Text based search or Xpath search

Hi,

    I have a text file that contains all the meta data of documents and folders like name, created date, modified date, owner etc. I generate this text file everytime a user wants to search for the meta data through python. I have a web interface where the user can input the necessary parameters he/she wants to search for. But, this operation of reading the information from the text file is slow. It becomes extremely slow, if there is quite a lot of information to go through, naturally.

Now, I was thinking that I can store the meta data as an XML file instead and use the XPATH option available in python to retrieve the necessary information.

Do you guys think that storing the data in an XML file and using the XPATH search will be faster than the text file search?

Thanks for you input in advance,

Ram