def trace_runpage(use_display_list, doc, number):
page = mupdf.Page(doc, number - 1)
mediabox = page.bound_page()
print(
f'<page number="{number}" mediabox="{mediabox.x0} {mediabox.y0} {mediabox.x1} {mediabox.y1}">'
)
output = mupdf.Output(mupdf.Output.Fixed_STDOUT)
dev = mupdf.Device(output)
if use_display_list:
list_ = mupdf.DisplayList(page)
list_.run_display_list(dev, mupdf.Matrix(mupdf.fz_identity),
mupdf.Rect(mupdf.fz_infinite_rect),
mupdf.Cookie())
else:
page.run_page(dev, mupdf.Matrix(mupdf.fz_identity), mupdf.Cookie())
output.close_output()
print('</page>')
def __init__(self):
self.num = 0
self.band = 0
self.list = None
self.ctm = None
self.tbounds = None
self.pix = None
self.bit = None
self.cookie = mupdf.Cookie()
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
document_out = mupdf.PdfDocument()
while rotate < 0:
rotate += 360
while rotate >= 360:
rotate -= 360
if rotate % 90 != 0: rotate = 0
for p in range(page_from, page_to + page_delta, page_delta):
page = document.load_page(p)
rect = page.bound_page()
dev, resources, contents = document_out.page_write(rect)
page.run_page(dev, mupdf.Matrix(), mupdf.Cookie())
pdf_obj = document_out.add_page(rect, rotate, resources, contents)
document_out.insert_page(-1, pdf_obj)
write_options = mupdf.PdfWriteOptions()
write_options.do_garbage = 4
write_options.do_compress = 1
write_options.do_compress_images = 1
write_options.do_compress_fonts = 1
write_options.do_sanitize = 1
write_options.do_incremental = 0
write_options.do_ascii = 0
write_options.do_decompress = 0
write_options.do_linear = 0
write_options.do_clean = 1
write_options.do_pretty = 0
buffer_ = mupdf.Buffer(8192)
output = mupdf.Output(buffer_)
document_out.write_document(output, write_options)
size, data = buffer_.buffer_extract_raw()
print(f'buffer_.buffer_extract() returned: {size, data}')
return data, size
def show_html(self):
'''
Convert to HTML using Extract, and show in new window using
PyQt5.QtWebKitWidgets.QWebView.
'''
buffer_ = self.page.new_buffer_from_page_with_format(
format="docx",
options="html",
transform=mupdf.Matrix(1, 0, 0, 1, 0, 0),
cookie=mupdf.Cookie(),
)
html_content = buffer_.buffer_extract().decode('utf8')
# Show in a new window using Qt's QWebView.
self.webview = PyQt5.QtWebKitWidgets.QWebView()
self.webview.setHtml(html_content)
self.webview.show()
def convert_runpage( doc, number, out):
page = mupdf.Page( doc, number - 1)
mediabox = page.bound_page()
dev = out.begin_page(mediabox)
page.run( dev, mupdf.Matrix(mupdf.fz_identity), mupdf.Cookie())
out.end_page()
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 drawpage(doc, pagenum):
list_ = None
cookie = mupdf.Cookie()
seps = None
features = ""
start = gettime() if state.showtime else 0
page = mupdf.Page(doc, pagenum - 1)
if state.spots != SPOTS_NONE:
seps = page.page_separations()
if seps.m_internal:
n = seps.count_separations()
if state.spots == SPOTS_FULL:
for i in range(n):
seps.set_separation_behavior(i, mupdf.FZ_SEPARATION_SPOT)
else:
for i in range(n):
seps.set_separation_behavior(i,
mupdf.FZ_SEPARATION_COMPOSITE)
elif page.page_uses_overprint():
# This page uses overprint, so we need an empty
# sep object to force the overprint simulation on.
seps = mupdf.Separations(0)
elif state.oi and state.oi.m_internal and state.oi.colorspace_n(
) != state.colorspace.colorspace_n():
# We have an output intent, and it's incompatible
# with the colorspace our device needs. Force the
# overprint simulation on, because this ensures that
# we 'simulate' the output intent too. */
seps = mupdf.Separations(0)
if state.uselist:
list_ = mupdf.DisplayList(page.bound_page())
dev = mupdf.Device(list_)
if state.lowmemory:
dev.enable_device_hints(FZ_NO_CACHE)
page.run(dev, mupdf.Matrix(), cookie)
dev.close_device()
if bgprint.active and state.showtime:
end = gettime()
start = end - start
if state.showfeatures:
# SWIG doesn't appear to handle the out-param is_color in
# mupdf.Device() constructor that wraps fz_new_test_device(), so we use
# the underlying mupdf function() instead.
#
dev, iscolor = mupdf.new_test_device(0.02, 0, None)
dev = mupdf.Device(dev)
if state.lowmemory:
dev.enable_device_hints(mupdf.FZ_NO_CACHE)
if list_:
list_.run_display_list(dev, mupdf.Matrix(mupdf.fz_identity),
mupdf.Rect(mupdf.fz_infinite_rect),
mupdf.Cookie())
else:
page.run(dev, fz_identity, cookie)
dev.close_device()
features = " color" if iscolor else " grayscale"
if state.output_file_per_page:
bgprint_flush()
if state.out:
state.out.close_output()
text_buffer = mupdf.format_output_path(state.output, pagenum)
state.out = mupdf.Output(text_buffer, 0)
if bgprint.active:
bgprint_flush()
if bgprint.active:
if not state.quiet or state.showfeatures or state.showtime or state.showmd5:
sys.stderr.write("page %s %d%s" %
(state.filename, pagenum, features))
bgprint.started = 1
bgprint.page = page
bgprint.list = list_
bgprint.seps = seps
bgprint.filename = state.filename
bgprint.pagenum = pagenum
bgprint.interptime = start
else:
if not state.quiet or state.showfeatures or state.showtime or state.showmd5:
sys.stderr.write("page %s %d%s" %
(state.filename, pagenum, features))
dodrawpage(page, list_, pagenum, cookie, start, 0, state.filename, 0,
seps)
def dodrawpage(page, list_, pagenum, cookie, start, interptime, filename, bg,
seps):
if state.output_file_per_page:
file_level_headers()
if list_:
mediabox = mupdf.Rect(list_)
else:
mediabox = page.bound_page()
if state.output_format == OUT_TRACE:
state.out.write_string(
"<page mediabox=\"%g %g %g %g\">\n" %
(mediabox.x0, mediabox.y0, mediabox.x1, mediabox.y1))
dev = mupdf.Device(state.out)
if state.lowmemory:
dev.enable_device_hints(mupdf.FZ_NO_CACHE)
if list_:
list_.run_display_list(dev, mupdf.Matrix(),
mupdf.Rect(mupdf.fz_infinite_rect), cookie)
else:
page.run(dev, fz_identity, cookie)
state.out.write_string("</page>\n")
dev.close_device()
dev = None
elif state.output_format == OUT_XMLTEXT:
state.out.write_string(
"<page mediabox=\"%g %g %g %g\">\n" %
(mediabox.x0, mediabox.y0, mediabox.x1, mediabox.y1))
dev = mupdf.Device.new_raw_device(state.out)
if list_:
list_.run_display_list(dev, mupdf.Matrix(),
mupdf.Rect(mupdf.fz_infinite_rect), cookie)
else:
page.run(dev, fz_identity, cookie)
state.out.write_string("</page>\n")
dev.close_device()
dev = None
elif state.output_format == OUT_BBOX:
bbox = mupdf.Rect(mupdf.Rect.Fixed_EMPTY)
dev = mupdf.Device(bbox)
if state.lowmemory:
dev.enable_device_hints(mupdf.FZ_NO_CACHE)
if list_:
list_.run_display_list(dev, fz_identity,
mupdf.Rect(mupdf.fz_infinite_rect), cookie)
else:
page.run(dev, fz_identity, cookie)
dev.close_device()
state.out.write_string(
"<page bbox=\"%s %s %s %s\" mediabox=\"%s %s %s %s\" />\n",
bbox.x0,
bbox.y0,
bbox.x1,
bbox.y1,
mediabox.x0,
mediabox.y0,
mediabox.x1,
