#
# lbike.py main program started from /etc/rc.local
##
import os
import subprocess
import wand
from wand.image import Image
# CONFIG
basedir = os.path.dirname(os.path.realpath(__file__))
def callSubprocess(cmd):
subprocess.call(cmd.split())
with Image(filename=basedir+'/tmp/qr.png') as qr:
qr.transform(resize='176x264')
qr.save(filename=basedir+'/tmp/qr.png')
with Image(filename=basedir+'/img/pay_invoice_176x264.png') as bg:
bg.composite(qr, 1, 1)
bg.save(filename=basedir+'/tmp/qr_176x264.png')
callSubprocess('convert '+basedir+'/tmp/qr_176x264.png -type Palette -depth 2 -compress none bmp3:'+basedir+'/tmp/qr_176x264.bmp')
def get_image(tsf_buff, headers):
m = bmp_re.search(tsf_buff)
if(m):
return Image(blob="BM" + m.group(2))
else:
return Image(width=mm2px(headers['Resolution'], headers['Size']['width']), height=mm2px(headers['Resolution'], headers['Size']['height']))
# Android - Supporting multiple screens
# http://developer.android.com/guide/practices/screens_support.html
MANIFEST = (
# dpi, ratio
('h', .75), # high-dpi
('m', .5), # medium-dpi
('l', .375) # low-dpi
)
if __name__ == "__main__":
# We start from here (2.0x size images)
files = os.listdir("res/drawable-xhdpi/")
# Filter out non-desired files
files = filter(lambda x: re.match(r"^\w+(.jpg|.png)$", x) is not None,
files)
for filename in files:
with Image(filename="res/drawable-xhdpi/%s" % filename) as img:
print("Resizing %s..." % filename)
width, height = img.size
for dpi, ratio in MANIFEST:
cimg = img.clone()
cimg.resize(int(width * ratio), int(height * ratio))
cimg.save(filename="res/drawable-%sdpi/%s" % (dpi, filename))
def load_image(self, load_processed=False):
if not self.image:
filename = self.processed_filename if load_processed else self.filename
self.image = Image(filename=Path('scans') / filename)
def create_asset(file_data,
friendly_name,
tags='',
url_path='',
optimize=False):
data = {"optimized": optimize}
if optimize:
try:
image = ImageProcessor(file_data)
image.optimize(allow_svg_errors=True)
file_data = image.data
except:
# If optimisation failed, just don't bother optimising
data["optimized"] = "Optimization failed"
if not url_path:
url_path = settings.FILE_MANAGER.generate_asset_path(
file_data, friendly_name)
try:
with Image(blob=file_data) as image_info:
data['width'] = image_info.width
data['height'] = image_info.height
except:
# Just don't worry if image reading fails
pass
# Create file
if not settings.DATA_MANAGER.exists(url_path):
try:
settings.FILE_MANAGER.create(file_data, url_path)
except:
raise file_error(
error_number=errno.EEXIST,
message="Failed to create asset {0}.".format(url_path),
filename=url_path)
else:
error_message = "Asset already exists at {0}.".format(url_path)
asset = settings.DATA_MANAGER.fetch_one(url_path)
image_data = {}
if "width" not in asset and "width" in data:
image_data["width"] = data["width"]
if "height" not in asset and "height" in data:
image_data["height"] = data["height"]
if image_data:
settings.DATA_MANAGER.update(url_path, asset['tags'], image_data)
error_message += " Updated image information."
raise file_error(error_number=errno.EEXIST,
message=error_message,
filename=url_path)
# Once the file is created, create file metadata
settings.DATA_MANAGER.update(url_path, tags, data)
return url_path
def overlay():
with Image(filename="booth_background.png") as overlay:
with Image(filename="0.jpg") as underlay:
underlay.composite_channel('default_channels', overlay, 'over', 0,
0)
underlay.save(filename="composited.jpg")
from wand.image import Image
from wand.drawing import Drawing
from wand.color import Color
# http://www.imagemagick.org/Usage/draw/#arcs
w = 100
h = 60
bgcolor = Color('skyblue')
# original imagemagick command:
# convert -size 100x60 xc:skyblue -fill white -stroke black \
# -draw "path 'M 30,40 A 3,2 45 0,0 70,20'" path_arc_x.gif
with Image(width=w, height=h, background=bgcolor) as img:
with Drawing() as draw:
draw.fill_color = Color('white')
draw.stroke_color = Color('black')
draw.path_start()
draw.path_move((30, 40))
draw.path_elliptic_arc(radius=(3, 2), to=(70, 20), rotation=45)
draw.path_finish()
draw(img)
img.save(filename='sample23a.png')
# convert -size 100x60 xc:skyblue -fill white -stroke black \
# -draw "path 'M 30,40 A 1,1 0 0,0 70,20'" path_hcircle.gif
with Image(width=w, height=h, background=bgcolor) as img:
with Drawing() as draw:
def draw_address_layout(position, element, text=None):
global qr_png_path
logger.debug('[draw_address_layout()]-qr_png_path: ' + qr_png_path)
try:
# 8 positions available on print layout
# Position argument is an integer from 1-8
if position == 1:
x_qr = addr_features_qr['left_x']
y_qr = addr_features_qr['row_one_y']
x_addr = addr_features_addr['column_four_x']
y_addr = addr_features_addr['top_y']
x_label = addr_features_label['left_x']
y_label = addr_features_label['row_one_y']
elif position == 2:
x_qr = addr_features_qr['right_x']
y_qr = addr_features_qr['row_one_y']
x_addr = addr_features_addr['column_four_x']
y_addr = addr_features_addr['bottom_y']
x_label = addr_features_label['right_x']
y_label = addr_features_label['row_one_y']
elif position == 3:
x_qr = addr_features_qr['left_x']
y_qr = addr_features_qr['row_two_y']
x_addr = addr_features_addr['column_three_x']
y_addr = addr_features_addr['top_y']
