Oracle releases Java JDK 18 with enhanced source code documentation

Oracle's Java 18 development environment has hit the streets, with Big Red promising nine enhancements including the ability to add sample source code to API documentation. Other new features include Simple Web Server (JEP 408) for prototyping and testing, two incubating modules, as well as the preview of Pattern Matching for …

[Anonymous Coward]

"With Java's consistent six-month release cadence Oracle is meeting those demands, helping to keep Java relevant for modern applications and workloads,"

Oh really? Demands from whom, exactly? We don't even test on these non-EOL releases because none of our customers use them. By contrast, we had an email just last month from a customer still running Java 1.4, which was released in 2002.

Re. the UTF-8 issue, I remember getting bitten by this back in about 2000, when some ASCII files I was blithely reading with a FileReader turned out not to be gibberish on an AS/400 - the native encoding was EBCDIC. A trap for young players (as I was then).

[DrXym]

I wouldn't feel sorry about getting bitten. Character encoding is something pretty much no programmer has a clue about until have to fix an entire broken codebase that doesn't work on non Latin chars.

These days the rule of thumb should be *always* use UTF-8 and isolate the exceptions to the periphery. Java & Win32 are slightly different because they use UTF-16 internally, but even so, UTF-8 should be used in external data and internal code be aware what UTF-16 means, i.e. no randomly slicing strings or iterating chars potentially straight through a code point.

This link says it best:

http://utf8everywhere.org/

[swm]

When writing Java code for file transfers from a UNIX-type machine the file names are not UTF8 - they can be anything except null characters or nulls. UTF8 has illegal byte sequences. So I was forced to use bytes.

[Warm Braw]

Enhanced source code documentation...

Per yesterday's article, presumably enhanced documentation of the licence terms is not on the roadmap.

[Anonymous Coward]

Re: Enhanced source code documentation...

Enhancement to documentation rejected. Reason: working as intended.

[Steve Channell]

Still no EOL for RMI class loader vulnerabilities

Once upon a time, the idea of a passing any object over a remote connection, that would load missing remote classes seemed like a good idea. Back when Java was seen as a technology for set-top-boxes or applets for HotJava browser it was part of the "develop anywhere, deploy everywhere" marketing.

Fast forward to the new millennium (20 years ago) the danger was clear, but bloody-mindedness prevented its deletion - triggering the STRUTS and then Log4J code-injection disasters.

GraalVM and IKVM could never fully support the RMI legacy: it's disappointing that Oracle is choosing to extend compatibility with a vulnerability rather than deprecating it. Since when was debug symbols considered an "enterprise" feature?

[DrXym]

Java is such a mess

I bet there are people who've not moved from Java 8 because of all the BS over licensing.

[F. Frederick Skitty]

Re: Java is such a mess

The place I work switched to OpenJDK about eight years ago, and haven't had any issues. We were probably quite conservative about making the change from Oracle's releases as I knew many other firms that had already switched. It was also one less dependency that came from outside the core Debian repos, so that was a plus as well.

[Anonymous Coward]

Re: Java is such a mess

Back in the old days the JVMs - IBM, Sun, JRockit - used to be considerably different, but since Java 11 on we haven't noticed any differences at all - we can substitute Oracle, OpenJDK, Graal or Liberica (our current favourite due to size) and see no difference across our entire testharness, which touches all sorts of dusty corners of the API.

[Plest]

Ker-ching!!

Oracle : "Crank up the audit machine guys! A lot of automatic updates going in and a lot of extra money about to roll in!"

[Fruit and Nutcase]

Bitter tasting Oracle Java

Ethical and Fare Trade? We're talking about Oracle. So I guess not

[Anonymous Coward]

This would be yet another version no one can use?

No automated install, because redistribution is banned (by marketing?).

It'll break anything that was working.

...so many things go wrong. I'd rather wait for a "free" (libre) version to get sorted. And then, perhaps, test.

But I probably won't. Anything that wants java will just continue to install the version it works with.

[matjaggard]

Who cares what oracle does?

I'm still a fan of Java and I like the new versions too. But even I'm struggling to care what oracle has to say or do on this. The more that oracle leaves Java to its own process the better as far as I'm concerned.