qubodup/Sound-Attribution-Tool-Freesound-Super-Credits.py

## Sound-Attribution-Tool-Freesound-Super-Credits.py
# The code is dedicated to the public domain https://creativecommons.org/publicdomain/mark/1.0/

# A very simple Flask Hello World app for you to get started with...

# SECURITY RISK START
FSKEY = "FREESOUND API KEY"
# SECURITY RISK OVER

html_header = """<!DOCTYPE HTML>
    <html>
    <head>
    <title>Freesound Super Credits HTML &amp; text author attribution &amp; license generator</title>
    <style>
    #slim { margin: 0 auto; width: 800px; }
    textarea { height: 200px; width: 100%%; }
    .error { background-color: orangered; color: black; }
    .warning { background-color: gold; color: black; }
    </style>
    </head>
    <body>
"""
html_footer = """
    </body>
    </html>
"""

# Creates license strings NEEDS TO BE UPDATED NOW THAT CCBY4.0 IS AVAILABLE?
def getLicenseName(url):
  name = "Unknown License"
  if url == "http://creativecommons.org/licenses/by/3.0/":
    name = "CC-BY 3.0"
  elif url == "http://creativecommons.org/publicdomain/zero/1.0/":
    name = "CC0"
  elif url == "http://creativecommons.org/licenses/by-nc/3.0/":
    name = "CC-BY-NC 3.0"
  elif url == "http://creativecommons.org/licenses/sampling+/1.0/":
    name = "CC-Sampling+"
  return name

# Remove "www." from links
def getShortLink(url):
  return url.replace("http://www.", "http://")

#from flask import Flask

from flask import Flask, request, url_for
import urllib.request
import json
import html

app = Flask(__name__)
app.secret_key = 'IDONTKNOWWHATTHISIS'

@app.route('/')
def hello_person():

    html_content = """
        <div id="slim">
        <h1>Freesound Super Credits</h1>
        <p>Freesound credits generator which formats author attribution and license information for HTML (webistes, blogs) &amp; plaintext (YouTube descriptions and readme files).</p>
        <p>If this tool is inconvenient, please write how you would prefer it to qubodup@gmail.com, it might take just a minute to make using this site comfortable for you.</p>
        <h2>Form</h2>
        <p>Please insert a newline-divided list of Freesound sound IDs (11), URLs (http://freesound.org/people/Bram/sounds/11/) or filenames (11__bram__sweep-log.wav):</p>
        <p class="warning">Warning: you might want to edit your input if there are errors but your browser might forget the data, so save it in a text file using gedit (Linux), TextEdit (OS X) or notepad.exe (Windows).</p>
        <form method="POST" action="%s"><textarea name="ids"></textarea><input type="submit" value="Submit" /></form>
        <h2>Sample data set</h2>
        <p>You can use this text to test both attribution generation as well as error handling:</p>
        <pre>http://www.freesound.org/people/Woodylein/sounds/149818/
http://www.freesound.org/people/rjonesxlr8/sounds/213783/
http://www.freesound.org/people/Dpoggioli/sounds/213608/
http://www.freesound.org/people/Ears68/sounds/185150/
http://www.freesound.org/people/Woodylein/sounds/149817/
http://www.freesound.org/people/BROKENTEST/sounds/1zz1/
11__bram__sweep-log.wav
http://www.freesound.org/people/jediduke/sounds/126518/
http://www.freesound.org/people/Philip%%20Goddard/sounds/172163/
211985
http://www.freesound.org/people/qubodup/sounds/211985/
http://www.freesound.org/people/qubodup/sounds/211983/""" % (url_for('credits'))
    html_content = html_content + html.escape("""
&asd&asd&amp;
<br>
<p style="background-color: red;">hm</p>""") + "</pre></div>"

    return html_header + html_content + html_footer

@app.route('/credits', methods=['POST'])
def credits():

    content = html.escape(request.form["ids"]).split("\n")

    sounds = []
    sounds_failed = []
    sounds_duplicate = []

    for i in content:
      tmp = i.split("\n")[0].strip() # from text file input times, leaving just in case
      #tmp = i
      tmp2 = tmp.split("__")[0]
      tmp3 = tmp.split("/sounds/")
      if len(tmp3) > 1:
        tmp_final = tmp3[1].split("/")[0]
      else:
        tmp_final = tmp2