x_label = addr_features_label['left_x']
y_label = addr_features_label['row_two_y']
elif position == 4:
x_qr = addr_features_qr['right_x']
y_qr = addr_features_qr['row_two_y']
x_addr = addr_features_addr['column_three_x']
y_addr = addr_features_addr['bottom_y']
x_label = addr_features_label['right_x']
y_label = addr_features_label['row_two_y']
elif position == 5:
x_qr = addr_features_qr['left_x']
y_qr = addr_features_qr['row_three_y']
x_addr = addr_features_addr['column_two_x']
y_addr = addr_features_addr['top_y']
x_label = addr_features_label['left_x']
y_label = addr_features_label['row_three_y']
elif position == 6:
x_qr = addr_features_qr['right_x']
y_qr = addr_features_qr['row_three_y']
x_addr = addr_features_addr['column_two_x']
y_addr = addr_features_addr['bottom_y']
x_label = addr_features_label['right_x']
y_label = addr_features_label['row_three_y']
elif position == 7:
x_qr = addr_features_qr['left_x']
y_qr = addr_features_qr['row_four_y']
x_addr = addr_features_addr['column_one_x']
y_addr = addr_features_addr['top_y']
x_label = addr_features_label['left_x']
y_label = addr_features_label['row_four_y']
elif position == 8:
x_qr = addr_features_qr['right_x']
y_qr = addr_features_qr['row_four_y']
x_addr = addr_features_addr['column_one_x']
y_addr = addr_features_addr['bottom_y']
x_label = addr_features_label['right_x']
y_label = addr_features_label['row_four_y']
if element == 'qr':
logger.info('Adding QR code to address card overlay.')
# QR Code
with Image(filename=addr_file) as layout:
with Image(filename=qr_png_path) as img:
img.resize(addr_features_qr['square_elements'][0],
addr_features_qr['square_elements'][1])
layout.composite(img, left=x_qr, top=y_qr)
layout.save(filename=addr_file)
elif element == 'address':
logger.info('Adding public address to address card overlay.')
# Public Address
with Drawing() as draw:
draw.font_family = 'Ubuntu'
draw.font_size = addr_features_addr['font_size']
draw.text_alignment = 'center'
draw.text(x_addr, y_addr, text)
with Image(filename=addr_file) as layout:
layout.rotate(90)
draw.draw(layout)
layout.rotate(270)
layout.save(filename=addr_file)
elif element == 'label':
logger.info('Adding label to address card overlay.')
# Label
with Drawing() as draw:
draw.font_family = 'Ubuntu'
draw.font_style = 'oblique'
draw.font_size = addr_features_label['font_size']
draw.text_alignment = 'left'
draw.text(x_label, y_label, text)
with Image(filename=addr_file) as layout:
draw.draw(layout)
layout.save(filename=addr_file)
except Exception as e:
logger.exception('Exception while drawing info on address layout.')
logger.exception(e)
raise
def main(argv):
parser = argparse.ArgumentParser(
description=
'Extract images from MS Word .docx file(s) and save in JPEG format')
parser.add_argument("infiles",
metavar="file",
type=argparse.FileType(),
nargs="+",
help="input .docx file(s)")
parser.add_argument(
"-o",
"--outdir",
type=Path,
#default = Path(__file__).absolute().parent,
default=Path.cwd(),
help="output directory for image files")
args = parser.parse_args()
# check for directory existence and create/error out as needed
if args.outdir.exists() and not args.outdir.is_dir():
print("output directory name already exists as a file, please rename")
exit(1)
elif not args.outdir.exists():
args.outdir.mkdir()
for infile in args.infiles:
try:
indocx = zf.ZipFile(infile.name)
except zf.BadZipfile:
print(infile.name, "is an invalid / corrupt .docx file")
exit(1)
for f in indocx.namelist():
if "word/media/" in f:
with indocx.open(f) as mediafile:
print("processing", mediafile.name, "from", infile.name,
"...")
# Rename file w/format: x-image-uuid.jpeg as x in x.docx:
outfile = args.outdir / str(infile.name + "-image-" +
str(uuid.uuid4()) + ".jpeg")
# test if jpeg format and one-to-one copy:
if (".jpeg" or ".jpg") in f:
print("writing to", outfile)
# pathlib.Path.write_bytes takes care of opening & closing file...
outfile.write_bytes(mediafile.read())
elif ".png" in f:
print("converting .png to", outfile)
im = Image(file=mediafile)
im.format = 'jpeg'
outfile.write_bytes(im.make_blob())
elif ".emf" in f:
print("converting .emf to", outfile)
# need to use unoconv, then ImageMagick, also fu@$!in tempfiles again!
tmp = args.outdir / "tmp_file.emf"
tmp.write_bytes(mediafile.read())
pdf = subprocess.run([
"unoconv", "--format=pdf", "--stdout",
tmp.resolve()
],
capture_output=True,
check=True)
jpeg = subprocess.run([
"convert", "-density", "150", "-trim",
"-bordercolor", "white", "-border", "5", "-",
"jpeg:-"
],
capture_output=True,
input=pdf.stdout,
check=True)
outfile.write_bytes(jpeg.stdout)
tmp.unlink()
elif ".wmf" in f:
print("converting .wmf to", outfile)
# need to use unoconv, then ImageMagick, also fu@$!in tempfiles again!
tmp = args.outdir / "tmp_file.wmf"
tmp.write_bytes(mediafile.read())
pdf = subprocess.run([
"unoconv", "--format=pdf", "--stdout",
tmp.resolve()
],
capture_output=True,
check=True)
jpeg = subprocess.run([
"convert", "-density", "150", "-trim",
"-bordercolor", "white", "-border", "5", "-",
"jpeg:-"
],
capture_output=True,
input=pdf.stdout,
check=True)
outfile.write_bytes(jpeg.stdout)
tmp.unlink()
elif ".wdp" in f:
print("converting .wdp to", outfile)
#ImageMagick JXR handler needs an actual file; won't take bytestream...
