嵌套循环和DataFrame-Python和Pandas

Question

I am having trouble writing a loop that returns what I need. I have two CSV files. For the values in a column in CSV 1, I need to find if there are matching values in CSV 2 and if there are matching values, return a dataframe for the row of the matching values. When I try to create a loop, I cannot get the right values in the loop. For example:

import pandas as pd

csv2 = pd.read_csv('/users/jamesh/documents/asiopods/asicrawlconcat.csv', header = 1)
csv1 = pd.read_csv('/users/jamesh/documents/asiopods/asiconcat.csv', header = 0)
h1s = csv1['Recommended_H1']


h1 = h1s
h1[0:3] #test
subject = csv2['H1_1']

for x in h1:
    for y in subject:
        if x == y:
            print y

The code above returns the values I need, but in string form. I need to return the dataframe for the values of y, from CSV2

Any help or direction is greatly appreciated!

Edit - with some offline help, I have been able get the correct information from the loop. However, I still can't figure out how to get the data into a pandas.dataframe. Instead the data is returned in a vertical manner. Here is the new loop:

def foogaiz():
    for k1, v1 in h1.iteritems():
        for k2, v2 in subject.iteritems():
            if v1 == v2:
                data = csv2.irow(k2)
                return data

Answer 1

It's a little unclear if the values you're matching on ("Recommend_H1" in your example) are unique and only appear once in asiconcat.csv. If so, then I recommend naming the two columns that have matching values the same ('H1_1' in my example syntax below) and doing a df.merge()

matched_df = df.merge(crawldf,on="H1_1",how="left")

The left join option is in order to keep rows that don't have matches on crawldf.

You can read the documentation for merge here:

http://pandas.pydata.org/pandas-docs/stable/merging.html