      # check if result is a duplicate
      duplicate = False
      for sound in sounds:
        if sound[1] == tmp_final:
          duplicate = True
          # keep track of duplicates
          sounds_duplicate.append("(\"" + i + "\" : \"" + tmp_final + "\")")

    # add item to sounds table if it's not a dupliacte
      if not duplicate:
        # include original item for error feedback
        sounds.append([i, tmp_final])

    database = []

    allUserNames = set()
    allLicenseNames = set()

    for sound in sounds:

      url="http://freesound.org/apiv2/sounds/" + sound[1] + "/?fields=name,url,username,license&format=json" + "&token=" + FSKEY
      try:
        freesoundResponse = urllib.request.urlopen(urllib.request.Request(url))#, headers={'User-Agent': 'Mozilla/5.0'})) # I tried user agent because I first got 403 but it was actually a restriction by pythonanywhere. might not be needed but I'm afraid to break the request for now
      except Exception as e:
        sounds_failed.append("(\"" + str(sound[0]) + "\" : \"" + str(sound[1]) + "\") " + str(e)) # + str(url) + " "
      else:
        jsonResponse = json.loads(freesoundResponse.read().decode("utf-8"))

        # metadata one-liners
        database.append({
          'soundName'  : jsonResponse['name'],
          'soundLink'  : jsonResponse['url'],
          'userName'   : jsonResponse['username'],
          'userLink'   : "http://"+"/".join(jsonResponse['url'].split("/")[2:5]),
          'licenseName': getLicenseName(jsonResponse['license']),
          'licenseLink': jsonResponse['license']
        })

        # keep track of users and licenses involved
        allUserNames.add(jsonResponse['username'])
        allLicenseNames.add(getLicenseName(jsonResponse['license']))

    # for all licenses, in order of unknown, by-nc, sampling+, by, cc0
    # NEEDS UPDATING PROBABLY FOR CCBY4.0
    orderedLicenseNames = [
      "Unknown License",
      "CC-BY-NC 3.0",
      "CC-Sampling+",
      "CC-BY 3.0",
      "CC0"
    ]

    credits = ""
    creditsHTML = ""

    # order credits by most strict licenses at the top
    for orderedLicenseName in orderedLicenseNames:

      # for each existing license
      if orderedLicenseName in allLicenseNames:

        firstLicense = True
        userNeeded = False #since we loop over all users, we need to know when one actually matches for ending user credits
        tmpUserName = ""
        tmpUserLink = ""

        # for each existing user
        for userName in allUserNames:

          # for each sound
          for item in database:

            if ((item['userName'] == userName) and (item['licenseName'] == orderedLicenseName)):

              userNeeded = True
              tmpUserName = item['userName']
              tmpUserLink = item['userLink']
              if firstLicense:
                credits = credits + "<br>Sounds under " + item['licenseName'] + " ( " + item['licenseLink'] + " ):<br>"
                creditsHTML = creditsHTML + "<br>Sounds under <a href='" + item['licenseLink'] + "'>" + item['licenseName'] + "</a>:<br>"
                firstLicense = False

              credits = credits + "\n" + item['soundName'] + " ( " + getShortLink(item['soundLink']) + " ), "
              creditsHTML = creditsHTML + "\n<a href='" + getShortLink(item['soundLink']) + "'>" + item['soundName'] + "</a>, "

          if userNeeded:
            creditsHTML = creditsHTML[:-2] + " by <a href='" + getShortLink(tmpUserLink) + "'>" + tmpUserName + "</a>\n<br>\n"
            credits = credits[:-2] + " by " + tmpUserName + " ( " + getShortLink(tmpUserLink) + " )\n<br>\n"
            userNeeded = False