tmp = args.outdir / "tmp_file.wdp"
tmp.write_bytes(mediafile.read())
im = Image(filename=tmp.resolve())
im.format = 'jpeg'
#im.save(filename=outfile.name)
outfile.write_bytes(im.make_blob())
#delete tmp_file.wdp file
tmp.unlink()
else:
print(mediafile.name, "is not a recognized image type")
mediafile.close()
import datetime
import os
from wand.image import Image
PDF_SOURCE = '2018_code_calendar.pdf[{}]'
BACKGROUND_SOURCE = 'paper.jpg'
PAGE_OFFSET = 6
MARGIN_LEFT = 100
MARGIN_TOP = 10
current_week = datetime.datetime.now().isocalendar()[1]
OUTPUT = 'turing' + str(current_week) + '.jpg'
page = PAGE_OFFSET + current_week
with Image(filename=PDF_SOURCE.format(page), resolution=160) as calendar:
with Image(filename=BACKGROUND_SOURCE) as background:
background.composite_channel('default_channels', calendar, 'blend',
MARGIN_LEFT, MARGIN_TOP)
background.save(filename=OUTPUT)
address = os.getcwd() + "/" + OUTPUT
os.system(
"osascript -e \"tell application \\\"Finder\\\" to set desktop picture to POSIX file \\\""
+ address + "\\\"\"")
def main():
mc = redis.StrictRedis()
while True:
hash_value = mc.rpop('_incoming')
if not hash_value:
time.sleep(0.25)
continue
desc = mc.get(hash_value)
if not desc:
continue
desc = cPickle.loads(desc)
try:
desc = desc._replace(status='Processing')
mc.set(hash_value, cPickle.dumps(desc))
im = Image(filename=os.path.join('upload', hash_value + '.' +
desc.ext))
dt_string = im.metadata.get('exif:DateTimeOriginal')
uniq = im.metadata.get('exif:UniqueCameraModel')
mm = im.metadata.get('exif:Make'), im.metadata.get('exif:Model')
if dt_string is None:
raise Exception("Image date undefined")
dt = datetime.strptime(dt_string, '%Y:%m:%d %H:%M:%S')
if (datetime.now() - dt).days > 365:
raise Exception("Image too old")
im.depth = 8
im.format = 'RGB'
blob = im.make_blob()
im = PIL.Image.frombytes('RGB', im.size, blob)
imgif = {}
model_string = ''
if uniq:
model_string = uniq
imgif[piexif.ImageIFD.UniqueCameraModel] = uniq
else:
if mm[0]:
imgif[piexif.ImageIFD.Make] = mm[0]
if mm[1]:
imgif[piexif.ImageIFD.Model] = mm[1]
model_string = ' '.join([v for v in mm if v])
imgif[piexif.ImageIFD.DateTime] = desc.upload_date
exif = {
piexif.ExifIFD.DateTimeOriginal: dt_string,
piexif.ExifIFD.UserComment: desc.name
}
exif_dict = {"0th": imgif, "Exif": exif}
exif_bytes = piexif.dump(exif_dict)
fp = io.BytesIO()
im.save(fp, "JPEG", exif=exif_bytes)
saved = fp.getvalue()
size_string = '%i' % len(saved)
desc = ImageDesc(hash_value=hash_value,
name=desc.name,
ext=desc.ext,
upload_date=desc.upload_date,
creation_date=dt_string,
camera=model_string,
size=size_string,
status='OK')
os.remove(os.path.join('upload', hash_value + '.' + desc.ext))
with open(os.path.join('static', hash_value + '.jpg'), 'wb') as f:
f.write(saved)
im.thumbnail((32, 32), PIL.Image.ANTIALIAS)
im.save(os.path.join('static', hash_value + '_thumb.jpg'), "JPEG")
mc.set(hash_value, cPickle.dumps(desc))
mc.rpush('_images', hash_value)
except Exception as e:
desc = desc._replace(status=str(e))
mc.setex(hash_value, 60 * 24, cPickle.dumps(desc))
#!/usr/bin/env python
from wand.image import Image
from wand.drawing import Drawing
from wand.color import Color
# http://www.imagemagick.org/Usage/fonts/
# original imagemagick command:
# convert -size 320x100 xc:lightblue -font Candice -pointsize 72 \
# -fill Navy -draw "translate 28,68 skewX -20 text 0,0 'Anthony'" \
# font_slanted.jpg
w = 320
h = 100
with Image(width=w, height=h, background=Color('lightblue')) as img:
with Drawing() as draw:
text = 'Anthony'
draw.font = 'Candice'
draw.font_size = 72
draw.gravity = 'forget'
draw.fill_color = Color('navy')
draw.translate(28, 68)
draw.skew(-20, 0)
draw.text(0, 0, text)
draw(img)
img.save(filename='sample06.png')
def image_upload():
error = ''
APP_RELATED = 'static/tmp/' + session['username'] + str(get_milliseconds())
tmp_target = os.path.join(APP_ROOT, APP_RELATED)
target = IMAGES_PATH + session['username']
try:
if request.method == "POST":
# access to database
cnx = get_db()
cursor = cnx.cursor()
s3 = get_s3bucket()
# fetch the user ID of user
cursor.execute("SELECT userID FROM users WHERE username = (%s)",
(escape_string(session['username'])))
uID = cursor.fetchone()[0]
# file path of images
if not os.path.isdir(os.path.join(APP_ROOT, 'static/tmp/')):
os.mkdir(os.path.join(APP_ROOT, 'static/tmp/'))
# create a tmp folder for the user if it does not exist
if not os.path.isdir(tmp_target):
os.mkdir(tmp_target)
# check if file exists in the request
if 'file' not in request.files:
error = "file does not exist"
cursor.close()
cnx.close()
return render_template("image-upload.html",
title="upload images",
error=error)
# get the list for uploaded files
for file in request.files.getlist("file"):
