def youtube_detector(timestamp):
# Downloading all the YouTube Sources
for vid in ytSources:
prettyPrint(
bcolors.ENDC,
os.popen("youtube-dl -o \"{}\" -f mp4 {}".format(
fullPath("data/rawVideos/%(id)s.mp4"), fullPath(vid))).read())
# Editing each of the videos
for targetVid in os.listdir("data/rawVideos"):
cut_video(targetVid)
def cut_video(video_path):
scenes = find_scenes("data/rawVideos/{}".format(video_path))
prettyPrint(
bcolors.OKBLUE,
"{0} number of scenes detected: {1}".format(video_path, len(scenes)))
# Finally splits the video into clips
split_video(
video_path,
scenes,
"data/rawClips/{}\\Scene.mkv".format(video_path),
video_path,
suppress_output=False,
)
def newUpload():
with open(fullPath("data/auth.json")) as jsonfile:
auth = json.load(jsonfile)
opts = webdriver.ChromeOptions()
opts.binary_location = "/usr/bin/google-chrome"
opts.accept_untrusted_certs = True
opts.assume_untrusted_cert_issuer = True
for arg in arguments:
opts.add_argument(arg)
driver = webdriver.Chrome(executable_path="data/browser/chromedriver",
options=opts)
# Authenticate into Youtube Studio
driver.get(
"https://accounts.google.com/o/oauth2/v2/auth/oauthchooseaccount?redirect_uri=https%3A%2F%2Fdevelopers.google.com%2Foauthplayground&prompt=consent&response_type=code&client_id=407408718192.apps.googleusercontent.com&scope=email&access_type=offline&flowName=GeneralOAuthFlow"
)
# Get to upload screen
CREATE = driver.find_element_by_id("create-icon").click()
UPLOAD = driver.find_element_by_id("text-item-0").click()
content = os.listdir(fullPath("data/output"))[0]
selectFiles = driver.find_element_by_id("select-files-button").click()
selectFiles.send_keys(content)
# Publish video to youtube
title = "{} memes for {}".format(random.choice(titles),
names.get_first_name())
titleBox = driver.find_element_by_id("textbox").send_keys(title)
nextButon = driver.find_element_by_id("next-button")
nextButon.click()
nextButon.click()
# Finally uploads the video
doneButton = driver.find_element_by_id("done-button").click()
driver.close()
prettyPrint(bcolors.OKGREEN,
"Video finished uploading, closing connection")
def VideoFormatter(target):
targetPath = fullPath("data/rawClips/{}.mp4".format(target))
outputPath = fullPath("data/sync_data/{}.mp4".format(target))
prettyPrint(bcolors.OKBLUE,
"Formatting data/rawClips/{}.mp4".format(target))
# This is a really long command so here are a bunch of comments explaining it
# Step #1: FFMPEG makes two copies, [original] and [copy]
# Step #2: Take [copy] and scale it to 16/9 so all videos are the same size, and apply a blur effect to it
# Step #3: Append [original] on top of [copy] and output the finished video to data/sync_data
prettyPrint(
bcolors.ENDC,
os.popen(
"ffmpeg -i {} -vf 'split[original][copy];[copy]scale=ih*16/9:-1,crop=h=iw*9/16,gblur=sigma=20[blurred];[blurred][original]overlay=(main_w-overlay_w)/2:(main_h-overlay_h)/2' -video_track_timescale 29971 -ac 1 {}"
.format(targetPath, outputPath)).read())
# Checks to make sure the video are valid files
try:
float(
os.popen(
"ffprobe -v error -show_entries format=duration -of default=noprint_wrappers=1:nokey=1 {}"
.format(outputPath)).read())
except Exception as e:
prettyPrint(
bcolors.WARNING,
"Non-viable file, file will be removed. Error code {}".format(e))
os.remove(outputPath)
# Cleans up the files and exist the thread
os.remove(targetPath)
sys.exit("Exiting the thread")
def split_video(input_video_paths,
scene_list,
output_file_template,
video_name,
suppress_output=False):
if not input_video_paths or not scene_list:
return
logging.info(
'Splitting input video%s using mkvmerge, output path template:\n %s',
's' if len(input_video_paths) > 1 else '', output_file_template)
ret_val = None
# mkvmerge automatically appends '-$SCENE_NUMBER'.
output_file_name = output_file_template.replace('-${SCENE_NUMBER}', '')
output_file_name = output_file_template.replace('-$SCENE_NUMBER', '')
output_file_template = Template(output_file_name)
output_file_name = output_file_template.safe_substitute(
VIDEO_NAME=video_name, SCENE_NUMBER='')
# Actually make the system call to MKVMerge
try:
call_list = ['mkvmerge']
call_list += [
'-o', output_file_name, '--split',
'parts:%s' % ','.join([
'%s-%s' % (start_time.get_timecode(), end_time.get_timecode())
for start_time, end_time in scene_list
]), "data/rawVideos/{}".format(input_video_paths)
]
total_frames = scene_list[-1][1].get_frames(
) - scene_list[0][0].get_frames()
processing_start_time = time.time()
ret_val = invoke_command(call_list)
print('')
prettyPrint(
bcolors.OKBLUE,
('Average processing speed %.2f frames/sec.', float(total_frames) /
(time.time() - processing_start_time)))
# Basically just some error handling
except CommandTooLong:
prettyPrint(bcolors.WARNING,
"COMMAND_TOO_LONG_STRING: {}".format(call_list))
except OSError:
prettyPrint(
bcolors.WARNING, 'mkvmerge could not be found on the system.'
