Notebook

これは日々の作業を通して学んだことや毎日の生活で気づいたことをを記録しておく備忘録である。

HTML ファイル生成日時: 2024/10/16 21:52:27.872 (台灣標準時)

NHK ラジオの「聞き逃し」の変更点 (2024 年 06 月)

NHK ラジオの「聞き逃し」を使ってラジオ番組を聞くために作っておいた Python スクリプトが期待通りに動かなくなっていたでござる。どうしたので ござろうか、と思い、調べてみると、番組情報が書かれていた https://www.nhk.or.jp/radioondemand/json/index_v3/index.json というファイルがなくなっていたでござる。

新たなファイルはどこにあるのでござろうか、と思い、探してみると、以下に 有用な情報があったでござる。

新たな JSON ファイルは、 https://www.nhk.or.jp/radio-api/app/v1/web/ondemand/corners/new_arrivals にあるようでござる。

また、 JSON ファイルの構造も変更されているようでござる。以前は、 https://www.nhk.or.jp/radioondemand/json/index_v3/index.json の中に m3u8 のアドレスも書いてあったのでござるが、新しい JSON ファイル では、 https://www.nhk.or.jp/radio-api/app/v1/web/ondemand/corners/new_arrivals の中には m3u8 のアドレスは書いていないようでござる。

では、どうすればよいか、というと、 https://www.nhk.or.jp/radio-api/app/v1/web/ondemand/corners/new_arrivals を見て、聞きたい番組の series_site_id (番組の ID 番号) と corner_site_id (番組内のコーナーの ID 番号) を調べ、それらの情報を使っ て、番組毎に用意されている JSON ファイルをダウンロードして、その JSON ファイルから m3u8 アドレスを見つける、ということをする必要があるようで ござる。

番組毎に準備されている JSON ファイルをダウンロードして、 m3u8 のアドレ スを見つけたら、 ffmpeg などを使って番組の内容を聞くことができるでござ る。

試しに、 NHK ラジオの「聞き逃し」で公開されている番組を聞くための Python スクリプトを書いてみたでござる。


#!/usr/pkg/bin/python3.12

#
# Time-stamp: <2024/06/10 17:06:50 (UT+8) daisuke>
#

###########################################################################

#
# importing modules
#

# importing argparse module
import argparse

# importing os module
import os

# importing sys module
import sys

# importing pathlib module
import pathlib

# importing datetime module
import datetime

# importing ssl module
import ssl

# importing urllib module
import urllib.request

# importing json module
import json

# importing re module
import re

# importing subprocess module
import subprocess

# importing random module
import random

# importing time module
import time

# importing shutil module
import shutil

# setting for SSL
ssl._create_default_https_context = ssl._create_unverified_context

###########################################################################

###########################################################################

#
# come constants and parameters
#

# URL of JSON file
url_json_nhk \
    = 'https://www.nhk.or.jp/radio-api/app/v1/web/ondemand/corners/new_arrivals'

# program names
dic_programs = {
    'adventure': '青春アドベンチャー',
    'asianview': 'Asian View',
    'culture_art': 'カルチャーラジオ 芸術その魅力',
    'culture_chinese': 'カルチャーラジオ 漢詩をよむ',
    'culture_history': 'カルチャーラジオ 歴史再発見',
    'culture_literature': 'カルチャーラジオ 文学の世界',
    'culture_science': 'カルチャーラジオ 科学と人間',
    'culture_showa': 'カルチャーラジオ 保阪正康が語る昭和人物史',
    'culture_sunday': 'カルチャーラジオ 日曜カルチャー',
    'earthradio': 'ちきゅうラジオ',
    'gendaieigo': 'ニュースで学ぶ「現代英語」',
    'genichiro': '高橋源一郎の飛ぶ教室',
    'hoshizora': 'MISIA 星空のラジオ ~Sunday Sunset~',
    'jikutabi': '音で訪ねる ニッポン時空旅',
    'learnjapanese': 'Learn Japanese from the News',
    'meisakuza': '新日曜名作座',
    'nemurenai': '眠れない貴女へ',
    'nhkjournal': 'NHKジャーナル',
    'roudoku': '朗読',
    'roudokuworld': '朗読の世界',
    'theatre': 'FMシアター',
    'weekendsunshine': 'ウィークエンドサンシャイン',
    'yamacafe': '石丸謙二郎の山カフェ',
}

