How to add pages to the IGSR website

This repo contains the static content for the IGSR website hosted at http://www.1000genomes.org/

New general pages

To add a new general page to the website, create a file ending in with the extension .md in the top level directory. The file name does not influence the link or page name but should reflect the contents of the page. At the top of any new file you need to add the following header

---

layout: NAME OF SECTION LAYOUT (normally single_section)
title: "TITLE FOR PAGE"
permalink: URL FOR PAGE

---

The URL needs to be close in form to the page title e.g a page titled The 1000 Genomes Phase 3 publicaation might want a url /1000_genomes_phase3_publication/

The contents of the page are then written in markdown format. Please see https://github.com/adam-p/markdown-here/wiki/Markdown-Cheatsheet for more details

New FAQs

An FAQ should be created in the _faq directory. The filename needs to end with the extension .md. The filename is used as the URL for FAQs. Like normal pages the file needs a header. For an FAQ the header looks like

---

title: "TITLE FOR PAGE"
faq_tags:

• descriptive tag used to group pages (see https://github.com/igsr/gca_1000genomes_website/tree/master/_faq_categories for full list of current categories faq_related:
  -Name (-.md) of any related FAQ e.g there_corrupt_file-your-ftp-site

---

Please look at http://www.1000genomes.org/faq/can-i-find-genomic-position-list-dbsnp-rs-numbers-0/ and https://github.com/igsr/gca_1000genomes_website/blob/master/_faq/can-i-find-genomic-position-list-dbsnp-rs-numbers-0.md for a good exemplar FAQ

New Announcements

New announcements must be posted in the annoucements/_posts directory
The file name should follow the convention YYYY-MM-DD-name.md. The URL will be the name without the date in it e.g the file 2016-04-26-1000genomes-chry-publication.md has the link http://www.1000genomes.org/announcements/1000genomes-chry-publication/

The announcement file also needs a header

---

title: "ANNOUNCEMENT TITLE"
pinned: [true|false] - this indicates if the page is pinned to the front page. If you want to pin a new announcement you must unpin an old one

---

Currently these announcements are not automatically tweeted. This is functionality we plan to add but it doesn't exist yet. If you make a new announcement please tweet it from the 1000genomes account

Directory strucutre of this repo

non-underscored directories

Directories that do not start with underscore (_) are for url paths. For example, a file called vcf4.0.md in the directory /wiki/Analysis/ will be served as the url www.1000genomes.org/wiki/Analysis/vcf4.0

_elasticsearch_settings

We have an elasticsearch server at EBI which we use for the search feature in the 1000genomes website. This directory contains some files that we use to configure elasticsearch.

_faq

Any file in this directory will be served in the website FAQ section. The yaml front matter should contain:

1. title - the title gets displayed at the top of the FAQ page and in the list of FAQ questions at www.1000genomes.org/faq
2. faq_related - an array of other files from the _faq directory (without extension)
3. faq_tags - an array of files in the _faq_categories directory (without extension)

_faq_categories

Any file in this directory becomes a category for grouping together related faq pages. The yaml front matter should contain one thing only:

1. title - the title of the category gets displayed at the top of the faq pages belonging to that category

_includes

This contains any little fragment of html (or markdown) that you can include from anywhere in any page e.g. {% include google_analytics.html %}

An important one is _includes/homepage.html. This is included from both index.html and from home.html. We do this so that these two urls show identical content: www.1000genomes.org/ and www.1000genomes.org/home.

_layouts

Files in this directory define the layouts available for any page you create.

• single_section: You probably want to use this layout. It gives you the basic layout of a page on the 1000genomes website with a single white sheet on which to put content
• multi_section: This gives you a basic layout but you have the option to add multiple white sheets separated vertically by grey space. Insert a white sheet section like this: {% capture my_content %}

  blah blah blah

  {% endcapture %}{% include section.html content=my_content %}

_media

Any file in this directory gets listed on www.100genomes.org/media. Files should be named starting with the date e.g. 2016-02-19-beano.html. The yaml front matter should contain:

1. link_url - link to the external website displaying the article
2. publication - name of the newspaper / website displaying the article
3. date - format yyyy-mm-dd
4. title - title of the article

_mojo_templates

At EBI we serve the static pages from a mojolicious webserver. This directory contains some template files which mean something to mojolicious.

_plugins

Jekyll has a plugin system with hooks that allow you to create custom generated content specific to your site.. Note that because we use plugins our site will never work on github pages. Nevermind.

stub_name_generator.rb

This plugin does several things which were helpful in the faq pages and faq list pages for creating links to other faq.

1. Every faq and faq_category is automatically given an attribute 'stub_name' which is the faq/faq_category filename stripped of the extension
2. The site variable site.faq_map is a hash. Key is the stub_name of each faq. Value is the faq page itself.
3. The site variable site.faq_category_map is a hash. Key is the stub_name of each faq_category. Value is the faq_category itself.

_press_release

Any file in this directory gets listed on www.100genomes.org/media. Files should be named starting with the date e.g. 2016-02-19-i-love-igsr.html. The yaml front matter should contain:

1. link_url - link to the press release
2. date - format yyyy-mm-dd
3. title - title of the press release

Other files that mean something

Gemfile and Gemfile.lock

These are here so that jekyll can build the site properly. Gemfile contains a list of 3rd-party plugins we are using

_config.yml

Standard jekyll configuration explained on the jekyll website

_mojo_redirects

At EBI we serve the static pages from a mojolicious webserver. This file contains a list of urls that should give a 302 redirect. We have configured the webserver to read this file.

sitemap.xml

This file is important for google site indexing

robots.txt

This file is important for google site indexing. It tells google where to look for the sitemap.xml

search_index.json

This json file gets populated with the content of all pages. This all gets loaded into EBI's elasticsearch server and is used when you search the website from the search box.

_README.md

You're reading it.