def trace( argv):
password = ''
layout_w = mupdf.FZ_DEFAULT_LAYOUT_W
layout_h = mupdf.FZ_DEFAULT_LAYOUT_H
layout_em = mupdf.FZ_DEFAULT_LAYOUT_EM
layout_css = None
layout_use_doc_css = 1
use_display_list = 0
argv_i = 0
while 1:
arg = argv[ argv_i]
if arg == '-p':
password = next( opt)
elif arg == '-W':
argv_i += 1
layout_w = float( argv[argv_i])
elif arg == '-H':
argv_i += 1
layout_h = float( argv[argv_i])
elif arg == '-S':
argv_i += 1
layout_em = float( argv[argv_i])
elif arg == '-U':
argv_i += 1
layout_css = argv[argv_i]
elif arg == '-X':
layout_use_doc_css = 0
elif arg == '-d':
use_display_list = 1
else:
break
argv_i += 1
if argv_i == len( argv):
trace_usage()
if layout_css:
buffer_ = mupdf.Buffer( layout_css)
mupdf.mupdf_set_user_css( buffer_.string_from_buffer())
mupdf.set_use_document_css( layout_use_doc_css)
for argv_i in range( argv_i, len( argv)):
arg = argv[ argv_i]
doc = mupdf.Document( arg)
if doc.needs_password():
doc.authenticate_password( password)
doc.layout_document( layout_w, layout_h, layout_em)
print( f'<document filename="{arg}">')
count = doc.count_pages()
if argv_i + 1 < len( argv) and mupdf.is_page_range( argv[ argv_i+1]):
argv_i += 1
trace_runrange( use_display_list, doc, count, argv[ argv_i])
else:
trace_runrange( use_display_list, doc, count, '1-N')
print( '<document>')
def open_path(self, path):
path = os.path.abspath(path)
try:
self.document = mupdf.Document(path)
except Exception as e:
print(f'Failed to open path={path!r}: {e}')
return
self.setWindowTitle(path)
self.goto_page(page_number=0, zoom=0)
def convert( argv):
# input options
password = ''
alphabits = 8
layout_w = mupdf.FZ_DEFAULT_LAYOUT_W
layout_h = mupdf.FZ_DEFAULT_LAYOUT_H
layout_em = mupdf.FZ_DEFAULT_LAYOUT_EM
layout_css = None
layout_use_doc_css = 1
# output options
output = None
format_ = None
options = ''
items, argv = getopt.getopt( argv, 'p:A:W:H:S:U:Xo:F:O:')
for option, value in items:
if 0: pass
elif option == '-p': password = value
elif option == '-A': alphabits = int(value)
elif option == '-W': layout_w = float( value)
elif option == '-H': layout_h = float( value)
elif option == '-S': layout_em = float( value)
elif option == '-U': layout_css = value
elif option == '-X': layout_use_doc_css = 0
elif option == '-o': output = value
elif option == '-F': format_ = value
elif option == '-O': options = value
else: assert 0
if not argv or (not format_ and not output):
convert_usage()
mupdf.set_aa_level( alphabits)
if layout_css:
buf = mupdf.Buffer( layout_css)
mupdf.set_user_css( buf.string_from_buffer())
mupdf.set_use_document_css(layout_use_doc_css)
if format_:
out = mupdf.DocumentWriter( output, format_, options)
else:
out = mupdf.DocumentWriter( output, options)
i = 0
while 1:
if i >= len( argv):
break
arg = argv[i]
doc = mupdf.Document( arg)
if doc.needs_password():
if not doc.authenticate_password( password):
raise Exception( f'cannot authenticate password: {arg}')
doc.layout_document( layout_w, layout_h, layout_em)
count = doc.count_pages()
range_ = '1-N'
if i + 1 < len(argv) and mupdf.is_page_range(ctx, argv[i+1]):
i += 1
range_ = argv[i]
convert_runrange( doc, count, range_, out)
i += 1
out.close_document_writer()
#! /usr/bin/env python3
import mupdf
import os
import sys
path = sys.argv[1]
document = mupdf.Document(path)
if document.pdf_specifics().m_internal:
raise Exception('document is PDF already')
path_out = f'{path}.pdf'
print(f'Converting {path!r} to {path_out!r}')
def convert_to_pdf(document, page_from=0, page_to=-1, rotate=0):
num_pages = document.count_pages()
page_from = max(page_from, 0)
page_from = min(page_from, num_pages - 1)
if page_to < 0:
page_to = num_pages - 1
page_to = min(page_to, num_pages - 1)
page_delta = 1
if page_to < page_from:
p = page_to
page_to = page_from
page_from = p
page_delta = -1
def test(path):
'''
Runs various mupdf operations on <path>, which is assumed to be a file that
mupdf can open.
