P:o:nAD:M:L:W:O:t:c:s:')
except getopt.GetoptError:
return usage()
if not args: return usage()
# debug option
debug = 0
# input option
password = ''
pagenos = set()
maxpages = 0
# output option
outfile = None
outtype = None
outdir = None
codec = 'utf-8'
pageno = 1
scale = 1
showpageno = True
laparams = LAParams()
for (k, v) in opts:
if k == '-d': debug += 1
elif k == '-p': pagenos.update(int(x) - 1 for x in v.split(','))
elif k == '-m': maxpages = int(v)
elif k == '-P': password = v
elif k == '-o': outfile = v
elif k == '-n': laparams = None
elif k == '-A': laparams.all_texts = True
elif k == '-D': laparams.writing_mode = v
elif k == '-M': laparams.char_margin = float(v)
elif k == '-L': laparams.line_margin = float(v)
elif k == '-W': laparams.word_margin = float(v)
elif k == '-O': outdir = v
elif k == '-t': outtype = v
elif k == '-c': codec = v
elif k == '-s': scale = float(v)
#
CMapDB.debug = debug
PDFResourceManager.debug = debug
PDFDocument.debug = debug
PDFParser.debug = debug
PDFPageInterpreter.debug = debug
PDFDevice.debug = debug
#
rsrc = PDFResourceManager()
if not outtype:
outtype = 'text'
if outfile:
if outfile.endswith('.htm') or outfile.endswith('.html'):
outtype = 'html'
elif outfile.endswith('.xml'):
outtype = 'xml'
elif outfile.endswith('.tag'):
outtype = 'tag'
if outfile:
outfp = file(outfile, 'w')
else:
outfp = sys.stdout
if outtype == 'text':
device = TextConverter(rsrc, outfp, codec=codec, laparams=laparams)
elif outtype == 'xml':
device = XMLConverter(rsrc,
outfp,
codec=codec,
laparams=laparams,
outdir=outdir)
elif outtype == 'html':
device = HTMLConverter(rsrc,
outfp,
codec=codec,
scale=scale,
laparams=laparams,
outdir=outdir)
elif outtype == 'tag':
device = TagExtractor(rsrc, outfp, codec=codec)
else:
return usage()
for fname in args:
fp = file(fname, 'rb')
process_pdf(rsrc,
device,
fp,
pagenos,
maxpages=maxpages,
password=password)
fp.close()
device.close()
outfp.close()
return
from pdfminer.layout import LTTextBox
from string import punctuation
import sys
import io
from string import punctuation
import re
# pdf path
document = open('C:/Users/Vincent/Dropbox/writeProgram/python/20180509.pdf', 'rb')
#crate pdf manager
rsrcmgr = PDFResourceManager()
# Set parameters for analysis.
laparams = LAParams()
laparams.char_margin = 3.2 #5 4best
laparams.line_margin = 8 #5-8
laparams.word_margin = 10
laparams.boxes_flow = 0.5
# Create a PDF page aggregator object
device = PDFPageAggregator(rsrcmgr, laparams=laparams)
#create pdf interpreter
interpreter = PDFPageInterpreter(rsrcmgr , device)
checkPoint = 0 #初始狀態 檢查各個段落
partOne = "General Information"
partTwo = "Deal History"
partThree = "Investors ("
# 这里layout是一个LTPage对象 里面存放着 这个page解析出的各种对象 一般包括LTTextBox, LTFigure, LTImage,
# LTTextBoxHorizontal 等等 想要获取文本就获得对象的text属性，
for page in PDFPage.get_pages(document):
interpreter.process_page(page)
#fp = open(r'C:/Users/ashmaro1/Documents/GitHub/Text-Analysis/PDFnOCR/data/POxca-000078-AN168968.pdf', 'rb')
#fp = open(r'C:/Users/ashmaro1/Documents/GitHub/Text-Analysis/PDFnOCR/data/POxca-000510-BX425914.pdf', 'rb')
#fp = open(r'C:/Users/ashmaro1/Documents/GitHub/Text-Analysis/PDFnOCR/data/POxca-000078-AN168907.pdf', 'rb')
#fp = open(r'C:/Users/ashmaro1/Documents/GitHub/Text-Analysis/PDFnOCR/data/POxca-000078-AN168907.pdf', 'rb')
#fp = open(r'S:/Bhavani/Aaron/POxca-000052-R201631.pdf', 'rb')
fp = open(r'C:/Users/ashmaro1/Documents/_Projects/Glencore/POxca-000052-R201631.pdf','rb')
parser = PDFParser(fp)
doc = PDFDocument(parser)
#doc.initialize('password') # leave empty for no password
rsrcmgr = PDFResourceManager()
laparams = LAParams()
laparams.char_margin = float('1.1') #too small and it splits the description, too big and Quantity-Unit-Part number are not separated: 1.4 seems to work
laparams.line_margin = float('0.8')
device = PDFPageDetailedAggregator(rsrcmgr, laparams=laparams)
interpreter = PDFPageInterpreter(rsrcmgr, device)
for page in PDFPage.create_pages(doc):
interpreter.process_page(page)
# receive the LTPage object for this page
device.get_result()
print(device.rows)
df = pd.DataFrame(device.rows, columns=['Page', 'x', 'y', 'c1','c2','String'])
# create text rows from 'y' coordinate data
def main(argv):
import getopt
def usage():
print(
'usage: %s [-d] [-p pagenos] [-m maxpages] [-P password] [-o output]'
' [-C] [-n] [-A] [-V] [-M char_margin] [-L line_margin] [-W word_margin]'
' [-F boxes_flow] [-Y layout_mode] [-O output_dir] [-R rotation]'
' [-t text|html|xml|tag] [-c codec] [-s scale]'
' file ...' % argv[0])
return 100
try:
(opts, args) = getopt.getopt(argv[1:],
'dp:m:P:o:CnAVM:L:W:F:Y:O:R:t:c:s:')
except getopt.GetoptError:
return usage()
if not args: return usage()
# debug option
debug = 0
# input option
password = ''
pagenos = set()
maxpages = 0
# output option
outfile = None
outtype = None
imagewriter = None
rotation = 0
layoutmode = 'normal'
codec = 'utf-8'
pageno = 1
scale = 1
caching = True
showpageno = True
laparams = LAParams()
for (k, v) in opts:
if k == '-d': debug += 1
elif k == '-p': pagenos.update(int(x) - 1 for x in v.split(','))
elif k == '-m': maxpages = int(v)
elif k == '-P': password = v
elif k == '-o': outfile = v
elif k == '-C': caching = False
elif k == '-n': laparams = None
elif k == '-A': laparams.all_texts = True
elif k == '-V': laparams.detect_vertical = True
elif k == '-M': laparams.char_margin = float(v)
elif k == '-L': laparams.line_margin = float(v)
elif k == '-W': laparams.word_margin = float(v)
elif k == '-F': laparams.boxes_flow = float(v)
elif k == '-Y': layoutmode = v
elif k == '-O': imagewriter = ImageWriter(v)
elif k == '-R': rotation = int(v)
elif k == '-t': outtype = v
elif k == '-c': codec = v
elif k == '-s': scale = float(v)
#
PDFDocument.debug = debug
PDFParser.debug = debug
CMapDB.debug = debug
PDFResourceManager.debug = debug
PDFPageInterpreter.debug = debug
PDFDevice.debug = debug
#
rsrcmgr = PDFResourceManager(caching=caching)
if not outtype:
outtype = 'text'
if outfile:
if outfile.endswith('.htm') or outfile.endswith('.html'):
outtype = 'html'
elif outfile.endswith('.xml'):
outtype = 'xml'
elif outfile.endswith('.tag'):
outtype = 'tag'
if outfile:
outfp = file(outfile, 'w')
else:
outfp = sys.stdout
if outtype == 'text':
device = TextConverter(rsrcmgr,
outfp,
codec=codec,
laparams=laparams,
imagewriter=imagewriter)
elif outtype == 'xml':
device = XMLConverter(rsrcmgr,
outfp,
codec=codec,
laparams=laparams,
imagewriter=imagewriter)
elif outtype == 'html':
device = HTMLConverter(rsrcmgr,
outfp,
codec=codec,
scale=scale,
layoutmode=layoutmode,
laparams=laparams,
imagewriter=imagewriter)
elif outtype == 'tag':
device = TagExtractor(rsrcmgr, outfp, codec=codec)
else:
return usage()
for fname in args:
fp = file(fname, 'rb')
interpreter = PDFPageInterpreter(rsrcmgr, device)
for page in PDFPage.get_pages(fp,
pagenos,
maxpages=maxpages,
password=password,
caching=caching,
check_extractable=True):
page.rotate = (page.rotate + rotation) % 360
interpreter.process_page(page)
fp.close()
device.close()
outfp.close()
return
def main(argv):
import getopt #getopt 模块，它的功能是 获取执行命令行时附带的参数，关于getopt模块详细可参照http://www.16kan.com/post/207647.html
def usage(): #usage() 函数，用于在用户输入错误命令或者命令输入不规范时，输出py文件的使用范例。当参数不足或错误时，usage()被调用
print ('usage: %s [-d] [-p pagenos] [-m maxpages] [-P password] [-o output] [-C] '
'[-n] [-A] [-V] [-M char_margin] [-L line_margin] [-W word_margin] [-F boxes_flow] '
'[-Y layout_mode] [-O output_dir] [-t text|html|xml|tag] [-c codec] [-s scale] file ...' % argv[0])
return 100
try:
(opts, args) = getopt.getopt(argv[1:], 'dp:m:P:o:CnAVM:L:W:F:Y:O:t:c:s:')
'''
getopt函数的格式是getopt.getopt ( [命令行参数列表], "短选项", [长选项列表] )
短选项名后的冒号(:)表示该选项必须有附加的参数。p,m,P,o,M,L,W,F,Y,O,t,c,s均为必须参数
长选项名后的等号(=)表示该选项必须有附加的参数。
返回opts和args。
'''
except getopt.GetoptError:
return usage()
if not args: return usage()
# debug option
debug = 0
# input option
password = '' #参数P
pagenos = set() #参数p
maxpages = 0 #参数m
# output option
outfile = None #参数o output
outtype = None #参数t out type
outdir = None #参数O output directory
