Ever wonder what a Web page would look like if just the plain HTML were shown, as if it were 1996 again? Ah, but which HTML? There are actually two forms of HTML that reach your browser, and there can be quite a disparity between these two types.

The first is what you see after the raw HTML has been pummelled about by CSS and javascript and the browser’s interpretation. This is often referred to as the ‘DOM’ HTML. This code is what you see and navigate through if you ‘Inspect element’ in your browser, or if you block an element with uBlock’s Element Picker tool.

The second type is the HTML code that gets sent to the browser in the first place, and that original is kept pristine and effectively ‘under’ the Web page. It can be seen via: right-click on page / ‘View source’. This source can then be selected and copied with a Ctrl + A / Ctrl + C keyboard command. Or it can be saved out when you ‘Save page as…’ / ‘Save as HTML only’, and from there you can re-open the saved page in the browser. Some remote CSS, javascript and images may still be called, even then.

A quicker way to ‘see’ this original without its CSS and other ‘remote-code’ flibbertigibbets is to install the add-on disable-HTML in your browser…

The addon is quite simple to use, and though old still works fine. It was somewhat mis-named, as it can robustly block everything except the HTML of the actual ‘page source’. With CSS and javascript blocked, it appears to be blocking the DOM version of the HTML from emerging from the page source. So what you see displayed, on page re-load, is effectively the page source. As such it can be quite handy for the removal of some types of especially tough and obstreperous CSS-and-javascript -driven overlays, in a situation where you don’t much care about the fancy wrapping and just want the words in a readable and/or copy-able form. Such as on the vile overlays of the unherd.com site.

I’ve expanded my free PDF file, “Some useful regex commands for Notepad++”, which had been released here in May 2019. Here’s the new one…

Update: now updated again My Little Regex Cookbook, for Notepad++ (September 2020).

Situation: You need to cleanly extract just the usernames from a .JSON file, and place each name on a new line. The result should look like…

name1
name2
name3

The value being extracted from the JSON could just as easily be email addresses, or map co-ordinates, or suchlike.

Why you might do this: You can’t just do a simple “.JSON out, .JSON in”. For instance, let’s say you have a Web browser add-on that offers a blocklist function based on usernames. Perhaps it’s the deviantART-Filter. You want to port this 1,600 name blocklist of scum-and-villainy over to a similar browser add-on. Perhaps it’s the dA_ignore UserScript. The old deviantART-Filter usefully exports a .JSON file of your blocked users. But…. the new dA_ignore can only import a pasted-in list of usernames, one per line.

Solution 1:

There is a working Regex for doing this, which only requires a recent copy of Notepad++ and a suitable .JSON file for the process…

It’s been tested and works. The use of the * wildcard will enable the extraction of a list in the form…

path_label”:”any_value_here

After this regex has processed the code you are thus left with a list that looks like…

username”:”name1
username”:”name2
username”:”name3

… and you need only to do a basic search/replace to clear username”:” to obtained your cleaned list of the various unique usernames.

In this use-case you go to your own DeviantArt Settings page, and paste in the new ported-over blocklist.

Solution 2:

1. Open the .JSON with the genuine Windows freeware JSONedit.

2. Filter the JSON on field “username” by typing username into the search box. In ‘List view’ you should now only see a list of the “username” fields and the adjacent data entry.

3. Then go: top menu | ‘Tools’ | ‘Export as .CSV’.

4. Add a file extension .CSV to the saved file if needed, and then open it with Excel. Select and copy out all the “usernames” column to a new Notepad++ file.

5. In Notepad++, a quick search-replace will then remove the “” marks. You now have a clean portable list, with one username per line.

Solution 3:

There are, of course, cloud services in Whereizitagain that may well offer to do this. But the above methods use Windows freeware and are thus more secure.