Gradle buildSrcVersions VS Guava

Lately I've been using gradle with kotlin-js and a mix of npm and jvm dependencies. There's the https://splitties.github.io/refreshVersions/ plugin that I can recommend if you are using gradle (with or without kotlin-js).

It manages plugin dependencies, library dependencies, and version dependencies. It can use the new gradle version catalog or not if you prefer. On first use, you migrate your versions and it extracts these to a versions.properties (or your version catalog). Then whenever you run refreshVersions it indicates available new versions in comments in this file. It also indicates unused dependencies. Though for npms this is a bit harder. All you need to do is manually use the versions you want.

I run this frequently to stay on top of upstream changes. Few software engineers realize that the testing and integration overhead with version changes multiplies (just like with other forms of change). Libraries that you haven't updated amount to technical debt that you haven't addressed. So, the workload increases massively if you don't update for a while. Staying up to date minimizes the workload. This plugin makes that super easy.

Given that we are using kotlin-js, we have to deal with a rapidly evolving library ecosystem so we do have occasional issues that we need to work around by either downgrading or fixing some code. Whenever I can't update something, I document it in my versions.properties with a comment. Often you just have to wait for the next release or so for things to straighten out. The price of using cutting edge software.

Kotlin-js manages a yarn lock file as well. So it properly locks dependencies. Whenever you update npm dependencies, you have to run a command to upgrade the lock file. There's also a whole mess of webpack dependencies that comes along with kotlin-js.

{$% embed https://github.com/splitties/refreshVersions %}

I need a Gradle plugin for managing version upgrades for dependencies. I used refreshVersions; however, I don't like how those dependencies end up in multiple files, e.g., versions.properties vs lib.versions.toml. I want something simpler. So, what do you prefer?

As somebody with a deep personal hatred of TOML I recommend refreshVersions https://github.com/jmfayard/refreshVersions just a million times better and will support version catalogs soon maybe

refreshVersions, it is literally the best

Note that I'm using gradle refreshVersions to make it easy to keep the project up-to-date. Therefore, the versions are not defined in the build.gradle files, they are centralized in the versions.properties file. RefreshVersions is bootstrapped like this in settings.gradle.kts:

Also the official sample for gradle refreshVersions

The first problem is at the level of Type System, given that a situation more correct would allow us to distinguish through the Collection Type which abstraction we are operating with, species if definable as mutable or immutable. The JCF was born at a time when great care was taken to offer immediate operational data structures, and with attention to performance, but with less attention to constructs or uses that are now seen as common. These concepts have been taken up by other infrastructures from which we certainly cannot fail to mention: Eclipse Collection, Guava Collections, and VAVR.

Even better is getting Gradle/Maven to correctly pull "plain" vs "Android" versions of the package instead of them just publishing the diverging code base as two repository packages.

https://github.com/google/guava/issues/2914

I'll admit I'm surprised to see that BOMs have been documented on maven.apache.org since mid-2008. It looks like Spring, for example, didn't adopt them until mid-2014. I don't know how widely they caught on in other areas. The first discussion of them in the context of Guava may have been in 2018, as I don't see mention of them in the various issues from 2011-2015 (#605, #1329, #1471, #1954.

And a quick peek at the source code for ImmutableList seems to confirm this (https://github.com/google/guava/blob/master/guava/src/com/google/common/collect/ImmutableList.java - it goes via a bunch of methods, but ends up using Arrays.copyOf(), which creates a fixed-size array).

The guava library of Java has some of these data structures implemented: https://github.com/google/guava/wiki/ImmutableCollectionsExplained , but implementations of the above book in many languages can be found on github (say, this one for Haskell: https://github.com/aistrate/Okasaki )

Java Mask: Java Mask is a library that offers various string masking techniques for sensitive data such as credit card numbers, email addresses, and more. You can find the library at: https://github.com/miguelfreitas93/java-mask DataMasker: DataMasker is a Java library specifically designed for masking sensitive data, including credit card numbers, using customizable masking patterns. Visit the GitHub repository for more information and usage examples: https://github.com/GDSSecurity/DataMasker Maskify: Maskify is a simple Java library that can be used to mask credit card numbers, Social Security numbers, and other sensitive information. You can find the library at: https://github.com/jonathancarvalhoalves/maskify CreditCardUtils: This is a lightweight Java library that provides utility methods for validating, formatting, and masking credit card numbers. Visit the GitHub repository for more information: https://github.com/malkusch/creditcardutils Google Guava: Google Guava is a popular set of Java libraries containing a wealth of utilities for working with strings, collections, and more. While not specifically designed for masking credit card information, you can use Guava's string manipulation methods to mask sensitive data: https://github.com/google/guava

I mean, I guess, technically? If you define it like that, then Microsoft has people working for them for free, as does Google, as does Apple, etc. It's not that weird, and you can try to twist it to be weird, but those of us in the software industry largely regard this as a good thing.

So we have https://github.com/google/guava/issues/4011

Ok, so let's implement our lazy initialization more smartly to avoid acquiring the lock every time and use old fashion but still working double-checked locking. I've found it implemented by Suppliers.memoize in guava library.