layoutmode = 'normal' #参数Y
codec = 'utf-8' #参数c
pageno = 1
scale = 1 #参数s,暂缺M,L,F,Y四个参数
caching = True
showpageno = True
laparams = LAParams()
for (k, v) in opts:
if k == '-d': debug += 1
elif k == '-p': pagenos.update( int(x)-1 for x in v.split(',') )
elif k == '-m': maxpages = int(v)
elif k == '-P': password = v
elif k == '-o': outfile = v
elif k == '-C': caching = False
elif k == '-n': laparams = None
elif k == '-A': laparams.all_texts = True
elif k == '-V': laparams.detect_vertical = True
elif k == '-M': laparams.char_margin = float(v)
elif k == '-L': laparams.line_margin = float(v)
elif k == '-W': laparams.word_margin = float(v)
elif k == '-F': laparams.boxes_flow = float(v)
elif k == '-Y': layoutmode = v
elif k == '-O': outdir = v
elif k == '-t': outtype = v
elif k == '-c': codec = v
elif k == '-s': scale = float(v)
#
PDFDocument.debug = debug
PDFParser.debug = debug
CMapDB.debug = debug
PDFResourceManager.debug = debug
PDFPageInterpreter.debug = debug
PDFDevice.debug = debug
#
rsrcmgr = PDFResourceManager(caching=caching)
if not outtype: #确认输出文件格式
outtype = 'text'
if outfile:
if outfile.endswith('.htm') or outfile.endswith('.html'):
outtype = 'html'
elif outfile.endswith('.xml'):
outtype = 'xml'
elif outfile.endswith('.tag'):
outtype = 'tag'
if outfile:
outfp = file(outfile, 'w')
else:
outfp = sys.stdout
if outtype == 'text':
device = TextConverter(rsrcmgr, outfp, codec=codec, laparams=laparams) #TextConverter貌似不能指定outdir参数
elif outtype == 'xml':
device = XMLConverter(rsrcmgr, outfp, codec=codec, laparams=laparams, outdir=outdir)
elif outtype == 'html':
device = HTMLConverter(rsrcmgr, outfp, codec=codec, scale=scale,
layoutmode=layoutmode, laparams=laparams, outdir=outdir)
elif outtype == 'tag':
device = TagExtractor(rsrcmgr, outfp, codec=codec)
else:
return usage()
for fname in args:
fp = file(fname, 'rb')
process_pdf(rsrcmgr, device, fp, pagenos, maxpages=maxpages, password=password,
caching=caching, check_extractable=True)
fp.close()
device.close()
outfp.close()
return
def main(argv):
import getopt
def usage():
print ('usage: %s [-d] [-p pagenos] [-m maxpages] [-P password] [-o output]'
' [-C] [-n] [-A] [-V] [-M char_margin] [-L line_margin] [-W word_margin]'
' [-F boxes_flow] [-Y layout_mode] [-O output_dir] [-R rotation]'
' [-t text|html|xml|tag] [-c codec] [-s scale]'
' file ...' % argv[0])
return 100
try:
(opts, args) = getopt.getopt(argv[1:], 'dp:m:P:o:CnAVM:L:W:F:Y:O:R:t:c:s:')
except getopt.GetoptError:
return usage()
if not args: return usage()
# debug option
debug = 0
# input option
password = ''
pagenos = set()
maxpages = 0
# output option
outfile = None
outtype = None
imagewriter = None
rotation = 0
layoutmode = 'normal'
codec = 'utf-8'
pageno = 1
scale = 1
caching = True
showpageno = True
laparams = LAParams()
for (k, v) in opts:
if k == '-d': debug += 1
elif k == '-p': pagenos.update( int(x)-1 for x in v.split(',') )
elif k == '-m': maxpages = int(v)
elif k == '-P': password = v
elif k == '-o': outfile = v
elif k == '-C': caching = False
elif k == '-n': laparams = None
elif k == '-A': laparams.all_texts = True
elif k == '-V': laparams.detect_vertical = True
elif k == '-M': laparams.char_margin = float(v)
elif k == '-L': laparams.line_margin = float(v)
elif k == '-W': laparams.word_margin = float(v)
elif k == '-F': laparams.boxes_flow = float(v)
elif k == '-Y': layoutmode = v
elif k == '-O': imagewriter = ImageWriter(v)
elif k == '-R': rotation = int(v)
elif k == '-t': outtype = v
elif k == '-c': codec = v
elif k == '-s': scale = float(v)
#
PDFDocument.debug = debug
PDFParser.debug = debug
CMapDB.debug = debug
PDFResourceManager.debug = debug
PDFPageInterpreter.debug = debug
PDFDevice.debug = debug
#
rsrcmgr = PDFResourceManager(caching=caching)
if not outtype:
outtype = 'text'
if outfile:
if outfile.endswith('.htm') or outfile.endswith('.html'):
outtype = 'html'
elif outfile.endswith('.xml'):
outtype = 'xml'
elif outfile.endswith('.tag'):
outtype = 'tag'
if outfile:
outfp = file(outfile, 'w')
else:
outfp = sys.stdout
if outtype == 'text':
device = TextConverter(rsrcmgr, outfp, codec=codec, laparams=laparams,
imagewriter=imagewriter)
elif outtype == 'xml':
device = XMLConverter(rsrcmgr, outfp, codec=codec, laparams=laparams,
imagewriter=imagewriter)
elif outtype == 'html':
device = HTMLConverter(rsrcmgr, outfp, codec=codec, scale=scale,
layoutmode=layoutmode, laparams=laparams,
imagewriter=imagewriter)
elif outtype == 'tag':
device = TagExtractor(rsrcmgr, outfp, codec=codec)
else:
return usage()
for fname in args:
fp = file(fname, 'rb')
interpreter = PDFPageInterpreter(rsrcmgr, device)
for page in PDFPage.get_pages(fp, pagenos,
maxpages=maxpages, password=password,
caching=caching, check_extractable=True):
page.rotate = (page.rotate+rotation) % 360
interpreter.process_page(page)
fp.close()
device.close()
outfp.close()
return
def main(argv=None):
parser = argparse.ArgumentParser(description='Convert PDF into text.')
parser.add_argument('file', nargs='*', type=argparse.FileType('rb'), default=sys.stdin, help='file(s) to convert')
parser.add_argument('-C', '--nocache', dest='cache', action='store_false', help='prevent object caching (slower)')
parser.add_argument('-l', metavar='level', default='warn', help='logging level (warn, info, debug)')
parser.add_argument('-p', metavar='page', nargs='+', default=[], type=int, help='page number(s) (space separated)')
parser.add_argument('-m', metavar='maxpages', default=0, type=int, help='maximum number of pages to extract')
parser.add_argument('-P', metavar='password', default='', help='pdf password')
parser.add_argument('-o', metavar='outfile', type=argparse.FileType('w'), default=sys.stdout,
help='output file name (default: stdout)')
parser.add_argument('-O', metavar='directory', type=ImageWriter, help='extract images and save to directory')
parser.add_argument('-t', metavar='outtype', help='output type (text, html, xml, tag)')
parser.add_argument('-c', metavar='codec', default='utf-8', help='output text encoding (default: %(default)s)')
lagroup = parser.add_argument_group(title='layout analysis')
lagroup.add_argument('-n', action='store_true', help='disable layout analysis')
lagroup.add_argument('-A', action='store_true', help='force layout analysis on all text')
lagroup.add_argument('-V', action='store_true', help='detect vertical text')
lagroup.add_argument('-M', metavar='char_margin', type=float, help='custom character margin')
lagroup.add_argument('-L', metavar='line_margin', type=float, help='custom line margin')
lagroup.add_argument('-W', metavar='word_margin', type=float, help='custom word margin')
lagroup.add_argument('-F', metavar='boxes_flow', type=float, help='custom boxes flow')
lagroup.add_argument('-Y', metavar='layout_mode', default='normal', help='layout mode for HTML (normal, exact, loose)')
lagroup.add_argument('-s', metavar='scale', default=1, type=float, help='output scaling for HTML')
args = parser.parse_args(argv)
logging.basicConfig()
logging.getLogger('pdfminer').setLevel(args.l.upper())
laparams = LAParams()
if args.n:
laparams = None
else:
laparams.all_texts = args.A
laparams.detect_vertical = args.V
if args.M:
laparams.char_margin = args.M
if args.L:
laparams.line_margin = args.L
if args.W:
laparams.word_margin = args.W
if args.F:
laparams.boxes_flow = args.F
rsrcmgr = PDFResourceManager(caching=args.cache)
outtype = args.t
if not outtype:
if args.o:
if args.o.name.endswith('.htm') or args.o.name.endswith('.html'):
outtype = 'html'
elif args.o.name.endswith('.xml'):
outtype = 'xml'
elif args.o.name.endswith('.tag'):
outtype = 'tag'
if outtype == 'xml':
device = XMLConverter(rsrcmgr, args.o, codec=args.c, laparams=laparams, imagewriter=args.O)
elif outtype == 'html':
device = HTMLConverter(rsrcmgr, args.o, codec=args.c, scale=args.s, layoutmode=args.Y,
laparams=laparams, imagewriter=args.O)
elif outtype == 'tag':
device = TagExtractor(rsrcmgr, args.o, codec=args.c)
else:
device = TextConverter(rsrcmgr, args.o, codec=args.c, laparams=laparams, imagewriter=args.O)
