def update_iptc(tmp_img, orig_iptc):
info = IPTCInfo(tmp_img, force=True)
valids = ['keywords', 'caption/abstract', 'country/primary location name', 'city', 'sub-location',
'credit', 'copyright notice', 'writer/editor', 'by-line', 'headline']
for k in orig_iptc:
if k in valids:
info[k] = orig_iptc[k]
print('saving new copy with IPTC tags:', tmp_img)
info.save_as(tmp_img, {'overwrite': True})
def test_save_as_saves_as_new_file_with_new_info():
if os.path.isfile('fixtures/deleteme.jpg'): # pragma: no cover
os.unlink('fixtures/deleteme.jpg')
new_headline = b'test headline %d' % random.randint(0, 100)
info = IPTCInfo('fixtures/Lenna.jpg')
info['headline'] = new_headline
info.save_as('fixtures/deleteme.jpg')
info2 = IPTCInfo('fixtures/deleteme.jpg')
assert info2['headline'] == new_headline
def test_save_as_saves_as_new_file_with_info():
if os.path.isfile('fixtures/deleteme.jpg'): # pragma: no cover
os.unlink('fixtures/deleteme.jpg')
info = IPTCInfo('fixtures/Lenna.jpg')
info.save_as('fixtures/deleteme.jpg')
info2 = IPTCInfo('fixtures/deleteme.jpg')
# The files won't be byte for byte exact, so filecmp won't work
assert info._data == info2._data
with open('fixtures/Lenna.jpg', 'rb') as fh, open('fixtures/deleteme.jpg',
'rb') as fh2:
start, end, adobe = jpeg_collect_file_parts(fh)
start2, end2, adobe2 = jpeg_collect_file_parts(fh2)
# But we can compare each section
assert start == start2
assert end == end2
assert adobe == adobe2
def get_date(DATADIR_RAW, DATADIR_RAW_NEW):
'''Function wich takes as input path of dirs. in those dirs the function will irreterate throug all jpg, extract a date from their meta date and it a IPTC protocol'''
# get the image name from the dir DATADIR_RAW
for img in os.listdir(DATADIR_RAW):
image_path = os.path.join(DATADIR_RAW, img)
info = IPTCInfo(image_path, force=True)
print(img)
# Open the image and access the metadate. Use regex to extract date:
image = Image.open(image_path)
txt = str(image.info['exif']) # exif are mostly JPEG
r = '[\d]{4}:[\d]{2}:[\d]{2}' # format yyyy:mm:dd
date = re.search(r, txt).group().replace(':', '')
print(f'date of picture: {date}')
# Force open the IPTC protocol to insert date
info = IPTCInfo(image_path, force=True)
print(f"info before: {info['date created']}"
) #Before. Shows wheter or not the entry was empty before
info['date created'] = date
print(f"info after: {info['date created']}"
) # after. Shows the date extracted from the meta data
print('\n')
# Create the new dir DATADIR_RAW_NEW if it does not already exist
if not os.path.exists(DATADIR_RAW_NEW):
os.makedirs(DATADIR_RAW_NEW)
# Create the new path for the image
new_path = os.path.join(DATADIR_RAW_NEW, 'new' + img)
# If that image is already in the path, delete it.
if os.path.exists(DATADIR_RAW_NEW):
try:
os.remove(new_path)
except:
pass
# Save new image.
info.save_as(new_path)
def write_keywords(keywords, output_file):
info = IPTCInfo(output_file, force=True, inp_charset='utf8')
info['keywords'] = list(keywords)
info.save_as(output_file)
from iptcinfo3 import IPTCInfo
def remove_prefix(text, prefix):
if text.startswith(prefix):
return text[len(prefix):]
return text
def write_keywords(keywords, output_file):
info = IPTCInfo(output_file, force=True, inp_charset='utf8')
info['keywords'] = list(keywords)
info.save_as(output_file)
if __name__ == '__main__':
input_dir = sys.argv[1]
if not os.path.isdir(input_dir):
raise ValueError("Input directory path is not correct")
for filename in os.listdir(input_dir):
info = IPTCInfo(input_dir + "/" + filename,
force=True,
inp_charset='utf_8')
new_keywords = set()
for each in info['keywords']:
new_keywords.add(remove_prefix(each, " ").lower())
info['keywords'] = list(new_keywords)
info.save_as(input_dir + "2/" + filename)
print()
