depdf package

Submodules

depdf.api module

depdf.api.api_load_pdf(api_func)
depdf.api.convert_page_to_html(pdf, pid, **kwargs)
Parameters:
  • pdf – pdf file path
  • pid – page number start from 1
  • kwargs – config keyword arguments
Returns:

page html string
depdf.api.convert_pdf_to_html(pdf, **kwargs)
Parameters:
  • pdf – pdf file path
  • kwargs – config keyword arguments
Returns:

pdf html string
depdf.api.extract_page_paragraphs(pdf, pid, **kwargs)
Parameters:
  • pdf – pdf file path
  • pid – page number start from 1
  • kwargs – config keyword arguments
Returns:

page paragraphs list
depdf.api.extract_page_tables(pdf, pid, **kwargs)
Parameters:
  • pdf – pdf file path
  • pid – page number start from 1
  • kwargs – config keyword arguments
Returns:

page tables list

depdf.base module

class depdf.base.Base

Bases: object

html
refresh()
reset()
soup
to_dict
to_soup(parser)
write_to(file_name)
class depdf.base.Box

Bases: object

bbox
bottom = Decimal('0')
height
static normalize_bbox(bbox)
top = Decimal('0')
width
x0 = Decimal('0')
x1 = Decimal('0')
class depdf.base.InnerWrapper

Bases: depdf.base.Base

inner_objects
to_dict

depdf.config module

class depdf.config.Config(**kwargs)

Bases: object

add_horizontal_line_tolerance = Decimal('0.1')
add_horizontal_lines_flag = False
add_line_flag = False
add_vertical_lines_flag = False
char_overlap_size = 3
char_size_lower = Decimal('3')
char_size_upper = Decimal('30')
column_region_half_width = 4
copy(**kwargs)
curved_line_flag = False
debug_flag = False
default_char_size = Decimal('12')
default_head_tail_page_offset_percent = 0.1
dotted_line_flag = True
image_class = 'pdf-image'
image_flag = True
image_resolution = 300
log_level = 30
logo_flag = True
main_frame_tolerance = None
max_columns = 3
max_double_line_tolerance = 3
min_column_region_objects = 1
min_double_line_tolerance = Decimal('0.05')
min_image_size = 80
mini_page_class = 'pdf-mini-page'
multiple_columns_flag = True
page_class = 'pdf-page'
page_num_left_fraction = Decimal('0.44')
page_num_right_fraction = Decimal('0.56')
page_num_top_fraction = Decimal('0.75')
paragraph_class = 'pdf-paragraph'
paragraph_flag = True
pdf_class = 'pdf-content'
resolution = 144
skip_empty_table = False
snap_flag = False
span_class = 'pdf-span'
table_cell_merge_tolerance = 5
table_class = 'pdf-table'
table_flag = True
temp_dir_prefix = 'temp_depdf'
to_dict
unique_prefix = None
update(**kwargs)
verbose_flag = False
vertical_double_line_tolerance = Decimal('2')
x_tolerance = None
y_tolerance = None
depdf.config.check_config(func)
depdf.config.check_config_type(config)

depdf.error module

exception depdf.error.BoxValueError(value)

Bases: ValueError

exception depdf.error.ConfigTypeError(value)

Bases: TypeError

exception depdf.error.PDFTypeError(value)

Bases: TypeError

exception depdf.error.PageTypeError(value)

Bases: TypeError

depdf.log module

depdf.log.logger_init(name)

depdf.page module

class depdf.page.DePage(page, pid='1', same=None, logo=None, config=None, columns=1, mini=False)

