Jeff,
I think in an age where people seldom print out source code for code reviews (or other purposes) anymore, it rarely matters. I think it certainly matters much less than it did when the ESAPI Project first started (in 2007, I think). When I first started on ESAPI in 2009, I used vim and gvim and set the text width 80, in part because I was used to reviewing my own code on dead trees, and printing in portrait rather than landscape was my preferred page orientation. But as ESAPI evolved and more contributors started using IDEs, that 80 character width went by the wayside. Fast forward to today, when I now normally use terminal windows of 132 characters and I have ':set tw=132' in my ~/.exrc file.
So, I learned that times change and everyone has their own preferences. Now, when I edit any existing file in any repo, I try to honor the predominant style of what that source file already uses. In other words, I try not to be overly anal about stylistic choices like I was more in my youth. (Of course, my original anal tendencies can somewhat be excused as my first programming language was FORTRAN IV entered on an IBM 029 Key Punch on an 80 column cards, so it mattered a lot more back then. 😜 And, yes, I am that old! 🦖)
So, if you are doing a PR for ESAPI, my advice and personal preference is to follow the style that dominates in the file you are editing. And if you create a new file, choose whatever style you like (but please use spaces rather than tabs for indentation as that makes it simpler for us to change in the future if the code gets reorganized). However, one thing we prefer not to see is to see hundreds of lines split and/or combined for only stylistic reasons (and I'd consider reformatting Javadoc along those lines since when it's translated to HTML, it really shouldn't matter) that are without semantic significance. It makes the code review process of the PR more difficult than it needs to be.
-kevin
P.S.- In hindsight, it probably would have been a good idea to suggest, and maybe even enforce, some style guides when ESAPI started. This is something we should probably consider for ESAPI 3, but I think it's rather late to worry about the existing code for now. If we carry any of the ESAPI 2.x code forward to ESAPI 3, that is when we should consider it.
--
NSA: All your crypto bit are belong to us.