#!/usr/bin/env python3
-import pypdf
+import pypdf
import argparse
import io
from reportlab.lib.pagesizes import A4
a4_width, a4_height = A4
+points_per_cm = 10 * 72 / 25.4
+cut_depth = 1.95 * points_per_cm
+cut_width = 1.05 * points_per_cm
+middle_point_depth = 0.4 * points_per_cm
+spine_limit = 1 * points_per_cm
parser = argparse.ArgumentParser(description="build print-ready book PDF")
-parser.add_argument("-i", "--input", dest="input_file", required=True, help="input PDF file")
-parser.add_argument("-o", "--output", dest="output_file", required=True, help="output PDF file")
-parser.add_argument("-p", "--pages", dest="page_range", help="page range, e.g., '3-end'")
-parser.add_argument("-c", "--crop", dest="crop_range", help="crops left, bottom, right, top – e.g., '10,10,10,10'")
-parser.add_argument("-n", "--nup4", dest="nup4", action='store_true', help="puts 4 input pages onto 1 output page")
-parser.add_argument("-a", "--analyze", dest="analyze", action="store_true", help="print lines identifying spine, page borders")
-parser.add_argument("-t", "--symmetry", dest="symmetry", action="store_true", help="alternate horizontal crops between odd and even pages")
-parser.add_argument("-s", "--second", dest="second", help="append second file as input to append")
-parser.add_argument("-r", "--rotate", dest="rotate", help="rotate page of number by 90°")
+parser.add_argument("-i", "--input_file", action="append", required=True, help="input PDF file")
+parser.add_argument("-o", "--output_file", required=True, help="output PDF file")
+parser.add_argument("-p", "--page_range", action="append", help="page range, e.g., '3-end'")
+parser.add_argument("-c", "--crop_range", action="append", help="cm crops left, bottom, right, top – e.g., '10,10,10,10'; prefix with ':'-delimited page range to limit effect")
+parser.add_argument("-s", "--symmetry", action="store_true", help="alternate horizontal crops between odd and even pages")
+parser.add_argument("-r", "--rotate", dest="rotate", type=int, action="append", help="rotate page of number by 90° (usable multiple times on same page!)")
+parser.add_argument("-n", "--nup4", action='store_true', help="puts 4 input pages onto 1 output page")
+parser.add_argument("-a", "--analyze", action="store_true", help="in --nup4, print lines identifying spine, page borders")
+parser.add_argument("-m", "--margin", type=float, default=0.43, help="print margin for --nup4 in cm (default 0.43)")
args = parser.parse_args()
-with open(args.input_file, 'rb') as file:
- reader = pypdf.PdfReader(file)
- # determine page range
+# select pages from input files
+def parse_page_range(range_string, pages):
start_page = 0
- end_page = len(reader.pages)
- if args.page_range:
- start, end = args.page_range.split('-')
+ end_page = len(pages)
+ if range_string:
+ start, end = range_string.split('-')
if not (len(start) == 0 or start == "start"):
- start_page = int(start) - 1
+ start_page = int(start) - 1
if not (len(end) == 0 or end == "end"):
end_page = int(end)
- pages_to_add = []
- for page_num in range(start_page, end_page):
- page = reader.pages[page_num]
+ return start_page, end_page
+pages_to_add = []
+opened_files = []
+new_page_num = 0
+for i, input_file in enumerate(args.input_file):
+ file = open(input_file, 'rb')
+ opened_files += [file]
+ reader = pypdf.PdfReader(file)
+ range_string = None
+ if args.page_range and len(args.page_range) > i:
+ range_string = args.page_range[i]
+ start_page, end_page = parse_page_range(range_string, reader.pages)
+ for old_page_num in range(start_page, end_page):
+ new_page_num += 1
+ page = reader.pages[old_page_num]
pages_to_add += [page]
- print("read in page number", page_num+1)
+ print("-i, -p: read in %s page number %d as new page %d" % (input_file, old_page_num+1, new_page_num))
- # add pages of second PDF
- if args.second:
- file2 = open(args.second, 'rb')
