Closed GoogleCodeExporter closed 9 years ago
Attached diff for review
Original comment by bioinformed@gmail.com
on 25 Sep 2008 at 3:57
Attachments:
I like the new APIs. One additional comment:
Can we merge table_reader with table_rows_reader? In that case, we could wrap
the
checking of the first argument either being an iterator(rows) or
string(filename)
inside the common new API.
Zhaoming
Original comment by zwangm...@yahoo.com
on 25 Sep 2008 at 7:49
I decided to split table_row_reader from table_reader to allow other functions
to
select columns from pre-existing tabular data. As such, table_reader is
intended as
an input function and is so named (currently supporting delimited and XLS, but
more
are certainly possible-- I'm playing with a SQLite adapter now).
table_row_reader is
used in, e.g., lib.association to further process row data that was previously
read.
Rather than merge them again, maybe we just need to rename table_row_reader to
something clearer. I like having this functionality decoupled from the input
format,
though it is still convenient to call from table_reader. How about one of the
following names:
table_columns
table_select_columns
table_generate_columns
?
Or can you think of something better?
Original comment by bioinformed@gmail.com
on 25 Sep 2008 at 8:29
The grand renaming has been applied and released as of 1.0a5.
Original comment by bioinformed@gmail.com
on 17 Oct 2008 at 2:37
Original issue reported on code.google.com by
bioinformed@gmail.com
on 25 Sep 2008 at 3:56