this site is 18+ and its visible: https://hentaihaven.xxx/

release criteria:
chrome extension
android

title

Goal: Add the nsfw subreddits to PorNo! (note: this list of subreddits will 99.9999% grow in the future).

Strategy 1:
Copy paste everything from (you might want to open this in incognito...) here via console commands (please don't copy paste by hand, learn a bit of JavaScript and save the time).
This method will be effective at the beginning for an "initial grab of URLs" but is very time consuming in the long run (how will you know what links are new?)

Strat 2:
While this method will allow the subreddit to load some content, this is as "dynamic capture" as we can get.
Unless....you know of a way to scan DOM before page load?

Programmatically find the nsfw tag that NSFW subreddits keep in their "About" section.
If the subreddit has the nsfw tag:

• Add the domain to the banlist (eventually push the list to the cloud database)
• Redirect the user

window.onload = () => {
  let nsfwClassName = '_1wzhGvvafQFOWAyA157okr';

  // Reddit check
  // ! Elements dont load until after the page loads.
  if (domain === 'reddit.com') {
    if (document.getElementsByClassName(nsfwClassName).length) {
      window.stop();
      // ...and so on
    }
  }
}

Users have asked for personal safe/banlists...
PorNo! is only supposed to block porn, so for now Im more interested in implementing a "Report porn url" feature instead of custom filters. I'm curious what feedback users will give because im still open to the idea of custom lists...

Process, kind of:

Add a REPORT THIS SITE button next to the EMERGENCY button
onReportClicked() => currentURL is put in local banlist and added to firestore "REPORT suggestions" list

For now, I'll check the suggestions periodically and add em to the core lists if they are indeed porn sites and not like
a user wanted to block twitter or something so they "reported it" for their own benefit...tho this could be problematic cuz
the user would need the ability to edit and delete entries from their blocked list ugh

WAIT

onReportClicked => CurrentURL is sent to IBM api that checks category of site. 
isPorn ? redirect and update databases : do nothing

What about when the ibm api hits the limit every month?
Hmm good question I dont know lmao ugh i guess custom lists is the move to allow both user freedom while providing
developer flexibility for "finding new" URLs that need to be blocked.

o Add button
o Add logic to button
o Add interface to CRUD custom ban / safe lists (how are we even gonna ensure ppl dont put porn sites on safe lists though...ah, maybe a "report site as not porn button" idk w/e)
o Publish
o Awesome

PorNo! is a URL-based porn blocker. As a result, PorNo! needs to get the URL of a website.

1. Get the URL of a site
2. Construct any additional URLs because some sites have curious NSWF hrefs (ie. hostname + path, for subreddits; "boards.4chan/abc123/")

Great, now we need to check that our URLs are porn sites (or not).

3. Verify our URLs are in the database via pornMap[...]
4. pornMap[...] ? redirect() : postWindowLoad() // tba

--

Just because a site isn't in our database doesn't immediately qualify the site as non-pornographic.

postWindowLoad() is a function that will parse the contents of a webpage after it loads. If the webpage seems NSFW, call a web category-checker API on that website for "official evaluation". (Also....add a way for users to report websites because they will surely encounter porn sites that should be blocked LOL).
Another issue will be opened for this at some point in the future.

Idk what demand there is, but I think it would be a helpful package.

Sample usage:

isPorn(MyURL) ? abc() : xyz()

Can we make this happen?

Urban Dictionary is on the list and it shouldn't be.

http://www.bigporn.com

See #20 for Chrome extension implementation

The XFORCE api we were using now requires api keys for access -- before, we could just visit the url constructed and receive the desired data.

I would prefer a free to use API as opposed to having to pay for api calls should we ever reach X calls a month, so let's find some free web DB apis and implement those ^_^

then find a way to push to iOS devices bwahahaha

Baidu is a search engine like Google -- let's put it on a safe list, not sure why it was blacklisted lol
Thx Ted

Users will regularly leave reviews that they don't understand English and therefore cannot use the app :(

Google Translate is a great start tbh

fightthenewdrug.org is not an adult website, we'll add an exclude for this site. Please add issues for other non-adult sites that get flagged!