for fp in args.file:
process_pdf(rsrcmgr, device, fp, [i-1 for i in args.p], maxpages=args.m, password=args.P,
caching=args.cache, check_extractable=True)
fp.close()
device.close()
if args.o is not sys.stdout:
args.o.close()
def pdf2txt(argv):
import getopt
(opts, args) = getopt.getopt(argv[0:], 'dp:m:P:o:CnAVM:L:W:F:Y:O:t:c:s:')
# debug option
debug = 0
# input option
password = ''
pagenos = set()
maxpages = 0
# output option
outfile = None
outtype = None
outdir = None
layoutmode = 'normal'
codec = 'utf-8'
pageno = 1
scale = 1
caching = True
showpageno = True
laparams = LAParams()
for (k, v) in opts:
if k == '-d': debug += 1
elif k == '-p': pagenos.update( int(x)-1 for x in v.split(',') )
elif k == '-m': maxpages = int(v)
elif k == '-P': password = v
elif k == '-o': outfile = v
elif k == '-C': caching = False
elif k == '-n': laparams = None
elif k == '-A': laparams.all_texts = True
elif k == '-V': laparams.detect_vertical = True
elif k == '-M': laparams.char_margin = float(v)
elif k == '-L': laparams.line_margin = float(v)
elif k == '-W': laparams.word_margin = float(v)
elif k == '-F': laparams.boxes_flow = float(v)
elif k == '-Y': layoutmode = v
elif k == '-O': outdir = v
elif k == '-t': outtype = v
elif k == '-c': codec = v
elif k == '-s': scale = float(v)
#
PDFDocument.debug = debug
PDFParser.debug = debug
CMapDB.debug = debug
PDFResourceManager.debug = debug
PDFPageInterpreter.debug = debug
PDFDevice.debug = debug
#
rsrcmgr = PDFResourceManager(caching=caching)
if not outtype:
outtype = 'text'
if outfile:
if outfile.endswith('.htm') or outfile.endswith('.html'):
outtype = 'html'
elif outfile.endswith('.xml'):
outtype = 'xml'
elif outfile.endswith('.tag'):
outtype = 'tag'
if outfile:
outfp = file(outfile, 'w')
else:
outfp = sys.stdout
if outtype == 'text':
device = TextConverter(rsrcmgr, outfp, codec=codec, laparams=laparams)
elif outtype == 'xml':
device = XMLConverter(rsrcmgr, outfp, codec=codec, laparams=laparams, outdir=outdir)
elif outtype == 'html':
device = HTMLConverter(rsrcmgr, outfp, codec=codec, scale=scale,
layoutmode=layoutmode, laparams=laparams, outdir=outdir)
elif outtype == 'tag':
device = TagExtractor(rsrcmgr, outfp, codec=codec)
else:
return usage()
for fname in args:
fp = file(fname, 'rb')
process_pdf(rsrcmgr, device, fp, pagenos, maxpages=maxpages, password=password,
caching=caching, check_extractable=True)
fp.close()
device.close()
outfp.close()
return
def main(files=None):
if files is None:
files = get_datafiles()
# debug option level
debug = 0
# input option
password = ''
pagenos = set()
# pagenos.update( int(x)-1 for x in v.split(',') )
maxpages = 0
# output option
rotation = 0
stripcontrol = False
layoutmode = 'normal'
codec = 'utf-8'
pageno = 1
scale = 1
caching = True
rsrcmgr = PDFResourceManager(caching=caching)
showpageno = True
# Line Agumentation ? Parameters
laparams = LAParams()
laparams.all_texts = True
laparams.detect_vertical = True
laparams.line_overlap = 0.3 # Line overlap
laparams.char_margin = 2.0 # Letter Spacing
laparams.line_margin = 0.5 # Line Spacing
laparams.word_margin = 0.1 # Word spacing
laparams.boxes_flow = 0.5 # +-1.0 how much hor vs. vertical matters
# position maters for line continuation
#
PDFDocument.debug = debug
PDFParser.debug = debug
CMapDB.debug = debug
PDFPageInterpreter.debug = debug
#
for fname in files:
fname = str(fname)
imagedir = os.path.abspath(os.path.join(os.path.dirname(fname), 'img'))
# print(imagedir)
imagewriter = None
imagewriter = ImageWriter(imagedir) # output folder for images
name = os.path.splitext(os.path.basename(fname))[0]
print(name)
outfile = fname[:-4] + '.txt'
device = TextCon(rsrcmgr,
laparams=laparams,
imagewriter=imagewriter,
imagename=name)
interpreter = PDFPageInterpreter(rsrcmgr, device)
fp = file(fname, 'rb')
try:
for page in PDFPage.get_pages(fp,
pagenos,
maxpages=maxpages,
password=password,
caching=caching,
check_extractable=True):
page.rotate = (page.rotate + rotation) % 360
interpreter.process_page(page)
except:
continue
rows = [list(row) for row in device.rows]
pages = max([row[0] for row in rows])
max_y = max([row[4] for row in rows])
min_y = min([row[2] for row in rows])
list_0 = [int(row[4]) for row in rows]
list_1 = []
[
list_1.append(obj) for obj in list_0
if obj not in list_1 and list_0.count(obj) > pages - 1
]
max_y2 = max(list_1)
list_0 = [int(row[2]) for row in rows]
list_1 = []
[
list_1.append(obj) for obj in list_0
if obj not in list_1 and list_0.count(obj) > pages - 1
]
min_y2 = min(list_1)
print('max_ys:', max_y - max_y2)
print('min_ys:', min_y - min_y2)
# Get max and min the hard way because of stupid headers
list_0 = [int(row[3]) for row in rows]
list_1 = []
[
list_1.append(obj) for obj in list_0
if obj not in list_1 and list_0.count(obj) > 10
]
if list_1:
max_x = max(list_1)
else:
max_x = max([int(row[3]) for row in device.rows])
list_0 = [int(row[1]) for row in rows]
list_1 = []
[
list_1.append(obj) for obj in list_0
if obj not in list_1 and list_0.count(obj) > 10
]
if list_1:
min_x = min(list_1)
else:
min_x = min([int(row[3]) for row in device.rows])
# Errors if more pics on one side then other
# mid_x = (sum([(float(row[1]) + float(row[3]))/2 for row in
# device.rows])/len(device.rows))
mid_x = (max_x + min_x) / 2
# mid_x = 595/2 # center of A4 at 72px/in Letter would be 612/2
l_height = sum([row[4] - row[2] for row in rows]) / len(rows)
# print('max_x:', max_x)
# print('min_x:', min_x)
# print('mid_x:', mid_x)
print('l_height:', l_height)
column2 = []
lines = []
pagenumber = 0
table_caps = ['\n']
table_data = []
table = False
for i, row in enumerate(rows):
#l_height = row[4]-row[2]
l_space = rows[i - 1][2] - row[4]
#print(l_height, l_space, rows[i-1][2], rows[i][4], str(row[5]))
if row[0] == pagenumber + 1:
lines += column2
column2 = []
pagenumber += 1
if row[0] == pagenumber:
if (max_y - min_y) * 0.95 > l_space > 0.8 * l_height:
# capture Table (assuming tables will span all columns)
if re.match(r"^table", str(row[5]), re.I):
table = True
table_caps.append(str(row[5]))
table_data.append('\n')
table_data.append(str(row[5]))
table_data.append('\n')
continue
else:
table = False
# capture table captions multi lines
elif (table_caps[-1] == str(rows[i - 1][5])
and -2 * l_height < l_space < 0.5 * l_height):
table_caps[-1] += str(row[5])
table_data[-2] += str(row[5])
continue
if table:
# capture table data
if int(rows[i - 1][2]) == int(rows[i][2]):
table_data[-1] += '\t' + str(row[5])
continue
else:
table_data.append(str(row[5]))
continue
elif int(row[1]) > mid_x and ((int(rows[i - 1][1]) < mid_x and
int(rows[i - 1][3]) < mid_x) or
(int(rows[i - 1][1]) > mid_x
and int(rows[i - 1][3]) > mid_x)
or rows[i - 1][3] > max_x * 0.9
or l_space > 2.5 * l_height):
"""
r_space > c_space or
previous[3] > max_x * 0.9 or
l_space > 2 * l_height):"""
if len(column2) > 0:
if 1 > (row[2] - column2[-1][2]) > -1:
# join if on same line
if int(row[1]) < int(column2[-1][1]):
column2[-1][5] = row[5] + " " + column2[-1][5]
else:
column2[-1][5] = column2[-1][5] + " " + row[5]
else:
column2.append(row)
else:
column2.append(row)
# print(2, str(row[5]))
else:
if len(lines) > 0:
if 1 > (row[2] - lines[-1][2]) > -1:
# join if on same line
if int(row[1]) < int(lines[-1][1]):
lines[-1][5] = row[5] + " " + lines[-1][5]
else:
lines[-1][5] = lines[-1][5] + " " + row[5]
else:
lines.append(row)
else:
lines.append(row)
# print(3, str(row[5]))
# add final column
lines += column2
fig_caps = ['\n']
headers = ['\n']
footers = ['\n']
supp_info = ['\n']
new_lines = []
supp_re = re.compile(
r"Corresponding author|Electronic mail|email"
"|E-mail|^doi|doi:|^keywords|^pacs|^apc", re.I)
for i, line in enumerate(lines):
#l_height = lines[i][4]-lines[i][2]
l_space = lines[i - 1][2] - lines[i][4]
l_space_below = 0
l_space_2below = 0
if i + 1 < len(lines):
l_space_below = lines[i][2] - lines[i + 1][4]
if i + 2 < len(lines):
l_space_2below = lines[i + 1][2] - lines[i + 2][4]
fig = fig_caps[-1]
print(l_space, l_space_below, l_space_2below, lines[i][2],
