4 ms·

first answer on https://stackoverflow.com/questions/1732348/regex-match-open-tags-except-xhtml-self-contained-tags https://stackoverflow.com/questions/1732348/r

• schoen 7y ago
  And in context, the idea is that when we use classic Unix text manipulation tools, we make implicit and sometimes explicit assumptions about the nature and structure of the data. Those assumptions can be wrong and then our pipelines are also wrong. By contrast, a strongly statically typed environment might force us to prove our assumptions in the context of the interactions among the different parts of our code, and so there will be fewer corner cases where we have unexpected or unspecified behavior. The only barely held back part is presumably that when we notice a corner case, we might add a little more logic to handle that specific case, but we never really approach the goal of handling all possible inputs with valid or useful behavior. (We might even declare that a non-goal.)