Closed ss-is-master-chief closed 6 years ago
@ss-is-master-chief, thanks for that idea, however this cannot be generic. It is a step wise process that only returns usage per level at which you are messing up. let me know if we are on the same page.
@henrykironde I agree. What about [-path]
?
Good question. So in the CSV engine we don't have path
as part of the required_opts
required_opts = [
("table_name",
"Format of table name",
os.path.join(DATA_DIR, "{db}_{table}.csv")),
]
@henrykironde Got it. Thanks for the clarification. I'll close this.
If we use
retriever install
, we are prompted withusage: retriever install [-h] [--compile] [--debug] [--not-cached] {mysql,postgres,sqlite,msaccess,csv,json,xml} ...
and on using, for exampleretriever install csv
, prompted withusage: retriever install csv [-h] [--table_name [TABLE_NAME]] dataset
.Could we change the usage alert into something common to both the incomplete command? How about
usage: retriever install {mysql,postgres,sqlite,msaccess,csv,json,xml} [-h] [--compile] [--debug] [--not-cached] [--table_name [TABLE_NAME]] dataset
?[-path]
could be a part of the usage alert as well.