# double check if the file exists
if file == None or file.filename == '':
error = "file does not exist"
cursor.close()
cnx.close()
return render_template("image-upload.html",
title="upload images",
error=error)
# give a pID to the new image
cursor.execute("SELECT max(pID) FROM images")
x = cursor.fetchone()
if x[0] == None:
pID = 1
else:
pID = x[0] + 1
# give a new image name accepted by database
filename = str(file.filename).split('.')[-1]
filename = escape_string(
str(get_milliseconds()) + '.' + filename)
# insert the image info into the database
cursor.execute(
"INSERT INTO images (pName, users_userID) VALUES (%s, %s)",
(filename, int(uID)))
cnx.commit()
# save the image file
destination = "/".join([target, filename])
tmp_dest = "/".join([tmp_target, filename])
file.save(tmp_dest)
file.seek(0)
s3.put_object(Key=destination, Body=file, ACL='public-read')
# get pID of the new image
cursor.execute("SELECT pID FROM images WHERE pName = (%s)",
(filename))
x = cursor.fetchone()
pID = x[0]
# apply image transformation
for i in range(3):
# give a tpID to the transformed image
# cursor.execute("SELECT max(tpID) FROM trimages")
# x = cursor.fetchone()
# if x[0] == None:
# tpID = 1
# else:
# tpID = x[0] + 1
# give a file name to the transformed image
tfilename = escape_string("tr" + str(i) + "_" + filename)
# insert the image info into the database
cursor.execute(
"INSERT INTO trimages (tpName, images_pID) VALUES (%s, %s)",
(tfilename, int(pID)))
cnx.commit()
# apply transformation
img = Image(filename=tmp_dest)
with img.clone() as tfile:
image_transfer(tfile, i)
# save the image file
tmp_tdest = "/".join([tmp_target, tfilename])
tfile.save(filename=tmp_tdest)
tdestination = "/".join([target, tfilename])
s3.put_object(Key=tdestination,
Body=open(tmp_tdest, 'rb'),
ACL='public-read')
# database commit, cleanup and garbage collect
shutil.rmtree(tmp_target)
cursor.close()
cnx.close()
gc.collect()
flash("upload successful")
return redirect(url_for("gallery"))
return render_template("image-upload.html", title="upload images")
except Exception as e:
if os.path.isdir(tmp_target):
shutil.rmtree(tmp_target)
teardown_db(e)
return str(e)
def start_scanning(file_path):
tool = pyocr.get_available_tools()[0]
req_image = []
final_text = []
path_to_pdf = file_path
file_name = path_to_pdf.split('/')[-1]
print(file_name)
print('This is path to pdf: ' + path_to_pdf)
path_to_new_pdf = 'scanned_pdf_for_email'
image_pdf = Image(filename=path_to_pdf, resolution=300)
image_jpeg = image_pdf.convert('jpeg')
for img in image_jpeg.sequence:
img_page = Image(image=img)
req_image.append(img_page.make_blob('jpeg'))
for img in req_image:
txt = tool.image_to_string(PI.open(io.BytesIO(img)),
lang='eng',
builder=pyocr.builders.TextBuilder())
final_text.append(txt)
with open('./output_text_file/listfile.txt', 'w') as f:
for item in final_text:
f.write('%s\n' % item)
company_name = []
sender_names = ['triumph']
found_company = ''
list_text_file = []
with open('./output_text_file/listfile.txt', 'r') as f:
list_text_file = f.read().split()
list_to_string = ' '.join(map(str, list_text_file))
# uding regex to get all caps words
result = filter(
None, [x.strip() for x in re.findall(r"\b[A-Z\s]+\b", list_to_string)])
print(*result) # printing the list using * operator separated by space
# values = keywords(text=list_to_string, split='\n', scores=True)
# data = pd.DataFrame(values, columns=['keyword','score'])
# data = data.sort_values('score', ascending=False)
# print(data.head(10))
print('this is text in string: ' + list_to_string)
# r = Rake()
# r.extract_keywords_from_text(list_to_string)
# phrases = r.get_ranked_phrases_with_scores()
# table = pd.DataFrame(phrases, columns=['score', 'Phrase'])
# table = table.sort_values('score', ascending=False)
# print(table.head(10))
#print(list_text_file)
for word in list_text_file:
if word.lower() in company_name:
found_company = word.lower()
break
subject_str = str(found_company.capitalize()) + ' mail' + '.pdf'
print(subject_str)
#os.rename(path_to_pdf, path_to_new_pdf + subject_str)
copy('./scanned_pdf_files/' + file_name, path_to_new_pdf)
os.rename(path_to_new_pdf + '/' + file_name,
path_to_new_pdf + '/' + subject_str)
def test_getitem_slice(slice_name, slice_, fx_asset):
with Image(filename=str(fx_asset.join('apple.ico'))) as img:
assert list(img.sequence[slice_]) == list(img.sequence)[slice_]
def data_parse(msg_dict):
# print('###')
logging.propagate = False
logging.getLogger().setLevel(logging.ERROR)
# print(msg_dict.get('Subject'))
if not msg_dict.get('Subject')[0] == '您':
logging.info('# tr filter ')
return
# print(type(msg_dict.get('Subject')))
ticket_no = msg_dict.get('Subject').split(' ')[-1]
