Code bumming used to make sense in the 60s, when you directly wrote in assembler, and each line was an instruction that mapped directly to what was executed by the machine. In high-level languages this is not the case, and one liner can perform several orders of magnitude worse than 20 lines long algorithm. In fact, the longer version can be made more robust with error checking and such.
And anyway, any line count bum you might have achieved with an obfuscated, mystery one liner is completely invalidated by people having to carefully consider what it does, and what the intent was. It's usually better to write things more simply, even if it ends up more verbose. That, of course, does not apply to things like well-established idioms of the language used, or of the program itself.