################################################################################
################################################################################
## Canvas user page view retrieval                                            ##
##                                                                            ##
## Fill in the three variables directly below this header, then run with py   ##
## **Use at your own risk.  Please try changes in Test/Beta before Prod.**    ##
##                                                                            ##
## 2025 Christopher M. Casey (cmcasey79@hotmail.com)                          ##
################################################################################
################################################################################
script_version='2025.03.11.00002'

# Script mode setup
# set as '' for deault operation, 'unattended' to run without prompts using the info below, 'verbose' for more detailed screen progress output, 'silent' to run without screen output, or 'silentunattended' or 'verboseunattended' to combine the options.
mode=''

# Canvas API token setup
# Key should be omitted for versions sent to its, will be downoloaded from server later on if last three characters of script_version is 'its' and left blank here
canvas_api_token=''

# Canvas subdomain setup
# set this variable as your subdomain of instructure.com (ex: umich)
canvas_subdomain=''

# Canvas vanity domain setup
# set these variable as your relevant vanity domain(s) if you have them, otherwise, leave it as ''
canvas_production_vanity_domain=''
canvas_beta_vanity_domain=''
canvas_test_vanity_domain=''

# Canvas account id setup
# set this variable as the account/subaccount id you'd like to work in.
canvas_account_id=''

# Canvas environment setup
# set this variable as production, test, beta or leave as '' to prompt if not running in unattended mode
canvas_environment=''

################################################################################
################################################################################
## No changes to below code should be necessary.                              ##
################################################################################
################################################################################

import socket
import requests
import urllib3.exceptions
import json
import datetime
import csv
import collections
import re


scriptlog=''

#Added to suppress warnings
requests.packages.urllib3.disable_warnings(category=urllib3.exceptions.InsecureRequestWarning)

#Requests call with retry on server error
def requestswithretry(retries=3, backoff_factor=0.3, status_forcelist=(500, 502, 504), session=None, ):
    session = session or requests.Session()
    retry = requests.packages.urllib3.util.retry.Retry(
        total=retries,
        read=retries,
        connect=retries,
        backoff_factor=backoff_factor,
        status_forcelist=status_forcelist,
        allowed_methods=frozenset({'DELETE', 'GET', 'HEAD', 'OPTIONS', 'PUT', 'TRACE', 'POST'})
    )
    adapter = requests.adapters.HTTPAdapter(max_retries=retry)
    session.mount('http://', adapter)
    session.mount('https://', adapter)
    return session

# Paginated API call to Canvas
# Returns a tuple including list of all items from all pages of a paginated API call and the http status code and reason returned from the call(s)
# Last updated 2025-03-07
def canvas_get_allpages(url, headers):
    if not 'per_page=' in url:
        url=url+('&' if '?' in url else '?')+'per_page=100'
    return_data=[]
    return_key=None
    repeat=True
    while repeat:
        canvas_request=requestswithretry().get(url,headers=headers)
        if canvas_request.status_code!=200:
            return_data=None
            repeat=False
        else:
            canvas_request_responsedata=canvas_request.json()
            # if Canvas returned a single item dictionary instead of a list, try using the value of that dictionary as the list
            if type(canvas_request_responsedata) is dict:
                if len(canvas_request_responsedata.keys())==1:
                    return_key=list(canvas_request_responsedata.keys())[0]
                    canvas_request_responsedata=next(iter(canvas_request_responsedata.values()))
                else:
                    return_data=canvas_request_responsedata
                    repeat=False
            if type(canvas_request_responsedata) is list:
                # if a list was returned, add it to the existing list
                return_data+=canvas_request_responsedata
                url=canvas_request.links.get('current',{}).get('url',url)
                last_url=canvas_request.links.get('last',{}).get('url','')
                # if not on the last page of results, set up retrieval of next page
                if (url != last_url) and ('next' in canvas_request.links.keys()):
                    url=canvas_request.links['next']['url']
                else:
                    repeat=False
    return_data_namedtuple = collections.namedtuple('return_data_namedtuple', ['data', 'reason', 'status_code'])
    return return_data_namedtuple(return_data if (return_key==None) else {return_key:return_data},str(url)+': '+str(canvas_request.reason),canvas_request.status_code)

