Skip to content

Commit

Permalink
Update index.md
Browse files Browse the repository at this point in the history
  • Loading branch information
ProgrammerNomad committed Jan 6, 2021
1 parent 835dfb3 commit 8d65945
Showing 1 changed file with 11 additions and 26 deletions.
37 changes: 11 additions & 26 deletions pages/index.md
Original file line number Diff line number Diff line change
Expand Up @@ -4,38 +4,23 @@ title: Domaindumper API
permalink: /
---

# Welcome to Docsy Jekyll
![CI](https://github.com/rundocs/jekyll-rtd-theme/workflows/CI/badge.svg?branch=develop)

This is a starter template for a docsy jekyll theme.
Domain Dumper daily states of domains

![assets/img/docsy-jekyll.png](assets/img/docsy-jekyll.png)
API for stats for daily registered domains. Here you can get daily updates for how many domains registered on every TLD. we are updating our files in every 24 hours,

## Purpose
# How to use
It's simple, we update states in JSON files, you only need to curl from its data directory.

GitHub pages uses Jekyll natively, so when I make documentation, I typically
look for Jekyll templates. Why? Using Jekyll means that I can use markdown,
and allow for users to easily contribute, and build automatically just by
way of pushing to a master branch (or general GitHub pages).
I found Docsy, a beautiful Hugo template, but it requires hugo with GoLang
which doesn't render natively on GitHub pages. For this reason, I've spent
some time creating a custom Jekyll template that is (almost) as beautiful,
and includes all the features that I might want.
# Example:

## Features
Get domains stats from 15 July 2020

What are these features? You should see the {% include doc.html name="Getting Started" path="getting-started" %}
guide for a complete summary. Briefly:
`curl https://api.domaindumper.com/database/stats/2020/07/15/stats.json`

- *User interaction* including consistent permalinks, links to ask questions via GitHub issues, and edit the file on GitHub directly.
- *Search* across posts, documentation, and other site pages, with an ability to exclude from search.
- *External Search* meaning an ability to link any page tag to trigger an external search.
- *Documentation* A documentation collection that was easy to organize on the filesystem, render with nested headings for the user, and refer to in markdown.
- *Pages* A separate folder for more traditional pages (e.g, about).
- *Navigation*: Control over the main navigation on the left of the page, and automatic generation of table of contents for each page on the right.
- *News* A posts feed for news and updates, along with an archive (organized by year).
- *Templates* or specifically, "includes" that make it easy to create an alert, documentation link, or other content.
- *Continuous Integration* recipes to preview the site
Please change your year, month and date according to your need. The file name will be always same.

# Need premium services

For features, getting started with development, see the {% include doc.html name="Getting Started" path="getting-started" %} page. Would you like to request a feature or contribute?
[Open an issue]({{ site.repo }}/issues)
if you need premium services like domains Whois OR Free daily registered domain names, then you can find this information our other website: https://www.whoisextractor.in/

0 comments on commit 8d65945

Please sign in to comment.