SinTan1729/TvTimeToTrakt
1
0
Fork 0
mirror of https://github.com/SinTan1729/TvTimeToTrakt.git synced 2024-12-25 04:48:38 -06:00
Code Issues Projects Releases Packages Wiki Activity

Merge pull request #1 from BapRx/fix/tvtime-import

Browse source 
Fix/tvtime import
This commit is contained in:
Baptiste 2022-02-19 14:16:22 +01:00 committed by GitHub
commit f33b2bfb2e
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
2 changed files with 213 additions and 164 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

375
TimeToTrackt.py
View file

@ -1,37 +1,55 @@
# main.py
from logging import error
import sys
from trakt import *
import trakt.core
import os
#!/usr/bin/env python3
import csv
from datetime import datetime
import time
from tinydb import TinyDB, Query
import json
import logging
import os
import re
import sys
import time
from datetime import datetime
from pathlib import Path
import trakt.core
from tinydb import Query, TinyDB
from trakt import init
from trakt.tv import TVShow
# Setup logger
logging.basicConfig(
format="%(asctime)s [%(levelname)7s] :: %(message)s",
level=logging.INFO,
datefmt="%Y-%m-%d %H:%M:%S",
)
# Adjust this value to increase/decrease your requests between episodes.
# Make sure it's above 1 seconds to remain within the rate limit.
DELAY_BETWEEN_EPISODES_IN_SECONDS = 5
# Make to remain within the rate limit: https://trakt.docs.apiary.io/#introduction/rate-limiting
DELAY_BETWEEN_EPISODES_IN_SECONDS = 0.75
# Create a database to keep track of completed processes
database = TinyDB('localStorage.json')
syncedEpisodesTable = database.table('SyncedEpisodes')
userMatchedShowsTable = database.table('TvTimeTraktUserMatched')
database = TinyDB("localStorage.json")
syncedEpisodesTable = database.table("SyncedEpisodes")
userMatchedShowsTable = database.table("TvTimeTraktUserMatched")
class Expando(object):
pass
def isAuthenticated():
with open(f"{Path.home()}/.pytrakt.json") as f:
data = json.load(f)
daysBeforeExpiration = (
datetime.fromtimestamp(data["OAUTH_EXPIRES_AT"]) - datetime.now()
).days
if daysBeforeExpiration < 1:
return False
return True
def getConfiguration():
configEx = Expando()
with open('config.json') as f:
with open("config.json") as f:
data = json.load(f)
configEx.TRAKT_USERNAME = data["TRAKT_USERNAME"]
@ -58,9 +76,17 @@ def getFollowedShowsPath():
def initTraktAuth():
if isAuthenticated():
return True
# Set the method of authentication
trakt.core.AUTH_METHOD = trakt.core.OAUTH_AUTH
return init(config.TRAKT_USERNAME, store=True, client_id=config.CLIENT_ID, client_secret=config.CLIENT_SECRET)
return init(
config.TRAKT_USERNAME,
store=True,
client_id=config.CLIENT_ID,
client_secret=config.CLIENT_SECRET,
)
# With a given title, check if it contains a year (e.g Doctor Who (2005))
# and then return this value, with the title and year removed to improve
@ -75,18 +101,19 @@ def getYearFromTitle(title):
yearSearch = re.search(r"\(([A-Za-z0-9_]+)\)", title)
yearValue = yearSearch.group(1)
# Then, get the title without the year value included
titleValue = title.split('(')[0].strip()
titleValue = title.split("(")[0].strip()
# Put this together into an object
ex.titleWithoutYear = titleValue
ex.yearValue = int(yearValue)
return ex
except:
except Exception:
# If the above failed, then the title doesn't include a year
# so return the object as is.
ex.titleWithoutYear = title
ex.yearValue = -1
return ex
# Shows in TV Time are often different to Trakt.TV - in order to improve results and automation,
# calculate how many words are in the title, and return true if more than 50% of the title is a match,
# It seems to improve automation, and reduce manual selection....
@ -116,6 +143,7 @@ def checkTitleNameMatch(tvTimeTitle, traktTitle):
# then return the title as a possibility to use
return percentage > 50
# Using TV Time data (Name of Show, Season No and Episode) - find the corresponding show
# in Trakt.TV either by automation, or asking the user to confirm.
@ -145,7 +173,7 @@ def getShowByName(name, seasonNo, episodeNo):
if checkTitleNameMatch(name, show.title):
# If the title included the year of broadcast, then we can be more picky in the results
# to look for a show with a broadcast year that matches
if doesTitleIncludeYear == True:
if doesTitleIncludeYear:
# If the show title is a 1:1 match, with the same broadcast year, then bingo!
if (name == show.title) and (show.year == titleObj.yearValue):
# Clear previous results, and only use this one
@ -168,7 +196,7 @@ def getShowByName(name, seasonNo, episodeNo):
if nameFromSearch.title == name:
completeMatchNames.append(nameFromSearch)
if (len(completeMatchNames) == 1):
if len(completeMatchNames) == 1:
showsWithSameName = completeMatchNames
# If the search contains multiple results, then we need to confirm with the user which show
@ -178,8 +206,7 @@ def getShowByName(name, seasonNo, episodeNo):
# Query the local database for existing selection
userMatchedQuery = Query()
queryResult = userMatchedShowsTable.search(
userMatchedQuery.ShowName == name)
queryResult = userMatchedShowsTable.search(userMatchedQuery.ShowName == name)
# If the local database already contains an entry for a manual selection
# then don't bother prompting the user to select it again!
@ -187,12 +214,12 @@ def getShowByName(name, seasonNo, episodeNo):
# Get the first result from the query
firstMatch = queryResult[0]
# Get the value contains the selection index
firstMatchSelectedIndex = int(firstMatch.get('UserSelectedIndex'))
firstMatchSelectedIndex = int(firstMatch.get("UserSelectedIndex"))
# Check if the user previously requested to skip the show
skipShow = firstMatch.get('SkipShow')
skipShow = firstMatch.get("SkipShow")
# If the user did not skip, but provided an index selection, get the
# matching show
if skipShow == False:
if not skipShow:
return showsWithSameName[firstMatchSelectedIndex]
# Otherwise, return None, which will trigger the script to skip
# and move onto the next show
@ -202,22 +229,25 @@ def getShowByName(name, seasonNo, episodeNo):
# then prompt the user to make a selection
else:
print(
f"INFO - MANUAL INPUT REQUIRED: The TV Time data for Show '{name}' (Season {seasonNo}, Episode {episodeNo}) has {len(showsWithSameName)} matching Trakt shows with the same name.")
f"INFO - MANUAL INPUT REQUIRED: The TV Time data for Show '{name}' (Season {seasonNo}, Episode {episodeNo}) has {len(showsWithSameName)} matching Trakt shows with the same name."
)
# Output each show for manual selection
for idx, item in enumerate(showsWithSameName):
# Display the show's title, broadcast year, amount of seasons and a link to the Trakt page.
# This will provide the user with enough information to make a selection.
print(
f" ({idx + 1}) {item.title} - {item.year} - {len(item.seasons)} Season(s) - More Info: https://trakt.tv/{item.ext}")
f" ({idx + 1}) {item.title} - {item.year} - {len(item.seasons)} Season(s) - More Info: https://trakt.tv/{item.ext}"
)
while(True):
while True:
try:
# Get the user's selection, either a numerical input, or a string 'SKIP' value
indexSelected = (input(
f"Please make a selection from above (or enter SKIP):"))
indexSelected = input(
"Please make a selection from above (or enter SKIP):"
)
if indexSelected != 'SKIP':
if indexSelected != "SKIP":
# Since the value isn't 'skip', check that the result is numerical
indexSelected = int(indexSelected) - 1
# Exit the selection loop
@ -229,17 +259,19 @@ def getShowByName(name, seasonNo, episodeNo):
except KeyboardInterrupt:
sys.exit("Cancel requested...")