# Prompt and/or validate canvas envornment information.  Canvas_environment is a dict which will be populated by the function.
# Prompts and/or validates the working environment, domains, api token, account_id.
# Only validates the working environment by default, but an enviromnet list in the form of ['production','beta','test'] can be passed if validation of additional environments is needed.
# Returns true if everything was properly validated, false if some information could not be valitated (signaling the script should end)
# Last updated 2025-03-07.
def canvas_validate_environment(canvas_environment, validate_environments=None):
    global scriptlog
    if mode[0:6]!='silent': print('--------------------------------------------------------------------------------\n')
    if mode[-10:]=='unattended': scriptlog+='--------------------------------------------------------------------------------\n\n'
    # Validate Canvas environment selection and domains
    domain_validated=False
    while not domain_validated:
        # Prompt for Canvas environment selection
        env_selection=False
        while (not env_selection):
            if mode[-10:]=='unattended' or canvas_environment['working_environment']!='':
                env=canvas_environment['working_environment'][0]
            else:
                env=input('Which Canvas environment would you like to work in: [P]roduction, [T]est, or [B]eta? ')
            env_selection=True
            if env.lower()=='p':
                canvas_environment['working_environment']='production'
            elif env.lower()=='t':
                canvas_environment['working_environment']='test'
            elif env.lower()=='b':
                canvas_environment['working_environment']='beta'
            else:
                if mode[-10:]=='unattended':
                    print('Invalid canvas_environment selection.')
                else:
                    print('Invalid selection, please try again.')
                    env_selection=False
                canvas_environment['working_environment']=''
        if validate_environments==None:
            validate_environments=[canvas_environment['working_environment']]
        if canvas_environment['subdomain']=='' and mode[-10:]!='unattended':
            canvas_environment['subdomain']=input('Please enter your instructure.com subdomain (ex: umich): ')
        for environment in validate_environments:
            if canvas_environment['domains'][environment]=='':
                if mode[-10:]!='unattended':
                    canvas_environment['domains'][environment]=input(f'Please enter your {environment} Canvas vanity domain (hit enter/return if you do not have one): ')
                if canvas_environment['domains'][environment]=='':
                    canvas_environment['domains'][environment]=f'{canvas_environment['subdomain']}{'.'+environment if (environment!='' and environment!='production') else ''}.instructure.com'
        domain_validated=True
        for environment in validate_environments:
            environment_validated=False
            url=f'https://{canvas_environment['domains'][environment]}'
            if mode[0:6]!='silent': print(url)
            try:
                r = requests.get(url)
            except requests.exceptions.RequestException as e:
                print(f'{datetime.datetime.now().isoformat()}: Connection error {e}')
                scriptlog+=f'{datetime.datetime.now().isoformat()}: Connection error {e}\n'
            else:
                if r.status_code==200:
                    if mode[0:6]!='silent': print(f'{datetime.datetime.now().isoformat()}: Connection test to {canvas_environment['domains'][environment]} successful, proceeding...')
                    if mode[-10:]=='unattended': scriptlog+=f'{datetime.datetime.now().isoformat()}: Connection test to {canvas_environment['domains'][environment]} successful, proceeding...\n'
                    environment_validated=True
                elif r.status_code==401:       
                    if mode[0:6]!='silent': print(f'{datetime.datetime.now().isoformat()}: Connection test to {canvas_environment['domains'][environment]} successful but not validated, vanity domain or authentication may be needed, proceeding...')
                    if mode[-10:]=='unattended': scriptlog+=f'{datetime.datetime.now().isoformat()}: Connection test to {canvas_environment['domains'][environment]} successful but not validated, vanity domain or authentication may be needed, proceeding...\n'
                    environment_validated=True
                else:
                    print(f'{datetime.datetime.now().isoformat()}: Error {r.status_code} - connection to {canvas_environment['domains'][environment]} failed.')
                    if mode[-10:]=='unattended': scriptlog+=f'{datetime.datetime.now().isoformat()}: Error {r.status_code} - connection to {canvas_environment['domains'][environment]} failed.\n'
            if not environment_validated:
                if canvas_environment['domains'][environment].split('.')[0]==canvas_environment['subdomain']:
                    canvas_environment['subdomain']=''
                canvas_environment['domains'][environment]=''
            domain_validated=domain_validated and environment_validated
        domain_validated=domain_validated or mode[-10:]=='unattended'
    # Validate API Token
    apitoken_validated=False
    while (not apitoken_validated) and (canvas_environment['domains'][environment]!=''):
        # Prompt for API Token if not given
        if canvas_environment['api_token']=='' and mode[-10:]!='unattended':
            canvas_environment['api_token']=str(input('Please enter your Canvas administrative API token: '))
        #Set HTTP headers for API calls
        url=f'https://{canvas_environment['domains'][canvas_environment['working_environment']]}/api/v1/accounts'
        canvas_accounts = canvas_get_allpages(url,headers={'Authorization' : f'Bearer {canvas_environment['api_token']}'})
        if canvas_accounts.status_code==200:
            canvas_environment['url_headers']={'Authorization' : f'Bearer {canvas_environment['api_token']}'}
            canvas_accounts_dict={}
            for canvas_account in canvas_accounts.data:
                canvas_accounts_dict[canvas_account['id']]=canvas_account
                url=f'https://{canvas_environment['domains'][canvas_environment['working_environment']]}/api/v1/accounts/{canvas_account['id']}/sub_accounts?recursive=true'
                canvas_subaccounts=canvas_get_allpages(url,canvas_environment['url_headers'])
                if canvas_subaccounts.status_code==200:
                    for canvas_subaccount in canvas_subaccounts.data:
                        canvas_accounts_dict[canvas_subaccount['id']]=canvas_subaccount