mediabox.y1,
)
elif state.output_format in (OUT_TEXT, OUT_HTML, OUT_XHTML, OUT_STEXT):
zoom = state.resolution / 72
ctm = mupdf.Matrix(
mupdf.pre_scale(mupdf.rotate(state.rotation), zoom, zoom))
stext_options = mupdf.StextOptions()
stext_options.flags = mupdf.FZ_STEXT_PRESERVE_IMAGES if (
state.output_format == OUT_HTML
or state.output_format == OUT_XHTML) else 0
text = mupdf.StextPage(mediabox)
dev = mupdf.Device(text, stext_options)
if state.lowmemory:
fz_enable_device_hints(dev, FZ_NO_CACHE)
if list_:
list_.run_display_list(dev, ctm,
mupdf.Rect(mupdf.fz_infinite_rect), cookie)
else:
page.run(dev, ctm, cookie)
dev.close_device()
dev = None
if state.output_format == OUT_STEXT:
state.out.print_stext_page_as_xml(text, pagenum)
elif state.output_format == OUT_HTML:
state.out.print_stext_page_as_html(text, pagenum)
elif state.output_format == OUT_XHTML:
state.out.print_stext_page_as_xhtml(text, pagenum)
elif state.output_format == OUT_TEXT:
state.out.print_stext_page_as_text(text)
state.out.write_string("\f\n")
elif state.output_format == OUT_SVG:
zoom = state.resolution / 72
ctm = mupdf.Matrix(zoom, zoom)
ctm.pre_rotate(state.rotation)
tbounds = mupdf.Rect(mediabox, ctm)
if not state.output or state.output == "-":
state.out = mupdf.Output(mupdf.Output.Fixed_STDOUT)
else:
buf = mupdf.format_output_path(state.output, pagenum)
state.out = mupdf.Output(buf, 0)
dev = mupdf.Device(state.out, tbounds.x1 - tbounds.x0,
tbounds.y1 - tbounds.y0, mupdf.FZ_SVG_TEXT_AS_PATH,
1)
if state.lowmemory:
dev.enable_device_hints(dev, mupdf.FZ_NO_CACHE)
if list_:
list_.run_display_list(dev, ctm, tbounds, cookie)
else:
page.run(dev, ctm, cookie)
dev.close_device()
state.out.close_output()
else:
zoom = state.resolution / 72
m = mupdf.rotate(state.rotation)
ctm = mupdf.Matrix(
mupdf.pre_scale(mupdf.rotate(state.rotation), zoom, zoom))
tbounds = mupdf.Rect(mediabox, ctm)
ibounds = tbounds.round_rect()
# Make local copies of our width/height
w = state.width
h = state.height
# If a resolution is specified, check to see whether w/h are
# exceeded; if not, unset them. */
if state.res_specified:
t = ibounds.x1 - ibounds.x0
if w and t <= w:
w = 0
t = ibounds.y1 - ibounds.y0
if h and t <= h:
h = 0
# Now w or h will be 0 unless they need to be enforced.
if w or h:
scalex = w / (tbounds.x1 - tbounds.x0)
scaley = h / (tbounds.y1 - tbounds.y0)
if state.fit:
if w == 0:
scalex = 1.0
if h == 0:
scaley = 1.0
else:
if w == 0:
scalex = scaley
if h == 0:
scaley = scalex
if not state.fit:
if scalex > scaley:
scalex = scaley
else:
scaley = scalex
scale_mat = mupdf.Matrix.scale(scalex, scaley)
ctm = mupdf.Matrix(
mupdf.concat(ctm.internal(), scale_mat.internal()))
tbounds = mupdf.Rect(mediabox, ctm)
ibounds = tbounds.round_rect()
tbounds = ibounds.rect_from_irect()
band_ibounds = ibounds
bands = 1
totalheight = ibounds.y1 - ibounds.y0
drawheight = totalheight
if state.band_height != 0:
# Banded rendering; we'll only render to a
# given height at a time.
drawheight = state.band_height
if totalheight > state.band_height:
band_ibounds.y1 = band_ibounds.y0 + state.band_height
bands = (totalheight + state.band_height - 1) / state.band_height
tbounds.y1 = tbounds.y0 + state.band_height + 2
#DEBUG_THREADS(("Using %d Bands\n", bands));
if state.num_workers > 0:
for band in range(min(state.num_workers, bands)):
state.workers[band].band = band
state.workers[band].ctm = ctm
state.workers[band].tbounds = tbounds
state.workers[band].cookie = mupdf.Cookie()
state.workers[band].list = list_
state.workers[band].pix = mupdf.Pixmap(state.colorspace,
band_ibounds, seps,
state.alpha)
state.workers[band].pix.set_pixmap_resolution(
state.resolution, state.resolution)
ctm.f -= drawheight
pix = state.workers[0].pix
else:
pix = mupdf.Pixmap(state.colorspace, band_ibounds, seps,
state.alpha)
pix.set_pixmap_resolution(int(state.resolution),
int(state.resolution))
# Output any page level headers (for banded formats)
if state.output:
state.bander = None
if state.output_format == OUT_PGM or state.output_format == OUT_PPM or state.output_format == OUT_PNM:
state.bander = mupdf.BandWriter(state.out,
mupdf.BandWriter.PNM)
elif state.output_format == OUT_PAM:
state.bander = mupdf.BandWriter(state.out,
mupdf.BandWriter.PAM)
elif state.output_format == OUT_PNG:
state.bander = mupdf.BandWriter(state.out,
mupdf.BandWriter.PNG)
elif state.output_format == OUT_PBM:
state.bander = mupdf.BandWriter(state.out,
mupdf.BandWriter.PBM)
elif state.output_format == OUT_PKM:
state.bander = mupdf.BandWriter(state.out,
mupdf.BandWriter.PKM)
elif state.output_format == OUT_PS:
state.bander = mupdf.BandWriter(state.out, mupdf.BandWriter.PS)
elif state.output_format == OUT_PSD:
state.bander = mupdf.BandWriter(state.out,
mupdf.BandWriter.PSD)
elif state.output_format == OUT_PWG:
if state.out_cs == CS_MONO:
state.bander = mupdf.BandWriter(state.out,
mupdf.BandWriter.MONO,
mupdf.PwgOptions())
else:
state.bander = mupdf.BandWriter(state.out,
mupdf.BandWriter.COLOR,
mupdf.PwgOptions())
elif state.output_format == OUT_PCL:
if state.out_cs == CS_MONO:
state.bander = mupdf.BandWriter(state.out,
mupdf.BandWriter.MONO,
mupdf.PclOptions())
else:
state.bander = mupdf.BandWriter(state.out,
mupdf.BandWriter.COLOR,
mupdf.PclOptions())
if state.bander:
state.bander.write_header(pix.w(), totalheight, pix.n(),
pix.alpha(), pix.xres(),
pix.yres(), state.output_pagenum,
pix.colorspace(), pix.seps())
state.output_pagenum += 1
for band in range(bands):
if state.num_workers > 0:
w = state.workers[band % state.num_workers]
pix = w.pix
bit = w.bit
w.bit = None
cookie.increment_errors(w.cookie.errors())
else:
bit = drawband(page, list_, ctm, tbounds, cookie,
band * state.band_height, pix)
if state.output:
if state.bander:
if bit:
state.bander.write_band(bit.stride(), drawheight,
bit.samples())
else:
state.bander.write_band(pix.stride(), drawheight,
pix.samples())
bit = None
if state.num_workers > 0 and band + state.num_workers < bands:
w = state.workers[band % state.num_workers]
w.band = band + state.num_workers
w.ctm = ctm
w.tbounds = tbounds
w.cookie = mupdf.Cookie()
ctm.f -= drawheight
# FIXME
if state.showmd5:
digest = pix.md5_pixmap()
sys.stderr.write(' ')
for i in range(16):
sys.stderr.write('%02x', digest[i])
if state.output_file_per_page:
file_level_trailers()
if state.showtime:
end = gettime()
diff = end - start
if bg:
if diff + interptime < timing.min:
timing.min = diff + interptime
timing.mininterp = interptime
timing.minpage = pagenum
timing.minfilename = filename
if diff + interptime > timing.max:
timing.max = diff + interptime
timing.maxinterp = interptime
timing.maxpage = pagenum
timing.maxfilename = filename
timing.count += 1
sys.stderr.write(
" %dms (interpretation) %dms (rendering) %dms (total)" %
(interptime, diff, diff + interptime))
else:
if diff < timing.min:
timing.min = diff
timing.minpage = pagenum
timing.minfilename = filename
if diff > timing.max:
timing.max = diff
timing.maxpage = pagenum
timing.maxfilename = filename
timing.total += diff
timing.count += 1
sys.stderr.write(" %dms" % diff)
if not state.quiet or state.showfeatures or state.showtime or state.showmd5:
sys.stderr.write("\n")
if state.lowmemory:
mupdf.empty_store()
if state.showmemory:
mupdf.dump_glyph_cache_stats(mupdf.stderr_())
mupdf.flush_warnings()
if cookie.get_errors():
state.errored = 1