Clothing website

unflag

Write a script to generate the Java file with the list of URLs (and keywords, though I have no idea what we will do with the keywords at the moment...).
See java.py for current progress.

Why?
Code generators for the lists will make it a lot easier and faster to release list updates in the future:

• JavaScript => Chrome Extension
• Java => Android app
• ??? => Your favorite environment

dont block vox news

triggered

Idea: convert google searches into safe mode searches...

idk...........

http://www.ixxx.wtf/

Browser market share

No support for Safari thru the Android app. iOS PorNo!, then?

wtf is going on

Resolve w next store release

ipchicken.com is just a webste that tells you your IP address. No porn here! but im not allowed to access
Thanks for this plugin its really great.

French internet operator, blocked for some weird reasons.

As noted in the README, performance is a crucial element to the success of this application. This application also employs the use of large hard coded arrays. Due to the nature of the search, we can potentially implement a broad-phase for discarding obvious, non-matching urls.

Solution:
You can transform your list to have an array of websites indexed by the first character each url starts with. The transformation of this list would likely have to be done with a script or by hand, because we (assumedly) don't have the time to manipulate it during runtime.
ex:

[
     ["a", ["awebsite","anotherwebsite"]],
     ["b", ["bxxxxx","byyyy"]]
     ["c", ["c.dummyurl"]]
]

When determining if a string exists in the list, you may trim your potential matches significantly, by grabbing the list that corresponds to the first character in the url being checked.
In the above example, if we were checking the existence of the url "bzyz" in the list, we need only check the list ["bxxxxx","byyyy"]. This drastically improves our worst case scenario for url searches.

Disclaimer:
I have very limited knowledge of JS so the details of the implementation would be up to you, whether it's nested arrays or a hashmap. Ideally we could make use of a Trie, but the creation of such a datastructure with so many elements would cause more trouble than it's worth in such a small and runtime-oriented application.

Let me know if there's anything I can clarify,
Congrats on the release brother

https://niceporn.tv

these sites should not be flagged, gotta remove em

Will have to test again later, was just surprised to see the site load after I "thought" this link was in the list...

add this site

see now biqle.ru does serve porn without changing the url path -- as a result, i can't detect pornography unless i parse the webpage, but at that point is a bit too late
ill investigate biqle.ru at some point.

Great pull request: add code that can parse a webpage's content. If the page is determined to be pornographic / serving pornographic media, redirect from that page WITHOUT adding that domain to the list.

• remove cda.pl from lists.js

Resolve w next store release

Is possible? I've seen it done before....I think disable incognito app can prevent its own uninstall?

Resolve with next store updates

Found more lists:
https://block.energized.pro/extensions/porn-lite/formats/domains.txt
https://block.energized.pro/porn/formats/domains.txt
https://energized.pro/support/

--

Thoughts:
As the static lists keep growing, this will only take up more space on the release builds. Additionally, I think the chrome browser loads the extension's code from scratch each time you visit a website (no source, just from what I've seen on inline alert statement executions) and i wonder if adding more to the file size of the list will slow down chrome's loading of porNo chrome extension

Finding a way to dynamically capture a site is important, but right now i gotta get a nice and solid roadmap going to have something to focus on laisjflajfjiof

As a developer of a porn-blocking software that is used on Android devices at api level 15*,
I want to use the findAccessibilityNodeInfosByViewId (api level 18) to identify the user's current URL in addition to keeping the existing detection methods
So that we can keep the software functional for the margin of users on api15 while providing users at and above api18 a much better detection system.

Work description:

o Maintain the minLevelSDK or whatever flag in the gradle and elsewhere to api 15.
o Abstract the existing detection method to its own function and call this function as follows:

deviceApiLevel >= 18 ? use findAccessibilityNodeInfosByViewId : use existingDFSMethod

o Publish the update to Google Play

* = approximately 2.8% of PorNo! users. This statistic is subject to change, of course lol