            if mode[0:6]!='silent': print(f'{datetime.datetime.now().isoformat()}: API token validated, proceeding...')
            if mode[-10:]=='unattended': scriptlog+=f'{datetime.datetime.now().isoformat()}: API token validated, proceeding...\n'
            apitoken_validated=True
        else:
            print(f'{datetime.datetime.now().isoformat()}: API token validation failed.  Token or domain settings are incorrect.')
            if mode[-10:]=='unattended': scriptlog+=f'{datetime.datetime.now().isoformat()}: API token validation failed.  Token or domain settings are incorrect.\n'
            canvas_environment['api_token']=''
            canvas_environment['url_headers']=''
            if mode[-10:]=='unattended': apitoken_validated=True
    # Prompt for Canvas Account ID
    # Get account list from Canvas
    account_id_validated=False
    if canvas_environment['domains'][canvas_environment['working_environment']]!='':
        all_accounts_ids_validated=True
        all_root_account_ids=set()
        for account in (set(canvas_environment['account_ids'].keys())-set(['root'])):
            account_id_validated=False
            while (not account_id_validated and all_accounts_ids_validated):
                if canvas_environment['account_ids'][account]=='' and mode[-10:]=='unattended':
                    canvas_environment['account_ids'][account]=canvas_environment['account_ids']['root']
                    account_id_validated=True
                else:
                    # Prompt for ID selection
                    if canvas_environment['account_ids'][account]=='' and mode[-10:]!='unattended':
                        canvas_environment['account_ids'][account]=input(f'What is the id number for the {account} account: # or [L]ist)? ')
                    if str(canvas_environment['account_ids'][account]).lower()=='l':
                        for account_id in canvas_accounts_dict.keys():
                            print(str(account_id)+': '+canvas_accounts_dict[account_id]['name'])
                        canvas_environment['account_ids'][account]=input(f'What is the id number for the {account} account? ')
                    if (type(canvas_environment['account_ids'][account])==int or (type(canvas_environment['account_ids'][account])==str and (canvas_environment['account_ids'][account].isdigit()))) and (int(canvas_environment['account_ids'][account]) in canvas_accounts_dict.keys()):
                        canvas_environment['account_ids'][account]=int(canvas_environment['account_ids'][account])
                        if mode[0:6]!='silent': print(f'{datetime.datetime.now().isoformat()}: {account} account selection validated (account_id: {canvas_environment['account_ids'][account]}, account_name: {canvas_accounts_dict[canvas_environment['account_ids'][account]]['name']}), proceeding...')
                        if mode[-10:]=='unattended': scriptlog+=f'{datetime.datetime.now().isoformat()}: {account} account selection validated (account_id: {canvas_environment['account_ids'][account]}, account_name: {canvas_accounts_dict[canvas_environment['account_ids'][account]]['name']}), proceeding...\n'
                        account_id_validated=True
                        account_root_id=canvas_environment['account_ids'][account]
                        while canvas_accounts_dict[account_root_id]['parent_account_id']!=None:
                            account_root_id=canvas_accounts_dict[account_root_id]['parent_account_id']
                        all_root_account_ids.add(account_root_id)
                    else:
                        print(f'{datetime.datetime.now().isoformat()}: invalid {account} account selection (account_id: {canvas_environment['account_ids'][account]}).')
                        if mode[-10:]=='unattended':scriptlog+=f'{datetime.datetime.now().isoformat()}: invalid {account} account selection (account_id: {canvas_environment['account_ids'][account]}).\n'
                        canvas_environment['account_ids'][account]=''
                        if mode[-10:]=='unattended':
                            account_id_validated=True
            if canvas_environment['account_ids'][account]=='':
                all_accounts_ids_validated=False
        canvas_environment['account_ids']['root']=min(all_root_account_ids)
        if len(all_root_account_ids)==1:
            if mode[0:6]!='silent': print(f'{datetime.datetime.now().isoformat()}: root account selection validated (account_id: {canvas_environment['account_ids']['root']}, account_name: {canvas_accounts_dict[canvas_environment['account_ids']['root']]['name']}), proceeding...')
            if mode[-10:]=='unattended': scriptlog+=f'{datetime.datetime.now().isoformat()}: root account selection validated (account_id: {canvas_environment['account_ids']['root']}, account_name: {canvas_accounts_dict[canvas_environment['account_ids']['root']]['name']}), proceeding...\n'
        else:
            print('Specified accounts belong to different root accounts, unable to continue...')
            if mode[-10:]=='unattended': scriptlog+='Specified accounts belong to different root accounts, unable to continue...\n'
            all_accounts_ids_validated=False
    if canvas_environment['domains'][canvas_environment['working_environment']]!='':
        url=f'https://{canvas_environment['domains'][canvas_environment['working_environment']]}/api/v1/users/self'
        r=requestswithretry().get(url,headers=canvas_environment['url_headers'])
        if r.status_code==200:
            canvas_user=r.json()
            if mode[0:6]!='silent': print(f'Running script as {canvas_user['name']}\n  login_id: {canvas_user['login_id']}\n  sis_user_id: {canvas_user['sis_user_id']}\n  email: {canvas_user['email']}')
            if mode[-10:]=='unattended': scriptlog+=f'Running script as {canvas_user['name']}\n  login_id: {canvas_user['login_id']}\n  sis_user_id: {canvas_user['sis_user_id']}\n  email: {canvas_user['email']}\n\n'
        else:
            print(f'{datetime.datetime.now().isoformat()}: Could not retrieve script user infomration.')
            if mode[-10:]=='unattended': scriptlog+=f'{datetime.datetime.now().isoformat()}: Could not retrieve script user infomration..\n'
    if mode[0:6]!='silent': print('\n--------------------------------------------------------------------------------\n')
    if mode[-10:]=='unattended': scriptlog+='\n--------------------------------------------------------------------------------\n\n'
    return (canvas_environment['working_environment']!='') and (canvas_environment['domains'][canvas_environment['working_environment']]!='') and (canvas_environment['subdomain']!='') and (canvas_environment['api_token']!='') and all_accounts_ids_validated and domain_validated;

