josch/img2pdf
3
6
Fork 10
Code Issues 22 Pull requests 5 Projects Releases 3 Wiki Activity

add --rotation argument overwriting exif data (closes: #100)

Browse source
This commit is contained in:
Johannes Schauer Marin Rodrigues 2021-05-14 11:30:23 +02:00
parent 114d7270a2
commit b4c8aa1a5f
Signed by: josch
GPG key ID: F2CBA5C78FBD83E1
1 changed files with 81 additions and 27 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

108
src/img2pdf.py
View file

@ -79,6 +79,8 @@ papernames = {
Engine = Enum("Engine", "internal pdfrw pikepdf") Engine = Enum("Engine", "internal pdfrw pikepdf")
Rotation = Enum("Rotation", "auto none ifvalid 0 90 180 270")
FitMode = Enum("FitMode", "into fill exact shrink enlarge") FitMode = Enum("FitMode", "into fill exact shrink enlarge")
PageOrientation = Enum("PageOrientation", "portrait landscape") PageOrientation = Enum("PageOrientation", "portrait landscape")
@ -1155,7 +1157,9 @@ class pdfdoc(object):
raise ValueError("unknown engine: %s" % self.engine) raise ValueError("unknown engine: %s" % self.engine)
def get_imgmetadata(imgdata, imgformat, default_dpi, colorspace, rawdata=None): def get_imgmetadata(
imgdata, imgformat, default_dpi, colorspace, rawdata=None, rotreq=None
):
if imgformat == ImageFormat.JPEG2000 and rawdata is not None and imgdata is None: if imgformat == ImageFormat.JPEG2000 and rawdata is not None and imgdata is None:
# this codepath gets called if the PIL installation is not able to # this codepath gets called if the PIL installation is not able to
# handle JPEG2000 files # handle JPEG2000 files
@ -1206,25 +1210,44 @@ def get_imgmetadata(imgdata, imgformat, default_dpi, colorspace, rawdata=None):
logger.debug("input dpi = %d x %d", *ndpi) logger.debug("input dpi = %d x %d", *ndpi)
rotation = 0 rotation = 0
if hasattr(imgdata, "_getexif") and imgdata._getexif() is not None: if rotreq in (None, Rotation.auto, Rotation.ifvalid):
for tag, value in imgdata._getexif().items(): if hasattr(imgdata, "_getexif") and imgdata._getexif() is not None:
if TAGS.get(tag, tag) == "Orientation": for tag, value in imgdata._getexif().items():
# Detailed information on EXIF rotation tags: if TAGS.get(tag, tag) == "Orientation":
# http://impulseadventure.com/photo/exif-orientation.html # Detailed information on EXIF rotation tags:
if value == 1: # http://impulseadventure.com/photo/exif-orientation.html
rotation = 0 if value == 1:
elif value == 6: rotation = 0
rotation = 90 elif value == 6:
elif value == 3: rotation = 90
rotation = 180 elif value == 3:
elif value == 8: rotation = 180
rotation = 270 elif value == 8:
elif value in (2, 4, 5, 7): rotation = 270
raise ExifOrientationError( elif value in (2, 4, 5, 7):
"Unsupported flipped rotation mode (%d)" % value if rotreq == Rotation.ifvalid:
) logger.warning(
else: "Unsupported flipped rotation mode (%d)", value
raise ExifOrientationError("Invalid rotation (%d)" % value) )
else:
raise ExifOrientationError(
"Unsupported flipped rotation mode (%d)" % value
)
else:
if rotreq == Rotation.ifvalid:
logger.warning("Invalid rotation (%d)", value)
else:
raise ExifOrientationError("Invalid rotation (%d)" % value)
elif rotreq in (Rotation.none, Rotation["0"]):
rotation = 0
elif rotreq == Rotation["90"]:
rotation = 90
elif rotreq == Rotation["180"]:
rotation = 180
elif rotreq == Rotation["270"]:
rotation = 270
else:
raise Exception("invalid rotreq")
logger.debug("rotation = %d°", rotation) logger.debug("rotation = %d°", rotation)
@ -1344,7 +1367,7 @@ def parse_png(rawdata):
return pngidat, palette return pngidat, palette
def read_images(rawdata, colorspace, first_frame_only=False): def read_images(rawdata, colorspace, first_frame_only=False, rot=None):
im = BytesIO(rawdata) im = BytesIO(rawdata)
im.seek(0) im.seek(0)
imgdata = None imgdata = None
@ -1386,7 +1409,7 @@ def read_images(rawdata, colorspace, first_frame_only=False):
# JPEG and JPEG2000 can be embedded into the PDF as-is # JPEG and JPEG2000 can be embedded into the PDF as-is
if imgformat == ImageFormat.JPEG or imgformat == ImageFormat.JPEG2000: if imgformat == ImageFormat.JPEG or imgformat == ImageFormat.JPEG2000:
color, ndpi, imgwidthpx, imgheightpx, rotation, iccp = get_imgmetadata( color, ndpi, imgwidthpx, imgheightpx, rotation, iccp = get_imgmetadata(
imgdata, imgformat, default_dpi, colorspace, rawdata imgdata, imgformat, default_dpi, colorspace, rawdata, rot
) )
if color == Colorspace["1"]: if color == Colorspace["1"]:
raise JpegColorspaceError("jpeg can't be monochrome") raise JpegColorspaceError("jpeg can't be monochrome")
@ -1443,7 +1466,7 @@ def read_images(rawdata, colorspace, first_frame_only=False):
