Generate machine descriptions from PDFs (or maybe XML).
Will have a purely machine-generated instruction definition:
“Read in PDF, write out an assembler configuration”.
Already deployed for the disassemblers.
I’ve seen other projects that do something similar to this, and I cringe at it. PDF doesn’t seem like a sane format for storing data like this. Surely maintaining XML, or another language would be saner? Or am I wrong?
I think their main motivation is the format hardware manufacturers (Intel, AMD etc.) ship their platform specifications. They do mention XML as they expect ARM to give them XML instead of PDF.
So while I agree that it’s not a great format. If that’s the only thing that’s possible to get out of those companies then I also prefer a tool that can slurp in the important bits from it instead of manually doing the job each time a new hardware platform appears.
I’ve seen other projects that do something similar to this, and I cringe at it. PDF doesn’t seem like a sane format for storing data like this. Surely maintaining XML, or another language would be saner? Or am I wrong?
I think their main motivation is the format hardware manufacturers (Intel, AMD etc.) ship their platform specifications. They do mention XML as they expect ARM to give them XML instead of PDF.
So while I agree that it’s not a great format. If that’s the only thing that’s possible to get out of those companies then I also prefer a tool that can slurp in the important bits from it instead of manually doing the job each time a new hardware platform appears.
Agreed, it isn’t about what is sane, it is about what exists. Also, the PDFs Intel release are very technical, well formatted and laden with parseable tables of data: http://www.intel.com/content/dam/www/public/us/en/documents/manuals/64-ia-32-architectures-software-developer-instruction-set-reference-manual-325383.pdf
As long as it is PDF-A (archival) should be decently parseable now and into the future.