def main():
    global scriptlog, canvas_api_token, canvas_subdomain, canvas_environment, canvas_production_vanity_domain, canvas_test_vanity_domain, canvas_beta_vanity_domain, canvas_account_id
    hostname=socket.gethostname()
    IPAddr=socket.gethostbyname(hostname)
    print(f'Running script version {script_version} on {hostname}, {IPAddr}')
    scriptlog+=f'Running script version {script_version} on {hostname}, {IPAddr}\n'
    start_time = datetime.datetime.now()

    canvas_environment = {'api_token':str(canvas_api_token), 'url_headers':{'Authorization' : 'Bearer '+str(canvas_api_token)},'subdomain':str(canvas_subdomain), 'working_environment':canvas_environment, 'domains':{'production':canvas_production_vanity_domain, 'test':canvas_test_vanity_domain, 'beta':canvas_beta_vanity_domain}, 'account_ids':{'root':'', 'working':canvas_account_id}}
    if (not canvas_validate_environment(canvas_environment)):
        print('Canvas environment validation failed.  Exiting Script.')
        scriptlog+='Canvas environment validation failed.  Exiting Script.\n'
    else:
        canvas_target_user_id=input('Please enter the canvas_user_id of the user which you wish to see the page view history: ')
        url=f'https://{canvas_environment['domains'][canvas_environment['working_environment']]}/api/v1/users/{canvas_target_user_id}'
        canvas_target_user_request=canvas_get_allpages(url,canvas_environment['url_headers'])
        if canvas_target_user_request.status_code!=200:
            print(f'{datetime.datetime.now().isoformat()}: Could not target canvas user infomration.')
        else:
            print(f'Retrieving pageviews for {canvas_target_user_request.data['name']}\n  login_id: {canvas_target_user_request.data['login_id']}\n  sis_user_id: {canvas_target_user_request.data['sis_user_id']}\n  email: {canvas_target_user_request.data['email']}\n')
            pageview_start_datetime_localstring=input('Enter start date, formatted yyyy-mm-dd: ')
            while re.match(r'\d{4}-[0-2][1-9]-[0-3][1-9]',pageview_start_datetime_localstring)==None:
                pageview_start_datetime_localstring=input('Enter start date, formatted yyyy-mm-dd: ')
            pageview_start_datetime=datetime.datetime.fromisoformat(pageview_start_datetime_localstring).astimezone()
            pageview_start_datetime_zulustring=pageview_start_datetime.astimezone(datetime.timezone.utc).isoformat().replace("+00:00", "Z")
            print('Start date/time zulustring:', pageview_start_datetime_zulustring,'\n')
            pageview_end_datetime_localstring=input('Enter end date, formatted yyyy-mm-dd: ')
            while re.match(r'\d{4}-[0-2][1-9]-[0-3][1-9]',pageview_end_datetime_localstring)==None:
                pageview_end_datetime_localstring=input('Enter end date, formatted yyyy-mm-dd: ')
            pageview_end_datetime=datetime.datetime.fromisoformat(pageview_end_datetime_localstring+'T23:59:59').astimezone()
            pageview_end_datetime_zulustring=pageview_end_datetime.astimezone(datetime.timezone.utc).isoformat().replace("+00:00", "Z")
            print('End_date/time zulustring:', pageview_end_datetime_zulustring,'\n')