# file_name = 'Itinerary - ' + ticket_no + '.pdf'
file_name = 'Itinerary – ' + ticket_no + '.pdf'
# file_name = 'Itinerary - V44DFI.pdf'
# print(file_name)
try:
data = msg_dict.get('attach').get(file_name)
if not data:
return
with open('parse/tr_data/test.pdf', 'wb') as f:
f.write(data)
tools = pyocr.get_available_tools()
if len(tools) == 0:
print("No OCR tool found")
sys.exit(1)
tool = tools[0]
print("Will use tool '%s'" % (tool.get_name()))
print(tool.get_available_languages())
langs = tool.get_available_languages()
print("Available languages: %s" % ", ".join(langs))
lang = langs[0]
print("Will use lang '%s'" % (lang))
req_image = []
final_text = []
image_pdf = Image(filename="parse/tr_data/test.pdf", resolution=400)
image_jpeg = image_pdf.convert('jpeg')
image_jpeg.save(filename='parse/tr_data/test.jpeg')
img_page = Image(image=image_jpeg.sequence[0])
req_image.append(img_page.make_blob('jpeg'))
# for img in image_jpeg.sequence:
# img_page = Image(image=img)
# req_image.append(img_page.make_blob('jpeg'))
img = req_image[0]
txt = tool.image_to_string(PI.open(io.BytesIO(img)),
lang=lang,
builder=pyocr.builders.TextBuilder())
# 全部数据
final_text.append(txt)
# for img in req_image:
# txt = tool.image_to_string(
# PI.open(io.BytesIO(img)),
# lang=lang,
# builder=pyocr.builders.TextBuilder()
# )
# final_text.append(txt)
# print(final_text)
a = 0
for text in final_text[0].replace(' ', '').replace(' ',
'').split(' '):
# print('#' * 66, a)
# print(text)
a = a + 1
print('long: ', len(final_text))
print('# 107', final_text)
flight_number_list = re.findall(r'\((.*?)\)', final_text[0])
flight_number = ''
for i in flight_number_list:
if i[0:2] == 'TR':
if flight_number == '':
flight_number = i
else:
flight_number = flight_number + ',' + i
print('flight_number: ', flight_number)
gender_dict = {
'先生': 'M',
'女士': 'F',
'夫人': 'F',
}
name_str_list = final_text[0].split('Umber')
gender = ''
name_list = []
p = 0
name_str = ''
# 处理乘客姓名
for i in name_str_list[1:]:
# print('# i: ', i, p)
# p = p + 1
# 列表，全部名字
name_str = i.replace('\n', ' ').replace(' ', '').replace(
' ', '').replace('-', '').replace('“',
' ').strip().split(' ')
# name_str = i.replace(' ', '').replace(' ', '').replace('-', '').strip().split(' ')
print('# 121 namestr: ', name_str)
count = 0
# 单人
for j in range(len(name_str)):
# print('### 131: ', name_str[j], len(name_str[j]))
if len(name_str[j]) > 8:
# print(name_str[j])
if name_str[j][:-2] == 'KrisFlyer':
# print('###################### KrisFlyer')
count = j
break
if name_str[j] == '' or name_str[j] == '-' or name_str[j][-2:] == 'kg' or name_str[j][
:-2] == 'KrisFlyer' or \
name_str[j] == ' ' or name_str[j] == '-' or len(name_str[j]) == 0:
name_str.remove(name_str[j])
if len(name_str[j]) > 20:
count = j
break
# ['女士', 'THANYALAK', '-', 'SRIJUMPA', '先生', 'YONG', 'JIN', 'CRUZ']
name_str = name_str[:count]
# print('### 139: ', name_str, len(name_str))
# # 多人
# for j in range(len(name_str)):
# if name_str[j] == '' or name_str[j] == '-' or len(name_str[j]) == 0:
# name_str.remove(name_str[j])
# if len(name_str[j]) > 20:
# count = j
# break
name_str = name_str[:count]
print('### 148: ', name_str, len(name_str))
# 不统计返程
break
# 去空
for i in name_str:
if len(i) == 0:
name_str.remove(i)
temp_name_list = []
temp_a = []
temp_b = []
# 获取乘客姓名
for i in range(len(name_str)):
# _temp
if i != 0 and name_str[i] in gender_dict:
for j in name_str[i:]:
temp_b.append(j)
break
temp_a.append(name_str[i])
temp_name_list.append(temp_a)
if len(temp_b) != 0:
temp_name_list.append(temp_b)
print('# 179', temp_name_list)
name = ''
for i in temp_name_list:
if name == '':
for j in i[1:-1]:
name = name + j
name = i[-1] + '/' + name
gender = gender_dict.get(i[0])
else:
temp_name = ''
for j in i[1:-1]:
temp_name = temp_name + j
name = name + ',' + i[-1] + '/' + temp_name
gender = gender + ',' + gender_dict.get(i[0])
print('# name', name, gender)
# 出发时间
name_str_list = name_str_list[0].replace('\n', ' ').replace(
' ', '').replace(' ',
'').replace('-',
'').replace('“',
' ').strip().split(' ')
print('# 221 deptime ', name_str_list)
dep_time = ''
dep_date = ''
ret_time = 0
ret_date = ''
for i in range(len(name_str_list)):
if name_str_list[i] == '值机时间':
if dep_time != '':
ret_time = name_str_list[i - 1].split(')')[1]
# break
if dep_time == '':
dep_time = name_str_list[i - 1].split(')')[1]
if name_str_list[i] == 'Fare':
# if dep_date == '':
# dep_date = name_str_list[i - 1].replace('年', '-').replace('月', '-').replace('日', '')
# ret_date = name_str_list[i - 1].replace('年', '-').replace('月', '-').replace('日', '')
# print('# 226 ',name_str_list[i-1])
if dep_date != '':
ret_date = name_str_list[i - 1].replace('年', '-').replace(
'月', '-').replace('日', '')
break
