Alternatives to DataFrames.readtable?

General Usage
1

I know that DataFrames.readtable is being depracated but the new CSV.read does not work for me. I have to read fixed-column width data, like this:

    Time (years)          a            e            i          peri         node           M          mass     
          0.0000000      3.00000     0.050000       1.0000     226.0000      18.0000     259.0000  1.000000E-03
       1000.0000000      2.97875     0.078517       0.3972      99.0831     249.7255     231.2168  1.000000E-03
       1999.9999997      2.98094     0.113592       0.9031     296.3859     128.8332     235.2106  1.000000E-03
       2999.9999996      2.92248     0.121087       1.9000     246.4872      47.8044     316.7246  1.000000E-03
       4000.0000006      2.92469     0.107297       1.9323      50.0935       2.5184     216.6421  1.000000E-03
       5000.0000015      2.93979     0.106321       1.6020      31.8273     321.6915     104.4990  1.000000E-03
       6000.0000024      2.91009     0.062727       0.9986     344.8856     277.3698     320.8678  1.000000E-03
...

CSV.read cannot handle this because this is not a CSV file. The old DataFrames.readtable could handle the idea of data columns being separated by any number of spaces. Is this functionality just gone forever? Do I have to roll my own parser using readline, split, and parse?

Daniel.

1 Like
2

On CSV master, you can do CSV.read(file; delim=" ", ignorerepeated=true). New release w/ this functionality is coming in a few days.

4 Likes
3

Thanks!