'''
log(f'testing path={path}')
assert os.path.isfile(path)
global g_test_n
g_test_n += 1
# See notes in mupdfwrap.py:build_swig() about buffer_extract() and
# buffer_storage().
#
assert getattr(mupdf.Buffer, 'buffer_storage_raw')
assert getattr(mupdf.Buffer, 'buffer_storage', None) is None
assert getattr(mupdf.Buffer, 'buffer_extract_raw')
assert getattr(mupdf.Buffer, 'buffer_extract')
# Test operations using functions:
#
log('Testing functions.')
log(f' Opening: %s' % path)
document = mupdf.open_document(path)
log(f' mupdf.needs_password(document)={mupdf.needs_password(document)}')
log(f' mupdf.count_pages(document)={mupdf.count_pages(document)}')
log(f' mupdf.document_output_intent(document)={mupdf.document_output_intent(document)}'
)
# Test operations using classes:
#
log(f'Testing classes')
document = mupdf.Document(path)
log(f'Have created mupdf.Document for {path}')
log(f'document.needs_password()={document.needs_password()}')
log(f'document.count_pages()={document.count_pages()}')
if 0:
log(f'stext info:')
show_stext(document)
for k in (
'format',
'encryption',
'info:Author',
'info:Title',
'info:Creator',
'info:Producer',
'qwerty',
):
v = document.lookup_metadata(k)
log(f'document.lookup_metadata() k={k} returned v={v!r}')
if k == 'qwerty':
assert v is None, f'v={v!r}'
else:
pass
zoom = 10
scale = mupdf.Matrix.scale(zoom / 100., zoom / 100.)
page_number = 0
log(f'Have created scale: a={scale.a} b={scale.b} c={scale.c} d={scale.d} e={scale.e} f={scale.f}'
)
colorspace = mupdf.Colorspace(mupdf.Colorspace.Fixed_RGB)
log(f'{colorspace.m_internal.key_storable.storable.refs}')
if 0:
c = colorspace.clamp_color([3.14])
log('colorspace.clamp_color returned c={c}')
pixmap = mupdf.Pixmap(document, page_number, scale, colorspace, 0)
log(f'Have created pixmap: {pixmap.m_internal.w} {pixmap.m_internal.h} {pixmap.m_internal.stride} {pixmap.m_internal.n}'
)
filename = f'mupdf_test-out1-{g_test_n}.png'
pixmap.save_pixmap_as_png(filename)
log(f'Have created {filename} using pixmap.save_pixmap_as_png().')
# Print image data in ascii PPM format. Copied from
# mupdf/docs/examples/example.c.
#
samples = pixmap.samples()
stride = pixmap.stride()
n = pixmap.n()
filename = f'mupdf_test-out2-{g_test_n}.ppm'
with open(filename, 'w') as f:
f.write('P3\n')
f.write('%s %s\n' % (pixmap.m_internal.w, pixmap.m_internal.h))
f.write('255\n')
for y in range(0, pixmap.m_internal.h):
for x in range(pixmap.m_internal.w):
if x:
f.write(' ')
offset = y * stride + x * n
if hasattr(mupdf, 'bytes_getitem'):
# swig
f.write('%3d %3d %3d' % (
mupdf.bytes_getitem(samples, offset + 0),
mupdf.bytes_getitem(samples, offset + 1),
mupdf.bytes_getitem(samples, offset + 2),
))
else:
# cppyy
f.write('%3d %3d %3d' % (
samples[offset + 0],
samples[offset + 1],
samples[offset + 2],
))
f.write('\n')
log(f'Have created {filename} by scanning pixmap.')
# Generate .png and but create Pixmap from Page instead of from Document.
#
page = mupdf.Page(document, 0)
separations = page.page_separations()
log(f'page_separations() returned {"true" if separations else "false"}')
pixmap = mupdf.Pixmap(page, scale, colorspace, 0)
filename = f'mupdf_test-out3-{g_test_n}.png'
pixmap.save_pixmap_as_png(filename)
log(f'Have created {filename} using pixmap.save_pixmap_as_png()')
# Show links
log(f'Links.')
page = mupdf.Page(document, 0)
link = mupdf.load_links(page.m_internal)
log(f'{link}')
if link:
for i in link:
log(f'{i}')
# Check we can iterate over Link's, by creating one manually.
#
link = mupdf.Link(mupdf.Rect(0, 0, 1, 1), "hello")
log(f'items in <link> are:')
for i in link:
log(f' {i.m_internal.refs} {i.m_internal.uri}')
# Check iteration over Outlines.
#
log(f'Outlines.')
outline = mupdf.Outline(document)
log(f'outline.m_internal={outline.m_internal}')
if outline.m_internal:
log(f'{outline.uri()} {outline.page()} {outline.x()} {outline.y()} {outline.is_open()} {outline.title()}'
)
log(f'items in outline tree are:')
for o in outline:
log(f' {o.uri()} {o.page()} {o.x()} {o.y()} {o.is_open()} {o.title()}'
)
# Check iteration over StextPage.
#
log(f'StextPage.')
