There are a multitude of different coding styles – and they usually differ widely between languages and environments. But it all comes down to two extremes: short and long.
In a Unix (read: ANSI or pre-ANSI C) environment the dominating style is the short one. It usually reads like a strange version of English written in a Latin-transcribed Hebrew alphabet: strtr, creat, malloc, yhwh (well, maybe not yhwh, but you get the point). Additionally, Unix and C are more or less free of upper-case letters.
The other extreme is the OOP world: their camelCase or MixedCase style looks like… well, a camel? Uh… yeah, sure looks like a camel. And as you know, those guys tend to use really long names: getLengthOfStringAsInteger and so on.
Ok, introduction done – main point:
No matter what style you think is better: blend in and use the style of the existing code for heaven’s sake!
Mixing these styles sucks. Hard time.
lotat = getListOfThisAndThatByFooId(foo.getId());
Really, even if listOfThisAndThat feels too long to write it out each and every time (which it does, at least for me): do not shorten it.
It is called listOfThisAndThat and that is how you must name it.
Or else, everyone coming along afterwards – with the longCamelCaseNames parsing mode activated – will take lotat for a proper name or something – and will have to waste time going back through the code, trying to find out what the hell it really is.
The worst style is no style at all.
Leave a Reply