    # chop off leading and tailing break
    creditsHTML = creditsHTML[4:][:-5]
    credits = credits[4:][:-5]

    # simple credits
    allUserNamesSingle = list(set(allUserNames))
    allUserNamesString = ', '.join(allUserNamesSingle)
    allFailedString = '<br>'.join(sounds_failed)
    allDuplicateString = '<br>'.join(sounds_duplicate)

    html_content = """
        <h1>Freesound Super Credits Results</h1>
        <p>Should there be issues with the results, please report to qubodup@gmail.com</p>
        <p>Use your browser "back" button/function to return to your input (might not work in any browser) or <a href="%s">start new query</a></p>
    """ % (url_for('hello_person'))

    if (len(sounds_failed) > 0):
        html_content = html_content + """
        <div class='error'>
        <h2>Failed items</h2>
        <p>%s</p>
        </div>
        """ % (allFailedString)


    if (len(sounds_duplicate) > 0):
        html_content = html_content + """
        <div class='warning'>
        <h2>Duplicate items</h2>
        <p>%s</p>
        </div>
        """ % (allDuplicateString)

    html_content = html_content + """
        <h2>Plaintext credits</h2>
        <p>%s</p>
        <h2>HTML credits (WYSIWYG)</h2>
        <p>%s</p>
        <h2>HTML credits (code)</h2>
        <textarea>%s</textarea>
        <h2>Additional minimal credits</h2>
        <p>Simple comma-separated list of all usernames:</p>
        <p>%s</p>
        """ % (credits, creditsHTML, creditsHTML, allUserNamesString)

    return html_header + html_content + html_footer

# Creates attribution html text
def attribution_construct_html(item):
  return (
    "<a href='" +
    item.soundLink +
    "'>" +
    item.soundName +
    "</a> by <a href='" +
    item.userLink +
    "'>" +
    item.userName +
    "</a>" +
    "\n" +
    item.licenseName +
    "<br>" # since this is for HTML output
  )

# Creates attribution plain text
def attribution_construct(item):
  item.sndAttribution = (
    "By " +
    item.userName +
    " " +
    item.sndUserUrl +
    " under " +
    item.sndLicense +
    "\n" +
    item.sndTitleOrg +
    ": " +
    item.licenseName +
    "<br>" # since this is for HTML output
  )
	# The code is dedicated to the public domain https://creativecommons.org/publicdomain/mark/1.0/

	# A very simple Flask Hello World app for you to get started with...

	# SECURITY RISK START
	FSKEY = "FREESOUND API KEY"
	# SECURITY RISK OVER

	html_header = """<!DOCTYPE HTML>
	<html>
	<head>
	<title>Freesound Super Credits HTML & text author attribution & license generator</title>
	<style>
	#slim { margin: 0 auto; width: 800px; }
	textarea { height: 200px; width: 100%%; }
	.error { background-color: orangered; color: black; }
	.warning { background-color: gold; color: black; }
	</style>
	</head>
	<body>
	"""
	html_footer = """
	</body>
	</html>
	"""

	# Creates license strings NEEDS TO BE UPDATED NOW THAT CCBY4.0 IS AVAILABLE?
	def getLicenseName(url):
	name = "Unknown License"
	if url == "http://creativecommons.org/licenses/by/3.0/":
	name = "CC-BY 3.0"
	elif url == "http://creativecommons.org/publicdomain/zero/1.0/":
	name = "CC0"
	elif url == "http://creativecommons.org/licenses/by-nc/3.0/":
	name = "CC-BY-NC 3.0"
	elif url == "http://creativecommons.org/licenses/sampling+/1.0/":
	name = "CC-Sampling+"
	return name

	# Remove "www." from links
	def getShortLink(url):
	return url.replace("http://www.", "http://")

	#from flask import Flask

	from flask import Flask, request, url_for
	import urllib.request
	import json
	import html

	app = Flask(__name__)
	app.secret_key = 'IDONTKNOWWHATTHISIS'