            if pageview_start_datetime>=pageview_end_datetime:
                print('Error: start date/time must be before end date/time')
            else:
                start_time = datetime.datetime.now()
                url=f'https://{canvas_environment['domains'][canvas_environment['working_environment']]}/api/v1/users/{canvas_target_user_id}/page_views?start_time={pageview_start_datetime_zulustring}&end_time={pageview_end_datetime_zulustring}'
                canvas_user_page_views_response=canvas_get_allpages(url,canvas_environment['url_headers'])
                print('User had',len(canvas_user_page_views_response.data),'total page views in the specified period\n')
                if canvas_target_user_request.status_code==200 and len(canvas_user_page_views_response.data)>0:
                    with open('pageviews-all.csv', 'w', newline='') as output_file:
                        writer = csv.DictWriter(output_file, canvas_user_page_views_response.data[0].keys())
                        writer.writeheader()
                        writer.writerows(canvas_user_page_views_response.data)
                    canvas_user_participation_pageviews = [pageview for pageview in canvas_user_page_views_response.data if pageview['participated']==True]
                    print('User had',len(canvas_user_participation_pageviews),'partipation page views in the specified period\n')
                    with open('pageviews-participations.csv', 'w', newline='') as output_file:
                        writer = csv.DictWriter(output_file, canvas_user_page_views_response.data[0].keys())
                        writer.writeheader()
                        if len(canvas_user_participation_pageviews)>0:
                            writer.writerows(canvas_user_participation_pageviews)
                print(f'Finished!  Run time: {datetime.datetime.now() - start_time}')
                scriptlog+=(f'\nFinished!  Run time: {datetime.datetime.now() - start_time}\n')

if __name__=='__main__':
    main()