Web Accessibility is not only for people with impairments; instead, it's for everybody. Web accessibility results in better user interaction with the website by enabling the utilization of tools and technologies that aid the site's perception, understanding, operability, contribution, and navigation.
There are more than 6 billion mobile subscribers worldwide. Recent research from the World Health Organization says that about 15% of the world's population is in some way 'specially-abled.' And it’s usually believed that people with impairments own significantly fewer mobile devices than people without impairments. But this number is increasing exponentially.
A few months ago, Aaron Crosman posted What I Brought from Drupal to Salesforce. While I've done a lot of CMS/CRM integration work, I've only been integrating Drupal with Salesforce and its related services for a few years. I still consider myself new to Salesforce development and I'm still learning about the open source side of Salesforce at events like Salesforce Community Sprints.
If I do get something wrong, please let me know and I'll update the posts.
Another reason for writing a series highlighting some of the similarities between Salesforce and Drupal is to respond to a recent post by Jacob Rockowitz questioning whether his Blueprint project that leverages Schema.org had a future.
I briefly mentioned Blueprints in a presentation at BADCamp, but I wanted to dive in deeper into Blueprint as well as some of the other features starting to mature in "modern Drupal" where I've seen similar approaches working well in Salesforce.
To keep myself sane, I'm breaking this up into 3 parts;
The UI in the work coming out of the Drupal Association's Project Browser Initiative is very similar to Salesforce's AppExchange (and MetaDeploy, AppExchange for open source and Commons supported Salesforce packages).
Visually, the UX of the Project Browser and AppExchange are very similar.
Out of the box, Project Browser isn't really that exciting for developers.
So I can search for modules inside the application I'm building and then go to Composer to
composer require drupal/[PROJECT NAME]? Why would anyone get excited about this?
The exciting part of Project Browser isn't using it to browse the same projects on Drupal.org in a different UX, it's being able to customize that experience for a specific use case or infrastructure. A feature that gives users a list of projects they can install directly on a test/sandbox version of their site is a game changer in a higher ed use case.
On the CMS side, we wrote something similar to Project Browser the University of Colorado in Drupal 7 we called Profile Module Manager. While the colorado.edu sites are now run from a monorepo/custom upstream approach on Pantheon, the original on-prem infrastructure paired Profile Module Manager with a custom devops solution to add a "bundle" of code to a site's codebase. The user experience of Profile Module Manager within Web Express in D7 and Project Browser with a customized project feed in D10 will be very similar.
Project Browser won't really be exciting until it can be combined with the Auto Update Initiative work. That work requires Composer 2.3.5 or later which many hosts (including Pantheon) do not support yet.
On the CRM side, when browsing packages and clicking Get It Now of free packages will bring up a prompt asking you where to install the package based on instances you have registered with the account you are authenticated with... or to spin up a new sandbox to test just this package.
While Drupal's Project Browser UX is designed to be used with the CMS instance you are planning to install the package on, it's not hard to imagine large, Drupal centric hosts like Acquia or Pantheon offering customized Project Browser feeds that list platform friendly/approved modules.
Modern Drupal still has to define a way to install front end dependencies required by PHP projects. Salesforce solves this to a certain extent with Lightning Web Components, their open source Web Component foundation.
While there is some traction around #2873160 to use NodeJS installer for Composer and #3340712 to get single directory components into Core, this is still going to be a challenge.
It is also important to acknowledge that most packages you can install through the different Salesforce project browsing services are NOT free or open source. Between the AppExchange and MetaInstall services, you'll find 4 different types of packages.
Paid plugins are more common in the WordPress ecosystem, but the underlying plugin code is considered a derivative of WordPress where distribution triggers the GPL-2.0 or later licensing requirement.
Where I think this is going to get interesting is the potential for more commercial Drupal packages in SaaS offerings. We've already seen some large hosts charge for value added services for sites hosted on their infrastructure like Acquia Site Studio. We may see more groups exploit the GPL SaaS Loophole enabling customers to install commercial modules and themes from customized Project Browsers.
