lesson-02.md

Search Congressmembers by last name

Source code: examples/02-congress-names

The front page:

The results view:

This is slightly more complicated than the Hello world, lesson, but not many more new concepts.

The end result of this app is to let a user search for congressmembers by last name (partial strings are OK). So, this app reads an external dataset -- the good ol' Congressmembers spreadsheet and parses it, which requires more code than a simple Hello, World.

For this example, I've just crammed it all into app.py. Note how the data reading and data filtering functions are wrapped up in their own functions.

Here's the function to read the data from the given filename and then to do a little wrangling -- filtering only for people who are in office, and then adding a fullname field:

def get_data():
    # open data file, filter for in_office,
    # add fullname field
    # then return list of dicts
    with open('./static/data/legislators.csv', 'r') as f:
        newrows = []
        for row in csv.DictReader(f):
            if row['in_office'] == '1':
                row['fullname'] = ' '.join([row['firstname'], row['middlename'], row['lastname']])
                newrows.append(row)
        return newrows

If you were to clone the [source-code folder]((examples/02-congress-names), you could run this function without running any other part of the web app.

That is, you could open iPython, then run this code:

from app import get_data
data = get_data()
print("There are", len(data), 'rows in the dataset.')
# There are 540 rows in the dataset.

Sorting data

So...sorting lists and other collections is its own bag of fun. Here's a separate lesson on that:

Sorting Python collections with the sorted method

Here's the relevant code in app.py:

from operator import itemgetter


def filter_data(lastname='', sortby=None):
    # first, select only peeps that match by a name, then sort them
    lowname = lastname.lower()
    rows = [d for d in get_data() if lowname in  d['lastname']]
    if sortby == 'oldest':
        return sorted(rows, key=itemgetter('birthdate'), reverse=True)
    elif sortby == 'youngest':
        return sorted(rows, key=itemgetter('birthdate'))
    else:
        # i.e. 'alpha' or any value...just sort by last name, first name
        return sorted(rows, key=itemgetter('lastname'))

It's a little hard to see, but filter_data() actually calls get_data(). If you're new to sorting and so forth (such as, what is that itemgetter?), you might want to go easy.

Here's a simpler version that just filters and then sorts the data in reverse-alphabetical order:

def filter_data(lastname=''):
    # first, select only peeps that match by a name, then sort them
    lowname = lastname.lower()
    rows = [d for d in get_data() if lowname in  d['lastname']]
    s_rows = sorted(rows, key=itemgetter('lastname'), reverse=True)
    return s_rows

Here's another way to write that, using the lambda construction instead of itemgetter():

def filter_data(lastname=''):
    # first, select only peeps that match by a name, then sort them
    lowname = lastname.lower()
    rows = [d for d in get_data() if lowname in  d['lastname']]
    s_rows = sorted(rows, key=lambda r: r['lastname'], reverse=True)
    return s_rows

And here's a version without using a list comprehension:

def filter_data(lastname=''):
    # first, select only peeps that match by a name, then sort them
    lowname = lastname.lower()
    rows = []
    for d in get_data():
        if lowname in d['lastname']:
            rows.append(d)
    s_rows = sorted(rows, key=lambda r: r['lastname'], reverse=True)
    return s_rows

Basically, this is all stuff that's fundamental to Python collections -- not with Flask. Re-read examples on how to sort things in Python until they feel a little more natural to you. Adding them to a Flask app is not much of a deal in comparison.

Again, the source code for an app for this lesson can be found here: examples/02-congress-names