table.read_where - can condition field name be a variable?

[Ken Walker]

I am working with HDF5 files that have the same data, but newer versions have different Table Field names. Imagine 'col_1' in version 1 is called 'Column_1' in version 2. Same data, different name. (I receive this data downstream from another process, so can't control the names.) I want to do a table.read_where() with a condition that references this column.

I found one way to do it (below):

if version == 1:
    arr = dtable.read_where('col_1 == id')
else:  
    arr = dtable.read_where('Column_1 == id')

In anticipation of future changes, I'd like to use a variable to do the same thing.

Something like this:

if version == 1:
    read_col = 'col_1'
else:  
    read_col = 'Column_1'
# Note: This line does not work as-is:
arr = dtable.read_where( read_col == id )

Is there a way to write this table.read_where() condition with a variable?

If so, I'm missing something.

Thanks.

--Ken

[Kyle Robertson]

You could just use string formatting to create the conditional ad a string using the value of the variable. So

if version == 1:
read_col = 'col_1'
else:
read_col = 'Column_1'
# Note: This line does not work as-is:

condition = “{} == id”.format(read_col)
arr = dtable.read_where( condition)

[Ken Walker]

Hi Kyle,

Thanks for the suggestion. I'm laughing @ me for the simplicity of the solution.

I tinkered with all kinds of Pytables methods to do this, and missed the obvious Python solution!

Again, thanks.

-Ken