As long as the end-user is interacting with your software over a network and you control the hardware / infrastructure the software is running on, that is not considered distribution.
While different than traditional, pure GPL Drupal sites, after seeing the quality in competing packages in Salesforce, I personally think a tier of commercial, closed source modules would be good for Drupal.
On 14th December 2022 Drupal 10 was released. We can't say we immediately set about the task of upgrading all our Drupal 9 sites, but we did start thinking and planning to move all our sites in 2023.
I thought it might be fun to take a look at how we do an upgrade and specifically the upgrade of this very website.
So over the next few weeks we'll be slowly getting this site ready and as up to date as possible before making the jump up to Drupal 10.
We've got a few challenges ahead, this site was originally a Drupal 8 site and has a fair amount of technical debt, we're going to need to pay down some of that debt and make some fairly major changes along the way.
We're going to need to make some 'infrastructure' level changes along the way too and ensure our hosting will cope with Drupal 10, and we're going to need to test out the new site to make sure that everything works as it should etc.
Keep an eye out over the next few weeks for more articles in the series and follow along and apply the techniques here to your own sites.
Jeff Geerling recently announced that the DrupalVM project is officially archived. DrupalVM was crucial in helping provide standardized local development for Drupal developers. DrupalVM was a customized Vagrant setup with batteries-included tooling powered by Jeff's many Ansible roles.
For me, it's wild to think back to 2014 when DrupalVM was first created. The ways we handled local development and all of the emerging technologies. Virtual machines ruled the land, and Vagrant empowered a new way to manage headless virtual machine instances. To use Docker on macOS, you had to use boot2docker to create a virtual machine.
A search feature enhances the user experience of a website by allowing the user to find what they’re looking for easily and quickly. More so for large websites, e-commerce sites, and sites with dynamic content (news sites, blogs).
Apache Solr is one of the most popular search platforms used by websites of all sizes. It is an open-source search engine based on Java that lets you search through large amounts of data, like articles, products, customer reviews and more. Take a deeper look into Apache Solr in this article.
Check out this article to learn how to configure Apache Solr in Drupal
Apache Solr is fast and flexible and allows for full-text search, hit highlighting (highlights the matching search term), faceted search (a more refined search), real-time indexing (allows new content to be indexed immediately), dynamic clustering (organizes search results into groups), database integration, NoSQL features (non-relational database) and rich document handling (to index a wide variety of document formats like PDF, MS Office, Open office).
Some good-to-know facts about Apache Solr:
Apache Lucene is an open-source Java search library that lets you easily add search or information retrieval to the application. It is versatile, powerful, accurate, and works on an efficient search algorithm.
Although known for its full-text search capabilities, Lucene can also be used for document classification, data analysis and information retrieval. It also supports many languages other than English like German, French, Spanish, Chinese, Japanese, and more.
All search engines begin with indexing. Indexing is the processing of original data into highly efficient cross-reference lookup to facilitate rapid search.
Search engines don't index data directly. The texts are first broken into tokens (atomic elements). Searching is the process of consulting the search index and retrieving the doc matching the query.
First, the document will be analyzed and split into tokens. All those tokens will be indexed to the inverted index. Inverted index is a way in which Solr builds the index.
Lets consider we have 3 documents:
The way it is tokenized is as shown in the 2nd column of the below table.
“Chocolate” is available in D1 and D2
“Cake” is available in D2 and D3
“Big” is available in D3
“Ordered” is available in D2
“Prepared” is available in D3
“Vanilla” is available in D3
You will notice that words like “I”, “love” are not tokenized. These are called Stop words which will not be indexed or searchable by Solr.
So when someone searches for the term “Chocolate Cake”, the engine looks into the index. Instead of looking for the document, it first looks into the index to see which documents do the words “Chocolate” and “Cake” fall under. This makes it easy and faster to fetch the particular document only. This is called inverted indexing.