stext_options = mupdf.StextOptions(0)
page_num = 40
try:
stext_page = mupdf.StextPage(document, page_num, stext_options)
except Exception:
log(f'no page_num={page_num}')
else:
device_stext = mupdf.Device(stext_page, stext_options)
matrix = mupdf.Matrix()
page = mupdf.Page(document, 0)
cookie = mupdf.Cookie()
page.run_page(device_stext, matrix, cookie)
log(f' stext_page is:')
for block in stext_page:
log(f' block:')
for line in block:
line_text = ''
for char in line:
line_text += chr(char.m_internal.c)
log(f' {line_text}')
device_stext.close_device()
# Check copy-constructor.
log(f'Checking copy-constructor')
document2 = mupdf.Document(document)
del document
page = mupdf.Page(document2, 0)
scale = mupdf.Matrix()
pixmap = mupdf.Pixmap(page, scale, colorspace, 0)
pixmap.save_pixmap_as_png('mupdf_test-out3.png')
stdout = mupdf.Output(mupdf.Output.Fixed_STDOUT)
log(f'{type(stdout)} {stdout.m_internal.state}')
mediabox = page.bound_page()
out = mupdf.DocumentWriter(filename, 'png', '',
mupdf.DocumentWriter.FormatPathType_DOCUMENT)
dev = out.begin_page(mediabox)
page.run_page(dev, mupdf.Matrix(mupdf.fz_identity), mupdf.Cookie())
out.end_page()
# Check out-params are converted into python return value.
bitmap = mupdf.Bitmap(10, 20, 8, 72, 72)
bitmap_details = bitmap.bitmap_details()
log(f'{bitmap_details}')
assert list(bitmap_details) == [10, 20, 8,
12], f'bitmap_details={bitmap_details!r}'
log(f'finished test of %s' % path)
def draw(argv):
password = ''
info = trace_info()
items, argv = getopt.getopt(
argv, 'qp:o:F:R:r:w:h:fB:c:e:G:Is:A:DiW:H:S:T:U:XLvPl:y:NO:a')
for option, value in items:
if 0: pass
elif option == '-q': state.quiet = 1
elif option == '-p': password = value
elif option == '-o': state.output = value
elif option == '-F': state.format_ = value
elif option == '-R': state.rotation = float(value)
elif option == '-r':
state.resolution = float(value)
state.res_specified = 1
elif option == '-w':
state.width = float(value)
elif option == '-h':
state.height = float(value)
elif option == '-f':
state.fit = 1
elif option == '-B':
state.band_height = int(value)
elif option == '-c':
state.out_cs = parse_colorspace(value)
elif option == '-e':
state.proof_filename = value
elif option == '-G':
state.gamma_value = float(value)
elif option == '-I':
state.invert += 1
elif option == '-W':
state.layout_w = float(value)
elif option == '-H':
state.layout_h = float(value)
elif option == '-S':
state.layout_em = float(value)
elif option == '-U':
state.layout_css = value
elif option == '-X':
state.layout_use_doc_css = 0
elif option == '-O':
state.spots = float(value)
if not mupdf.FZ_ENABLE_SPOT_RENDERING:
sys.stderr.write(
'Spot rendering/Overprint/Overprint simulation not enabled in this build\n'
)
state.spots = SPOTS_NONE
elif option == '-s':
if 't' in value: state.showtime += 1
if 'm' in value: state.showmemory += 1
if 'f' in value: state.showfeatures += 1
if '5' in value: state.showmd5 += 1
elif option == '-A':
state.alphabits_graphics = int(value)
sep = value.find('/')
if sep >= 0:
state.alphabits_text = int(value[sep + 1:])
else:
state.alphabits_text = state.alphabits_graphics
elif option == '-D':
state.uselist = 0
elif option == '-l':
state.min_line_width = float(value)
elif option == '-i':
state.ignore_errors = 1
elif option == '-N':
state.no_icc = 1
elif option == '-T':
state.num_workers = int(value)
elif option == '-L':
state.lowmemory = 1
elif option == '-P':
bgprint.active = 1
elif option == '-y':
state.layer_config = value
elif option == '-a':
state.useaccel = 0
elif option == '-v':
sys.stderr.write(f'mudraw version {mupdf.FZ_VERSION}\n')
if not argv:
usage()
if state.num_workers > 0:
if state.uselist == 0:
sys.stderr.write(
'cannot use multiple threads without using display list\n')
sys.exit(1)
if state.band_height == 0:
sys.stderr.write(
'Using multiple threads without banding is pointless\n')
if bgprint.active:
if state.uselist == 0:
sys.stderr.write('cannot bgprint without using display list\n')
sys.exit(1)
if state.proof_filename:
