def robot():
try:
print('\n>>> [Image Robot] Starting...')
content = state.load()
content = fetch_images_of_all_sentences(content)
content = download_all_images(content)
state.save(content)
print('>>> [Image Robot] Stopping. Work done!')
except Exception as e:
print(f'\n> [Image Robot] Unexpected Error: {e}\n')
exit_video_pymaker()
def robot():
try:
print('\n>>> [YouTube Robot] Starting...')
content = state.load()
service = get_authenticated_service()
video_id = upload_video(content, service)
upload_thumbnail(video_id, service)
state.save(content)
print('>>> [YouTube Robot] Stopping. Work done!')
except Exception as e:
print(f'\n> [YouTube Robot] Unexpected Error: {e}\n')
exit_video_pymaker()
def robot():
""" Text Robot
This function gives life to the text robot
"""
print("Pre loading data from JSON...")
content = load()
fetch_content_from_wikipedia(content)
clear_content(content)
break_content_in_sentences(content)
limit_max_sentences(content)
fetch_sentences_keywords(content)
save(content.__dict__)
def robot():
try:
print('\n>>> [Video Robot] Starting...')
content = state.load()
convert_all_images(content)
create_all_sentences_images(content)
create_youtube_thumbnail()
create_video_initial_images(content)
create_video_with_moviepy()
state.save(content)
print('>>> [Video Robot] Stopping. Work done!')
except Exception as e:
print(f'\n> [Video Robot] Unexpected Error: {e}\n')
exit_video_pymaker()
def robot():
try:
print('\n>>> [Text Robot] Starting...')
content = state.load()
content = fetch_content_from_wikipedia(content)
content = sanitize_content(content)
content = break_content_into_sentences(content)
content = fetch_keywords_of_all_sentences(content)
state.save(content)
print('>>> [Text Robot] Stopping. Work done!')
except Algorithmia.errors.AlgorithmException as algorithm_exception:
print(
f'\n> [Text Robot] Error: video-pymaker was unable to fetch Wikipedia with this search term:',
f'{content.search_term}.')
print(f'> [Text Robot] See Algorithmia error below: ')
print(f'> [Text Robot] Algorithmia Error: {algorithm_exception}')
exit_video_pymaker()
except Exception as e:
print(f'\n> [Text Robot] Unexpected Error: {e}\n')
exit_video_pymaker()
def robot():
current_directory = os.path.dirname(__file__)
content = load()
def google_search(search_term, api_key, cse_id, **kwargs):
service = build("customsearch", "v1", developerKey=api_key)
response = service.cse().list(q=search_term, cx=cse_id,
**kwargs).execute()
return response['items']
def fetch_google_and_return_images_links(query):
images_url = []
results = google_search(query,
settings.GOOGLE_SEARCH_API_KEY,
settings.SEARCH_ENGINE_ID,
searchType="image",
imgSize="huge",
num=2)
for result in results:
images_url.append(result['link'])
return images_url
def fetch_images_of_all_sentences(content):
print("> Getting image URLs from Google Images...")
for sentence in content.sentences:
query = f"{ content.search_term } { sentence.keywords[0] }"
sentence.images = fetch_google_and_return_images_links(query)
sentence.google_search_query = query
def download_and_save_image(image_url, file_name):
response = requests.get(image_url, stream=True)
response.raise_for_status()
response.raw.decode_content = True # Required to decompress gzip/deflate compressed responses
with Image.open(response.raw) as img:
img.save(f"{ current_directory }/../static/{ file_name }")
def download_all_images(content):
downloaded_images = []
sentence_index = 0
print("> Downloading images...")
for sentence in content.sentences:
images = sentence.images
for image in images:
image_url = image
try:
if image_url in str(downloaded_images):
raise FileExistsError("Image already downloaded!")
