- 1. One interface, one set of tools for all searches
- 2. General Search examples
- 3. Vertical Search Examples
In this article we will take you through the search capabilities of Pagehop.
Using Pagehop, you no longer depend on the mouse/trackpad, tabbing or key shortcuts specific for every page’s implementation in order to navigate through the results of your searches.
No matter if you search horizontally (Google, Bing, DuckDuckGo, etc.) or vertically (querying specific resource like StackOverflow or Wikipedia), you use 1 interface, one set of powerful tools and you can stay on the keyboard, 100% of the time.
Here at Universless, we believe that, to prove a point, there is no better argument than results. That’s why this section of the documentation (as many others, too) will contain only examples that should be very useful for you and solve problems you are having with the current state of search engines and The Web as a whole.
When you search in Google, most of the time, there are a lot more than 10 results (most of the time results are more than 1000).
But if you want to find a text in the results in the browser (
⌘ + F), you always get to search only on the 10 results displayed on the current page you are viewing.
Pagehop, by default, returns the first 30 results (configurable for more in the Settings -> Number of results). These are the first 3 pages (3 * 10) of results. This covers > 99.7% of searches in Google. That’s right. Did you know that less than 1% go to the 3rd page of results - check this study.
Here is Pagehop configured for 90 results (9 pages) searching for “Java web frameworks”:
If we want to filter the results we got in the last example, with, lest say, a regex to match all results having the singular “framework” form all that have the plural “frameworks“, we can do just that using the :r (Regex) tool.
Here is how:
What if we want to match results pointing to domains of the type [something]framework.[extension].
We can do this by first using the :a (Addresses) tool, to show us the addresses of the results (thus making search on them instead on titles) and then piping a :r filtering at the end:
If you use Fuzzy matching somewhere on your computer, you probably know the frustration of going to the browser and searching only through absolute-matching of strings.
If you search for a very common personal name in Google, the :f (Fuzzy) tool will be a nice choice to filter the results:
Although horizontal search is at a pretty descent state, sometimes we know what source we want to get results from. Here are some examples in using vertical search engines with Pagehop.
If you use Google (or pretty much any other horizonatal search engine) and you want to search in Wikipedia, what you probably do is:
- Open up a browser (if not in the browser, yet);
- Open-up a new tab;
- Focus the address bar (if not done automatically for you);
- Write your query ending with “ wiki”;
- Press Enter.
In Pagehop you have to:
- Open-up Pagehop;
- Write your query starting with “wiki “ (right now “w “ is enough with the default set of recipes).
Both of these procedures end up with results for your query.
Even if we turn a blind-eye to the awesome tooling (:r, :f, :a, etc.), Pagehop still gives you 2 big advantages:
- You get faster to the results (smaller procedure);
- All results are coming from Wikipedia, not some (potentially none when searching through horizontal engine) all of them.
Here is an example:
Using sophisticated algorithms for natural language processing, Wolfram Alpha is able to understand very complex search queries. From the Examples section of the WolframAlpha recipe’s documentation:
- wolf plot x^2
- wolf second derivative of x^2
- wolf what is the population of UK
- wolf what is the distance between new york and san francisco
- wolf distance between new york and san francisco
- wolf what year is it
- wolf which is the biggest state in the U.S.
Here are a couple of examples:
With the CodeSearch recipe, you can search for real-world code samples. The recipe is using the searchcode.com public API. searchcode.com has indexed over 5 000 000 projects on Github, Bitbucket, Google Code, Codeplex, Sourceforge, Fedora Project and more.
Here is how you can search for usages of the readFileSync function in the fs package in NodeJS:
You probably guessed that already - writing “s [your-query]” will search in StackOverflow.
There probably isn’t a website you search in, where Pagehop wouldn’t be an option. We believe that Pagehop’s development will explode in a rich set of Recipes, Tools and probably other pieces of beautifully crafted code we haven’t even dreamed about, yet.
So, if you use a resource that has no recipe for Pagehop - get onboard, write it yourself and share it with the community (or not).-