lines[i][4], str(line[5]))
# capture figure captions multi lines
if (fig_caps[-1] == str(lines[i - 1][5])
and -2 * l_height < l_space < 0.5 * l_height):
fig_caps.append(str(line[5]))
continue
# capture headers (up to two lines)
if (lines[i][2] > max_y * 0.95
and (l_space_below > 0.5 * l_height
or l_space_2below > 0.5 * l_height)):
headers.append('\n')
headers.append(str(line[5]))
if supp_re.search(str(line[5])):
headers.append('\n')
headers.append(str(line[5]))
else:
continue
# capture supporting info
if supp_re.search(str(line[5])):
print(str(line[5]))
supp_info.append('\n')
supp_info.append(str(line[5]))
continue
if (max_y - min_y) * 0.95 > l_space > 0.5 * l_height:
# capture figure captions
if re.match(r"^fig", str(line[5]), re.I):
fig_caps.append('\n')
fig_caps.append(str(line[5]))
continue
# capture footers
elif lines[i][2] < min_y + max_y * 0.015:
footers.append('\n')
footers.append(str(line[5]))
continue
else:
string = str(lines[i - 1][5])
if (any(string in s for s in fig_caps)
or any(string in s for s in headers)): # or
#string == footers[-1] or string == supp_info[-1]):
pass
else:
new_lines.append('\n')
new_lines.append(str(line[5]))
with open(outfile, 'w') as f:
f.write(' '.join(new_lines))
f.write('\n\nFigures')
f.write(' '.join(fig_caps))
f.write('\n\nTables')
#f.write(' '.join(table_caps))
f.write('\n'.join(table_data))
f.write('\n\nHeaders')
f.write(' '.join(headers))
f.write('\n\nFooters')
f.write(' '.join(footers))
f.write('\n\nSupporting Info')
f.write(' '.join(supp_info))
# the histogram of the data
# n, bins, patches = plt.hist(x_data, 50)
# plt.show()
device.close()
print('Done')
return
def extTxt(fname, outfile):
# debug option
debug = 0
# input option
password = ''
pagenos = set()
maxpages = 0
# output option
# outfile = None
outtype = None
imagewriter = None
rotation = 0
layoutmode = 'normal'
codec = 'utf-8'
pageno = 1
scale = 1
caching = True
showpageno = True
# '-d': debug += 1
laparams = LAParams()
laparams.line_margin = float(30)
laparams.word_margin = float(0.1)
# '-n': laparams = None
# '-A': laparams.all_texts = True
# '-V': laparams.detect_vertical = True
# '-M': laparams.char_margin = float(v)
# '-F': laparams.boxes_flow = float(v)
# '-Y': layoutmode = v
#
PDFDocument.debug = debug
PDFParser.debug = debug
CMapDB.debug = debug
PDFResourceManager.debug = debug
PDFPageInterpreter.debug = debug
PDFDevice.debug = debug
#
rsrcmgr = PDFResourceManager(caching=caching)
outtype = 'text'
# outfile = 'SAMPLE/Output.tmp'
if outfile:
outfp = file(outfile, 'w')
else:
outfp = sys.stdout
outLns = ''
device = TextConverter(rsrcmgr,
outfp,
codec=codec,
laparams=laparams,
imagewriter=imagewriter)
# fname = 'SAMPLE/sample.pdf'
fp = file(fname, 'rb')
interpreter = PDFPageInterpreter(rsrcmgr, device)
for page in PDFPage.get_pages(fp,
pagenos,
maxpages=maxpages,
password=password,
caching=caching,
check_extractable=True):
page.rotate = (page.rotate + rotation) % 360
interpreter.process_page(page)
fp.close()
device.close()
outfp.close()
return
def main(argv):
import getopt
def usage():
print ('usage: %s [-d] [-p pagenos] [-m maxpages] [-P password] [-o output] '
'[-n] [-A] [-D writing_mode] [-M char_margin] [-L line_margin] [-W word_margin] '
'[-c codec] file ...' % argv[0])
return 100
try:
(opts, args) = getopt.getopt(argv[1:], 'dp:m:P:o:nAD:M:L:W:c:')
except getopt.GetoptError:
return usage()
if not args: return usage()
# debug option
debug = 0
# input option
password = ''
pagenos = set()
maxpages = 0
# output option
outfile = None
codec = 'utf-8'
pageno = 1
showpageno = True
laparams = LAParams()
for (k, v) in opts:
if k == '-d': debug += 1
elif k == '-p': pagenos.update( int(x)-1 for x in v.split(',') )
elif k == '-m': maxpages = int(v)
elif k == '-P': password = v
elif k == '-o': outfile = v
elif k == '-n': laparams = None
elif k == '-A': laparams.all_texts = True
elif k == '-D': laparams.writing_mode = v
elif k == '-M': laparams.char_margin = float(v)
elif k == '-L': laparams.line_margin = float(v)
elif k == '-W': laparams.word_margin = float(v)
elif k == '-c': codec = v
#
CMapDB.debug = debug
PDFResourceManager.debug = debug
PDFDocument.debug = debug
PDFParser.debug = debug
PDFPageInterpreter.debug = debug
PDFDevice.debug = debug
#
rsrcmgr = PDFResourceManager()
if outfile:
outfp = file(outfile, 'w')
else:
outfp = sys.stdout
device = CourseRegisterParser(rsrcmgr, outfp, codec=codec, laparams=laparams)
for fname in args:
fp = file(fname, 'rb')
process_pdf(rsrcmgr, device, fp, pagenos, maxpages=maxpages, password=password)
fp.close()
device.close()
outfp.close()
return
def main(argv=None):
parser = argparse.ArgumentParser(description='Convert PDF into text.')
parser.add_argument('file',
nargs='*',
type=argparse.FileType('rb'),
default=sys.stdin,
help='file(s) to convert')
parser.add_argument('-C',
'--nocache',
dest='cache',
action='store_false',
help='prevent object caching (slower)')
parser.add_argument('-l',
metavar='level',
default='warn',
help='logging level (warn, info, debug)')
parser.add_argument('-p',
metavar='page',
nargs='+',
default=[],
type=int,
help='page number(s) (space separated)')
parser.add_argument('-m',
metavar='maxpages',
default=0,
type=int,
help='maximum number of pages to extract')
parser.add_argument('-P',
metavar='password',
default='',
help='pdf password')
parser.add_argument('-o',
metavar='outfile',
type=argparse.FileType('w'),
default=sys.stdout,
help='output file name (default: stdout)')
parser.add_argument('-O',
metavar='directory',
type=ImageWriter,
help='extract images and save to directory')
parser.add_argument('-t',
metavar='outtype',
help='output type (text, html, xml, tag)')
parser.add_argument('-c',
metavar='codec',
default='utf-8',
help='output text encoding (default: %(default)s)')
lagroup = parser.add_argument_group(title='layout analysis')
lagroup.add_argument('-n',
action='store_true',
help='disable layout analysis')
lagroup.add_argument('-A',
action='store_true',
help='force layout analysis on all text')
lagroup.add_argument('-V',
action='store_true',
help='detect vertical text')
lagroup.add_argument('-M',
metavar='char_margin',
type=float,
help='custom character margin')
lagroup.add_argument('-L',
metavar='line_margin',
type=float,
help='custom line margin')
lagroup.add_argument('-W',
metavar='word_margin',
type=float,
help='custom word margin')
lagroup.add_argument('-F',
metavar='boxes_flow',
type=float,
help='custom boxes flow')
lagroup.add_argument('-Y',
metavar='layout_mode',
default='normal',
help='layout mode for HTML (normal, exact, loose)')
lagroup.add_argument('-s',
metavar='scale',
default=1,
type=float,
help='output scaling for HTML')
args = parser.parse_args(argv)
logging.basicConfig()
logging.getLogger('pdfminer').setLevel(args.l.upper())
laparams = LAParams()
if args.n:
laparams = None
else:
laparams.all_texts = args.A
laparams.detect_vertical = args.V
if args.M:
laparams.char_margin = args.M
if args.L:
laparams.line_margin = args.L
if args.W:
laparams.word_margin = args.W
if args.F:
laparams.boxes_flow = args.F
rsrcmgr = PDFResourceManager(caching=args.cache)
outtype = args.t
if not outtype:
if args.o:
if args.o.name.endswith('.htm') or args.o.name.endswith('.html'):
outtype = 'html'
elif args.o.name.endswith('.xml'):
outtype = 'xml'
elif args.o.name.endswith('.tag'):
outtype = 'tag'
if outtype == 'xml':
device = XMLConverter(rsrcmgr,
args.o,
codec=args.c,
laparams=laparams,
imagewriter=args.O)
elif outtype == 'html':
device = HTMLConverter(rsrcmgr,
args.o,
codec=args.c,
scale=args.s,
layoutmode=args.Y,
laparams=laparams,
imagewriter=args.O)
elif outtype == 'tag':
device = TagExtractor(rsrcmgr, args.o, codec=args.c)
else:
device = TextConverter(rsrcmgr,
args.o,
codec=args.c,
laparams=laparams,
imagewriter=args.O)
for fp in args.file:
process_pdf(rsrcmgr,
device,
fp, [i - 1 for i in args.p],
maxpages=args.m,
password=args.P,
caching=args.cache,
check_extractable=True)
fp.close()
device.close()
if args.o is not sys.stdout:
args.o.close()
from pdfminer.pdfparser import PDFParser