dep_date = name_str_list[i - 1].replace('年', '-').replace(
'月', '-').replace('日', '')
# print('# deptime: ', dep_date, dep_time, ret_date, ret_time)
dep_time = time.mktime(
time.strptime(dep_date + dep_time, '%Y-%m-%d%H:%M'))
flight_type = 1
if ret_date != '' and ret_time != '':
ret_time = time.mktime(
time.strptime(ret_date + ret_time, '%Y-%m-%d%H:%M'))
flight_type = 2
# ticket_no = 123
data = {
"name": name.strip(',').replace(' ', '').upper(),
"dep_time": dep_time,
"ret_time": ret_time,
"flight_number": flight_number,
"ticketNo": ticket_no,
"addtime": time.time(),
"flight_type": flight_type,
"gender": gender.strip(','),
"email": msg_dict['To'][1:-1].lower()
# "email": 123
}
except:
print(traceback.print_exc())
return
print('# data: ', data)
return data
def pdf_preview(tmp_file_path, tmp_dir):
if use_generic_pdf_cover:
return None
else:
if use_PIL:
try:
input1 = PdfFileReader(open(tmp_file_path, 'rb'), strict=False)
page0 = input1.getPage(0)
xObject = page0['/Resources']['/XObject'].getObject()
for obj in xObject:
if xObject[obj]['/Subtype'] == '/Image':
size = (xObject[obj]['/Width'], xObject[obj]['/Height'])
data = xObject[obj]._data # xObject[obj].getData()
mode = "P"
if xObject[obj]['/ColorSpace'] == '/DeviceRGB':
mode = "RGB"
if xObject[obj]['/ColorSpace'] == '/DeviceCMYK':
mode = "CMYK"
if '/Filter' in xObject[obj]:
if xObject[obj]['/Filter'] == '/FlateDecode':
img = Image.frombytes(mode, size, data)
cover_file_name = os.path.splitext(tmp_file_path)[0] + ".cover.png"
img.save(filename=os.path.join(tmp_dir, cover_file_name))
return cover_file_name
# img.save(obj[1:] + ".png")
elif xObject[obj]['/Filter'] == '/DCTDecode':
cover_file_name = os.path.splitext(tmp_file_path)[0] + ".cover.jpg"
img = open(cover_file_name, "wb")
img.write(data)
img.close()
if mode == 'CMYK':
img2 = Image.open(cover_file_name)# .convert('RGB')
img2 = CMYKInvert(img2)
img2.save(cover_file_name)
return cover_file_name
elif xObject[obj]['/Filter'] == '/JPXDecode':
cover_file_name = os.path.splitext(tmp_file_path)[0] + ".cover.jp2"
img = open(cover_file_name, "wb")
img.write(data)
img.close()
if mode == 'CMYK':
img2 = Image.open(cover_file_name)# .convert('RGB')
img2 = CMYKInvert(img2)
img2.save(cover_file_name)
return cover_file_name
else:
img = Image.frombytes(mode, size, data)
cover_file_name = os.path.splitext(tmp_file_path)[0] + ".cover.png"
img.save(filename=os.path.join(tmp_dir, cover_file_name))
return cover_file_name
# img.save(obj[1:] + ".png")
except Exception as ex:
print(ex)
try:
cover_file_name = os.path.splitext(tmp_file_path)[0] + ".cover.jpg"
with Image(filename=tmp_file_path + "[0]", resolution=150) as img:
img.compression_quality = 88
img.save(filename=os.path.join(tmp_dir, cover_file_name))
return cover_file_name
except PolicyError as ex:
log.warning('Pdf extraction forbidden by Imagemagick policy: %s', ex)
return None
except Exception as ex:
log.warning('Cannot extract cover image, using default: %s', ex)
return None
def test_length(fx_asset):
with Image(filename=str(fx_asset.join('apple.ico'))) as img:
assert len(img.sequence) == 4
def process(path, height_each):
image = Image(filename=path)
crop_square_center(image)
image.resize(width=1040, height=960)
return image
def test_validate_position_error(fx_asset):
with Image(filename=str(fx_asset.join('apple.ico'))) as img:
with raises(TypeError):
img.sequence.validate_position(0.001)
mid_url = "http://dba.bn-ent.net/character/images/" + midchar + "/portrait.png"
right_url = "http://dba.bn-ent.net/character/images/" + rightchar + "/portrait.png"
character_distance_multiplier = 0.5
character_height_multiplier = 3
character_distance_multiplier = float(sys.argv[4])
character_height_multiplier = float(sys.argv[5])
left_char_bigger = 0
mid_char_bigger = 0
right_char_bigger = 0
mid_char_bigger = int(float(sys.argv[6]))
with Image(filename=bg_url) as img:
character_height = int(img.height * character_height_multiplier)
with Image(filename=mid_url) as mid:
if mid_char_bigger == 1:
resizeImgToHeight(mid, int(character_height * 1.2))
else:
resizeImgToHeight(mid, character_height)
with Image(filename=left_url) as left:
resizeImgToHeight(left, character_height)
with Image(filename=right_url) as right:
resizeImgToHeight(right, character_height)
with Image(filename="http://i.imgur.com/QvZ9Ce0.png"
) as watermark:
#print(img.size)
#print(left.size)
def test_validate_slice_error(fx_asset):
with Image(filename=str(fx_asset.join('apple.ico'))) as img:
with raises(ValueError):
img.sequence.validate_slice(slice(0, 10, 3))
import sys
from wand.image import Image
if len(sys.argv)<2:
print('{0} <ORIGINAL PATH>'.format(sys.argv[0]))
sys.exit()
original_path=sys.argv[1]
#이미지 파일을 Image 객체로 열 때는 생성자의 filename 매개변수에 파일 경로를 입력.