proof_buffer = mupdf.Buffer(state.proof_filename)
state.proof_cs = mupdf.Colorspace(FZ_COLORSPACE_NONE, 0, None,
proof_buffer)
mupdf.set_text_aa_level(state.alphabits_text)
mupdf.set_graphics_aa_level(state.alphabits_graphics)
mupdf.set_graphics_min_line_width(state.min_line_width)
if state.no_icc:
mupdf.disable_icc()
else:
mupdf.enable_icc()
if state.layout_css:
buf = mupdf.Buffer(state.layout_css)
mupdf.set_user_css(buf.string_from_buffer())
mupdf.set_use_document_css(state.layout_use_doc_css)
# Determine output type
if state.band_height < 0:
sys.stderr.write('Bandheight must be > 0\n')
sys.exit(1)
state.output_format = OUT_PNG
if state.format_:
for i in range(len(suffix_table)):
if state.format_ == suffix_table[i].suffix[1:]:
state.output_format = suffix_table[i].format
if state.spots == SPOTS_FULL and suffix_table[i].spots == 0:
sys.stderr.write(
f'Output format {suffix_table[i].suffix[1:]} does not support spot rendering.\nDoing overprint simulation instead.\n'
)
state.spots = SPOTS_OVERPRINT_SIM
break
else:
sys.stderr.write(f'Unknown output format {format}\n')
sys.exit(1)
elif state.output:
suffix = state.output
i = 0
while 1:
if i == len(suffix_table):
break
s = suffix.find(suffix_table[i].suffix)
if s != -1:
suffix = suffix_table[i].suffix[s + 1:]
state.output_format = suffix_table[i].format
if state.spots == SPOTS_FULL and suffix_table[i].spots == 0:
sys.stderr.write(
'Output format {suffix_table[i].suffix[1:]} does not support spot rendering\nDoing overprint simulation instead.\n'
)
state.spots = SPOTS_OVERPRINT_SIM
i = 0
else:
i += 1
if state.band_height:
if state.output_format not in (OUT_PAM, OUT_PGM, OUT_PPM, OUT_PNM,
OUT_PNG, OUT_PBM, OUT_PKM, OUT_PCL,
OUT_PCLM, OUT_PS, OUT_PSD):
sys.stderr.write(
'Banded operation only possible with PxM, PCL, PCLM, PS, PSD, and PNG outputs\n'
)
sys.exit(1)
if state.showmd5:
sys.stderr.write('Banded operation not compatible with MD5\n')
sys.exit(1)
for i in range(len(format_cs_table)):
if format_cs_table[i].format == state.output_format:
if state.out_cs == CS_UNSET:
state.out_cs = format_cs_table[i].default_cs
for j in range(len(format_cs_table[i].permitted_cs)):
if format_cs_table[i].permitted_cs[j] == state.out_cs:
break
else:
sys.stderr.write('Unsupported colorspace for this format\n')
sys.exit(1)
state.alpha = 1
if state.out_cs in (CS_MONO, CS_GRAY, CS_GRAY_ALPHA):
state.colorspace = mupdf.Colorspace(mupdf.Colorspace.Fixed_GRAY)
state.alpha = (state.out_cs == CS_GRAY_ALPHA)
elif state.out_cs in (CS_RGB, CS_RGB_ALPHA):
state.colorspace = mupdf.Colorspace(mupdf.Colorspace.Fixed_RGB)
state.alpha = (state.out_cs == CS_RGB_ALPHA)
elif state.out_cs in (CS_CMYK, CS_CMYK_ALPHA):
state.colorspace = mupdf.Colorspace(mupdf.Colorspace.Fixed_CMYK)
state.alpha = (state.out_cs == CS_CMYK_ALPHA)
elif state.out_cs == CS_ICC:
try:
icc_buffer = mupdf.Buffer(state.icc_filename)
state.colorspace = Colorspace(mupdf.FZ_COLORSPACE_NONE, 0, None,
icc_buffer)
except Exception as e:
sys.stderr.write('Invalid ICC destination color space\n')
sys.exit(1)
if state.colorspace.m_internal is None:
sys.stderr.write('Invalid ICC destination color space\n')
sys.exit(1)
state.alpha = 0
else:
sys.stderr.write('Unknown colorspace!\n')
sys.exit(1)
if state.out_cs != CS_ICC:
state.colorspace = mupdf.Colorspace(state.colorspace)
else:
# Check to make sure this icc profile is ok with the output format */
okay = 0
for i in range(len(format_cs_table)):
if format_cs_table[i].format == state.output_format:
for j in range(len(format_cs_table[i].permitted_cs)):
x = format_cs_table[i].permitted_cs[j]
if x in (CS_MONO, CS_GRAY, CS_GRAY_ALPHA):
if state.colorspace.colorspace_is_gray():
okay = 1
elif x in (CS_RGB, CS_RGB_ALPHA):
if state.colorspace.colorspace_is_rgb():
okay = 1
elif x in (CS_CMYK, CS_CMYK_ALPHA):
if state.colorspace.colorspace_is_cmyk():
okay = 1
if not okay:
sys.stderr.write(
'ICC profile uses a colorspace that cannot be used for this format\n'
)
sys.exit(1)
if state.output_format == OUT_SVG:
# SVG files are always opened for each page. Do not open "output".
pass
elif state.output and (not state.output.startswith('-') or
len(state.output) >= 2) and len(state.output) >= 1:
if has_percent_d(state.output):
state.output_file_per_page = 1
else:
state.out = mupdf.Output(state.output, 0)
else:
state.quiet = 1 # automatically be quiet if printing to stdout
if 0:
# Windows specific code to make stdout binary.
if state.output_format not in (OUT_TEXT, OUT_STEXT, OUT_HTML,
OUT_XHTML, OUT_TRACE, OUT_XMLTEXT):
setmode(fileno(stdout), O_BINARY)
state.out = mupdf.Output(mupdf.Output.Fixed_STDOUT)
state.filename = argv[0]
if not state.output_file_per_page:
file_level_headers()
timing.count = 0
timing.total = 0
timing.min = 1 << 30
timing.max = 0
timing.mininterp = 1 << 30
timing.maxinterp = 0
timing.minpage = 0
timing.maxpage = 0
timing.minfilename = ""
timing.maxfilename = ""
timing.layout = 0
timing.minlayout = 1 << 30
timing.maxlayout = 0
timing.minlayoutfilename = ""
timing.maxlayoutfilename = ""
if state.showtime and bgprint.active:
timing.total = gettime()
fz_optind = 0
try:
while fz_optind < len(argv):
try:
accel = None
state.filename = argv[fz_optind]
fz_optind += 1
state.files += 1
if not state.useaccel:
accel = None
# If there was an accelerator to load, what would it be called?
else:
accelpath = get_accelerator_filename(state.filename)
# Check whether that file exists, and isn't older than
# the document.
atime = stat_mtime(accelpath)
dtime = stat_mtime(state.filename)
if atime == 0:
# No accelerator
pass
elif atime > dtime:
accel = accelpath
else:
# Accelerator data is out of date
os.unlink(accelpath)
accel = None # In case we have jumped up from below
# Unfortunately if accel=None, SWIG doesn't seem to think of it
# as a char*, so we end up in fz_open_document_with_stream().
#
# If we try to avoid this by setting accel='', SWIG correctly
# calls Document(const char *filename, const char *accel) =>
# fz_open_accelerated_document(), but the latter function tests
# for NULL not "" so fails.
#
# So we choose the constructor explicitly rather than leaving
# it up to SWIG.
#
if accel:
doc = mupdf.Document(state.filename, accel)
else:
doc = mupdf.Document(state.filename)
if doc.needs_password():
if not doc.authenticate_password(password):
raise Exception(
f'cannot authenticate password: {state.filename}')
# Once document is open check for output intent colorspace
state.oi = doc.document_output_intent()
if state.oi.m_internal:
# See if we had explicitly set a profile to render
if state.out_cs != CS_ICC:
# In this case, we want to render to the output intent
# color space if the number of channels is the same
if state.oi.colorspace_n(
) == state.colorspace.colorspace_n():
state.colorspace = state.oi
layouttime = time.time()
doc.layout_document(state.layout_w, state.layout_h,
state.layout_em)
doc.count_pages()
layouttime = time.time() - layouttime
timing.layout += layouttime
if layouttime < timing.minlayout:
timing.minlayout = layouttime
timing.minlayoutfilename = state.filename
if layouttime > timing.maxlayout:
timing.maxlayout = layouttime
timing.maxlayoutfilename = state.filename
if state.layer_config:
apply_layer_config(doc, state.layer_config)
if fz_optind == len(argv) or not mupdf.is_page_range(
argv[fz_optind]):
drawrange(doc, "1-N")
if fz_optind < len(argv) and mupdf.is_page_range(
argv[fz_optind]):
drawrange(doc, argv[fz_optind])
fz_optind += 1
bgprint_flush()
if state.useaccel:
save_accelerator(doc, state.filename)
except Exception as e:
if not state.ignore_errors:
raise
bgprint_flush()
sys.stderr.write(f'ignoring error in {state.filename}\n')
except Exception as e:
bgprint_flush()
sys.stderr.write(
f'error: cannot draw \'{state.filename}\' because: {e}\n')
state.errored = 1
if not state.output_file_per_page:
