F frontend #1275
F frontend #1275
Conversation
| outpath = args.out | ||
|
|
||
| # Load SDFG | ||
| sdfg = fortran_parser.create_sdfg_from_fortran_file(filepath) |
There was a problem hiding this comment.
A bit sad that fcdc has to live separate from sdfgcc, since they seem functionally and programatically identical in all but one (or two, if you count the import) line.
What do you think about just using sdfgcc and detecting whether something is an SDFG or Fortran file? The name sdfgcc may be sub-optimal in that case, maybe it makes sense to have a dace-cc that autodetects (or is told explicitly what it is with --source-lang, and have sdfgcc alias dace-cc --source-lang=sdfg and fcdc alias dace-cc --source-lang=fortran.
Not a crucial point, but the code duplication is a bit sad.
There was a problem hiding this comment.
Agreed - this was done by request of @tbennun . What do you think?
There was a problem hiding this comment.
Not sure about this. We can discuss
There was a problem hiding this comment.
Seconding the idea of dace-cc with autodetection and additional parameters.
| outpath = args.out | ||
|
|
||
| # Load SDFG | ||
| sdfg = fortran_parser.create_sdfg_from_fortran_file(filepath) |
There was a problem hiding this comment.
Seconding the idea of dace-cc with autodetection and additional parameters.
Currently, the core Fortran frontend is ready for comment.
Tests have also been added
There are many features and improvements possible, however, I want to focus on merging the current working prototype as soon as it is DaCe compliant to keep the size of the PR to a minimum as it is already very large.