- reader2 = pypdf.PdfReader(file2)
- page_num = 1
- for page in reader2.pages:
- pages_to_add += [page]
- print("read second PDF's page number", page_num)
- page_num += 1
+# if necessary, pad pages to multiple of 8
+if args.nup4:
+ mod_to_8 = len(pages_to_add) % 8
+ if mod_to_8 > 0:
+ print("-n: number of input pages %d not multiple of 8, padding to that" % len(pages_to_add))
+ for _ in range(8 - mod_to_8):
+ new_page = pypdf.PageObject.create_blank_page(width=a4_width, height=a4_height)
+ pages_to_add += [new_page]
- # rotate page canvas
- if args.rotate:
- page = pages_to_add[int(args.rotate) - 1]
+# rotate page canvas
+if args.rotate:
+ for rotate in args.rotate:
+ page = pages_to_add[rotate - 1]
page.add_transformation(pypdf.Transformation().translate(tx=-a4_width/2, ty=-a4_height/2))
page.add_transformation(pypdf.Transformation().rotate(-90))
page.add_transformation(pypdf.Transformation().translate(tx=a4_width/2, ty=a4_height/2))
+ print("-r: rotating (by 90°) page", rotate)
- # normalize all pages to portrait A4
- for page in pages_to_add:
- if "/Rotate" in page:
- page.rotate(360 - page["/Rotate"])
- page.mediabox.left = 0
- page.mediabox.bottom = 0
- page.mediabox.top = a4_height
- page.mediabox.right = a4_width
- page.cropbox = page.mediabox
+# normalize all pages to portrait A4
+for page in pages_to_add:
+ if "/Rotate" in page:
+ page.rotate(360 - page["/Rotate"])
+ page.mediabox.left = 0
+ page.mediabox.bottom = 0
+ page.mediabox.top = a4_height
+ page.mediabox.right = a4_width
+ page.cropbox = page.mediabox
- # determine page crop
- crop_left, crop_bottom, crop_right, crop_top = 0, 0, 0, 0
- if args.crop_range:
- crop_left, crop_bottom, crop_right, crop_top = [float(x) for x in args.crop_range.split(',')]
- cropped_width = a4_width - crop_left - crop_right
- cropped_height = a4_height - crop_bottom - crop_top
- zoom = 1
- if args.crop_range:
- zoom_horizontal = a4_width / (a4_width - crop_left - crop_right)
- zoom_vertical = a4_height / (a4_height - crop_bottom - crop_top)
- if (zoom_horizontal > 1 and zoom_vertical < 1) or (zoom_horizontal < 1 and zoom_vertical > 1):
- print("Error: opposing zooms.")
- exit(1)
- elif zoom_horizontal + zoom_vertical > 2:
- zoom = min(zoom_horizontal, zoom_vertical)
- else:
- zoom = max(zoom_horizontal, zoom_vertical)
+# determine page crops, zooms, crop symmetry
+crops_at_page = [(0,0,0,0)]*len(pages_to_add)
+zoom_at_page = [1]*len(pages_to_add)
+if args.crop_range:
+ for crop_range in args.crop_range:
+ initial_split = crop_range.split(':')
+ if len(initial_split) > 1:
+ page_range = initial_split[0]
+ crops = initial_split[1]
+ else:
+ page_range = None
+ crops = initial_split[0]
+ start_page, end_page = parse_page_range(page_range, pages_to_add)
+ crop_left_cm, crop_bottom_cm, crop_right_cm, crop_top_cm = [float(x) for x in crops.split(',')]
+ crop_left = crop_left_cm * points_per_cm
+ crop_bottom = crop_bottom_cm * points_per_cm
+ crop_right = crop_right_cm * points_per_cm
+ crop_top = crop_top_cm * points_per_cm
+ if args.symmetry:
+ print("-c, -t: to pages %d to %d applying crops: left %.2fcm, bottom %.2fcm, right %.2fcm, top %.2fcm (but alternating left and right crop between even and odd pages)" % (start_page + 1, end_page, crop_left_cm, crop_bottom_cm, crop_right_cm, crop_top_cm))
+ else:
+ print("-c: to pages %d to %d applying crops: left %.2fcm, bottom %.2fcm, right %.2fcm, top %.2fcm" % (start_page + 1, end_page, crop_left_cm, crop_bottom_cm, crop_right_cm, crop_top_cm))
+ cropped_width = a4_width - crop_left - crop_right
+ cropped_height = a4_height - crop_bottom - crop_top
+ zoom = 1
+ zoom_horizontal = a4_width / (a4_width - crop_left - crop_right)
+ zoom_vertical = a4_height / (a4_height - crop_bottom - crop_top)
+ if (zoom_horizontal > 1 and zoom_vertical < 1) or (zoom_horizontal < 1 and zoom_vertical > 1):
+ print("Error: opposing zooms.")
