There is a database of .dssp files which contain secondary structure assignments, and can be downloaded following the instructions. You would have to figure out a way of parsing the files for your species of interest.

Yet another way that would require a lot more work is to use advanced search at RCSB. Select Biology -> Source Organism Taxonomy name from the drop-down menu, enter your species of interest and download all the PDB files. From those files you can extract secondary structures, but beware that there is great redundancy of structures, including mutant proteins. For example, searching for Escherichia coli gives 10206 structure files, which would cover the whole genome couple of times over if not for redundancy.

Both suggestions assume that you want to deal with real secondary structures rather than predictions. If you are OK with predictions, for reasonably-sized genomes that can be done in several hours at 70-75% accuracy. It would take at least several weeks and lots of computational power to do predictions at 80-85% accuracy.