Computational analysis of emotions has been considered a challenging and interesting task. Researchers rely on various cues such as physiological sensors and facial expressions to identify human emotions. However, there are few prior works who work with textual input to analyse these emotions. This survey attempts to summarize these diverse approaches, datasets and resources that have been reported for emotional analysis from text. We feel that there is an essential need to have a collective understanding of the research in this area. Therefore, we report trends in emotion analysis research. We also present a research matrix that summarizes past work, and list pointers to future work.