It's a messy, confusing screen.

1. By default, the services should be hidden on the options page, with a button to toggle "Show Advanced Options". That will reduce the confusion for new users, so they can easily see simple options to choose the default service and mode.

2. There is a lot of wasted space around the form elements in the advanced options (Bible Services). Convert the layout into a table, and use a content-editable functionality that will convert the table cells into input fields when clicked.

When changing preferences in the options, the user needs to refresh any page that is already open before the generated links are updated.

We need to send a message when options change, and the content script needs to listen for the message, then update all links.

Also, let's implement a cache-breaker in the data-scripture-tips-processed attribute, to keep track of whether our links need to be updated.

So, we first increment and store the cache breaker value (use timestamp of options last update), then any of our links that do not match that cache breaker need to be updated.

Need to convert Quoran references to links in the same way as Bible references.

Need to identify common WYSIWYG editors and implement a hook where available, to intercept the user's form submission with a prompt: "Would you like Scripture Tips to automatically convert the references to links?".

The prompt shouldn't show if there are no scripture references in the post.

Also, for other popular services:

Facebook.
Twitter.
Youtube.
... etc

The present search methodology is quite restrictive and doesn't catch the variety of shorthand spellings that people use. For example, when a person references the book of Matthew, they might say

Mt 13:12 or Matt 13.12 or Mat. 13:12, or even Matthew 13:12.

Furthermore, sometimes a bible service website will have their own shorthand, as for example BibleHub.com does not support Camel Case, so a text-transform to lowercase is needed in the urls.

1. A CSV upload and processing feature:

First column is the official book name that is used in the generated urls. Any subsequent column on that row is a valid name that maps to that name. Eg:

Matthew,Mt,Mt.,mt,mt.,matt,matt.,Matt,Matt.
Songs,Songs of Solomon,Songs Of Solomon,songs of solomon,sos,SoS,SOS,songs,sgs

There should be one master CSV record, that is used for all services, and the user should be able to download, edit and upload that file.

Each service must also allow a custom CSV to be used in preference to the master CSV.

Note: no locale-specific functionality is required in this feature. If a user wishes to match Matthew in Spanish as Mateo, they can achieve that by adding Mateo and mateo to the spreadsheet. If the website service they are using requires Mateo, then it should be spelled that way in the first column of the spreadsheet and Matthew should be placed in another cell on that row.

Create a comprehensive CSV as a starting point, that is permanently bundled in the extension and is processed as an upload to the master CSV record when the extension is installed.

2. Update the search algorithm to support period (.) separator or colon (:) separator, where presently it will only match a colon pattern. Eg:

Matthew 13.12 will not match, but it should, and Matthew 13:12 does match, as it should.

The configuration specifies that all divs that have no child nodes should be processed. This is intended to prevent recursive processing, but it is causing the text content of a div to not be procesed at all, when that div contains child nodes.

Need to change the configuration to operate on divs that contain no child divs, even if that div contains other child nodes.