download_and_save_image(image_url,
f"{ sentence_index }_raw.png")
downloaded_images.append(image_url)
print(f"Image download successfuly! URL: { image_url }")
break
except Exception as error:
print(
f"Error: Couldn't download image. URL: { image_url } - { error }"
)
sentence_index += 1
content.downloaded_images = downloaded_images
fetch_images_of_all_sentences(content)
download_all_images(content)
save(content.__dict__)
def robot():
current_directory = os.path.dirname(__file__)
content = load()
def convert_image(sentence_index):
img_original = False
input_image_path = f"{ current_directory }/../static/{ sentence_index }_raw.png"
output_image_path = f"{ current_directory }/../static/{ sentence_index }_converted.png"
width = 1920
height = 1080
try:
img_original = Image.open(input_image_path).convert('RGBA')
except FileNotFoundError as err:
print(
f"Image '{ sentence_index }_raw.png' doesn't exist on file system"
)
print(input_image_path)
if img_original:
try:
img_resized = img_original.resize((width, height))
img_resized.filter(
ImageFilter.GaussianBlur(30)).save(output_image_path)
print(">Successfuly converted")
except AttributeError as err:
print(
f"Couldn't convert image '{ sentence_index }_raw.png': { err }"
)
try:
img_original.thumbnail((height, height))
img_w, img_h = img_original.size
background_img = Image.open(output_image_path).convert('RGBA')
bkg_w, bkg_h = background_img.size
offset = ((bkg_w - img_w) // 2, (bkg_h - img_h) // 2)
background_img.paste(img_original, offset, img_original)
background_img.save(output_image_path)
print(">Successfuly merged")
except AttributeError as err:
print(
f"Couldn't blur and merge image '{ sentence_index }_raw.png': { err }"
)
def convert_all_images(content):
print("> Transforming images...")
for index in range(len(content.sentences)):
convert_image(index)
def create_sentence_image(sentence_index, sentence_text):
img_path = f"{ current_directory }/../static/{ sentence_index }_converted.png"
output_image_path = f"{ current_directory }/../static/{ sentence_index }_sentence.png"
img = False
# It wraps a whole string into separated lines as a list for every 80 chars long
text_lines = textwrap.wrap(sentence_text, width=75)
try:
img = Image.open(img_path)
except FileNotFoundError as err:
print(
f"Image '{ sentence_index }_raw.png' doesn't exist on file system: { err }"
)
if img:
try:
img_w, img_h = img.size
# Template for positioning text on image using 2 different positions for every index iteration
template = {
'0': {
'position': (100, 50)
},
'1': {
'position': (100, img_h - 350)
},
'2': {
'position': (100, 50)
},
'3': {
'position': (100, img_h - 350)
},
'4': {
'position': (100, 50)
},
'5': {
'position': (100, img_h - 350)
},
'6': {
'position': (100, 50)
},
'7': {
'position': (100, img_h - 350)
}
}
draw = ImageDraw.Draw(img)
font = ImageFont.truetype("arial.ttf", 48)
position = template[str(sentence_index)]['position']
width_pos, height_pos = position
shadow_offset = 2
for line in text_lines:
font_width, font_height = font.getsize(line)
# For every line of the sentence add the corresponding font size height value to the
# Height position of the text, so it doesn't overlap on the past added line
# Draws text into the image (position tuple, text, color and font)
draw.text((width_pos + shadow_offset,
height_pos + shadow_offset),
line, (0, 0, 0),
font=font)
draw.text((width_pos - shadow_offset,
height_pos - shadow_offset),
line, (0, 0, 0),
font=font)
draw.text((width_pos + shadow_offset,
height_pos - shadow_offset),
line, (0, 0, 0),
font=font)
draw.text((width_pos - shadow_offset,
height_pos + shadow_offset),
line, (0, 0, 0),
font=font)
draw.text((width_pos, height_pos),
line, (255, 255, 255),
font=font)
height_pos += font_height
img.save(output_image_path)
except AttributeError:
print(
f"Couldn't add text to image { sentence_index }_converted.png"
)
def create_all_sentence_images(content):
print("> Drawing sentences text into related images...")
sentence_index = 0
for sentence in content.sentences:
create_sentence_image(sentence_index, sentence.text)
sentence_index += 1
def create_youtube_thumbnail():
img_path = f"{ current_directory }/../static/0_converted.png"
output_img = f"{ current_directory }/../static/youtube_thumbnail.jpg"
print("> Creating Youtube Thumbnail...")
try:
img = Image.open(img_path).convert('RGB')
img.save(output_img)
except FileNotFoundError as err:
print(f"Couldn't generate Youtube Thumbnail: { err }")
convert_all_images(content)
create_all_sentence_images(content)
create_youtube_thumbnail()