# environmental variables
dir_home = os.environ['HOME']

# process ID
pid = os.getpid ()

###########################################################################

###########################################################################

#
# date/time
#

# date/time
datetime_now = datetime.datetime.now ()
YYYY         = datetime_now.year
MM           = datetime_now.month
DD           = datetime_now.day
hh           = datetime_now.hour
mm           = datetime_now.minute
ss           = datetime_now.second
date_str     = f'{YYYY:04d}{MM:02d}{DD:02d}'
time_str     = f'{hh:02d}{mm:02d}{ss:02d}'
datetime_str = f'{date_str}_{time_str}'

###########################################################################

###########################################################################

#
# command-line arguments analysis using argparse
#

# default parameters
choices_programs = list (dic_programs.keys ())
default_programs = 'adventure'
help_programs    = f'program names (default: {default_programs})'

default_useragent \
    = 'Mozilla/5.0 (X11; NetBSD amd64; rv:96.0) Gecko/20100101 Firefox/96.0'
help_useragent \
    = f'user agent for HTTP retrieval (default: {default_useragent})'

default_dir_radio = f'{dir_home}/audio/radio'
help_dir_radio \
    = f'directory to store recorded file (default: {default_dir_radio})'

default_dir_tmp = f'/tmp/radio_{datetime_str}_{pid}'
help_dir_tmp \
    = f'directory to store temporary file (default: {default_dir_tmp})'

default_ffmpeg = '/usr/pkg/bin/ffmpeg6'
help_ffmpeg    = f'location of ffmpeg command (default: {default_ffmpeg})'

default_sleep  = 60
help_sleep \
    = f'max sleep time between file retrieval (default: {default_sleep} sec)'

default_verbose = 0
help_verbose    = f'verbosity level (default: {default_verbose})'

# construction of parser object
desc = 'NHK on-demand radio program recording script (2024 version)'
parser = argparse.ArgumentParser (description=desc)

# adding arguments
parser.add_argument ('programs', nargs='+', choices=choices_programs, \
                     help=help_programs)
parser.add_argument ('-r', '--radio-dir', default=default_dir_radio, \
                     help=help_dir_radio)
parser.add_argument ('-t', '--temporary-dir', default=default_dir_tmp, \
                     help=help_dir_tmp)
parser.add_argument ('-u', '--user-agent', default=default_useragent, \
                     help=help_useragent)
parser.add_argument ('-f', '--ffmpeg', default=default_ffmpeg, \
                     help=help_ffmpeg)
parser.add_argument ('-s', '--sleep', default=default_sleep, \
                     help=help_sleep)
parser.add_argument ('-v', '--verbose', action='count', \
                     default=default_verbose, help=help_verbose)

# command-line argument analysis
args = parser.parse_args ()

# parameters
list_programs  = args.programs
dir_radio      = args.radio_dir
dir_tmp        = args.temporary_dir
user_agent     = args.user_agent
command_ffmpeg = args.ffmpeg
max_sleep_time = args.sleep
verbosity      = args.verbose

###########################################################################

###########################################################################

#
# files and directories
#

# files
file_json_nhk = f'{dir_tmp}/index_{datetime_str}.json'
if (verbosity):
    print (f'file_json_nhk = {file_json_nhk}')

# existence check of commands
list_commands = [command_ffmpeg]
for command in list_commands:
    # making a pathlib object
    path_command = pathlib.Path (command)
    # if command does not exist, then stop the script
    if not (path_command.exists ()):
        # printing message
        print (f'The command "{command}" does not exist!')
        print (f'Install "{command}" and then run the command again.')
        # exit
        sys.exit ()

# existence check of directories
list_dir = [dir_radio, dir_tmp]
for directory in list_dir:
    # making pathlib object
    path_dir = pathlib.Path (directory)
    # if directory does not exist
    if not (path_dir.exists ()):
        # printing message
        if (verbosity):
            print (f'# making directory "{directory}"...')
        # making directory
        path_dir.mkdir (parents=True, exist_ok=True)
        # printing message
        if (verbosity):
            print (f'# finished making directory "{directory}"!')