with Image(filename=original_path) as image:
print("Original : {0}, {1}".format(image.format, image.size))
#clone()은 메모리상에 복제본을 생성합니다.
clone=image.clone()
print("Clone {0}, {1}".format(clone.format, clone.size))
#save()는 Image 객체의 내용을 이미지 파일에 저장합니다.
clone.save(filename='clone.{0}'.format(original_path))
def test_container(fx_asset):
with Image(filename=str(fx_asset.join('apple.ico'))) as img:
for single in img.sequence:
assert single.container is img
def rendertask(task):
global args
# in debug mode we have no thread-worker which prints its progress
if debug:
print("generating {0} from {1}".format(task.outfile, task.infile))
if args.skip_frames and 'only_rerender_frames_after' not in task.parameters:
if os.path.isdir(os.path.join(task.workdir, '.frames')):
shutil.rmtree(os.path.join(task.workdir, '.frames'))
# make sure a .frames-directory exists in out workdir
ensurePathExists(os.path.join(task.workdir, '.frames'))
# open and parse the input file
with open(os.path.join(task.workdir, task.infile), 'r') as fp:
svgstr = fp.read()
for key in task.parameters.keys():
svgstr = svgstr.replace(key, xmlescape(str(task.parameters[key])))
parser = etree.XMLParser(huge_tree=True)
svg = etree.fromstring(svgstr.encode('utf-8'), parser)
# if '$subtitle' in task.parameters and task.parameters['$subtitle'] == '':
# child = svg.findall(".//*[@id='subtitle']")[0]
# child.getparent().remove(child)
# frame-number counter
frameNr = 0
# iterate through the animation seqence frame by frame
# frame is a ... tbd
cache = {}
for frame in task.sequence(task.parameters):
skip_rendering = False
# skip first n frames, to speed up rerendering during debugging
if 'only_rerender_frames_after' in task.parameters:
skip_rendering = (frameNr <=
task.parameters['only_rerender_frames_after'])
if args.skip_frames:
skip_rendering = (frameNr <= args.skip_frames)
if args.only_frame:
skip_rendering = (frameNr != args.only_frame)
# print a line for each and every frame generated
if debug and not skip_rendering:
print("frameNr {0:3d} => {1}".format(frameNr, frame))
frame = tuple(frame)
if frame in cache:
if debug:
print("cache hit, reusing frame {0}".format(cache[frame]))
framedir = task.workdir + "/.frames/"
shutil.copyfile("{0}/{1:04d}.png".format(framedir, cache[frame]),
"{0}/{1:04d}.png".format(framedir, frameNr))
frameNr += 1
continue
elif not skip_rendering:
cache[frame] = frameNr
# apply the replace-pairs to the input text, by finding the specified xml-elements by thier id and modify thier css-parameter the correct value
for replaceinfo in frame:
(id, type, key, value) = replaceinfo
for el in svg.findall(".//*[@id='" + id.replace("'", "\\'") +
"']"):
if type == 'style':
style = cssutils.parseStyle(
el.attrib['style'] if 'style' in el.attrib else '')
style[key] = str(value)
el.attrib['style'] = style.cssText
elif type == 'attr':
el.attrib[key] = str(value)
elif type == 'text':
el.text = str(value)
if not skip_rendering:
# open the output-file (named ".gen.svg" in the workdir)
with open(os.path.join(task.workdir, '.gen.svg'), 'w') as fp:
# write the generated svg-text into the output-file
fp.write(etree.tostring(svg, encoding='unicode'))
if task.outfile.endswith('.ts') or task.outfile.endswith('.mov'):
width = 1920
height = 1080
else:
width = 1024
height = 576
if args.imagemagick:
# invoke imagemagick to convert the generated svg-file into a png inside the .frames-directory
infile = '{0}/.gen.svg'.format(task.workdir)
outfile = '{0}/.frames/{1:04d}.png'.format(
task.workdir, frameNr)
with Image(filename=infile) as img:
with img.convert('png') as converted:
converted.save(filename=outfile)
else:
# invoke inkscape to convert the generated svg-file into a png inside the .frames-directory
cmd = 'cd {0} && inkscape --export-background=white --export-background-opacity=0 --export-width={2} --export-height={3} --export-png=$(pwd)/.frames/{1:04d}.png $(pwd)/.gen.svg 2>&1 >/dev/null'.format(
task.workdir, frameNr, width, height)
errorReturn = subprocess.check_output(cmd,
shell=True,
universal_newlines=True,
stderr=subprocess.STDOUT)
if errorReturn != '':
print("inkscape exitted with error\n" + errorReturn)
# sys.exit(42)
# increment frame-number
frameNr += 1
if args.only_frame:
task.outfile = '{0}.frame{1:04d}.png'.format(task.outfile,
args.only_frame)
# remove the dv/ts we are about to (re-)generate
ensureFilesRemoved(os.path.join(task.workdir, task.outfile))
if task.outfile.endswith('.png'):