file_level_trailers()
if state.out:
state.out.close_output()
state.out = None
if state.showtime and timing.count > 0:
if bgprint.active:
timing.total = gettime() - timing.total
if state.files == 1:
sys.stderr.write(
f'total {timing.total:.0f}ms ({timing.layout:.0f}ms layout) / {timing.count} pages for an average of {timing.total / timing.count:.0f}ms\n'
)
if bgprint.active:
sys.stderr.write(
f'fastest page {timing.minpage}: {timing.mininterp:.0f}ms (interpretation) {timing.min - timing.mininterp:.0f}ms (rendering) {timing.min:.0f}ms(total)\n'
)
sys.stderr.write(
f'slowest page {timing.maxpage}: {timing.maxinterp:.0f}ms (interpretation) {timing.max - timing.maxinterp:.0f}ms (rendering) {timing.max:.0f}ms(total)\n'
)
else:
sys.stderr.write(
f'fastest page {timing.minpage}: {timing.min:.0f}ms\n')
sys.stderr.write(
f'slowest page {timing.maxpage}: {timing.max:.0f}ms\n')
else:
sys.stderr.write(
f'total {timing.total:.0f}ms ({timing.layout:.0f}ms layout) / {timing.count} pages for an average of {timing.total / timing.count:.0f}ms in {state.files} files\n'
)
sys.stderr.write(
f'fastest layout: {timing.minlayout:.0f}ms ({timing.minlayoutfilename})\n'
)
sys.stderr.write(
f'slowest layout: {timing.maxlayout:.0f}ms ({timing.maxlayoutfilename})\n'
)
sys.stderr.write(
f'fastest page {timing.minpage}: {timing.min:.0f}ms ({timing.minfilename})\n'
)
sys.stderr.write(
f'slowest page {timing.maxpage}: {timing.max:.0f}ms ({timing.maxfilename})\n'
)
if state.showmemory:
sys.stderr.write(
f'Memory use total={info.total} peak={info.peak} current={info.current}\n'
)
return state.errored != 0
#! /usr/bin/env python3
import mupdf
import os
import sys
assert len(sys.argv) == 7
filename, page_num, zoom, rotate, output, needle = sys.argv[1:]
page_num = int(page_num)
zoom = int(zoom)
rotate = int(rotate)
document = mupdf.Document(filename)
for key in mupdf.metadata_keys:
value = document.lookup_metadata(key)
if value is not None:
print(f'{key}: {value}')
def draw(argv):
global alphabits_graphics
global alphabits_text
global band_height
global colorspace
global files
global fir
global format_
global gamma_value
global height
global ignore_errors
global invert
global layer_config
global layout_css
global layout_em
global layout_h
global layout_use_doc_css
global layout_w
global lowmemory
global min_line_width
global no_icc
global num_workers
global num_workers
global out
global out_cs
global output
global output_file_per_page
global output_format
global password
global proof_filename
global quiet
global resolution
global res_specified
global rotation
global showfeatures
global showmd5
global showmemory
global showtime
global spots
global uselist
global width
info = trace_info()
quiet = 0
items, argv = getopt.getopt(
argv, 'qp:o:F:R:r:w:h:fB:c:e:G:Is:A:DiW:H:S:T:U:XLvPl:y:NO:')
for option, value in items:
if 0: pass
elif option == '-q': quiet = 1
elif option == '-p': password = value
elif option == '-o': output = value
elif option == '-F': format_ = value
elif option == '-R': rotation = float(value)
elif option == '-r':
resolution = float(output)
res_specified = 1
elif option == '-w':
width = float(value)
elif option == '-h':
height = float(value)
elif option == '-f':
fir = 1
elif option == '-B':
band_height = int(value)
elif option == '-c':
out_cs = parse_colorspace(value)
elif option == '-e':
proof_filename = value
elif option == '-G':
gamma_value = float(value)
elif option == '-I':
invert += 1
elif option == '-W':
layout_w = float(value)
elif option == '-H':
layout_h = float(value)
elif option == '-S':
layout_em = float(value)
elif option == '-U':
layout_css = value
elif option == '-X':
layout_use_doc_css = 0
elif option == '-O':
spots = float(value)
if not mupdf.FZ_ENABLE_SPOT_RENDERING:
jlib.log(