###########################################################################

###########################################################################

#
# retrieving of JSON file
#

# retrieval of JSON file from NHK website
req_nhk = urllib.request.Request (url=url_json_nhk)
req_nhk.add_header ('User-Agent', user_agent)
with urllib.request.urlopen (req_nhk) as www:
    data_json_nhk = www.read ()

# decoding JSON file content
text_json_nhk = data_json_nhk.decode ('utf8')

# writing JSON data into file
with open (file_json_nhk, 'w') as fh:
    fh.write (text_json_nhk)

# decoding JSON data
dic_nhk = json.loads (text_json_nhk)

#for key in dic_nhk.keys ():
#    for i in range ( len (dic_nhk[key]) ):
#        print (f'ID = {dic_nhk[key][i]["id"]}')
#        print (f'  Channel   = {dic_nhk[key][i]["radio_broadcast"]}')
#        print (f'  Series ID = {dic_nhk[key][i]["series_site_id"]}')
#        print (f'  Corner ID = {dic_nhk[key][i]["corner_site_id"]}')

###########################################################################

###########################################################################

#
# retrieving of audio data
#

# processing each program
for program in list_programs:
    # printing status
    if (verbosity):
        print (f'Now, processing the program "{program}"...')

    # finding channel, series ID, and corner ID
    for key in dic_nhk.keys ():
        for i in range ( len (dic_nhk[key]) ):
            if (dic_nhk[key][i]["title"] == dic_programs[program]):
                channel   = dic_nhk[key][i]["radio_broadcast"]
                series_id = dic_nhk[key][i]["series_site_id"]
                corner_id = dic_nhk[key][i]["corner_site_id"]

    # printing channel, series ID, and cornder ID
    if (verbosity):
        print (f'  Channel   = {channel}')
        print (f'  Series ID = {series_id}')
        print (f'  Corner ID = {corner_id}')
    
    # url of JSON file for given program
    url_json_program \
        = f'https://www.nhk.or.jp/radio-api/app/v1/web/ondemand/series' \
        + f'?site_id={series_id}&corner_site_id={corner_id}'

    # printing status
    if (verbosity):
        print (f'  Now, fetching JSON file...')
        print (f'  {url_json_program}')
    
    # fetching JSON file for the program
    req_program = urllib.request.Request (url=url_json_program)
    req_program.add_header ('User-Agent', user_agent)
    with urllib.request.urlopen (req_program) as www:
        data_json_program = www.read ()

    # printing status
    if (verbosity):
        print (f'  Finished fetching JSON file!')

    # decoding JSON file content
    text_json_program = data_json_program.decode ('utf8')

    # name for JSON file
    file_json_program = f'{dir_tmp}/{series_id}_{corner_id}.json'
    
    # writing JSON file into file
    with open (file_json_program, 'w') as fh:
        fh.write (text_json_program)

    # decoding JSON data
    dic_program = json.loads (text_json_program)
    
    #print (f'{dic_program["episodes"][0]["stream_url"]}')

    # retrieving audio data
    for i in range ( len (dic_program["episodes"]) ):
        # URL of m3u8
        url_m3u8 = dic_program["episodes"][i]["stream_url"]
        # onair date
        contents_id = dic_program["episodes"][i]["aa_contents_id"]
        
        # pattern matching of date/time of start of program
        pattern_datetime \
            = re.compile (r'(\d{4})-(\d{2})-(\d{2})T(\d{2}):(\d{2}):(\d{2})\+09:00_') 
        match_datetime = re.search (pattern_datetime, contents_id)
        if (match_datetime):
            start_YYYY = int (match_datetime.group (1))
            start_MM   = int (match_datetime.group (2))
            start_DD   = int (match_datetime.group (3))
            start_hh   = int (match_datetime.group (4))
            start_mm   = int (match_datetime.group (5))
            start_ss   = int (match_datetime.group (6))
        else:
            if (verbosity):
                print (f'WARNING: pattern matching failed!')
            continue