Bases: depdf.base.Base

analyze_lines()
analyze_main_frame()
analyze_page_attributes()
analyze_paragraph_border()
ave_cs = 0
ave_lh = 3
border = (0, 0, 0, 0)
chars
check_if_toc_page()
check_multi_column_page()
config
debug = False
extract_images()
extract_paragraph()
extract_phrases()
extract_tables()
frame_bottom = 0
frame_top = 0
h_edges = []
height
html
images
images_raw
min_cs = 0
new_para_end_flag = None
new_para_start_flag = None
object_key_list = ['_tables', '_paragraphs', '_images']
objects
orientation = ''
page
pagination_phrases = []
paragraphs
phrases = None
pid
prefix = UUID('6f4176ec-ac20-4d25-a08e-ade1feaec45a')
process_mini_page()
process_page()
refresh()
save_html()
screenshot
set_global()
tables
tables_raw
temp_dir = 'temp'
to_html
to_screenshot()
toc_flag = False
v_edges = []
verbose = False
width
x_tolerance = 3
y_tolerance = 3
class depdf.page.MiniDePage(page, pid='1', same=None, logo=None, config=None, columns=1, mini=False)

Bases: depdf.page.DePage

save_html()
to_html
depdf.page.check_page_type(page)
depdf.page.convert_plumber_table(pdf_page, table, pid='1', tid=1, config=None, min_cs=1)
depdf.page.extract_cell_region(cell_region, bbox, config=None, pid='1', tid=1, cid=1)

depdf.page_tools module

depdf.page_tools.add_horizontal_lines(v_lines, h_lines, vlts_tolerance=0.1)
depdf.page_tools.add_vertical_lines(v_lines, h_lines, page_rects, page, ave_cs)
depdf.page_tools.analyze_char_size(chars, char_size_upper=30, char_size_lower=3, default_char_size=12)
depdf.page_tools.analyze_page_num_word(phrases, page_height, page_width, top_fraction=Decimal('0.7'), left_fraction=Decimal('0.4'), right_fraction=Decimal('0.6'))
depdf.page_tools.analyze_page_orientation(plumber_page)
Parameters:plumber_page – pdfplumber.page.Page class
Returns:
depdf.page_tools.calculate_paragraph_border(depdf_page_object)
depdf.page_tools.curve_to_lines(curves)
depdf.page_tools.edges_to_lines(edges)
depdf.page_tools.format_text(text)
depdf.page_tools.merge_page_figures(pdf_page, tables_raw=None, logo=None, min_width=3, min_height=3, pid='1')
depdf.page_tools.remove_duplicate_chars(chars, overlap_size=3)
depdf.page_tools.remove_single_lines(lines, max_double=3, min_double=0.05, vertical_double=2, m='h')

depdf.pdf module

class depdf.pdf.DePDF(pdf, config=None, **kwargs)

Bases: depdf.base.Base

close()
config
extract_html_pages()
generate_pages()
get_prefix()
html
html_pages
classmethod load(file_name, config=None, **kwargs)
classmethod open(*args, **kwargs)
page_num
pages
pdf
refresh()
same
save_html()
to_html
depdf.pdf.check_pdf_type(pdf)

depdf.pdf_tools module

depdf.pdf_tools.check_page_orientation(pdf, pid)
Parameters:
  • pdf – pdfplumber class
  • pid – page number starts from 0
Returns:
depdf.pdf_tools.pdf_head_tail(pdf, config=None)
Parameters:
  • pdf – plumber pdf object
  • config – depdf config class
Returns:

PDF 文件的页眉和页脚

depdf.settings module

depdf.utils module

depdf.utils.calc_bbox(objects)
depdf.utils.calc_overlap(a, b)

检查两个线段的重叠部分长度 :param a: [a_lower, a_upper] :param b: [b_lower, b_upper] :return: overlapping length

depdf.utils.construct_style(style=None)
depdf.utils.convert_html_to_soup(html, parser='html.parser')
depdf.utils.convert_soup_to_html(soup)
depdf.utils.repr_str(text, max_length=5)

depdf.version module

Module contents

depdf

An ultimate pdf file disintegration tool. DePDF is designed to extract tables and paragraphs into structured markup language [eg. html] from embedding pdf pages. You can also use it to convert page/pdf to html.