from pdfminer.pdfdocument import PDFDocument
from pdfminer.pdfinterp import PDFResourceManager, PDFPageInterpreter
from pdfminer.layout import LAParams
from pdfminer.converter import PDFPageAggregator
fp = open("Lista_samurai_x.pdf", "rb")
parser = PDFParser(fp)
doc = PDFDocument(parser)
parser.set_document(doc)
doc.set_parser(parser)
doc.initialize("")
rsrcmgr = PDFResourceManager()
laparamns = LAParams()
laparamns.line_margin = 0.3
laparamns.word_margin = 0.3
device = PDFPageAggregator(rsrcmgr, laparamns=laparamns)
interpreter = PDFPageInterpreter(rsrcmgr, device)
for page in doc.get_pages():
interpreter.process_page(page)
layout = device.get_result()
for ltobject in layout:
print(ltobject.get_text())
def main(argv):
import getopt
def usage():
print 'Syntax:\npdf2htm.exe SourcePDF\n where the parameter is either a file name or\na wildcard spec like\n*.pdf\nEnclose it with quotes if it contains a space\n\nAdditional options are supported with named command line parameters as follows:'
print ('usage: %s [-d] [-p pagenos] [-m maxpages] [-P password] [-o output]'
' [-C] [-n] [-A] [-V] [-M char_margin] [-L line_margin] [-W word_margin]'
' [-F boxes_flow] [-Y layout_mode] [-O output_dir] [-R rotation]'
' [-t text|html|xml|tag] [-c codec] [-s scale]'
' file ...' % argv[0])
return 100
try:
(opts, args) = getopt.getopt(argv[1:], 'dp:m:P:o:CnAVM:L:W:F:Y:O:R:t:c:s:')
except getopt.GetoptError:
return usage()
if not args: return usage()
# debug option
debug = 0
# input option
password = ''
pagenos = set()
maxpages = 0
# output option
outfile = None
outtype = 'tag'
imagewriter = None
rotation = 0
layoutmode = 'normal'
codec = 'utf-8'
pageno = 1
scale = 1
caching = True
showpageno = False
laparams = LAParams()
for (k, v) in opts:
if k == '-d': debug += 1
elif k == '-p': pagenos.update( int(x)-1 for x in v.split(',') )
elif k == '-m': maxpages = int(v)
elif k == '-P': password = v
elif k == '-o': outfile = v
elif k == '-C': caching = False
elif k == '-n': laparams = None
elif k == '-A': laparams.all_texts = True
elif k == '-V': laparams.detect_vertical = True
elif k == '-M': laparams.char_margin = float(v)
elif k == '-L': laparams.line_margin = float(v)
elif k == '-W': laparams.word_margin = float(v)
elif k == '-F': laparams.boxes_flow = float(v)
elif k == '-Y': layoutmode = v
elif k == '-O': imagewriter = ImageWriter(v)
elif k == '-R': rotation = int(v)
elif k == '-t': outtype = v
elif k == '-c': codec = v
elif k == '-s': scale = float(v)
#
PDFDocument.debug = debug
PDFParser.debug = debug
CMapDB.debug = debug
PDFResourceManager.debug = debug
PDFPageInterpreter.debug = debug
PDFDevice.debug = debug
#
rsrcmgr = PDFResourceManager(caching=caching)
if not outtype:
outtype = 'tag'
if outfile:
if outfile.endswith('.htm') or outfile.endswith('.html'):
outtype = 'html'
elif outfile.endswith('.xml'):
outtype = 'xml'
elif outfile.endswith('.tag'):
outtype = 'tag'
if outfile:
outfp = file(outfile, 'w')
else:
outfp = sys.stdout
for fname in args:
l = glob.glob(fname)
count = len(l)
print 'Converting ' + str(count) + ' from ' + fname + ' to ' + outtype + ' format'
for pdf in l:
# print pdf
d = {'html' : 'htm', 'tag' : 'tag', 'text' : 'txt', 'xml' : 'xml'}
ext = '.' + d[outtype]
outfile = pdf[0:-4] + ext
print outfile
outfp = file(outfile, 'wb')
if outtype == 'text':
device = TextConverter(rsrcmgr, outfp, codec=codec, laparams=laparams,
imagewriter=imagewriter)
device.showpageno = False
elif outtype == 'xml':
device = XMLConverter(rsrcmgr, outfp, codec=codec, laparams=laparams,
imagewriter=imagewriter)
device.showpageno = False
elif outtype == 'html':
device = HTMLConverter(rsrcmgr, outfp, codec=codec, scale=scale,
layoutmode=layoutmode, laparams=laparams,
imagewriter=imagewriter)
device.showpageno = False
elif outtype == 'tag':
device = TagExtractor(rsrcmgr, outfp, codec=codec)
device.showpageno = False
else:
return usage()
fp = file(pdf, 'rb')
interpreter = PDFPageInterpreter(rsrcmgr, device)
for page in PDFPage.get_pages(fp, pagenos,
maxpages=maxpages, password=password,
caching=caching, check_extractable=True):
page.rotate = (page.rotate+rotation) % 360
interpreter.process_page(page)
fp.close()
device.close()
outfp.close()
print 'Done'
return
def parse_pdf(PDF_path, TEX_Path, is_annotation, is_train_image, is_main):
filename = os.path.basename(PDF_path).split('.pdf')[0]
page_counter = 0
titles_counter = 0
titles_coordinates = []
images_counter = 0
images_coordinates = []
lists_coordinates = []
tables_coordinates = []
text_coordinates = []
all_train_objects_coordinates = []
with_annotations = is_annotation
if is_main: generate_images(PDF_path, filename, is_train_image)
if is_train_image:
tex_instances = find_tex_istances(TEX_Path)
if not tex_instances:
return all_train_objects_coordinates
# Open a PDF file.
fp = open(PDF_path, 'rb')
# Create a PDF parser object associated with the file object.
parser = PDFParser(fp)
# Create a PDF document object that stores the document structure.
# Supply the password for initialization.
try:
document = PDFDocument(parser)
except PSSyntaxError:
print('Invalid PDF structure')
return all_train_objects_coordinates
# Check if the document allows text extraction. If not, abort.
if not document.is_extractable:
raise PDFTextExtractionNotAllowed
# Create a PDF resource manager object that stores shared resources.
rsrcmgr = PDFResourceManager()
# Create a PDF device object.
# Set parameters for analysis.
laparams = LAParams()
laparams.line_margin = 0.4
# Create a PDF page aggregator object.
device = PDFPageAggregator(rsrcmgr, laparams=laparams)
interpreter = PDFPageInterpreter(rsrcmgr, device)
for page in PDFPage.create_pages(document):
page_counter += 1
page_length = page.mediabox[3]
interpreter.process_page(page)
# receive the LTPage object for the page.
layout = device.get_result()
print(
'##########################################################################################################')
print('PAGE NUMBER: ', page_counter)
print(
'##########################################################################################################',
'\n')
for x in layout:
# TEXTS (TITLES AND LISTS)
if isinstance(x, LTTextBoxHorizontal):
lines = x._objs
if '' in lines: lines.remove('')
for i in range(2 if len(lines) > 2 else len(
lines)): # iterate over the first lines of a texbox since titles are always on the top
pdf_line_result = lines[i].get_text().split('\n')[0].lower()
pdf_line_result = ''.join([i for i in pdf_line_result if not i.isdigit()])
for instance in tex_instances[0]:
tex_title = instance[2].lower()
if are_similar(tex_title, pdf_line_result) and pdf_line_result != '':
titles_counter += 1
titles_coordinates.append(
calculate_object_coordinates(page_counter, lines[i].bbox, page_length, 'title'))
lines_counter = 0
for i in range(len(lines)):
pdf_line_result = lines[i].get_text().split('\n')[0].lower()
pdf_line_result = ''.join([i for i in pdf_line_result if not i.isdigit()])
for instance in tex_instances[2]:
tex_list_item = instance[3]
if are_similar(tex_list_item[0:50], pdf_line_result[0:50]) and pdf_line_result != '':
lists_coordinates.append(
calculate_object_coordinates(page_counter, lines[i].bbox, page_length, 'list'))
elif lines_counter > 2:
text_coordinates.append(
calculate_object_coordinates(page_counter, lines[i].bbox, page_length, 'text'))
lines_counter += 1
# FIGURES
elif isinstance(x, LTImage) or isinstance(x, LTFigure):
if (x.width / x.height > 5) or (x.height / x.width > 5):
pass
else:
images_counter += 1
images_coordinates.append(
calculate_object_coordinates(page_counter, x.bbox, page_length, 'image'))
elif isinstance(x, LTLine):
if (x.height == 0 and x.width < 30) or x.width <= 0:
pass
else:
tables_coordinates.append(
calculate_object_coordinates(page_counter, x.bbox, page_length, 'table'))
extracted_tables_coordinates = extract_tables_coordinates(tables_coordinates)
extracted_lists_coordinates = extract_lists_coordinates(lists_coordinates)
if with_annotations == 'yes':
print('Generating annotations...')
