What ensures that Relativity does not re-extract text from previously discovered files?

The reason the choice indicating that Relativity does not re-extract text from previously discovered files is correct is based on the system's design regarding text extraction errors. If there is an extraction error, the process is triggered to re-examine and potentially re-extract text from those specific files. This mechanism is essential because it ensures integrity in the data handling process, allowing users to resolve issues that might have occurred during the initial extraction without unnecessarily duplicating work for files that have already been processed correctly. Thus, when an extraction has been successfully completed, Relativity tracks this status and avoids re-extracting to maintain efficiency.

Other presented options, while relevant to data processing, do not directly control this specific aspect of extraction management. File type settings can determine how files are processed initially, and file modification dates can indicate whether content has changed since the last extraction. Manual overrides can allow a user to force a re-extraction, but they do not represent the default operational behavior of the system concerning already processed files.