	@app.route('/')
	def hello_person():

	html_content = """
	<div id="slim">
	<h1>Freesound Super Credits</h1>
	<p>Freesound credits generator which formats author attribution and license information for HTML (webistes, blogs) & plaintext (YouTube descriptions and readme files).</p>
	<p>If this tool is inconvenient, please write how you would prefer it to qubodup@gmail.com, it might take just a minute to make using this site comfortable for you.</p>
	<h2>Form</h2>
	<p>Please insert a newline-divided list of Freesound sound IDs (11), URLs (http://freesound.org/people/Bram/sounds/11/) or filenames (11__bram__sweep-log.wav):</p>
	<p class="warning">Warning: you might want to edit your input if there are errors but your browser might forget the data, so save it in a text file using gedit (Linux), TextEdit (OS X) or notepad.exe (Windows).</p>
	<form method="POST" action="%s"><textarea name="ids"></textarea><input type="submit" value="Submit" /></form>
	<h2>Sample data set</h2>
	<p>You can use this text to test both attribution generation as well as error handling:</p>
	<pre>http://www.freesound.org/people/Woodylein/sounds/149818/
	http://www.freesound.org/people/rjonesxlr8/sounds/213783/
	http://www.freesound.org/people/Dpoggioli/sounds/213608/
	http://www.freesound.org/people/Ears68/sounds/185150/
	http://www.freesound.org/people/Woodylein/sounds/149817/
	http://www.freesound.org/people/BROKENTEST/sounds/1zz1/
	11__bram__sweep-log.wav
	http://www.freesound.org/people/jediduke/sounds/126518/
	http://www.freesound.org/people/Philip%%20Goddard/sounds/172163/
	211985
	http://www.freesound.org/people/qubodup/sounds/211985/
	http://www.freesound.org/people/qubodup/sounds/211983/""" % (url_for('credits'))
	html_content = html_content + html.escape("""
	&asd&asd&
	<br>
	<p style="background-color: red;">hm</p>""") + "</pre></div>"

	return html_header + html_content + html_footer

	@app.route('/credits', methods=['POST'])
	def credits():

	content = html.escape(request.form["ids"]).split("\n")

	sounds = []
	sounds_failed = []
	sounds_duplicate = []

	for i in content:
	tmp = i.split("\n")[0].strip() # from text file input times, leaving just in case
	#tmp = i
	tmp2 = tmp.split("__")[0]
	tmp3 = tmp.split("/sounds/")
	if len(tmp3) > 1:
	tmp_final = tmp3[1].split("/")[0]
	else:
	tmp_final = tmp2

	# check if result is a duplicate
	duplicate = False
	for sound in sounds:
	if sound[1] == tmp_final:
	duplicate = True
	# keep track of duplicates
	sounds_duplicate.append("(\"" + i + "\" : \"" + tmp_final + "\")")

	# add item to sounds table if it's not a dupliacte
	if not duplicate:
	# include original item for error feedback
	sounds.append([i, tmp_final])

	database = []

	allUserNames = set()
	allLicenseNames = set()

	for sound in sounds:

	url="http://freesound.org/apiv2/sounds/" + sound[1] + "/?fields=name,url,username,license&format=json" + "&token=" + FSKEY
	try:
	freesoundResponse = urllib.request.urlopen(urllib.request.Request(url))#, headers={'User-Agent': 'Mozilla/5.0'})) # I tried user agent because I first got 403 but it was actually a restriction by pythonanywhere. might not be needed but I'm afraid to break the request for now
	except Exception as e:
	sounds_failed.append("(\"" + str(sound[0]) + "\" : \"" + str(sound[1]) + "\") " + str(e)) # + str(url) + " "
	else:
	jsonResponse = json.loads(freesoundResponse.read().decode("utf-8"))

	# metadata one-liners
	database.append({
	'soundName' : jsonResponse['name'],
	'soundLink' : jsonResponse['url'],
	'userName' : jsonResponse['username'],
	'userLink' : "http://"+"/".join(jsonResponse['url'].split("/")[2:5]),
	'licenseName': getLicenseName(jsonResponse['license']),
	'licenseLink': jsonResponse['license']
	})

	# keep track of users and licenses involved
	allUserNames.add(jsonResponse['username'])
	allLicenseNames.add(getLicenseName(jsonResponse['license']))