if len(titles_coordinates) != 0: annotate_img(filename, titles_coordinates, titles_coordinates[0][0],
(0, 0, 255), 3)
if len(images_coordinates) != 0: annotate_img(filename, images_coordinates, images_coordinates[0][0],
(0, 255, 0), 3)
if len(extracted_lists_coordinates) != 0: annotate_img(filename, extracted_lists_coordinates,
extracted_lists_coordinates[0][0], (255, 0, 0), 3)
if len(extracted_tables_coordinates) != 0: annotate_img(filename, extracted_tables_coordinates,
extracted_tables_coordinates[0][0], (230, 255, 102),
3)
all_train_objects_coordinates.extend(titles_coordinates)
all_train_objects_coordinates.extend(images_coordinates)
all_train_objects_coordinates.extend(extracted_lists_coordinates)
all_train_objects_coordinates.extend(extracted_tables_coordinates)
all_train_objects_coordinates = sorted(all_train_objects_coordinates, key=itemgetter(0))
return all_train_objects_coordinates
def main(argv):
# debug option
debug = 0
# input option
password = ''
pagenos = set()
maxpages = 0
# output option
outfile = None
outtype = None
imagewriter = None
rotation = 0
stripcontrol = False
layoutmode = 'normal'
codec = 'utf-8'
pageno = 1
scale = 1
caching = True
showpageno = False
laparams = LAParams()
using_optparse = False
parser = ArgumentParser(prog='pdf2txt.py',
description='Convert pdf to txt',
formatter_class=ArgumentDefaultsHelpFormatter)
if using_optparse:
DEBUG(3, 'using optparse')
parser.add_argument = parser.add_option
parser.parse_known_args = parser.parse_args
parser.disable_interspersed_args()
parser.add_argument('-d', dest='debuglevel', action='count',
default = 0,
help='Debug (repeat for more verbose debugging)')
parser.add_argument('-p', '--pages', dest='pagenos', action='store',
type=str,
default = '',
help='Specifies the comma-separated list of the page numbers to be extracted. Page numbers start at one. By default, it extracts text from all the pages.')
parser.add_argument('-c', '--codec', dest='codec', action='store',
type=str,
default='utf-8',
help='Specifies the output codec.')
parser.add_argument('-t', '--type', dest='outtype', action='store',
type=str,
default='shape',
choices = ['text', 'html', 'xml', 'tag', 'shape'],
help='Specifies the output format, one of: shape, text, html, xml, tag')
parser.add_argument('-m', dest='maxpages', action='store',
type=int,
default=0,
help='Specifies the maximum number of pages to extract. By default (0), it extracts all the pages in a document.')
parser.add_argument('-P', '--password', dest='password', action='store',
type=str,
default='',
help='Provides the user password to access PDF contents.')
parser.add_argument('-o', '--output', dest='outfile', action='store',
type=str,
default=None,
help='Specifies the output file name. By default, it prints the extracted contents to stdout in text format.')
parser.add_argument('-C', '--no-caching', dest='caching', action='store_false',
default=True,
help='Suppress object caching. This will reduce the memory consumption but also slows down the process.')
parser.add_argument('-n', '--no-layout', dest='layout', action='store_false',
default=True,
help='Suppress layout analysis.')
parser.add_argument('--show-pageno', dest='show_pageno', action='store_true',
default=False,
help='Show page numbers.')
parser.add_argument('-A', '--analyze-all', dest='all_texts', action='store_true',
default=False,
help='Forces to perform layout analysis for all the text strings, including text contained in figures.')
parser.add_argument('-V', '--detect-vertical', dest='detect_vertical', action='store_true',
default=False,
help='Allows vertical writing detection.')
parser.add_argument('-M', dest='char_margin', action='store',
type=float,
default=2.0,
help='Two text chunks whose distance is closer than the char_margin (shown as M) is considered continuous and get grouped into one.')
parser.add_argument('-L', dest='line_margin', action='store',
type=float,
default=0.5,
help='Two lines whose distance is closer than the line_margin (L) is grouped as a text box, which is a rectangular area that contains a "cluster" of text portions.')
parser.add_argument('-W', dest='word_margin', action='store',
type=float,
default=0.1,
help='It may be required to insert blank characters (spaces) as necessary if the distance between two words is greater than the word_margin (W), as a blank between words might not be represented as a space, but indicated by the positioning of each word.')
parser.add_argument('-F', dest='boxes_flow', action='store',
type=float,
default=0.5,
help='Specifies how much a horizontal and vertical position of a text matters when determining a text order. The value should be within the range of -1.0 (only horizontal position matters) to +1.0 (only vertical position matters).')
parser.add_argument('-Y', '--layout-mode', dest='layoutmode', action='store',
type=str,
default='normal',
choices = ['exact', 'normal', 'loose'],
help='Specifies how the page layout should be preserved. (Currently only applies to HTML format.) One of: exact, normal, loose.')
parser.add_argument('-O', '--image-writer', dest='imagewriter', action='store',
type=str,
default=None,
help='imagewriter')
parser.add_argument('-R', '--rotation', dest='rotation', action='store',
type=int,
default=0,
help='rotation')
parser.add_argument('-S', '--strip-control', dest='stripcontrol', action='store_true',
default=False,
help='stripcontrol')
parser.add_argument('-s', dest='scale', action='store',
type=float,
default=1,
help='Specifies the output scale. Can be used in HTML format only.')
parser.add_argument('--draw-lines', dest='draw_lines', action='store_true',
help="Draw crude page representation, coloured TextLines (= short pieces of text). Valid only for the `shape' output.")
parser.add_argument('--draw-boxes', dest='draw_boxes', action='store_true',
help="Draw crude page representation, coloured TextBoxes (= grouped text lines). Valid only for the `shape' output.")
parser.add_argument('--draw-blocks', dest='draw_blocks', action='store_true',
help="Draw crude page representation, coloured TextBlocks (= grouped TextBoxes). Valid only for the `shape' output.")
parser.add_argument('--shear-limit', dest='shear_limit', action='store',
default=0.1,
type=float,
help="If the text is sheared above this limit, reject it. Valid only for the `shape' output.")
parser.add_argument('--rotation-limit', dest='rotation_limit', action='store',
default=2,
type=float,
help="If the text is rotated above this angle (in degrees), reject it. Valid only for the `shape' output.")
parser.add_argument('--line-height-diff', dest='line_height_diff', action='store',
type=float,
default=0.1,
help='Two lines whose vertical sizes differ more than this ratio are not to be considered of the same paragraph (but e.g. one of them is a heading).')
parser.add_argument('--heading-before', dest='heading_before', action='store',
type=str,
default='',
help='String to put before each heading, e.g. <h1>')
parser.add_argument('--heading-after', dest='heading_after', action='store',
type=str,
default='',
help='String to put after each heading, e.g. </h1>')
parser.add_argument('--box-separator', dest='box_separator', action='store',
type=str,
default=r'\n\n',
help=r'Separate boxes with this string. Use \n for new line, \t for TAB, other escape sequences are not recognized.')
parser.add_argument('--block-separator', dest='block_separator', action='store',
type=str,
default=r'\n\n',
help=r'Separate blocks with this string. Use \n for new line, \t for TAB, other escape sequences are not recognized.')
parser.add_argument('--indent-separator', dest='indent_separator', action='store',
type=str,
default=r'\n\n',
help=r'Separate indented lines with this string. Use \n for new line, \t for TAB, other escape sequences are not recognized.')
parser.add_argument('--indent-string', dest='indent_string', action='store',
type=str,
default=r'\t',
help=r'Put this string in front of indented lines. Use \n for new line, \t for TAB, other escape sequences are not recognized.')
parser.add_argument('--indent-limit', dest='indent_limit', action='store',
type=float,
default=3,
help='If the line is indented more then this (approximately characters), it will separated by --indent-separator from the previous one.')
parser.add_argument('--page-separator', dest='page_separator', action='store',
type=str,
default=r'\n\n',
help=r'Separate pages with this string. Use \n for new line, \t for TAB, other escape sequences are not recognized.')
parser.add_argument('--norm-whitespace', dest='norm_whitespace', action='store_true',
default=False,
help='Normalize whitespace (remove duplicate spaces, replace end of lines with spaces).')
parser.add_argument('--print-stats', dest='print_stats', action='store_true',
default=False,
help='Instead of the text, output some simple statistics about the file.')
parser.add_argument('--max-blocks', dest='max_blocks', action='store',
default=0,
type=int,
help='If there is more than this blocks per page, do not return any text. Use to discriminate abnormal files (run --print-stats first to find out the number of boxes per "normal" file). 0 means no limit. 50 is maybe a good value.')
parser.add_argument('--max-textlines', dest='max_textlines', action='store',
default=0,
type=int,
help='If there is more than this textlines per any block, do not return any text. Use to discriminate abnormal files (run --print-stats first to find out the number of boxes per "normal" page). 0 means no limit. 18 is maybe a good value.')
parser.add_argument('--line-height-method', dest='line_height_method', action='store',
type=str,
default='bbox',
choices = ['bbox', 'mean', 'median'],
help='Method to calculate height of line (relevant if there are characters with uneven height). bbox takes the bounding box (rectangle encompassing the line), mean the arithmetic mean of the height of all the characters, median is the median of the height of all the characters. Use mean or median if there are outlier characters, e.g. one big character at the beginning of line.')