cmd = 'cd {0} && cp ".frames/{1:04d}.png" "{2}"'.format(
task.workdir, args.only_frame, task.outfile)
# invoke avconv aka ffmpeg and renerate a lossles-dv from the frames
# if we're not in debug-mode, suppress all output
elif task.outfile.endswith('.ts'):
cmd = 'cd {0} && '.format(task.workdir)
cmd += 'ffmpeg -f image2 -i .frames/%04d.png '
if task.audiofile is None:
cmd += '-ar 48000 -ac 1 -f s16le -i /dev/zero -ar 48000 -ac 1 -f s16le -i /dev/zero '
else:
cmd += '-i {0} -i {0} '.format(task.audiofile)
cmd += '-map 0:0 -c:v mpeg2video -q:v 2 -aspect 16:9 '
if task.audiofile is None:
cmd += '-map 1:0 -map 2:0 '
else:
cmd += '-map 1:0 -c:a copy -map 2:0 -c:a copy '
cmd += '-shortest -f mpegts "{0}"'.format(task.outfile)
elif task.outfile.endswith('.mov'):
cmd = 'cd {0} && '.format(task.workdir)
cmd += 'ffmpeg -f lavfi -i anullsrc=channel_layout=stereo:sample_rate=44100 -f image2 -i .frames/%04d.png -r 25 -shortest -c:v qtrle -f mov "{0}"'.format(
task.outfile)
else:
cmd = 'cd {0} && ffmpeg -ar 48000 -ac 2 -f s16le -i /dev/zero -f image2 -i .frames/%04d.png -target pal-dv -aspect 16:9 -shortest "{1}"'.format(
task.workdir, task.outfile)
if debug:
print(cmd)
r = os.system(cmd + ('' if debug else '>/dev/null 2>&1'))
# as before, in non-debug-mode the thread-worker does all progress messages
if debug:
if r != 0:
sys.exit()
if not debug:
print("cleanup")
# remove the generated svg
ensureFilesRemoved(os.path.join(task.workdir, '.gen.svg'))
def test_delay(fx_asset):
with Image(filename=str(fx_asset.join('nocomments-delay-100.gif'))) as img:
for s in img.sequence:
assert s.delay == 100
for j, label in enumerate(cluster.label):
print(label)
selection.append(label)
os.system(f"cp assets/{label}* {cluster_dir}")
column = j + 1
plt.subplot(1, span, column)
plt.axis("off")
fpath = f"../coloured/{label}.png"
plt.imshow(io.imread(fpath))
selections.append(selection)
plt.savefig(f"results/cluster_{cluster_id}.png")
# Trim montages in place
run_paths = list(Path("results").glob("cluster_?.png"))
for source in run_paths:
with Image(filename=source.as_posix()) as img:
img.trim()
img.save(filename=source.as_posix())
i = 0
# choose three random images from each cluster
for s in selections:
a = random.sample(s, 3)
print(a)
span = 3
plt.subplots(1, span) #, figsize=(span, 4))
rnd = []
for j, label in enumerate(a):
print(label)
rnd.append(label)
def test_delitem_not_loaded(fx_asset):
with Image(filename=str(fx_asset.join('apple.ico'))) as img:
del img.sequence[1]
assert len(img.sequence) == 3
def draw_bg(width, height, filename):
with Image(width=width, height=height, background=Color('white')) as img:
img.save(filename=filename)
def convert_pdf_to_png(event, context):
"""
Simple use of ImageMagick to convert a 1-n page PDF into images in groups of 3 pages.
Takes in b64 encoded PDF files and turns them into b64 encoded PNG files to be sent to Google
Vision OCR. Print statements are used for CloudWatch logging.
:param event: JSON data POST'd to endpoint.
:param context: Unused param in all lambda functions.
:return: list of zlibcompressed/b64encoded PNG data.
"""
GROUP_SIZE = 3
img_data = []
pdf_data = event.get('pdf_bin')
if not pdf_data:
return {'error': 'No PDF data passed'}
decoded = b64decode(pdf_data)
del pdf_data
print 'Decoded'
#Creates a single large Image object from the PDF data.
with Image(blob=decoded, resolution=300) as pdf:
"""
pdf.sequence returns an iterable ie pictures of each page from the pdf.
When the iterable is 3 or less:
((<wand.sequence.SingleImage: dc1d561 (2470x3281)>, None, None))
When the iterable is more than 3:
(
(<wand.sequence>, (<wand.sequence>, (<wand.sequence>),
(<wand.sequence>, None, None)
)
"""
del decoded
grouped = grouper(pdf.sequence, GROUP_SIZE)
for group in grouped:
# Creates a new 'blank' Image that is smaller than the above pdf context.
trimmed = [item for item in group if item]
with Image(width=pdf.width, height=pdf.height * len(trimmed)) as image:
for i, sequence in enumerate(trimmed):
image.composite(
sequence,
top=pdf.height * i,
left=0
)
del trimmed
print 'Image composed'
blob = image.make_blob('png')
del image
print 'Blob made'
compressed = compress(blob)
del blob
print 'Compressed'
encoded = b64encode(compressed)
del compressed
print 'Encoded'
img_data.append(encoded)
del encoded
timeouts = xrange(45, 75)
new_timeout = random.choice(timeouts)
lambda_client.update_function_configuration(
FunctionName = 'imageMagick',
Timeout=new_timeout
)
return {'img_data': img_data}
