When exporting frequencies as csv the file contains info about the query in the header -- which is a very(!) useful feature for further semi-automatic handling of these files. But what is missing in the header is information about the composition of the "context" (in terms of the freqs.py script), i.e. the name of the columns!
E.g.
"corpus","coha01"
"subcorpus",""
"concordance size","3118"
"query","Query:[lc=""house"" | lemma=""house""]"
"1993","NF",262, 54.72091
"1896","NF",187, 39.05653
On the other hand: When exporting as xslx, names for the user-selected columns
are provided (though not for freqs and the percentage-columns) but it is otherwise "incomplete" as it is lacking the information about corpus, subcorpus and query.
It would be great if at least csv export could contain "context" / "columns-names" in its header.
It would be even greater, if both the csv and xslx could be enhanced to both uniformly supplying the full information on corpus, subcorpus, query and context.
This could in my opinion easily achieved by adding these as extra columns in csv and xslx.
This meta-info-in-columns-solution might look awkward for now, but would allow easy direct access to this information for applications like Excel.
cheers
Hannes