rotation, rotation,
iccp, iccp,
) = get_imgmetadata( ) = get_imgmetadata(
imframe, ImageFormat.JPEG, default_dpi, colorspace imframe, ImageFormat.JPEG, default_dpi, colorspace, rotreq=rot
) )
if color == Colorspace["1"]: if color == Colorspace["1"]:
raise JpegColorspaceError("jpeg can't be monochrome") raise JpegColorspaceError("jpeg can't be monochrome")
@ -1480,7 +1503,7 @@ def read_images(rawdata, colorspace, first_frame_only=False):
# must be the first chunk. # must be the first chunk.
if imgformat == ImageFormat.PNG and rawdata[28] == 0: if imgformat == ImageFormat.PNG and rawdata[28] == 0:
color, ndpi, imgwidthpx, imgheightpx, rotation, iccp = get_imgmetadata( color, ndpi, imgwidthpx, imgheightpx, rotation, iccp = get_imgmetadata(
imgdata, imgformat, default_dpi, colorspace, rawdata imgdata, imgformat, default_dpi, colorspace, rawdata, rot
) )
pngidat, palette = parse_png(rawdata) pngidat, palette = parse_png(rawdata)
# PIL does not provide the information about the original bits per # PIL does not provide the information about the original bits per
@ -1562,7 +1585,7 @@ def read_images(rawdata, colorspace, first_frame_only=False):
"group4 tiff: %d" % photo "group4 tiff: %d" % photo
) )
color, ndpi, imgwidthpx, imgheightpx, rotation, iccp = get_imgmetadata( color, ndpi, imgwidthpx, imgheightpx, rotation, iccp = get_imgmetadata(
imgdata, imgformat, default_dpi, colorspace, rawdata imgdata, imgformat, default_dpi, colorspace, rawdata, rot
) )
offset, length = ccitt_payload_location_from_pil(imgdata) offset, length = ccitt_payload_location_from_pil(imgdata)
im.seek(offset) im.seek(offset)
@ -1605,7 +1628,7 @@ def read_images(rawdata, colorspace, first_frame_only=False):
logger.debug("Converting frame: %d" % img_page_count) logger.debug("Converting frame: %d" % img_page_count)
color, ndpi, imgwidthpx, imgheightpx, rotation, iccp = get_imgmetadata( color, ndpi, imgwidthpx, imgheightpx, rotation, iccp = get_imgmetadata(
imgdata, imgformat, default_dpi, colorspace imgdata, imgformat, default_dpi, colorspace, rotreq=rot
) )
newimg = None newimg = None
@ -2022,6 +2045,7 @@ def convert(*images, **kwargs):
trimborder=None, trimborder=None,
artborder=None, artborder=None,
pdfa=None, pdfa=None,
rotation=None,
) )
for kwname, default in _default_kwargs.items(): for kwname, default in _default_kwargs.items():
if kwname not in kwargs: if kwname not in kwargs:
@ -2099,7 +2123,12 @@ def convert(*images, **kwargs):
depth, depth,
rotation, rotation,
iccp, iccp,
) in read_images(rawdata, kwargs["colorspace"], kwargs["first_frame_only"]): ) in read_images(
rawdata,
kwargs["colorspace"],
kwargs["first_frame_only"],
kwargs["rotation"],
):
pagewidth, pageheight, imgwidthpdf, imgheightpdf = kwargs["layout_fun"]( pagewidth, pageheight, imgwidthpdf, imgheightpdf = kwargs["layout_fun"](
imgwidthpx, imgheightpx, ndpi imgwidthpx, imgheightpx, ndpi
) )
@ -2402,6 +2431,13 @@ def input_images(path_expr):
return result return result
def parse_rotationarg(string):
for m in Rotation:
if m.name == string.lower():
return m
raise argparse.ArgumentTypeError("unknown rotation value: %s" % string)
def parse_fitarg(string): def parse_fitarg(string):
for m in FitMode: for m in FitMode:
if m.name == string.lower(): if m.name == string.lower():
@ -3560,6 +3596,23 @@ of the input image. If the orientation of a page gets flipped, then so do the
values set via the --border option. values set via the --border option.
""", """,
) )
sizeargs.add_argument(
"-r",
"--rotation",
"--orientation",
metavar="ROT",
type=parse_rotationarg,
default=Rotation.auto,
help="""
Specifies how input images should be rotated. ROT can be one of auto, none,
ifvalid, 0, 90, 180 and 270. The default value is auto and indicates that input
images are rotated according to their EXIF Orientation tag. The values none and
0 ignore the EXIF Orientation values of the input images. The value ifvalid
acts like auto but ignores invalid EXIF rotation values and only issues a
warning instead of throwing an error. The values 90, 180 and 270 perform a
clockwise rotation of the image.
""",
)
sizeargs.add_argument( sizeargs.add_argument(
"--crop-border", "--crop-border",
metavar="L[:L]", metavar="L[:L]",
@ -3797,6 +3850,7 @@ and left/right, respectively. It is not possible to specify asymmetric borders.
trimborder=args.trim_border, trimborder=args.trim_border,
artborder=args.art_border, artborder=args.art_border,
pdfa=args.pdfa, pdfa=args.pdfa,
rotation=args.rotation,
) )
except Exception as e: except Exception as e:
logger.error("error: " + str(e)) logger.error("error: " + str(e))