+ exit(1)
+ elif zoom_horizontal + zoom_vertical > 2:
+ zoom = min(zoom_horizontal, zoom_vertical)
+ else:
+ zoom = max(zoom_horizontal, zoom_vertical)
+ for page_num in range(start_page, end_page):
+ if args.symmetry and page_num % 2:
+ crops_at_page[page_num] = (crop_right, crop_bottom, crop_left, crop_top)
+ else:
+ crops_at_page[page_num] = (crop_left, crop_bottom, crop_right, crop_top)
+ zoom_at_page[page_num] = zoom
- writer = pypdf.PdfWriter()
- if not args.nup4:
- odd_page = True
- for page in pages_to_add:
- if args.symmetry and odd_page:
- page.add_transformation(pypdf.Transformation().translate(tx=-crop_right, ty=-crop_bottom))
- else:
- page.add_transformation(pypdf.Transformation().translate(tx=-crop_left, ty=-crop_bottom))
- page.add_transformation(pypdf.Transformation().scale(zoom, zoom))
- page.mediabox.right = cropped_width * zoom
- page.mediabox.top = cropped_height * zoom
- writer.add_page(page)
- odd_page = not odd_page
- else:
- n_pages_per_axis = 2
- points_per_mm = 2.83465
- printable_margin = 4.3 * points_per_mm
- printable_scale = (a4_width - 2*printable_margin)/a4_width
- spine_limit = 10 * points_per_mm
- half_width = a4_width / n_pages_per_axis
- half_height = a4_height / n_pages_per_axis
- section_scale_factor = 1 / n_pages_per_axis
- spine_part_of_page = (spine_limit / half_width) / printable_scale
- bonus_shrink_factor = 1 - spine_part_of_page
- new_page_order = []
- eight_pack = []
- mod_to_8 = len(pages_to_add) % 8
- if mod_to_8 > 0:
- for _ in range(8 - mod_to_8):
- new_page = pypdf.PageObject.create_blank_page(width=a4_width, height=a4_height)
- pages_to_add += [new_page]
- i = 0
- for page in pages_to_add:
- if i == 0:
- eight_pack = []
- eight_pack += [page]
- i += 1
- if i == 8:
- i = 0
- new_page_order += [eight_pack[3]] # page front, upper left
- new_page_order += [eight_pack[0]] # page front, upper right
- new_page_order += [eight_pack[7]] # page front, lower left
- new_page_order += [eight_pack[4]] # page front, lower right
- new_page_order += [eight_pack[1]] # page back, upper left
- new_page_order += [eight_pack[2]] # page back, upper right
- new_page_order += [eight_pack[5]] # page back, lower left
- new_page_order += [eight_pack[6]] # page back, lower right
- i = 0
- page_count = 0
- front_page = True
- for page in new_page_order:
- if i == 0:
- new_page = pypdf.PageObject.create_blank_page(width=a4_width, height=a4_height)
+writer = pypdf.PdfWriter()
+if not args.nup4:
+ # single-page output
+ print("building 1-input-page-per-output-page book")
+ odd_page = True
+ for i, page in enumerate(pages_to_add):
+ crop_left, crop_bottom, crop_right, crop_top = crops_at_page[i]
+ zoom = zoom_at_page[i]
+ page.add_transformation(pypdf.Transformation().translate(tx=-crop_left, ty=-crop_bottom))
+ page.add_transformation(pypdf.Transformation().scale(zoom, zoom))
+ cropped_width = a4_width - crop_left - crop_right
+ cropped_height = a4_height - crop_bottom - crop_top
+ page.mediabox.right = cropped_width * zoom
+ page.mediabox.top = cropped_height * zoom
+ writer.add_page(page)
+ odd_page = not odd_page
+ print("built page number %d (of %d)" % (i+1, len(pages_to_add)))
+
+else:
+ print("-n: building 4-input-pages-per-output-page book")
+ print("-m: applying printable-area margin of %.2fcm" % args.margin)
+ if args.analyze:
+ print("-a: drawing page borders, spine limits")
+ n_pages_per_axis = 2
+ printable_margin = args.margin * points_per_cm
