Skip to main content

Finding books in nearby library branches

  • Posted 10 October 2022
  • Tagged with books

I’m trying to make better use of my local public library. I want to read more books, and borrowing from the library keeps the habit sustainable. It also saves a fair bit of money, and I don’t have to decide what to do with books when I’m done.

Recently, I built a tool to help me find books to borrow, and it’s working pretty well.

It starts with my library’s online catalogue. When I find a book I might be interested in reading, I tap the bookmark icon in the upper right-hand corner of the search result:

A search result in my library’s online catalogue. The result includes the title, the author, and the publication year. In the top right-hand corner is a small green bookmark icon.

This saves the book to a list on the library site, which I can view on my account page. But the website makes it hard to work out what I can actually borrow. The example above says “92 copies”, which means 92 physical books floating around the library network – but they could be on loan, on hold for another reader, or in a branch halfway across the county.

If I tap “View availability”, I get a long list of every copy in the network, its location and status:

A modal dialog labelled ‘Availability’. It’s a long list. Each entry in the list has a location (e.g. Abbots Langley Library), a collection (e.g. Hot picks), a call number (e.g. General fiction) and a status/description (e.g. Available). Only two results are visible in the current scroll position.

Given there are dozens of books I’m interested in, scrolling lists like this gets tedious. There must be a better way!

I’ve written a Python script that scrapes the library website, fetches all this information, then presents it in a nicer way. The library website uses a platform called Spydus, and I found some existing code for logging into Spydus sites by pretending to be a browser. Then I use BeautifulSoup to parse the data from the library HTML, and Jinja to render it in a nicer way.

This is what my new page looks like:

A list of books. The first two books have large titles, a summary, and a list of branches where copies are available for immediate borrowing. There are two more books which are shown in smaller text and with greyed-out covers -- these aren't available nearby.

It shows the list of books I’m interested in, and highlights the copies which are actually available – and specifically, copies that are available in branches within walking distance of my home. I could walk in and borrow any of these books immediately.

Over time, I expect books to gradually move around as they’re borrowed by other readers – so even if a book isn’t in a nearby branch today, it might be sometime in the future. I can order books from a different branch to pick up locally, but right now I have plenty to choose from.

This filtered list is particularly useful when I’m in a hurry. There’s a library branch near the train station, so I can pop in and pick up a few books on the way to the office – but I can’t take too long, or I’ll be late for work. Having a list of what’s readily available means I can be in and out quickly.

The tool includes a pick list of branches. For example, if I’m visiting Ware to get my pictures framed, I can find out if there’s anything I want to borrow at the library branch just down the road:

A list of library branch names with tickboxes. A single branch is ticked ‘Ware Library’, and below is shown a book with a single copy in Ware Library.

This is some JavaScript that listens to the onchange event on the checkboxes; it re-sorts the list whenever I check or uncheck a branch.

I’ve already used this for several rounds of borrowing, and it’s working great. Having a steady supply of new and interesting books is encouraging me to read more, and I feel less guilt about abandoning a book I’m not enjoying – I’m not losing anything. (I can only read one book at a time, so a bad book really stops me in my tracks.)

Although I don’t expect anybody else to use this tool directly, I’ve put the code on GitHub. There may be ideas or techniques here that apply to a different problem you have – and while you read through that, I have a stack of library books to finish.