python: chaining complex conditional dataframe changes using pandas

Question

1. I have a csv that may or may not always have the same column order or headers (hence, the need to call columns by name).
2. I use pandas to convert the spreadsheet into a dataframe
3. i modify the dataframe based upon (hopefully) modular, scalable booleans

4. i reconstruct the csv file:

   df.to_csv('csv/csv_out.csv')

I have been able to load the csv into a dataframe and conditionally change a column header (using the following:

df = pd.read_csv('csv/csv_in.csv', encoding='utf-8', engine='python', na_values=['.'])

if 'Column Title A' in df.columns:
df.rename(columns = {'Column Title A':'Column Title B'}, inplace = False)

but i cant seem to understand the syntax involved when chaining multiple if/then statements involving both text and numerical values.

Ive attempted df.ix;

dfmod_swheader = df.ix[(df['Column A'] == '12345') | (df['Column A'] == '54321') & 'Column Title B' in df.columns:]

but consistently receive "invalid type comparison" that I can't quite understand enough to get around.

So, the ask:

can I get assistance on the proper syntax for a pandas dataframe transform that achieves the following:

take the loaded dataframe (from csv_in.csv):

Column Name  A,Column Name  B,Column Name Red,Column Name Blue
foo,234,bar,foobar
foo,Box,bar,foobar
foo,12,bar,foobar
foo,9,bar,foobar
foo,1,bar,foobar
foo,Frog,bar,foobar
foo,5567,bar,foobar
foo,12,bar,foobar
foo,a,bar,foobar

and transform it (into csv_out.csv):

Column Name  A,Column Name C,Column Name  B,Column Name Red,Column Name Blue
foo,,234,bar,foobar
foo,,Box,bar,foobar
foo,,12,bar,foobar
foo,,9,bar,foobar
,foo,1,bar,foobar
foo,,Frog,bar,foobar
foo,,5567,bar,foobar
foo,,12,bar,foobar
,foo,a,bar,foobar

using the following logic:

if 'column name a' exists and
'column name b has cell value' = 'numerical value 1' or 'text value a'
then move 'column name a cell value' to (new?) 'column name c',
placing 'column name c' directly to the right of 'column name a'

Answer 1

I think you can use something like this to move column A to a new column under a condition.

cond  = df_in['Column Name B'].isin(['1','a'])

df_in['Column Name C'] = df_in.loc[cond,'Column Name A']
df_in['Column Name A'] = df_in['Column Name A'].mask(cond)

Output:

  Column Name A Column Name B Column Name C
0           foo             1           NaN
1           bar             a           NaN
2           NaN             2        foobar
3           NaN             b        barfoo

My output:

  Column Name  A Column Name  B Column Name Blue Column Name Red Column Name  C
0            foo            234           foobar             bar            NaN
1            foo            Box           foobar             bar            NaN
2            foo             12           foobar             bar            NaN
3            foo              9           foobar             bar            NaN
4            NaN              1           foobar             bar            foo
5            foo           Frog           foobar             bar            NaN
6            foo           5567           foobar             bar            NaN
7            foo             12           foobar             bar            NaN
8            NaN              a           foobar             bar            foo