'Spot rendering/Overprint/Overprint simulation not enabled in this build'
)
spots = SPOTS_NONE
elif option == '-s':
if 't' in value: showtime += 1
if 'm' in value: showmemory += 1
if 'f' in value: showfeatures += 1
if '5' in value: showmd5 += 1
elif option == '-A':
alphabits_graphics = int(value)
sep = value.find('/')
if sep >= 0:
alphabits_text = int(value[sep + 1:])
else:
alphabits_text = alphabits_graphics
elif option == '-D':
uselist = 0
elif option == '-l':
min_line_width = float(value)
elif option == '-i':
ignore_errors = 1
elif option == '-N':
no_icc = 1
elif option == '-T':
num_workers = int(value)
elif option == '-L':
lowmemory = 1
elif option == '-P':
bgprint.active = 1
elif option == '-y':
layer_config = value
elif option == '-v':
print(f'mudraw version {mupdf.FZ_VERSION}')
if not argv:
usage()
if num_workers > 0:
if uselist == 0:
printf('cannot use multiple threads without using display list')
sys.exit(1)
if band_height == 0:
printf('Using multiple threads without banding is pointless')
if bgprint.active:
if uselist == 0:
printf('cannot bgprint without using display list')
sys.exit(1)
if proof_filename:
proof_buffer = mupdf.Buffer(proof_filename)
proof_cs = mupdf.Colorspace(FZ_COLORSPACE_NONE, 0, None, proof_buffer)
mupdf.set_text_aa_level(alphabits_text)
mupdf.set_graphics_aa_level(alphabits_graphics)
mupdf.set_graphics_min_line_width(min_line_width)
if no_icc:
mupdf.disable_icc()
else:
mupdf.enable_icc()
if layout_css:
buf = mupdf.Buffer(layout_css)
mupdf.set_user_css(buf.string_from_buffer())
mupdf.set_use_document_css(layout_use_doc_css)
# Determine output type
if band_height < 0:
print('Bandheight must be > 0')
sys.exit(1)
output_format = OUT_PNG
if format_:
jlib.log('{format=}')
for i in range(len(suffix_table)):
if format_ == suffix_table[i].suffix[1:]:
output_format = suffix_table[i].format
if spots == SPOTS_FULL and suffix_table[i].spots == 0:
print(
f'Output format {suffix_table[i].suffix[1:]} does not support spot rendering.'
)
print('Doing overprint simulation instead.')
spots = SPOTS_OVERPRINT_SIM
break
else:
print(f'Unknown output format {format}')
sys.exit(1)
elif output:
suffix = output
i = 0
while 1:
if i == len(suffix_table):
break
s = suffix.find(suffix_table[i].suffix)
if s != -1:
suffix = suffix_table[i].suffix[s + 1:]
output_format = suffix_table[i].format_
if spots == SPOTS_FULL and suffix_table[i].spots == 0:
print(
'Output format {suffix_table[i].suffix[1:]} does not support spot rendering'
)
print('Doing overprint simulation instead.')
spots = SPOTS_OVERPRINT_SIM
i = 0
else:
i += 1
if band_height:
if output_format not in (OUT_PAM, OUT_PGM, OUT_PPM, OUT_PNM, OUT_PNG,
OUT_PBM, OUT_PKM, OUT_PCL, OUT_PCLM, OUT_PS,
OUT_PSD):
print(
'Banded operation only possible with PxM, PCL, PCLM, PS, PSD, and PNG outputs'
)
sys.exit(1)
if showmd5:
print('Banded operation not compatible with MD5')
sys.exit(1)
for i in range(len(format_cs_table)):
if format_cs_table[i].format_ == output_format:
if out_cs == CS_UNSET:
out_cs = format_cs_table[i].default_cs
for j in range(len(format_cs_table[i].permitted_cs)):
if format_cs_table[i].permitted_cs[j] == out_cs:
break
else:
print('Unsupported colorspace for this format')
sys.exit(1)
alpha = 1
if out_cs in (CS_MONO, CS_GRAY, CS_GRAY_ALPHA):
colorspace = mupdf.Colorspace(mupdf.Colorspace.Fixed_GRAY)
alpha = (out_cs == CS_GRAY_ALPHA)
elif out_cs in (CS_RGB, CS_RGB_ALPHA):
colorspace = mupdf.Colorspace(mupdf.Colorspace.Fixed_RGB)
alpha = (out_cs == CS_RGB_ALPHA)
elif out_cs in (CS_CMYK, CS_CMYK_ALPHA):
colorspace = mupdf.Colorspace(mupdf.Colorspace.Fixed_CMYK)
alpha = (out_cs == CS_CMYK_ALPHA)
elif out_cs == CS_ICC:
icc_buffer = mupdf.Buffer(icc_filename)
colorspace = Colorspace(mupdf.FZ_COLORSPACE_NONE, 0, None, icc_buffer)
alpha = 0
else:
print('Unknown colorspace!')