class depdf.Config(**kwargs)

Bases: object

add_horizontal_line_tolerance = Decimal('0.1')
add_horizontal_lines_flag = False
add_line_flag = False
add_vertical_lines_flag = False
char_overlap_size = 3
char_size_lower = Decimal('3')
char_size_upper = Decimal('30')
column_region_half_width = 4
copy(**kwargs)
curved_line_flag = False
debug_flag = False
default_char_size = Decimal('12')
default_head_tail_page_offset_percent = 0.1
dotted_line_flag = True
image_class = 'pdf-image'
image_flag = True
image_resolution = 300
log_level = 30
logo_flag = True
main_frame_tolerance = None
max_columns = 3
max_double_line_tolerance = 3
min_column_region_objects = 1
min_double_line_tolerance = Decimal('0.05')
min_image_size = 80
mini_page_class = 'pdf-mini-page'
multiple_columns_flag = True
page_class = 'pdf-page'
page_num_left_fraction = Decimal('0.44')
page_num_right_fraction = Decimal('0.56')
page_num_top_fraction = Decimal('0.75')
paragraph_class = 'pdf-paragraph'
paragraph_flag = True
pdf_class = 'pdf-content'
resolution = 144
skip_empty_table = False
snap_flag = False
span_class = 'pdf-span'
table_cell_merge_tolerance = 5
table_class = 'pdf-table'
table_flag = True
temp_dir_prefix = 'temp_depdf'
to_dict
unique_prefix = None
update(**kwargs)
verbose_flag = False
vertical_double_line_tolerance = Decimal('2')
x_tolerance = None
y_tolerance = None
class depdf.DePDF(pdf, config=None, **kwargs)

Bases: depdf.base.Base

close()
config
extract_html_pages()
generate_pages()
get_prefix()
html
html_pages
classmethod load(file_name, config=None, **kwargs)
classmethod open(*args, **kwargs)
page_num
pages
pdf
refresh()
same
save_html()
to_html
class depdf.DePage(page, pid='1', same=None, logo=None, config=None, columns=1, mini=False)

Bases: depdf.base.Base

analyze_lines()
analyze_main_frame()
analyze_page_attributes()
analyze_paragraph_border()
ave_cs = 0
ave_lh = 3
border = (0, 0, 0, 0)
chars
check_if_toc_page()
check_multi_column_page()
config
debug = False
extract_images()
extract_paragraph()
extract_phrases()
extract_tables()
frame_bottom = 0
frame_top = 0
h_edges = []
height
html
images
images_raw
min_cs = 0
new_para_end_flag = None
new_para_start_flag = None
object_key_list = ['_tables', '_paragraphs', '_images']
objects
orientation = ''
page
pagination_phrases = []
paragraphs
phrases = None
pid
prefix = UUID('6f4176ec-ac20-4d25-a08e-ade1feaec45a')
process_mini_page()
process_page()
refresh()
save_html()
screenshot
set_global()
tables
tables_raw
temp_dir = 'temp'
to_html
to_screenshot()
toc_flag = False
v_edges = []
verbose = False
width
x_tolerance = 3
y_tolerance = 3
depdf.convert_pdf_to_html(pdf, **kwargs)
Parameters:
  • pdf – pdf file path
  • kwargs – config keyword arguments
Returns:

pdf html string
depdf.convert_page_to_html(pdf, pid, **kwargs)
Parameters:
  • pdf – pdf file path
  • pid – page number start from 1
  • kwargs – config keyword arguments
Returns:

page html string
depdf.extract_page_tables(pdf, pid, **kwargs)
Parameters:
  • pdf – pdf file path
  • pid – page number start from 1
  • kwargs – config keyword arguments
Returns:

page tables list
depdf.extract_page_paragraphs(pdf, pid, **kwargs)
Parameters:
  • pdf – pdf file path
  • pid – page number start from 1
  • kwargs – config keyword arguments
Returns:

page paragraphs list