+ printable_scale = (a4_width - 2*printable_margin)/a4_width
+ half_width = a4_width / n_pages_per_axis
+ half_height = a4_height / n_pages_per_axis
+ section_scale_factor = 1 / n_pages_per_axis
+ spine_part_of_page = (spine_limit / half_width) / printable_scale
+ bonus_shrink_factor = 1 - spine_part_of_page
+ new_page_order = []
+ new_i_order = []
+ eight_pack = []
+ i = 0
+ n_eights = 0
+ for page in pages_to_add:
+ if i == 0:
+ eight_pack = []
+ eight_pack += [page]
+ i += 1
+ if i == 8:
+ i = 0
+ new_i_order += [8 * n_eights + 3,
+ 8 * n_eights + 0,
+ 8 * n_eights + 7,
+ 8 * n_eights + 4,
+ 8 * n_eights + 1,
+ 8 * n_eights + 2,
+ 8 * n_eights + 5,
+ 8 * n_eights + 6]
+ n_eights += 1
+ new_page_order += [eight_pack[3]] # page front, upper left
+ new_page_order += [eight_pack[0]] # page front, upper right
+ new_page_order += [eight_pack[7]] # page front, lower left
+ new_page_order += [eight_pack[4]] # page front, lower right
+ new_page_order += [eight_pack[1]] # page back, upper left
+ new_page_order += [eight_pack[2]] # page back, upper right
+ new_page_order += [eight_pack[5]] # page back, lower left
+ new_page_order += [eight_pack[6]] # page back, lower right
+ i = 0
+ page_count = 0
+ front_page = True
+ for j, page in enumerate(new_page_order):
+ if i == 0:
+ new_page = pypdf.PageObject.create_blank_page(width=a4_width, height=a4_height)
- # in-section transformations: align pages on top, left-hand pages to left, right-hand to right
- page.add_transformation(pypdf.Transformation().translate(ty=(a4_height / zoom - (a4_height - crop_top))))
- if i == 0 or i == 2:
- if args.symmetry:
- page.add_transformation(pypdf.Transformation().translate(tx=-crop_left))
- else:
- page.add_transformation(pypdf.Transformation().translate(tx=-crop_right))
- elif i == 1 or i == 3:
- if args.symmetry:
- page.add_transformation(pypdf.Transformation().translate(tx=(a4_width / zoom - (a4_width - crop_left))))
- else:
- page.add_transformation(pypdf.Transformation().translate(tx=(a4_width / zoom - (a4_width - crop_right))))
- page.add_transformation(pypdf.Transformation().scale(zoom * bonus_shrink_factor, zoom * bonus_shrink_factor))
- if i == 2 or i == 3:
- page.add_transformation(pypdf.Transformation().translate(ty=-2*printable_margin/printable_scale))
+ # in-section transformations: align pages on top, left-hand pages to left, right-hand to right
+ new_i = new_i_order[j]
+ crop_left, crop_bottom, crop_right, crop_top = crops_at_page[new_i]
+ zoom = zoom_at_page[new_i]
+ page.add_transformation(pypdf.Transformation().translate(ty=(a4_height / zoom - (a4_height - crop_top))))
+ if i == 0 or i == 2:
+ page.add_transformation(pypdf.Transformation().translate(tx=-crop_left))
+ elif i == 1 or i == 3:
+ page.add_transformation(pypdf.Transformation().translate(tx=(a4_width / zoom - (a4_width - crop_right))))
+ page.add_transformation(pypdf.Transformation().scale(zoom * bonus_shrink_factor, zoom * bonus_shrink_factor))
+ if i == 2 or i == 3:
+ page.add_transformation(pypdf.Transformation().translate(ty=-2*printable_margin/printable_scale))
- # outer section transformations
- page.add_transformation(pypdf.Transformation().translate(ty=(1-bonus_shrink_factor)*a4_height))
- if i == 0 or i == 1:
- y_section = a4_height
- page.mediabox.bottom = half_height
- page.mediabox.top = a4_height
- if i == 2 or i == 3:
- y_section = 0
- page.mediabox.bottom = 0
- page.mediabox.top = half_height
- if i == 0 or i == 2:
