Large Deviations Principle for a Large Class of One-Dimensional Markov Processes