tipue_search abandoned¶

Element theme documentation suggests using the following configuration to enable search, which relies on the tipue_search plugin:

PLUGINS = ['tipue_search']
DIRECT_TEMPLATES = ['search']

Unfortunately, this plugin and the jquery code it relies on is abandoned, as stated on Pelican Plugin's page of the plugin:

pelican_search plugin should be used instead, but how?

pelican_search abandoned¶

pelican_search relies on stork and this tool should be installed prior to install and run the Pelican plugin:

Install Stork on Ubuntu:

cd ~/bin
wget -O stork https://files.stork-search.net/releases/v1.6.0/stork-ubuntu-20-04
chmod +x stork

Unfortunately, running stork fails on Ubuntu 22.04:

$ stork
stork: error while loading shared libraries: libssl.so.1.1: cannot open shared object file: No such file or directory

At this point, I started searching for a solution. Found one for 2-years old Ubuntu 20 (https://stackoverflow.com/a/72633324) that required manually installing a specific version of OpenSSL and that I would not have followed even if it applied to Ubuntu 22.

And then I found out the one maintainer of the project had stepped out a year ago:

• Maintainer's post on winding down from project

Use Google search instead¶

I don't have enough time to investigate alternative solutions to stork and the pelican_search plugin.

Change the search form¶

However, I can customize the Elegant Theme search field, that I have forked.

In file base.html, I can replace

<li><form class="navbar-search" action="{{ SITEURL }}/{{ SEARCH_URL }}" onsubmit="return validateForm(this.elements['q'].value);"> <input type="text" class="search-query" placeholder="Search" name="q" id="tipue_search_input"></form></li>

by the following and create a Google search in a new window with site:www.javatronic.fr as a filter:

{% if SEARCH_URL %}
<li><form class="navbar-search" action="{{SEARCH_URL}}" onsubmit="return validateForm(this.elements['q'][0].value);" target="_blank"> <input type="text" class="search-query" placeholder="Search" name="q" id="tipue_search_input"><input type="hidden" name="q" value="site:{{ SITEURL }}"></form></li>
{% endif %}

Enable indexing by Google¶

To appear in search results, the website must first be crawled and indexed by Google's robots.

This can be achieved by adding a robots.txt file to the root of the website and creating a site map.

Add robots.txt to Pelican

Create file extra/robots.txt, informing that no robot (not only Google) can crawl any page of the website but those indicated in the referred sitemap file.

User-agent: *
Disallow:

Sitemap: https://www.javatronic.fr/sitemap.xml

Have this file deployed to the generated website by adding the following to pelicanconf.py:

STATIC_PATHS.append("extra/robots.txt")
EXTRA_PATH_METADATA["extra/robots.txt"] = {"path": "robots.txt"}

Create a sitemap

There is a plugin to generate a sitemap within the official pelican plugin repository.

For the small additions it provides, I decided to use the pelican-extended-sitemap plugin.

1. Add pelican-extended-sitemap to requirements.txt
2. In pelicanconf.py, enable pelican-extended-sitemap with

   PLUGINS.append("extended_sitemap")
   

3. Since my resume is a page, I want to boost the priority of pages compared to the default configuration and I'll overwrite EXTENDED_SITEMAP_PLUGIN in pelicanconf.py:

   EXTENDED_SITEMAP_PLUGIN = {
       'priorities': {
           'index': 1.0,
           'articles': 0.8,
           'pages': 0.9,
           'others': 0.4
       },
       'changefrequencies': {
           'index': 'daily',
           'articles': 'weekly',
           'pages': 'monthly',
           'others': 'monthly',
       }
   }
   

Ask Google to reindex¶

Simply waiting a few days for Google to reindex the website didn't work for me.

I decided to use the submission of a sitemap in Search Console to get re-indexed.

verify website property

To add a website to your Google Account's Search Console, you first need to prove you own it.

1. Go to Google's search console: https://search.google.com/search-console/welcome?action=inspect
2. Select the Inspection method, I selected "URL prefix" as I have subdomains I don't need/want indexed
   
3. Select property method with HTML file (the default one)
   
4. Download the HTML file, store it in the content/extra directory
5. Keep note of the original name of the file and rename it to google_property_verification_file (or any other name without the .html suffix to avoid Pelican error not finding a title)
6. Add the following to pelicanconf.py, using the original name of your file:

   STATIC_PATHS.append("extra/google_property_verification_file")
   EXTRA_PATH_METADATA["extra/google_property_verification_file"] = {"path": "googled2a1d235c02ddb0d.html"}
   

7. Commit and publish the file and changes to pelicanconf.py
8. Click on "Validate"
9. After a few seconds, the Search Console for the website is accessible

submit the sitemap

1. In the Search Console, go to sitemaps
2. Input the URL to the sitemap.xml file
   
3. Validate and confirm, after a few seconds, whether Google successfully read the sitemap

verify indexing is in progress

In the search console, go to "Pages" and confirm indexing is pending