def __init__(self):
self.check = False
self.url = None
self.attempStartTimeStr = Easy.NowLocalStr()
self.workDir = None
self.dlFname = None
self.fname = None
self.txtFile = None
self.resultEx = None ## result obj from extracted info stage
self.resultDl = None ## result obj from dl stage
self.resultEntries = []
def _ytDownload(url,
location=None,
checkFunc=None,
resultsList=None,
maxEntries=-1):
"""
this is the lower level downloading func
it shouldn't be used directly
"""
dirAtFunStart = Easy.Cwd
if location is None:
location = Easy.Cwd
nowStr = Easy.NowLocalStr()
workDir = Easy.TmpDir(dir=location, prefix='01--working--' + nowStr)
Easy.Cd(workDir)
urlsExtarctedToDownload = []
if resultsList is None:
resultsList = []
ret = YtResult()
ret.attempStartTimeStr = nowStr
ret.url = url
ret.workDir = workDir
try:
## Check for early skip
if True: ##'v=' in url:
## easiest checks are for id as parts after 'v='
## or just as part after last slash
uEnd = url.split('/')[-1]
uEnd = uEnd.split('?v=')[-1]
uEnd = uEnd.split('&v=')[-1]
## if we've chopped off the end via slash or v=
## then if it's a simple video, it shouldn't have '?'
## in it
if checkFunc is not None and not '?' in uEnd:
if not checkFunc(uEnd):
#print( f'early skip: {url}' )
raise CheckerError('Early skip! check failed on id: ' +
str(uEnd))
tmpUrl = url
if '&' in url:
urlSplit = url.split('&')
loopI = len(urlSplit) - 1
while loopI >= 0:
listPart = urlSplit[loopI]
if loopI > 0 and urlSplit[loopI].startswith('list='):
del urlSplit[loopI]
loopI -= 1
tmpUrl = '&'.join(urlSplit)
print(tmpUrl)
ydl_opts = {
#'merge_output_format':'mkv',
#'extract_flat':True,
'extractaudio': True,
'noplaylist': True,
#'audio_only':True,
'restrict_filenames': True,
#'no_overwrites':True,
#'no_post_overwrites':True,
'keepvideo': False,
#'extract-audio' : True,
'quiet': False,
#'noplaylist':True,
#'get_filename':True,
#'progress_hooks': [my_hook],
#def my_hook(d):
}
ydl = youtube_dl.YoutubeDL(ydl_opts)
#pdb.set_trace()
result = ydl.extract_info(
tmpUrl, download=False) ## tmpUrl to ensure not a list!
#pdb.set_trace()
ret.resultEx = result
if 'entries' in result:
raise "handling of playlists/channels is disabled currently"
## old playlist handling code was here, moved to end of this file for reference
else:
fname = ydl.prepare_filename(result)
title = result['title']
id = result['id']
acodec = result['acodec']
ext = result['acodec']
if checkFunc is not None:
if not checkFunc(id):
raise CheckerError('Check failed on id: ' + str(id))
else:
ret.check = True
## This is a hack to prevent youtube_dl from running sys.exit
oldSysExit = sys.exit
def fakeExit(arg):
raise FakeExitError
sys.exit = fakeExit
try:
youtube_dl.main([
url,
'--extract-audio',
'--no-playlis',
])
except KeyboardInterrupt as err:
oldSysExit(1)
raise err
except FakeExitError:
'pass'
#print( 'sys.exit was prevented' )
sys.exit = oldSysExit
## if we were able to dl a file
## find name, fix name, move files
files = Easy.Ls(Easy.Cwd)
if not len(files):
print(f'Error, unable to download url: {url}')
else:
dlFname = files[0]
ret.dlFname = dlFname
#print( f'dlFname: {dlFname}' )
fixedName = Easy.Mods.re.sub(r'[^\x00-\x7f]', r'', dlFname)
if dlFname != fixedName:
os.rename(dlFname, fixedName) ## make Easy.Rename
os.rename(fixedName, os.path.join(location, fixedName))
ret.fname = fixedName
ret.txtFile = os.path.join(
location, fixedName) + '.alreadyDownloaded.txt'
if not os.path.exists(ret.txtFile):
with open(ret.txtFile, 'w') as txtFh:
import json
metadata = {}
metadata['date_of_dl'] = Easy.NowLocalStr()
metadata['title'] = title
json.dump(metadata, txtFh)
#txtFh.write( 'already downloaded' )
## This gets triggered if the checkFails,
## in which case most of the above code never runs
## since it would be after the CheckerError
except CheckerError as err:
pass
## Really important to handle keyboard errors! For quit with Ctrl-c
except KeyboardInterrupt as err:
raise err
## On most other errors we just continue after printing a warning
except:
Easy.PrintTraceback()
print('continuing anyway...')
finally:
## really important we go back to original directory!
Easy.Cd(dirAtFunStart)
try:
os.rmdir(workDir)
except:
## something is really wrong in this case, so we don't continue on
print(f"Couldn't delete tmp dir: {workDir}")
## regardless of whether our result has much useful data in it, we add the result
resultsList.append(ret)
## return the same list we were given, but modifed, with YtResult added
return resultsList