parser.add_argument(dest='pdffile', help='List of PDF files to go through', default=None, nargs='+')
args, rest = parser.parse_known_args()
global debuglevel
debuglevel = debug = args.debuglevel
DEBUG(3, 'args:', str(args))
DEBUG(3, 'rest:', str(rest))
DEBUG(3, 'optparse:', using_optparse)
if args.pagenos:
pagenos.update( int(x)-1 for x in args.pagenos.split(',') )
maxpages = args.maxpages
outfile = args.outfile
password = args.password
caching = args.caching
showpageno = args.show_pageno
if not args.layout:
laparams = None
if laparams and args.all_texts:
laparams.all_texts = True
if laparams and args.detect_vertical:
laparams.detect_vertical = True
if laparams:
laparams.char_margin = args.char_margin
laparams.line_margin = args.line_margin
laparams.word_margin = args.word_margin
laparams.boxes_flow = args.boxes_flow
layoutmode = args.layoutmode
if args.imagewriter:
imagewriter = ImageWriter(args.imagewriter)
rotation = args.rotation
stripcontrol = args.stripcontrol
outtype = args.outtype
codec = args.codec
scale = args.scale
args.box_separator = unescape_string(args.box_separator)
args.block_separator = unescape_string(args.block_separator)
args.indent_separator = unescape_string(args.indent_separator)
args.indent_string = unescape_string(args.indent_string)
args.page_separator = unescape_string(args.page_separator)
global options
options = args
PDFDocument.debug = debug
PDFParser.debug = debug
CMapDB.debug = debug
PDFPageInterpreter.debug = debug
rsrcmgr = PDFResourceManager(caching=caching)
if not outtype:
outtype = 'text'
if outfile:
if outfile.endswith('.htm') or outfile.endswith('.html'):
outtype = 'html'
elif outfile.endswith('.xml'):
outtype = 'xml'
elif outfile.endswith('.tag'):
outtype = 'tag'
if outfile:
outfp = file(outfile, 'w')
DEBUG(2, 'output goes to', outfile)
else:
outfp = sys.stdout
DEBUG(2, 'output goes to stdout')
if outtype == 'shape':
device = ShapeTextConverter(rsrcmgr, outfp, codec=codec, laparams=laparams,
showpageno=showpageno, imagewriter=imagewriter)
elif outtype == 'text':
device = TextConverter(rsrcmgr, outfp, codec=codec, laparams=laparams,
imagewriter=imagewriter)
elif outtype == 'xml':
device = XMLConverter(rsrcmgr, outfp, codec=codec, laparams=laparams,
imagewriter=imagewriter,
stripcontrol=stripcontrol)
elif outtype == 'html':
device = HTMLConverter(rsrcmgr, outfp, codec=codec, scale=scale,
layoutmode=layoutmode, laparams=laparams,
imagewriter=imagewriter, debug=debug)
elif outtype == 'tag':
device = TagExtractor(rsrcmgr, outfp, codec=codec)
else:
return usage()
for fname in options.pdffile:
DEBUG(2, 'processing', fname)
fp = file(fname, 'rb')
interpreter = PDFPageInterpreter(rsrcmgr, device)
for page in PDFPage.get_pages(fp, pagenos,
maxpages=maxpages, password=password,
caching=caching, check_extractable=True):
page.rotate = (page.rotate+rotation) % 360
interpreter.process_page(page)
fp.close()
device.close()
outfp.close()
DEBUG(2, 'finished.')
return
def main(argv):
import getopt
def usage():
print(
"usage: %s [-d] [-p pagenos] [-m maxpages] [-P password] [-o output] [-C] "
"[-n] [-A] [-V] [-M char_margin] [-L line_margin] [-W word_margin] [-F boxes_flow] "
"[-Y layout_mode] [-O output_dir] [-t text|html|xml|tag] [-c codec] [-s scale] file ..." % argv[0]
)
return 100
try:
(opts, args) = getopt.getopt(argv[1:], "dp:m:P:o:CnAVM:L:W:F:Y:O:t:c:s:")
except getopt.GetoptError:
return usage()
if not args:
return usage()
# debug option
debug = 0
# input option
password = ""
pagenos = set()
maxpages = 0
# output option
outfile = None
outtype = None
outdir = None
layoutmode = "normal"
codec = "utf-8"
pageno = 1
scale = 1
caching = True
showpageno = True
laparams = LAParams()
for (k, v) in opts:
if k == "-d":
debug += 1
elif k == "-p":
pagenos.update(int(x) - 1 for x in v.split(","))
elif k == "-m":
maxpages = int(v)
elif k == "-P":
password = v
elif k == "-o":
outfile = v
elif k == "-C":
caching = False
elif k == "-n":
laparams = None
elif k == "-A":
laparams.all_texts = True
elif k == "-V":
laparams.detect_vertical = True
elif k == "-M":
laparams.char_margin = float(v)
elif k == "-L":
laparams.line_margin = float(v)
elif k == "-W":
laparams.word_margin = float(v)
elif k == "-F":
laparams.boxes_flow = float(v)
elif k == "-Y":
layoutmode = v
elif k == "-O":
outdir = v
elif k == "-t":
outtype = v
elif k == "-c":
codec = v
elif k == "-s":
scale = float(v)
#
# PDFDocument.debug = debug
# PDFParser.debug = debug
CMapDB.debug = debug
PDFResourceManager.debug = debug
PDFPageInterpreter.debug = debug
PDFDevice.debug = debug
#
rsrcmgr = PDFResourceManager(caching=caching)
if not outtype:
outtype = "text"
if outfile:
if outfile.endswith(".htm") or outfile.endswith(".html"):
outtype = "html"
elif outfile.endswith(".xml"):
outtype = "xml"
elif outfile.endswith(".tag"):
outtype = "tag"
if outfile:
outfp = file(outfile, "w")
else:
outfp = sys.stdout
if outtype == "text":
device = TextConverter(rsrcmgr, outfp, codec=codec, laparams=laparams)
elif outtype == "xml":
device = XMLConverter(rsrcmgr, outfp, codec=codec, laparams=laparams, outdir=outdir)
elif outtype == "html":
device = HTMLConverter(
rsrcmgr, outfp, codec=codec, scale=scale, layoutmode=layoutmode, laparams=laparams, outdir=outdir
)
elif outtype == "tag":
device = TagExtractor(rsrcmgr, outfp, codec=codec)
else:
return usage()
for fname in args:
fp = file(fname, "rb")
process_pdf(
rsrcmgr, device, fp, pagenos, maxpages=maxpages, password=password, caching=caching, check_extractable=True
)
fp.close()
device.close()
outfp.close()
return
def convert_pdf_To_Txt(path,opts={}):
"""
this ALGO form pdfinterp modul documentation
"""
# debug option
debug = 0
# input option
password = ''
pagenos = set()
maxpages = 0
# output option
outfile = None
outtype = None
imagewriter = None
layoutmode = 'normal'
codec = 'utf-8'
pageno = 1
scale = 1
caching = True
showpageno = True
laparams = LAParams()
for (k, v) in opts:
if k == '-d': debug += 1
elif k == '-p': pagenos.update( int(x)-1 for x in v.split(',') )
elif k == '-m': maxpages = int(v)
elif k == '-P': password = v
elif k == '-o': outfile = v
elif k == '-C': caching = False
elif k == '-n': laparams = None
elif k == '-A': laparams.all_texts = True
elif k == '-V': laparams.detect_vertical = True
elif k == '-M': laparams.char_margin = float(v)
elif k == '-L': laparams.line_margin = float(v)
elif k == '-W': laparams.word_margin = float(v)
elif k == '-F': laparams.boxes_flow = float(v)
elif k == '-Y': layoutmode = v
elif k == '-O': imagewriter = ImageWriter(v)
elif k == '-t': outtype = v
elif k == '-c': codec = v
elif k == '-s': scale = float(v)
#
PDFDocument.debug = debug
PDFParser.debug = debug
CMapDB.debug = debug
PDFResourceManager.debug = debug
PDFPageInterpreter.debug = debug
PDFDevice.debug = debug
#
rsrcmgr = PDFResourceManager(caching=caching)
if not outtype:
outtype = 'text'
if outfile:
if outfile.endswith('.htm') or outfile.endswith('.html'):
outtype = 'html'
elif outfile.endswith('.xml'):
outtype = 'xml'
elif outfile.endswith('.tag'):
outtype = 'tag'
if outfile:
outfp = file(outfile, 'w')
else:
outfp = sys.stdout
retstr = StringIO()
if outtype == 'text':
device = TextConverter(rsrcmgr, retstr, codec=codec, laparams=laparams,
imagewriter=imagewriter)
fp = file(path, 'rb')
interpreter = PDFPageInterpreter(rsrcmgr, device)
for page in PDFPage.get_pages(fp, pagenos,
maxpages=maxpages, password=password,
caching=caching, check_extractable=True):
interpreter.process_page(page)
#print retstr.getvalue()
txt2Pdf=retstr.getvalue()
#print type(txt2Pdf)
#fp.close()
#device.close()
#outfp.close()
return txt2Pdf
def main(argv):
import getopt
def usage():
print(f'usage: {argv[0]} [-P password] [-o output] [-t text|html|xml|tag]'
' [-O output_dir] [-c encoding] [-s scale] [-R rotation]'
' [-Y normal|loose|exact] [-p pagenos] [-m maxpages]'
' [-S] [-C] [-n] [-A] [-V] [-M char_margin] [-L line_margin]'
' [-W word_margin] [-F boxes_flow] [-d] input.pdf ...')