' Please install mkvmerge to enable video output support.')
if ret_val is not None and ret_val != 0:
prettyPrint(bcolors.WARNING,
('Error splitting video (mkvmerge returned %d).', ret_val))
def reddit_detector(timestamp, source):
# Gets the creds from disk
with open(fullPath("data/auth.json")) as jsonfile:
auth = json.load(jsonfile)
# Configures PRAW creds
prettyPrint(bcolors.OKBLUE, "Authenticating provided credentials")
redditAuth = praw.Reddit(client_id=auth['reddit']['client_id'],
client_secret=auth['reddit']['client_secret'],
user_agent=auth['reddit']['user_agent'],
username=auth['reddit']['username'],
password=auth['reddit']['password'])
prettyPrint(bcolors.OKBLUE,
"Checking and downloading content from {}".format(source))
# Generates a list of moderators, this allows us to filter out their posts
mods = []
for moderator in redditAuth.subreddit(source).moderator():
mods.append(str(moderator))
# Downloads all of the videos
for targetSubmission in redditAuth.subreddit(source).hot():
if not (os.path.exists(
fullPath("data/rawClips/{}.mp4".format(
targetSubmission.id)))) and not (os.path.exists(
fullPath("data/rawClips/{}.mp4".format(
targetSubmission.id)))):
if not (str(targetSubmission.author) in mods):
prettyPrint(
bcolors.OKBLUE,
"Thread {} is now downloading {} from {} on Reddit.".
format(threading.currentThread().ident,
targetSubmission.id, source))
prettyPrint(
bcolors.ENDC,
os.popen("youtube-dl --no-continue -o \"{}\" {}".format(
fullPath("data/rawClips/{}.mp4".format(
targetSubmission.id)),
targetSubmission.url)).read())
fT = threading.Thread(target=VideoFormatter,
args=(targetSubmission.id, ))
fT.start()
def detectorDirector():
# Reddit Detector
startTime = time.perf_counter()
prettyPrint(bcolors.OKBLUE,
"Checking reddit sources and downloading any new content")
redditThreads = []
for source in redditSources:
t = threading.Thread(target=reddit_detector,
args=(datetime.utcnow(), source))
redditThreads.append(t)
t.start()
prettyPrint(bcolors.OKBLUE,
"All threads finished, closing all active processes")
for threads in redditThreads:
threads.join()
# A little alert to let the user know the function is over and how long it took
endTime = time.perf_counter()
prettyPrint(
bcolors.OKGREEN,
f"Reddit sources checked and new content downloaded. Process took {endTime - startTime:0.4f} seconds"
)
def GenerateVideo():
# Gets minimum length of video (between 11 and 14 minutes)
videoLength = random.randint(660, 841)
prettyPrint(
bcolors.OKBLUE, "Expected video length is {}".format(
datetime.timedelta(seconds=videoLength)))
# A list of which clips to use and how long the video is currently
clipPaths = []
totalLength = 0
# Randomly selecting clips to add until the video is long enough
while (totalLength < videoLength):
newChoice = fullPath("data/sync_data/{}".format(
random.choice(os.listdir(fullPath("data/sync_data")))))
if newChoice not in clipPaths:
try:
totalLength += float(
os.popen(
"ffprobe -v error -show_entries format=duration -of default=noprint_wrappers=1:nokey=1 {}"
.format(newChoice)).read())
clipPaths.append(newChoice)
except Exception as e:
prettyPrint(
bcolors.WARNING,
"Exception thrown, continuing with process. Exception: {}".
format(e))
# Just some output for the user
prettyPrint(
bcolors.OKBLUE,
"Actual video length is {} and contains {} videos. All file paths are listed below"
.format(datetime.timedelta(seconds=int(totalLength)), len(clipPaths)))
# Logs the video files for future reference
logString = ""
spacing = len(str(datetime.datetime.now())) + 4
for path in clipPaths:
logString += "{} || ".format(path)
logString += "\n"
for x in range(spacing):
logString += " "
prettyPrint(bcolors.ENDC, logString)
# Edits the video together
startTime = time.perf_counter()
# Creates a list of the input files for FFMPEG
with open(fullPath("inputPaths.txt"), "a") as inputPathFile:
ouputString = ""
for inputPath in clipPaths:
ouputString += "file '{}' \n".format(inputPath)
inputPathFile.write(ouputString)
# Finally edits the clips togeather
vidIndex = len(os.listdir(fullPath("/data/output"))) + 1
prettyPrint(
bcolors.ENDC,
os.popen("ffmpeg -f concat -safe 0 -i {} -c copy {}".format(
fullPath("inputPaths.txt"),
fullPath("data/output/video_{}.mp4".format(vidIndex)))).read())
os.remove(fullPath("inputPaths.txt"))
# A little alert to let the user know the function is over and how long it took
endTime = time.perf_counter()
prettyPrint(
bcolors.OKGREEN,
f"Collected clips and edited video. Process took {endTime - startTime:0.4f} seconds"
)