Priors are the entities in p-search that score documents according to some criteria. Some priors may look for the occurrence of specific strings while others are more broad, perhaps assigning a score based on when the file was last modified. The name “prior” comes from the fact that these are the assertions you make a priori, before looking at the results, and are meant to quantitatively reflect you believes via a probability (value between 0 and 1).
When you’re trying to find something with p-search, try to think beyond the particular string you want to search for. Think of everything you believe about the file and try to reflect it via a prior. If a prior doesn’t exist for the criteria you want, feel free to open an issue, or you can even try your hand at creating one yourself as explained in a later chapter.