return 100
try:
(opts, args) = getopt.getopt(argv[1:], 'dP:o:t:O:c:s:R:Y:p:m:SCnAVM:W:L:F:')
except getopt.GetoptError:
return usage()
if not args: return usage()
# debug option
debug = 0
# input option
password = b''
pagenos = set()
maxpages = 0
# output option
outfile = None
outtype = None
imagewriter = None
rotation = 0
stripcontrol = False
layoutmode = 'normal'
encoding = 'utf-8'
pageno = 1
scale = 1
caching = True
showpageno = True
laparams = LAParams()
for (k, v) in opts:
if k == '-d': debug += 1
elif k == '-P': password = v.encode('ascii')
elif k == '-o': outfile = v
elif k == '-t': outtype = v
elif k == '-O': imagewriter = ImageWriter(v)
elif k == '-c': encoding = v
elif k == '-s': scale = float(v)
elif k == '-R': rotation = int(v)
elif k == '-Y': layoutmode = v
elif k == '-p': pagenos.update( int(x)-1 for x in v.split(',') )
elif k == '-m': maxpages = int(v)
elif k == '-S': stripcontrol = True
elif k == '-C': caching = False
elif k == '-n': laparams = None
elif k == '-A': laparams.all_texts = True
elif k == '-V': laparams.detect_vertical = True
elif k == '-M': laparams.char_margin = float(v)
elif k == '-W': laparams.word_margin = float(v)
elif k == '-L': laparams.line_margin = float(v)
elif k == '-F': laparams.boxes_flow = float(v)
#
PDFDocument.debug = debug
PDFParser.debug = debug
CMapDB.debug = debug
PDFPageInterpreter.debug = debug
#
rsrcmgr = PDFResourceManager(caching=caching)
if not outtype:
outtype = 'text'
if outfile:
if outfile.endswith('.htm') or outfile.endswith('.html'):
outtype = 'html'
elif outfile.endswith('.xml'):
outtype = 'xml'
elif outfile.endswith('.tag'):
outtype = 'tag'
if outfile:
if sys.platform == 'linux':
outfp = open(outfile, 'w', encoding=encoding)
elif sys.platform == 'win32':
outfp = open(outfile, 'wb')
else:
outfp = sys.stdout
if outtype == 'text':
device = TextConverter(rsrcmgr, outfp, laparams=laparams,
imagewriter=imagewriter)
elif outtype == 'xml':
device = XMLConverter(rsrcmgr, outfp, laparams=laparams,
imagewriter=imagewriter,
stripcontrol=stripcontrol)
elif outtype == 'html':
device = HTMLConverter(rsrcmgr, outfp, scale=scale,
layoutmode=layoutmode, laparams=laparams,
imagewriter=imagewriter, debug=debug)
elif outtype == 'tag':
device = TagExtractor(rsrcmgr, outfp)
else:
return usage()
for fname in args:
with open(fname, 'rb') as fp:
interpreter = PDFPageInterpreter(rsrcmgr, device)
for page in PDFPage.get_pages(fp, pagenos,
maxpages=maxpages, password=password,
caching=caching, check_extractable=True):
page.rotate = (page.rotate+rotation) % 360
interpreter.process_page(page)
device.close()
outfp.close()
return
if m is None:
continue
if gname is not None:
logging.warn('skip %s: 2 groups in one doc: %s [%s] and %s [%s]' % (pdf_file, gname, gtype, m.groups()[2], m.groups()[0]))
return
gname = m.groups()[2].strip()
gtype = m.groups()[0].strip()
if u'五十音順' in gname:
# HACK: some docs are really borked..
return
logging.info(u'%s is a doc for %s' % (pdf_file, gname))
return gname
LAPARAMS = LAParams()
LAPARAMS.line_margin = 10.0
def extract_pdf_text(fname):
rsrcmgr = PDFResourceManager(caching=True)
codec = 'utf-8'
outfp = StringIO()
device = TextConverter(rsrcmgr, outfp, codec=codec, laparams=LAPARAMS,
imagewriter=None)
fp = file(fname, 'rb')
interpreter = PDFPageInterpreter(rsrcmgr, device)
for page in PDFPage.get_pages(fp, set(), maxpages=1,
caching=True, check_extractable=True):
interpreter.process_page(page)
fp.close()
device.close()
def main(argv):
def usage():
print((
'usage: %s [-d] [-p pagenos] [-m maxpages] [-P password] [-o output] [-C] '
'[-n] [-A] [-V] [-M char_margin] [-L line_margin] [-W word_margin] [-F boxes_flow] '
'[-Y layout_mode] [-O output_dir] [-t text|html|xml|tag] [-c codec] [-s scale] file ...'
% argv[0]))
return 100
try:
(opts, args) = getopt.getopt(argv[1:],
'dp:m:P:o:CnAVM:L:W:F:Y:O:t:c:s:')
except getopt.GetoptError:
return usage()
if not args: return usage()
debug = False
# input option
password = ''
pagenos = set()
maxpages = 0
# output option
outfile = None
outtype = None
outdir = None
layoutmode = 'normal'
codec = 'utf-8'
pageno = 1
scale = 1
caching = True
showpageno = True
laparams = LAParams()
for (k, v) in opts:
if k == '-d': debug = True
elif k == '-p': pagenos.update(int(x) - 1 for x in v.split(','))
elif k == '-m': maxpages = int(v)
elif k == '-P': password = v
elif k == '-o': outfile = v
elif k == '-C': caching = False
elif k == '-n': laparams = None
elif k == '-A': laparams.all_texts = True
elif k == '-V': laparams.detect_vertical = True
elif k == '-M': laparams.char_margin = float(v)
elif k == '-L': laparams.line_margin = float(v)
elif k == '-W': laparams.word_margin = float(v)
elif k == '-F': laparams.boxes_flow = float(v)
elif k == '-Y': layoutmode = v
elif k == '-O': outdir = v
elif k == '-t': outtype = v
elif k == '-c': codec = v
elif k == '-s': scale = float(v)
if debug:
set_debug_logging()
rsrcmgr = PDFResourceManager(caching=caching)
if not outtype:
outtype = 'text'
if outfile:
if outfile.endswith('.htm') or outfile.endswith('.html'):
outtype = 'html'
elif outfile.endswith('.xml'):
outtype = 'xml'
elif outfile.endswith('.tag'):
outtype = 'tag'
if outfile:
outfp = io.open(outfile, 'wt', encoding=codec, errors='ignore')
close_outfp = True
else:
outfp = sys.stdout
close_outfp = False
if outtype == 'text':
device = TextConverter(rsrcmgr, outfp, laparams=laparams)
elif outtype == 'xml':
device = XMLConverter(rsrcmgr, outfp, laparams=laparams, outdir=outdir)
elif outtype == 'html':
device = HTMLConverter(rsrcmgr,
outfp,
scale=scale,
layoutmode=layoutmode,
laparams=laparams,
outdir=outdir,
debug=debug)
elif outtype == 'tag':
device = TagExtractor(rsrcmgr, outfp)
else:
return usage()
for fname in args:
fp = io.open(fname, 'rb')
process_pdf(rsrcmgr,
device,
fp,
pagenos,
maxpages=maxpages,
password=password,
caching=caching,
check_extractable=True)
fp.close()
device.close()
if close_outfp:
outfp.close()
def main(argv):
import getopt
def usage():
print ('usage: %s [-d] [-p pagenos] [-m maxpages] [-P password] [-o output] [-C] '
'[-n] [-A] [-V] [-M char_margin] [-L line_margin] [-W word_margin] [-F boxes_flow] '
'[-Y layout_mode] [-O output_dir] [-t text|html|xml|tag] [-c codec] [-s scale] [-r] '
'[-S] [-f] file ...' % argv[0])
return 100
try:
(opts, args) = getopt.getopt(argv[1:], 'fSrdp:m:P:o:CnAVM:L:W:F:Y:O:t:c:s:')
except getopt.GetoptError:
return usage()
if not args: return usage()
# debug option
debug = 0
# input option
password = ''
pagenos = set()
maxpages = 0
# output option
outfile = None
outtype = None
imagewriter = None
layoutmode = 'normal'
codec = 'utf-8'
pageno = 1
scale = 1
caching = True
showpageno = True
roundCoords = False
simplifyOutput = False
formatOutput = False
laparams = LAParams()
for (k, v) in opts:
if k == '-d':
debug += 1
elif k == '-p':
pagenos.update(int(x) - 1 for x in v.split(','))
elif k == '-m':
maxpages = int(v)
elif k == '-P':
password = v
elif k == '-o':
outfile = v
elif k == '-C':
caching = False
elif k == '-n':
laparams = None
elif k == '-A':
laparams.all_texts = True
elif k == '-V':
laparams.detect_vertical = True
elif k == '-M':
laparams.char_margin = float(v)
elif k == '-L':
laparams.line_margin = float(v)
elif k == '-W':
laparams.word_margin = float(v)
elif k == '-F':
laparams.boxes_flow = float(v)
elif k == '-Y':
layoutmode = v
elif k == '-O':
imagewriter = ImageWriter(v)
elif k == '-t':
outtype = v
elif k == '-c':
codec = v
elif k == '-s':
scale = float(v)
elif k == '-r':
roundCoords = True
elif k == '-S':
simplifyOutput = True
elif k == '-f':
formatOutput = True
PDFDocument.debug = debug
PDFParser.debug = debug
CMapDB.debug = debug
PDFResourceManager.debug = debug
PDFPageInterpreter.debug = debug
PDFDevice.debug = debug
#
rsrcmgr = PDFResourceManager(caching=caching)
if not outtype:
outtype = 'text'
if outfile:
if outfile.endswith('.htm') or outfile.endswith('.html'):
outtype = 'html'
elif outfile.endswith('.xml'):
outtype = 'xml'
elif outfile.endswith('.tag'):
outtype = 'tag'
if formatOutput and outtype.endswith('ml'):
try:
from cStringIO import StringIO
except ImportError:
from StringIO import StringIO
outfp = StringIO()
else:
outfp = getRealOutput(outfile)
if outtype == 'text':
device = TextConverter(rsrcmgr, outfp, codec=codec, laparams=laparams,
imagewriter=imagewriter)
elif outtype == 'xml':
device = XMLConverter(rsrcmgr, outfp, codec=codec, laparams=laparams,
imagewriter=imagewriter, layoutmode=layoutmode,
scale=scale, roundCoords=roundCoords, simplifyOutput=simplifyOutput)
elif outtype == 'html':
device = HTMLConverter(rsrcmgr, outfp, codec=codec, scale=scale,
layoutmode=layoutmode, laparams=laparams,
imagewriter=imagewriter)
elif outtype == 'tag':
device = TagExtractor(rsrcmgr, outfp, codec=codec)
else:
return usage()
for fname in args:
fp = file(fname, 'rb')
process_pdf(rsrcmgr, device, fp, pagenos, maxpages=maxpages, password=password,
caching=caching, check_extractable=True)
fp.close()
device.close()
if formatOutput:
root = outfp.getvalue()
with getRealOutput(outfile) as realOutput:
try:
from bs4 import BeautifulSoup as bs
except ImportError:
bs = None
sys.stderr.write('Could not import BeautifulSoup, skipping output formatting')
realOutput.write(root)
else:
soup = bs(root)
prettyHTML = soup.prettify()
realOutput.write(prettyHTML)
outfp.close()
return