	# for all licenses, in order of unknown, by-nc, sampling+, by, cc0
	# NEEDS UPDATING PROBABLY FOR CCBY4.0
	orderedLicenseNames = [
	"Unknown License",
	"CC-BY-NC 3.0",
	"CC-Sampling+",
	"CC-BY 3.0",
	"CC0"
	]

	credits = ""
	creditsHTML = ""

	# order credits by most strict licenses at the top
	for orderedLicenseName in orderedLicenseNames:

	# for each existing license
	if orderedLicenseName in allLicenseNames:

	firstLicense = True
	userNeeded = False #since we loop over all users, we need to know when one actually matches for ending user credits
	tmpUserName = ""
	tmpUserLink = ""

	# for each existing user
	for userName in allUserNames:

	# for each sound
	for item in database:

	if ((item['userName'] == userName) and (item['licenseName'] == orderedLicenseName)):

	userNeeded = True
	tmpUserName = item['userName']
	tmpUserLink = item['userLink']
	if firstLicense:
	credits = credits + "<br>Sounds under " + item['licenseName'] + " ( " + item['licenseLink'] + " ):<br>"
	creditsHTML = creditsHTML + "<br>Sounds under <a href='" + item['licenseLink'] + "'>" + item['licenseName'] + "</a>:<br>"
	firstLicense = False

	credits = credits + "\n" + item['soundName'] + " ( " + getShortLink(item['soundLink']) + " ), "
	creditsHTML = creditsHTML + "\n<a href='" + getShortLink(item['soundLink']) + "'>" + item['soundName'] + "</a>, "

	if userNeeded:
	creditsHTML = creditsHTML[:-2] + " by <a href='" + getShortLink(tmpUserLink) + "'>" + tmpUserName + "</a>\n<br>\n"
	credits = credits[:-2] + " by " + tmpUserName + " ( " + getShortLink(tmpUserLink) + " )\n<br>\n"
	userNeeded = False

	# chop off leading and tailing break
	creditsHTML = creditsHTML[4:][:-5]
	credits = credits[4:][:-5]

	# simple credits
	allUserNamesSingle = list(set(allUserNames))
	allUserNamesString = ', '.join(allUserNamesSingle)
	allFailedString = '<br>'.join(sounds_failed)
	allDuplicateString = '<br>'.join(sounds_duplicate)

	html_content = """
	<h1>Freesound Super Credits Results</h1>
	<p>Should there be issues with the results, please report to qubodup@gmail.com</p>
	<p>Use your browser "back" button/function to return to your input (might not work in any browser) or <a href="%s">start new query</a></p>
	""" % (url_for('hello_person'))

	if (len(sounds_failed) > 0):
	html_content = html_content + """
	<div class='error'>
	<h2>Failed items</h2>
	<p>%s</p>
	</div>
	""" % (allFailedString)


	if (len(sounds_duplicate) > 0):
	html_content = html_content + """
	<div class='warning'>
	<h2>Duplicate items</h2>
	<p>%s</p>
	</div>
	""" % (allDuplicateString)

	html_content = html_content + """
	<h2>Plaintext credits</h2>
	<p>%s</p>
	<h2>HTML credits (WYSIWYG)</h2>
	<p>%s</p>
	<h2>HTML credits (code)</h2>
	<textarea>%s</textarea>
	<h2>Additional minimal credits</h2>
	<p>Simple comma-separated list of all usernames:</p>
	<p>%s</p>
	""" % (credits, creditsHTML, creditsHTML, allUserNamesString)

	return html_header + html_content + html_footer

	# Creates attribution html text
	def attribution_construct_html(item):
	return (
	"<a href='" +
	item.soundLink +
	"'>" +
	item.soundName +
	"</a> by <a href='" +
	item.userLink +
	"'>" +
	item.userName +
	"</a>" +
	"\n" +
	item.licenseName +
	"<br>" # since this is for HTML output
	)

	# Creates attribution plain text
	def attribution_construct(item):
	item.sndAttribution = (
	"By " +
	item.userName +
	" " +
	item.sndUserUrl +
	" under " +
	item.sndLicense +
	"\n" +
	item.sndTitleOrg +
	": " +
	item.licenseName +
	"<br>" # since this is for HTML output
	)