        # file names
        start_date     = f'{start_YYYY:04d}{start_MM:02d}{start_DD:02d}'
        start_time     = f'{start_hh:02d}{start_mm:02d}'
        start_datetime = f'{start_date}_{start_time}'
        file_aac_tmp   = f'{dir_tmp}/{program}_{start_datetime}.aac'
        file_aac       = f'{dir_radio}/{program}_{start_datetime}.aac'

        # sleeping for a short time
        sleep_time = random.randint (5, max_sleep_time)
        time.sleep (sleep_time)
            
        # command for fetching audio stream using ffmpeg command
        command_fetch = f'{command_ffmpeg} -http_seekable 0 -n' \
            + f' -i {url_m3u8} -vn -acodec aac {file_aac_tmp}'
        if (verbosity):
            print (f'#    {command_fetch}')
        subprocess.run (command_fetch, shell=True)

        # existence check of fetched audio file
        path_aac_tmp = pathlib.Path (file_aac_tmp)
        if not (path_aac_tmp.exists ()):
            # printing message
            print (f'The file "{file_aac_tmp}" does not exist!')
            print (f'Something is wrong with retrieval of data.')
            print (f'Exiting...')
            # exit
            sys.exit ()

        # file size of file_aac_tmp
        filesize_aac_tmp = path_aac_tmp.stat ().st_size

        # file size of file_aac
        path_aac = pathlib.Path (file_aac)
        if (path_aac.exists ()):
            filesize_aac = path_aac.stat ().st_size
        else:
            filesize_aac = 0

        # printing file sizes
        if (verbosity):
            print (f'#    file sizes')
            print (f'#      {file_aac_tmp}: {filesize_aac_tmp} byte')
            print (f'#      {file_aac}: {filesize_aac} byte')
    
        # copying AAC file
        if ( (path_aac.exists ()) and (filesize_aac >= filesize_aac_tmp) ):
            # if file exists and larger than new file, then not copying file
            if (verbosity):
                print (f'#    file "{file_aac_tmp}" is not copied to {dir_radio}')
        else:
            if (verbosity):
                print (f'#    copy: {file_aac_tmp} ==> {file_aac}')
            # copying file
            shutil.copy2 (file_aac_tmp, file_aac)
            if (verbosity):
                print (f'#    file copy finished')

###########################################################################



Frequently accessed files

  1. Computer___Python/20220518_0.html
  2. Computer___Network/20230726_00.html
  3. Computer___Network/20230516_00.html
  4. Misc___Taiwan/20240207_00.html
  5. Computer___FreeBSD/20220621_0.html
  6. Computer___Python/20220715_0.html
  7. Food___Taiwan/20220429_0.html
  8. Computer___Network/20230508_00.html
  9. Computer___NetBSD/20220817_3.html
  10. Computer___Python/20220410_0.html
  11. Computer___Network/20240416_00.html
  12. Computer___Debian/20210223_1.html
  13. Computer___Python/20210124_0.html
  14. Computer___Network/20240130_00.html
  15. Computer___NetBSD/20220818_1.html
  16. Computer___NetBSD/20220428_0.html
  17. Computer___NetBSD/20230119_00.html
  18. Computer___Python/20221013_0.html
  19. Science___Math/20220420_0.html
  20. Computer___NetBSD/20230515_00.html
  21. Computer___TeX/20230503_00.html
  22. Computer___NetBSD/20240101_02.html
  23. Science___Astronomy/20220503_0.html
  24. Computer___NetBSD/20220808_0.html
  25. Computer___Python/20240101_00.html
  26. Computer___NetBSD/20210127_0.html
  27. Computer___NetBSD/20210204_0.html
  28. Travel___Taiwan/20220809_2.html
  29. Computer___Python/20220816_1.html
  30. Computer___Network/20220413_1.html


HTML file generated by Kinoshita Daisuke.