- x_section = 0
- page.mediabox.left = 0
- page.mediabox.right = half_width
- if i == 1 or i == 3:
- page.add_transformation(pypdf.Transformation().translate(tx=(1-bonus_shrink_factor)*a4_width))
- x_section = a4_width
- page.mediabox.left = half_width
- page.mediabox.right = a4_width
- page.add_transformation(pypdf.Transformation().translate(tx=x_section, ty=y_section))
- page.add_transformation(pypdf.Transformation().scale(section_scale_factor, section_scale_factor))
- new_page.merge_page(page)
- page_count += 1
- print("merged page number", page_count)
- i += 1
- if i > 3:
- from reportlab.pdfgen import canvas
- if args.analyze:
- # borders
- packet = io.BytesIO()
- c = canvas.Canvas(packet, pagesize=A4)
- c.setLineWidth(0.1)
- c.line(0, a4_height, a4_width, a4_height)
- c.line(0, half_height, a4_width, half_height)
- c.line(0, 0, a4_width, 0)
- c.line(0, a4_height, 0, 0)
- c.line(half_width, a4_height, half_width, 0)
- c.line(a4_width, a4_height, a4_width, 0)
- c.save()
- new_pdf = pypdf.PdfReader(packet)
- new_page.merge_page(new_pdf.pages[0])
- printable_offset_x = printable_margin
- printable_offset_y = printable_margin * a4_height / a4_width
- new_page.add_transformation(pypdf.Transformation().scale(printable_scale, printable_scale))
- new_page.add_transformation(pypdf.Transformation().translate(tx=printable_offset_x, ty=printable_offset_y))
- x_left_spine_limit = half_width * bonus_shrink_factor
- x_right_spine_limit = a4_width - x_left_spine_limit
- if args.analyze or front_page:
- packet = io.BytesIO()
- c = canvas.Canvas(packet, pagesize=A4)
- if args.analyze:
- # # spine lines
- c.setLineWidth(0.1)
- c.line(x_left_spine_limit, a4_height, x_left_spine_limit, 0)
- c.line(x_right_spine_limit, a4_height, x_right_spine_limit, 0)
- if front_page:
- c.setLineWidth(0.2)
- cut_depth = 19.5 * points_per_mm
- cut_width = 10.5 * points_per_mm
- middle_point_depth = 4 * points_per_mm
+ # outer section transformations
+ page.add_transformation(pypdf.Transformation().translate(ty=(1-bonus_shrink_factor)*a4_height))
+ if i == 0 or i == 1:
+ y_section = a4_height
+ page.mediabox.bottom = half_height
+ page.mediabox.top = a4_height
+ if i == 2 or i == 3:
+ y_section = 0
+ page.mediabox.bottom = 0
+ page.mediabox.top = half_height
+ if i == 0 or i == 2:
+ x_section = 0
+ page.mediabox.left = 0
+ page.mediabox.right = half_width
+ if i == 1 or i == 3:
+ page.add_transformation(pypdf.Transformation().translate(tx=(1-bonus_shrink_factor)*a4_width))
+ x_section = a4_width
+ page.mediabox.left = half_width
+ page.mediabox.right = a4_width
+ page.add_transformation(pypdf.Transformation().translate(tx=x_section, ty=y_section))
+ page.add_transformation(pypdf.Transformation().scale(section_scale_factor, section_scale_factor))
+ new_page.merge_page(page)
+ page_count += 1
+ print("merged page number %d (of %d)" % (page_count, len(pages_to_add)))
+ i += 1
+ if i > 3:
+ from reportlab.pdfgen import canvas
+ if args.analyze:
+ # borders
+ packet = io.BytesIO()
+ c = canvas.Canvas(packet, pagesize=A4)
+ c.setLineWidth(0.1)
+ c.line(0, a4_height, a4_width, a4_height)
+ c.line(0, half_height, a4_width, half_height)
+ c.line(0, 0, a4_width, 0)
+ c.line(0, a4_height, 0, 0)
+ c.line(half_width, a4_height, half_width, 0)
+ c.line(a4_width, a4_height, a4_width, 0)
+ c.save()
+ new_pdf = pypdf.PdfReader(packet)
+ new_page.merge_page(new_pdf.pages[0])
+ printable_offset_x = printable_margin