# Otherwise, the user has entered an invalid value, warn the user to try again
except:
print(
f"Sorry! Please select a value between 0 to {len(showsWithSameName)}")
except Exception:
logging.error(
f"Sorry! Please select a value between 0 to {len(showsWithSameName)}"
)
# If the user entered 'SKIP', then exit from the loop with no selection, which
# will trigger the program to move onto the next episode
if (indexSelected == 'SKIP'):
if indexSelected == "SKIP":
# Record that the user has skipped the TV Show for import, so that
# manual input isn't required everytime
userMatchedShowsTable.insert(
{'ShowName': name, 'UserSelectedIndex': 0, 'SkipShow': True})
{"ShowName": name, "UserSelectedIndex": 0, "SkipShow": True}
)
return None
# Otherwise, return the selection which the user made from the list
@ -247,18 +279,24 @@ def getShowByName(name, seasonNo, episodeNo):
selectedShow = showsWithSameName[int(indexSelected)]
userMatchedShowsTable.insert(
{'ShowName': name, 'UserSelectedIndex': indexSelected, 'SkipShow': False})
{
"ShowName": name,
"UserSelectedIndex": indexSelected,
"SkipShow": False,
}
)
return selectedShow
else:
if (len(showsWithSameName) > 0):
if len(showsWithSameName) > 0:
# If the search returned only one result, then awesome!
# Return the show, so the import automation can continue.
return showsWithSameName[0]
else:
return None
# Since the Trakt.Py starts the indexing of seasons in the array from 0 (e.g Season 1 in Index 0), then
# subtract the TV Time numerical value by 1 so it starts from 0 as well. However, when a TV series includes
# a 'special' season, Trakt.Py will place this as the first season in the array - so, don't subtract, since
@ -279,7 +317,7 @@ def parseSeasonNo(seasonNo, traktShowObj):
# Otherwise, if the Trakt seasons start with no specials, then return the seasonNo,
# but subtracted by one (e.g Season 1 in TV Time, will be 0)
else:
# Only subtract is the TV Time season number is greater than 0.
# Only subtract if the TV Time season number is greater than 0.
if seasonNo != 0:
return seasonNo - 1
# Otherwise, the TV Time season is a special! Then you don't need to change the starting position
@ -291,157 +329,161 @@ def processWatchedShows():
# Total amount of rows which have been processed in the CSV file
rowsCount = 0
# Total amount of rows in the CSV file
rowsTotal = 0
# Total amount of errors which have occurred in one streak
errorStreak = 0
# Get the total amount of rows in the CSV file,
# which is helpful for keeping track of progress.
# However, if you have a VERY large CSV file (e.g above 100,000 rows)
# then it might be a good idea to remove this due to the performance
# overhead.
with open(getWatchedShowsPath()) as f:
rowsTotal = sum(1 for line in f)
# Open the CSV file within the GDPR exported data
with open(getWatchedShowsPath(), newline='') as csvfile:
with open(getWatchedShowsPath(), newline="") as csvfile:
# Create the CSV reader, which will break up the fields using the delimiter ','
showsReader = csv.reader(csvfile, delimiter=',')
showsReader = csv.DictReader(csvfile, delimiter=",")
# Get the total amount of rows in the CSV file,
rowsTotal = len(list(showsReader))
# Move position to the beginning of the file
csvfile.seek(0, 0)
# Loop through each line/record of the CSV file
for row in showsReader:
# Increment the row counter to keep track of progress completing the
# records during the import process.
rowsCount += 1
# Ignore the header row
next(showsReader, None)
for rowsCount, row in enumerate(showsReader):
# Get the name of the TV show
tvShowName = row[8]
tvShowName = row["tv_show_name"]
# Get the TV Time Episode Id
tvShowEpisodeId = row["episode_id"]
# Get the TV Time Season Number
tvShowSeasonNo = row["episode_season_number"]
# Get the TV Time Episode Number
tvShowEpisodeNo = row["episode_number"]
# Get the date which the show was marked 'watched' in TV Time
tvShowDateWatched = row["updated_at"]
# Parse the watched date value into a Python type
tvShowDateWatchedConverted = datetime.strptime(
tvShowDateWatched, "%Y-%m-%d %H:%M:%S"
)
# Ignore the header row
if rowsCount > 1:
# Get the TV Time Episode Id
tvShowEpisodeId = row[4]
# Get the TV Time Season Number
tvShowSeasonNo = row[5]
# Get the TV Time Episode Number
tvShowEpisodeNo = row[6]
# Get the date which the show was marked 'watched' in TV Time
tvShowDateWatched = row[7]
# Parse the watched date value into a Python type
tvShowDateWatchedConverted = datetime.strptime(
tvShowDateWatched, '%Y-%m-%d %H:%M:%S')
# Query the local database for previous entries indicating that
# the episode has already been imported in the past. Which will
# ease pressure on TV Time's API server during a retry of the import
# process, and just save time overall without needing to create network requests
episodeCompletedQuery = Query()
queryResult = syncedEpisodesTable.search(
episodeCompletedQuery.episodeId == tvShowEpisodeId
)
# Query the local database for previous entries indicating that
# the episode has already been imported in the past. Which will
# ease pressure on TV Time's API server during a retry of the import
# process, and just save time overall without needing to create network requests
episodeCompletedQuery = Query()
queryResult = syncedEpisodesTable.search(
episodeCompletedQuery.episodeId == tvShowEpisodeId)
# If the query returned no results, then continue to import it into Trakt
if len(queryResult) == 0:
# Create a repeating loop, which will break on success, but repeats on failures
while True:
# If more than 10 errors occurred in one streak, whilst trying to import the episode
# then give up, and move onto the next episode, but warn the user.
if (errorStreak > 10):
print(
f"WARNING: An error occurred 10 times in a row... skipping episode...")
# If the query returned no results, then continue to import it into Trakt
if len(queryResult) == 0:
# Create a repeating loop, which will break on success, but repeats on failures
while True:
# If more than 10 errors occurred in one streak, whilst trying to import the episode
# then give up, and move onto the next episode, but warn the user.
if errorStreak > 10:
logging.warning(
"An error occurred 10 times in a row... skipping episode..."
)
break
try:
# Sleep for a second between each process, before going onto the next watched episode.
# This is required to remain within the API rate limit, and use the API server fairly.
# Other developers share the service, for free - so be considerate of your usage.
time.sleep(DELAY_BETWEEN_EPISODES_IN_SECONDS)
# Search Trakt for the TV show matching TV Time's title value
traktShowObj = getShowByName(
tvShowName, tvShowSeasonNo, tvShowEpisodeNo
)
# If the method returned 'None', then this is an indication to skip the episode, and
# move onto the next one
if traktShowObj is None:
break
try:
# Sleep for a second between each process, before going onto the next watched episode.
# This is required to remain within the API rate limit, and use the API server fairly.
# Other developers share the service, for free - so be considerate of your usage.
time.sleep(DELAY_BETWEEN_EPISODES_IN_SECONDS)
# Search Trakt for the TV show matching TV Time's title value
traktShowObj = getShowByName(
tvShowName, tvShowSeasonNo, tvShowEpisodeNo)
# If the method returned 'None', then this is an indication to skip the episode, and
# move onto the next one
if traktShowObj == None:
break
# Show the progress of the import on-screen
print(
f"({rowsCount}/{rowsTotal}) Processing Show {tvShowName} on Season {tvShowSeasonNo} - Episode {tvShowEpisodeNo}")
# Get the season from the Trakt API
season = traktShowObj.seasons[parseSeasonNo(
tvShowSeasonNo, traktShowObj)]
# Get the episode from the season
episode = season.episodes[int(tvShowEpisodeNo) - 1]