Apache Solr uses a document-based storage schema and stores every piece of data as a separate document within a collection. This allows for efficient and flexible storage and retrieval of data.
In Drupal, each node is considered as a document. So when you index your node to Apache Solr, it is considered as a document. Each document can contain multiple fields. Lucene does not have common global schema. Which means you can index any type of field in each document in Apache Solr.
◦ bin/solr -e techproducts
This will create a dummy core for demonstration and it will also start the Solr server.
Once you have installed Solr, you will see many folders like:
Docs - contains documentation about Solr
Dist - Solr main .jar file
Contrib - contains add-on plugins and specialised features of Solr
Bin - scripts of Solr
Example - contains demonstrate solr capabilities
Server - heart of Solr. Contains Solr web application, logs, Solr core
To create a core, we need two files mandatory.
Now lets see how to query the Solr results in the Solr admin UI.
For example: cat: electronics
For example: cat: electronics id:TWINX2048-3200PRO with q.op AND
[OR]
cat: electronics AND id:TWINX2048-3200PRO
[OR]
A filter query helps narrow down the results of a search. A query can be specified by the fq parameter to restrict which documents are returned in the superset, without affecting the score.
The sort parameter arranges search results in either ascending (asc) or descending (desc) order. Depending on the content, the parameter can be used either numerically or alphabetically.
The rows parameter allows you to paginate results from a query.
The fl parameter limits the information included in a query response to a specified list of fields.
Default field parameter is the default field for query parameter.
The highlight feature in Solr enables the inclusion of fragments of documents that match a query.
Some of the most common highlight parameters are:
Facets enable users to explore and refine large sets of search results. They’re displayed in a UI as checkboxes, dropdowns or other controls. The two general parameters to control facets are:
Using the facet parameter, users can generate facets based on the values of one or more fields in their search index. In the search results, the facet parameter can be configured to control how facets are generated and displayed.
2. Facet.query paramater
When a user includes a facet.query parameter in their Solr query, Solr will generate a list of facet counts that correspond to the number of documents in the index that match each query. Facet.query is useful when you want to generate facets based on complex search criteria that can't be easily represented using a simple field value.
There are several other facet parameters like the facet.field (to specify the fields that should be used to generate facets), facet.limit (max number of facets to display for each field), facet.mincount (min number of document needed for the facet to be included in the response), facet.sort (specifies the order in which the facet values should display).
Apache Solr is a highly versatile search engine that comes with many interesting features which can be customized as per your requirements. Drupal works extremely well with Apache Solr. If you’re looking for Drupal experts to configure a powerful search engine for your new project, we would love to take it further!
Author: Saranya Ashok Kumar
Meet Saranya Ashok Kumar, Drupal Specialist, who’s extremely fond of coding and Drupal and likes sharing valuable content through her YouTube channel. Saranya likes tapping her toes to her favorite music and dreams of traveling to the Maldives.
Email Address Subscribe Leave this field blank Drupal Development Drupal Planet DrupalLeave us a Comment
Great Southern Homes, one of the fastest growing home builders in the US, sees greater results with Drupal
ExploreA reimagined digital solution for Abaco, a global leader in embedded computing systems for the defense industry
ExploreUpgrading the web presence of IEEE Information Theory Society, the most trusted voice for advanced technology
Explore View all Case StudiesToday we are talking about ChatGPT with Ezequiel Lanza.
For show notes visit: www.talkingDrupal.com/387
TopicsHey GitHub - Coding with your voice ChatGPT Wolfram Alpha
GuestsEzequiel Lanza - github.com/ezelanza @eze_lanza
HostsNic Laflin - www.nLighteneddevelopment.com @nicxvan John Picozzi - www.epam.com @johnpicozzi Katherine Druckman - katherinedruckman.com @katherined
MOTW CorrespondentMartin Anderson-Clutz - @mandclu Search API Solr Boost By User Term Allows your site to boost search results that share taxonomy term references with your users.