+ printable_offset_y = printable_margin * a4_height / a4_width
+ new_page.add_transformation(pypdf.Transformation().scale(printable_scale, printable_scale))
+ new_page.add_transformation(pypdf.Transformation().translate(tx=printable_offset_x, ty=printable_offset_y))
+ x_left_spine_limit = half_width * bonus_shrink_factor
+ x_right_spine_limit = a4_width - x_left_spine_limit
+ if args.analyze or front_page:
+ packet = io.BytesIO()
+ c = canvas.Canvas(packet, pagesize=A4)
+ if args.analyze:
+ # # spine lines
+ c.setLineWidth(0.1)
+ c.line(x_left_spine_limit, a4_height, x_left_spine_limit, 0)
+ c.line(x_right_spine_limit, a4_height, x_right_spine_limit, 0)
+ if front_page:
+ c.setLineWidth(0.2)
- start_up_left_left_x = x_left_spine_limit - 0.5 * cut_width
- start_up_left_right_x = x_left_spine_limit + 0.5 * cut_width
- middle_point_up_left_y = half_height + middle_point_depth
- end_point_up_left_y = half_height + cut_depth
- c.line(start_up_left_right_x, half_height, x_left_spine_limit, end_point_up_left_y)
- c.line(x_left_spine_limit, end_point_up_left_y, x_left_spine_limit, middle_point_up_left_y)
- c.line(x_left_spine_limit, middle_point_up_left_y, start_up_left_left_x, half_height)
+ start_up_left_left_x = x_left_spine_limit - 0.5 * cut_width
+ start_up_left_right_x = x_left_spine_limit + 0.5 * cut_width
+ middle_point_up_left_y = half_height + middle_point_depth
+ end_point_up_left_y = half_height + cut_depth
+ c.line(start_up_left_right_x, half_height, x_left_spine_limit, end_point_up_left_y)
+ c.line(x_left_spine_limit, end_point_up_left_y, x_left_spine_limit, middle_point_up_left_y)
+ c.line(x_left_spine_limit, middle_point_up_left_y, start_up_left_left_x, half_height)
- start_down_right_left_x = x_right_spine_limit - 0.5 * cut_width
- start_down_right_right_x = x_right_spine_limit + 0.5 * cut_width
- middle_point_down_right_y = half_height - middle_point_depth
- end_point_down_right_y = half_height - cut_depth
- c.line(start_down_right_left_x, half_height, x_right_spine_limit, end_point_down_right_y)
- c.line(x_right_spine_limit, end_point_down_right_y, x_right_spine_limit, middle_point_down_right_y)
- c.line(x_right_spine_limit, middle_point_down_right_y, start_down_right_right_x, half_height)
+ start_down_right_left_x = x_right_spine_limit - 0.5 * cut_width
+ start_down_right_right_x = x_right_spine_limit + 0.5 * cut_width
+ middle_point_down_right_y = half_height - middle_point_depth
+ end_point_down_right_y = half_height - cut_depth
+ c.line(start_down_right_left_x, half_height, x_right_spine_limit, end_point_down_right_y)
+ c.line(x_right_spine_limit, end_point_down_right_y, x_right_spine_limit, middle_point_down_right_y)
+ c.line(x_right_spine_limit, middle_point_down_right_y, start_down_right_right_x, half_height)
- if args.analyze or front_page:
- c.save()
- # packet.seek(0)
- new_pdf = pypdf.PdfReader(packet)
- new_page.merge_page(new_pdf.pages[0])
- writer.add_page(new_page)
- i = 0
- front_page = not front_page
+ if args.analyze or front_page:
+ c.save()
+ new_pdf = pypdf.PdfReader(packet)
+ new_page.merge_page(new_pdf.pages[0])
+ writer.add_page(new_page)
+ i = 0
+ front_page = not front_page
- with open(args.output_file, 'wb') as output_file:
- writer.write(output_file)
- if args.second:
- file2.close()
+# write and close
+for file in opened_files:
+ file.close()
+with open(args.output_file, 'wb') as output_file:
+ writer.write(output_file)
projects / misc / blobdiff