Introducing the new RapidSearch Live stack

After many months in development, I am pleased to unveil the all new RapidSearch Live stack.

RapidSearch Live came about because there was a distinct demand for a new search stack similar to RapidSearch Pro, but without the complexities of MySQL and suchlike. Although the original RapidSearch stack gives novice users a pathway towards getting a basic search setup, its strong ties to Google certainly are not to everybody’s liking. And although the SimpleSearch stack works great, it is more tedious for larger websites. There was evidently a gap in the market for a new search stack that could balance the demands of being both simple to setup and powerful in what it can do.

I designed and coded RapidSearch Live entirely myself, exclusively for RapidWeaver. I needed it for a client project and it seemed logical to continue development of RapidSearch Live into a fully-fledged stack for others to have access to.

With this in mind, we have finished with an incredible website search stack that is agile, reliable and fast. Very fast! Part of the reason for calling it RapidSearch Live is because the search results get returned to your site users almost instantaneously.

Key features at a glance:

  • No MySQL database or complicated configuration. RapidSearch Live constructs its own flat-file database in XML format.
  • Simply give RapidSearch Live your sitemap.xml file, and it will intelligently ‘spider’ all your existing webpages.
  • A simple (and client safe!) Admin Control Panel to access, for managing your website search.
  • Add pictures to your search results, by providing Open Graph images in your RapidWeaver 8 social media settings.
  • Various optimisations to improve search box accessibility and to help your website users with disabilities.
  • Show search results ‘live’ as a user types in the box, or only show search results when the ‘search’ button is clicked (the latter is the default behaviour).
  • Various style and colour settings for the search bar. Or use a search bar built with the free RapidSearch Helper stack on another webpage.
  • Control what information is shown in your search results and add optional labels against items like keywords.
  • Font styling and link colours get inherited from your RapidWeaver theme style settings, for greater design consistency and less setup work.
  • An optional setting to automatically redirect the user to the target webpage, if only a single ‘positive’ result is returned when the search is executed.
  • Support for 5 bespoke meta types, so you can create custom search criteria.
  • Make the entire search result item a clickable link or restrict clicking to only the image / title and URL.
  • Ability to test some aspects of RapidSearch Live locally in RapidWeaver Preview, prior to publishing your website online.
  • No adverts, no privacy blackholes, no competitor content, no sponsored links, no API keys to deal with! RapidSearch Live is entirely your OWN search engine.

A quick summary of how it works:
You drag and drop the RapidSearch Live stack into your website and publish your new search page. Press and hold the search bar for 5 seconds to open the Admin Control Panel. Click the button to begin an index of your website. During this period, RapidSearch Live goes to your sitemap.xml file and visits each of the webpages listed in your sitemap. From your webpages, RapidSearch Live ‘spiders’ the page link, page title, meta data and open graph meta tags. With this information, RapidSearch Live generates its own XML datafile to store all the search results within. After returning back to your search page, all this harvested data forms the basis for your site search.

RapidSearch Live in edit mode:

RapidSearch Live search results:

RapidSearch Live Admin Control Panel:

I invite you to have a play around with the live demo site and also download your FREE trial version of RapidSearch Live, to install and test on your own website.

This is a new stack. There are bound to be further updates to include new features and any required bug fixes.

7 Likes

@willwood This new stack seems to be great! Congratulations on the release.

Curious Will, if I already have RapidSearch Pro is there anything in RapidSearch Live (aside from the setup) that makes it better than RapidSearch Pro?

Curious if I can submit more than one sitemap, as example, TotalCms creates a separate site map for the blog.

I think in general you can create a sitemap of sitemaps, if that makes sense.
I was looking into it for TotalCMS too, and one approach (according to the sitemaps standard) was to create your own sitemap so you can reference the RW one and TCMS ones all in one master sitemap.

Found it Sitemap Index

Aside from easier setup, the main benefits are greater flexibility over the presentation of search results (controlling what data is shown to the searcher). The ability to get and display the RapidWeaver 8 Open Graph images against each page is also a significant new addition - because many users had been asking for image support long before I took over development of RapidSearch Pro from Jonathan Head in 2016.

I did seriously consider replacing RapidSearch Pro with this new stack. Hosting companies like GoDaddy have left me totally exasperated with their atrociously bad PHP / MySQL server configurations and the enormous amount of extra technical support that this so often entails.

However on the flip-side, there are still plenty of RapidSearch Pro users paying for genuine web hosting and very happy with their existing search setups. Therefore I’ve decided to continue making RapidSearch Pro available. However its codebase is not really well-suited towards being extended with lots of new features, so my thinking has been to preserve it as it is (maintenance fixes only going forwards) and make RapidSearch Live more of the new ‘flagship’ one we continue adding the new features to.

Not currently. For now, you are best merging the multiple sitemaps together and submitting them to RapidSearch Live as a single sitemap.xml. Treat it as if you were inviting a mainstream search engine (like Google) to index your website. Providing the listed pages are all public and on the same domain, there should not be an issue. But I admit it is an area that needs more testing and there may certainly be scope to fix / extend things a little further, to help you.

I have made sure from this very first release that we can support a custom sitemap.xml file, as opposed to the basic one RapidWeaver is already creating on each publish. This gives you a little more wiggle-room to submit a slightly more unique sitemap.xml file and we’ll try our best to spider it.

1 Like

Thanks Will,
Google allows you to submit multiple site maps, so there is no manual intervention needed when pages are added, removed etc. That is why I asked.