sys.exit(1)
if out_cs != CS_ICC:
colorspace = mupdf.Colorspace(colorspace)
else:
# Check to make sure this icc profile is ok with the output format */
okay = 0
for i in range(len(format_cs_table)):
if format_cs_table[i].format == output_format:
for j in range(len(format_cs_table[i].permitted_cs)):
x = format_cs_table[i].permitted_cs[j]
if x in (CS_MONO, CS_GRAY, CS_GRAY_ALPHA):
if colorspace.colorspace_is_gray():
okay = 1
elif x in (CS_RGB, CS_RGB_ALPHA):
if colorspace.colorspace_is_rgb():
okay = 1
elif x in (CS_CMYK, CS_CMYK_ALPHA):
if colorspace.colorspace_is_cmyk():
okay = 1
if not okay:
print(
'ICC profile uses a colorspace that cannot be used for this format'
)
sys.exit(1)
if output_format == OUT_SVG:
# SVG files are always opened for each page. Do not open "output".
pass
elif output and (not output.startswith('-')
or len(output) >= 2) and len(output) >= 1:
if has_percent_d(output):
output_file_per_page = 1
else:
out = mupdf.Output(output, 0)
else:
quiet = 1
# automatically be quiet if printing to stdout
if 0:
# Windows specific code to make stdout binary.
if output_format not in (OUT_TEXT, OUT_STEXT, OUT_HTML, OUT_XHTML,
OUT_TRACE):
setmode(fileno(stdout), O_BINARY)
out = mupdf.Output(mupdf.Output.Fixed_STDOUT)
filename = argv[0]
if not output_file_per_page:
file_level_headers()
timing.count = 0
timing.total = 0
timing.min = 1 << 30
timing.max = 0
timing.mininterp = 1 << 30
timing.maxinterp = 0
timing.minpage = 0
timing.maxpage = 0
timing.minfilename = ""
timing.maxfilename = ""
if showtime and bgprint.active:
timing.total = gettime()
fz_optind = 0
while fz_optind < len(argv):
filename = argv[fz_optind]
fz_optind += 1
files += 1
doc = mupdf.Document(filename)
if doc.needs_password():
if not doc.authenticate_password(password):
raise Exception(f'cannot authenticate password: {filename}')
# Once document is open check for output intent colorspace
oi = doc.document_output_intent()
if oi: # todo: fz_document_output_intent() can return NULL, and we don't handle this.
# See if we had explicitly set a profile to render
if out_cs != CS_ICC:
# In this case, we want to render to the output intent
# color space if the number of channels is the same
if oi.colorspace_n() == colorspace.colorspace_n():
colorspace = oi
doc.layout_document(layout_w, layout_h, layout_em)
if layer_config:
apply_layer_config(doc, layer_config)
if fz_optind == len(argv) or not mupdf.is_page_range(argv[fz_optind]):
drawrange(doc, "1-N")
if fz_optind < len(argv) and mupdf.is_page_range(argv[fz_optind]):
drawrange(doc, argv[fz_optind])
fz_optind += 1
bgprint_flush()
if not output_file_per_page:
file_level_trailers()
if out:
out.close_output()
out = None
if showtime and timing.count > 0:
if bgprint.active:
timing.total = gettime() - timing.total
if files == 1:
print(
f'total {timing.total}ms / {timing.count} pages for an average of {timing.total / timing.count}ms'
)
if bgprint.active:
printf(
f'fastest page {timing.minpage}: {timing.mininterp}ms (interpretation) {timing.min - timing.mininterp}ms (rendering) {timing.min}ms(total)'
)
print(
f'slowest page {timing.maxpage}: {timing.maxinterp}ms (interpretation) {timing.max - timing.maxinterp}ms (rendering) {timing.max}ms(total)'
)
else:
print(f'fastest page {timing.minpage}: {timing.min}ms')
print(f'slowest page {timing.maxpage}: {timing.max}ms')
else:
print(
f'total {timing.total}ms / {timing.count} pages for an average of {timing.total / timing.count}ms in {files} files'
)
print(
f'fastest page {timing.minpage}: {timing.min}ms ({timing.minfilename})'
)
print(
f'slowest page {timing.maxpage}: {timing.max}ms ({timing.maxfilename})'
)
if showmemory:
print(
f'Memory use total={info.total} peak={info.peak} current={info.current}'
)
return errored != 0
("XPS Files", "*.*xps"),
("Epub Files", "*.epub"),
("Fiction Books", "*.fb2"),
("Comic Books", "*.cbz"),
("HTML", "*.htm*"),
# add more document types here
),
)
else:
fname = sys.argv[1]
if not fname:
sg.Popup("Cancelling:", "No filename supplied")
raise SystemExit("Cancelled: no filename supplied")
doc = mupdf.Document(fname)
page_count = doc.count_pages()
# allocate storage for page display lists
dlist_tab = [None] * page_count
title = "PyMuPDF display of '%s', pages: %i" % (fname, page_count)
# ------------------------------------------------------------------------------
# read the page data
# ------------------------------------------------------------------------------
def get_page(pno, zoom=False, max_size=None):
"""Return a tkinter.PhotoImage or a PNG image for a document page number.
:arg int pno: 0-based page number
:arg zoom: top-left of old clip rect, and one of -1, 0, +1 for dim. x or y