# Mark the episode as watched!
episode.mark_as_seen(tvShowDateWatchedConverted)
# Add the episode to the local database as imported, so it can be skipped,
# if the process is repeated
syncedEpisodesTable.insert(
{'episodeId': tvShowEpisodeId})
# Clear the error streak on completing the method without errors
errorStreak = 0
break
# Catch errors which occur because of an incorrect array index. This occurs when
# an incorrect Trakt show has been selected, with season/episodes which don't match TV Time.
# It can also occur due to a bug in Trakt Py, whereby some seasons contain an empty array of episodes.
except IndexError:
print(
f"({rowsCount}/{rowsTotal}) WARNING: {tvShowName} Season {tvShowSeasonNo}, Episode {tvShowEpisodeNo} does not exist (season/episode index) in Trakt!")
break
# Catch any errors which are raised because a show could not be found in Trakt
except trakt.errors.NotFoundException:
print(
f"({rowsCount}/{rowsTotal}) WARNING: {tvShowName} Season {tvShowSeasonNo}, Episode {tvShowEpisodeNo} does not exist (search) in Trakt!")
break
# Catch errors because of the program breaching the Trakt API rate limit
except trakt.errors.RateLimitException:
print(
"WARNING: The program is running too quickly and has hit Trakt's API rate limit! Please increase the delay between " +
"episdoes via the variable 'DELAY_BETWEEN_EPISODES_IN_SECONDS'. The program will now wait 60 seconds before " +
"trying again.")
time.sleep(60)
# Show the progress of the import on-screen
logging.info(
f"({rowsCount+1}/{rowsTotal}) - Processing '{tvShowName}' Season {tvShowSeasonNo} / Episode {tvShowEpisodeNo}"
)
# Get the season from the Trakt API
season = traktShowObj.seasons[
parseSeasonNo(tvShowSeasonNo, traktShowObj)
]
# Get the episode from the season
episode = season.episodes[int(tvShowEpisodeNo) - 1]
# Mark the episode as watched!
episode.mark_as_seen(tvShowDateWatchedConverted)
# Add the episode to the local database as imported, so it can be skipped,
# if the process is repeated
syncedEpisodesTable.insert({"episodeId": tvShowEpisodeId})
# Clear the error streak on completing the method without errors
errorStreak = 0
break
# Catch errors which occur because of an incorrect array index. This occurs when
# an incorrect Trakt show has been selected, with season/episodes which don't match TV Time.
# It can also occur due to a bug in Trakt Py, whereby some seasons contain an empty array of episodes.
except IndexError:
tvShowSlug = traktShowObj.to_json()["shows"][0]["ids"]["ids"][
"slug"
]
logging.warning(
f"({rowsCount}/{rowsTotal}) - {tvShowName} Season {tvShowSeasonNo}, Episode {tvShowEpisodeNo} does not exist in Trakt! (https://trakt.tv/shows/{tvShowSlug}/seasons/{tvShowSeasonNo}/episodes/{tvShowEpisodeNo})"
)
break
# Catch any errors which are raised because a show could not be found in Trakt
except trakt.errors.NotFoundException:
logging.warning(
f"({rowsCount}/{rowsTotal}) - {tvShowName} Season {tvShowSeasonNo}, Episode {tvShowEpisodeNo} does not exist (search) in Trakt!"
)
break
# Catch errors because of the program breaching the Trakt API rate limit
except trakt.errors.RateLimitException:
logging.warning(
"The program is running too quickly and has hit Trakt's API rate limit! Please increase the delay between "
+ "episdoes via the variable 'DELAY_BETWEEN_EPISODES_IN_SECONDS'. The program will now wait 60 seconds before "
+ "trying again."