Hi @willwood
I am a complete newbie with ‘searches’ within websites. I am helping a charity with their WW1 heritage site - they have 109 soldiers and their stories. I want to have a search for only the website that allows visitors to look for the relevant information that will be shown on the site!!
Sorry if this isn’t clear - but for example to search for a regiment/name/address and be able to find some of the matches??
Is this something a search can do?? Sorry bit clueless. Thank you for any help

This project sounds really interesting. Off the top of my head, there are two potential ways you could accomplish this:-

You could setup an individual page for each soldier and include a full writeup for each. Within the RW Page Inspector, create some new meta tags like this:

<meta name="regiment" content="Royal Guards">
<meta name="name" content="John Smith">
<meta name="address" content="No. 6, Garden Cottages, Oxford">

(It would look a lot less cryptic in RW, because you have a proper interface to feed this data through).

Essentially these give you a method to ‘tag’ each page (each soldier) with additional information. RapidSearch Live can be configured to index this information, so that it becomes searchable in your search results. Continue to add the full body of content for each soldier in the main page area, including their stories, documents, papers etc. This offers users the ability to search for a soldier, but click a link and navigate to their full ‘page’. 109 pages would not be too much to ask of RapidSearch Live and potentially you could include the photograph of the soldier as the OG:image too - so it shows in the search results.

If you don’t think there is enough information to warrant creating individual pages for each soldier, another option is to use a stack like Filter. Again this would let you ‘tag’ each soldier with different information. But rather than storing the soldier information on separate webpages, it would all be stored on one page, as a long continual list. A user could use the search boxes or buttons to filter and reduce the number of entries shown.

Incidentally, I only learnt about a week ago (from someone in my own family doing some research) of a relative of mine who was killed in the Capture of Schwaben Redoubt and buried at Ancre British Cemetery. It’s very topical at the moment. I’d be happy to help you further.

1 Like

Hi Will,

Thank you for your detailed reply…and appreciate the offer of help/advice. I also had the thought of a filter? I have your sort stack…but maybe filter is stronger? I am not sure how much information my friends/researchers will be able to find out about each soldier…so my initial design thought was to use something like the expanse stack from 1LD…so each soldier would look the same at the front level even though there may not be much information…but it is very much a work in progress…and also the search question…not sure if appropriate…or overkill!!

Thanks again

Wendy

Hi Wendy,

SortStack would be great for more generalised sorting / filtering. For example if you had a list of regiment names to let users sort by, it would be good for that.

However for more precise searches or filtering (like typing in the exact name of a soldier) you would probably still need a true filtering / searching stack.

-Will.

Hi Will
Thanks…likely purchase your Filter stack, as you say it can give more precise filtering options…
…but likely purchase other stuff too - your support is second to none!!
W

1 Like

RapidSearch Live v1.0.2 is now available. This small update includes the following changes:

  • Improved the consistency of spacing spacing between search results.
  • The ‘no results’ message text is now customisable within the stack settings.
  • Updated a couple of informational tooltips.
  • The search bar placeholder text is now customisable within the stack settings.
  • Fixed a bug relating to button padding in Foundations.
  • The search cancel / reset button can be toggled on or off, for click-to-search configurations.

This update is available via automatic updates in Stacks or by downloading the latest version from your order lookup page. The free demo version has been updated to include these same changes too.

Feedback about RapidSearch Live so far I have received has been overwhelmingly positive. So many thanks for all the kind words. Ideas for the next update are already starting to get compiled.

1 Like

Just to ask @willwood is there a way it can be setup so that the user doesn’t leave the page the search box is on and the results are displayed on the same page floating above?
Because if I read the RapidSearch Live page properly, you said that for technical reasons it was better to use RapidSearch Helper, and set that up to redirect to a seperate search page rather than having RapidSearch Live on every page, or have I read it wrongly?

Hope that makes sense.

It might work as a Partial, but you would need to ensure there was adequate space for it to expand and display the search results. So you could try it in a theme sidebar or ExtraContent container. It would be less suitable in a smaller area, like a theme navigation bar.

Traditionally the normal approach is to have a dedicated ‘search’ page. Then configure all search boxes throughout the website to point to that webpage and display the results in full.

There is a clever option in RapidSearch Live marked Single Positive Result Redirect. That means that if only a single search result is returned, you can have RapidSearch Live automatically redirect the user to that webpage (without showing the search result first). Websites like John Lewis have something similar, if you search for a specific product code e.g 86010660.

I don’t understand what you mean about having pages “floating above”? The search results in RapidSearch Live are always returned below the search bar. Perhaps what you meant to describe was a modal window? If so, then yes you probably could iFrame your RapidSearch Live page within a modal. Although you would want to check you were still getting good accessibility standards and mobile responsive behaviour.

2 Likes

Is the Live Search Stack able to consume a dynamic sitemap like this one:

https://instacks.com/blog/index.php?feed=sitemap

Generated from Poster Stack:

2 Likes

HI @willwood
Have emailed you but unusually no reply (mind you, sure you are overwhelmed!!) I will definitely get your filter stack (have tried in demo mode and it seems to accept the 1LD expanse stack), but as regards a search…which one searches only the website the visitor is on??
Sorry too confusing for a newbie brain.
Thanks for any info/help

@instacks Crawling / spidering currently requires a physical sitemap.xml file and webpages on the server. If you cannot see the files via FTP, they cannot be seen by RapidSearch Live either.

@pigglypots I have already responded to your emails. Please check your spam or junk mail folders. You would normally configure RapidSearch Live to search the website a visitor is on. If you want to search multiple websites and merge results from different domains, you’ll probably want to use a stack like RapidSearch (powered by Google) or RapidSearch Go (which uses DuckDuckGo) for returning results from the domain names you list in the settings.

Great - purchased. Thanks!

1 Like

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.