The start of a new is a time for fresh starts and New Year’s Resolutions…
If you develop a library (in any language) over 10 years, you accumulate lots of ‘stuff’. You have code which used to do things or should have been refactored and tidied up. And all this makes it harder to maintain and develop the code base. The JPedal PDF library has evolved substantially in power and functionality over the years and is also now being used in ways we never even considered back then.
Backwards compatibility is very important but there is always a tension with going forward. We get the same problem with which Java version to support – should we still be supporting Java 1.0?
So we had a good look at our main PdfDecoder class over the holiday and noticed that we had (over the years) added a lot of code which should be moved down and methods which cluttered the API and in some cases no longer did anything. It had also become a convenient ‘dumping ground’ so things which should have been put elsewhere. We have tidied these up and documented any changes to methods on the build release note which we think are in general usage. None of the main methods have been altered. And all the changes have been regression tested against our large batch of automated tests.
Most people will not even be aware of most of the methods removed. But if you do find any problems, please let us know on the support forum and we will help with any changes we have made.
The result is a cleaner, faster Java library and a lot of clutter moved out of the way. So we can continue to improve the library over the next 10 years…
IDRsolutions develop a Java PDF library, a PDF forms to HTML5 converter, a PDF to HTML5 or SVG converter and a Java Image Library that doubles as an ImageIO replacement. On the blog our team post about anything interesting they learn about.