)
time.sleep(60)
# Mark the exception in the error streak
errorStreak += 1
# Catch a JSON decode error - this can be raised when the API server is down and produces a HTML page, instead of JSON
except json.decoder.JSONDecodeError:
print(
f"({rowsCount}/{rowsTotal}) WARNING: A JSON decode error occuring whilst processing {tvShowName} " +
f"Season {tvShowSeasonNo}, Episode {tvShowEpisodeNo}! This might occur when the server is down and has produced " +
"a HTML document instead of JSON. The script will wait 60 seconds before trying again.")
# Mark the exception in the error streak
errorStreak += 1
# Catch a JSON decode error - this can be raised when the API server is down and produces a HTML page, instead of JSON
except json.decoder.JSONDecodeError:
logging.warning(
f"({rowsCount}/{rowsTotal}) - A JSON decode error occuring whilst processing {tvShowName} "
+ f"Season {tvShowSeasonNo}, Episode {tvShowEpisodeNo}! This might occur when the server is down and has produced "
+ "a HTML document instead of JSON. The script will wait 60 seconds before trying again."
)
# Wait 60 seconds
time.sleep(60)
# Wait 60 seconds
time.sleep(60)
# Mark the exception in the error streak
errorStreak += 1
# Catch a CTRL + C keyboard input, and exits the program
except KeyboardInterrupt:
sys.exit("Cancel requested...")
# Skip the episode
else:
print(
f"({rowsCount}/{rowsTotal}) Skipping '{tvShowName}' Season {tvShowSeasonNo} Episode {tvShowEpisodeNo}. It's already been imported.")
# Mark the exception in the error streak
errorStreak += 1
# Catch a CTRL + C keyboard input, and exits the program
except KeyboardInterrupt:
sys.exit("Cancel requested...")
# Skip the episode
else:
logging.info(
f"({rowsCount}/{rowsTotal}) - Already imported, skipping '{tvShowName}' Season {tvShowSeasonNo} / Episode {tvShowEpisodeNo}."
)
def start():
# Create the initial authentication with Trakt, before starting the process
if initTraktAuth():
# Display a menu selection
print(f">> What do you want to do?")
print(f" 1) Import Watch History from TV Time")
print(">> What do you want to do?")
print(" 1) Import Watch History from TV Time")
while True:
try:
menuSelection = int(input(f"Enter your menu selection: "))
menuSelection = input("Enter your menu selection: ")
menuSelection = 1 if not menuSelection else int(menuSelection)
break
except ValueError:
print("Invalid input. Please enter a numerical number.")
logging.warning("Invalid input. Please enter a numerical number.")
# Start the process which is required
if menuSelection == 1:
# Invoke the method which will import episodes which have been watched
# from TV Time into Trakt
processWatchedShows()
else:
print("Sorry - that's an unknown menu selection")
logging.warning("Sorry - that's an unknown menu selection")
else:
print("ERROR: Unable to complete authentication to Trakt - please try again.")
logging.error(
"ERROR: Unable to complete authentication to Trakt - please try again."
)
if __name__ == "__main__":
@ -451,7 +493,12 @@ if __name__ == "__main__":
if os.path.isdir(config.GDPR_WORKSPACE_PATH):
start()
else:
print("Oops! The TV Time GDPR folder '" + config.GDPR_WORKSPACE_PATH +
"' does not exist on the local system. Please check it, and try again.")
logging.error(
"Oops! The TV Time GDPR folder '"
+ config.GDPR_WORKSPACE_PATH
+ "' does not exist on the local system. Please check it, and try again."
)
else:
print(f"ERROR: The 'config.json' file cannot be found - have you created it yet?")
logging.error(
"The 'config.json' file cannot be found - have you created it yet?"
)

2
requirements.txt Normal file
View file

@ -0,0 +1,2 @@
trakt==3.4.0
tinydb==4.6.1