The MET API

import requests

# the structure of our request: base_url, path, query

base_url = "https://collectionapi.metmuseum.org"
path = "/public/collection/v1/search"
query = "?q=woman"

# the request, saved to a variable

url = f'{base_url}{path}{query}'
women = requests.get(url)

# what type of object do we have?

type(women)

dir(women)

women.headers

women.raw

women.elapsed

women.ok

women

# why do we need to add parenthesis? 
parsed = women.json()

type(parsed)

dir(parsed)

parsed.keys()

instructor = {
    'name': ['filipa calado', 'patrick smyth', 'stephen zweibel'],
    'age': [35, 37, 38],
    'degree': ['literature', 'literature', 'library science'],
    'job': ['digital scholarship specialist', 'startup', 'digital scholarship librarian']
}

# see the keys

instructor.keys()

# access items through brackets containing keys

instructor['name']

import pandas as pd
df = pd.DataFrame(instructor)

df

# let's try with the first object

parsed.keys()

# just printing out the first ten
parsed['objectIDs'][:10]

type(parsed['objectIDs'])

parsed['objectIDs'][0]

first = parsed['objectIDs'][0]
path = '/public/collection/v1/objects/'

# the base_url stays the same, so we don't need to change it.

url = f"{base_url}{path}{first}"

# running the request

first_object = requests.get(url)

# checking the resulting object

first_object

dir(first_object)

# all of the data about this object
# look at the URL! 

first_object.json()

# now we can save this first object to its own variable. 
# will make it easier to do more things to it!

first_json = first_object.json()

# no result!

first_json['artistGender']

first_json['department']

first_json['culture']

# first, print out the first 10 items in the objectIDs list
parsed['objectIDs'][:10]

ids = parsed['objectIDs'][:10]

# remember we can slice our list, to get just the first ten

for item in ids:
    print(item)

type(ids)

len(ids)

first_ten = []
for item in ids:
    # passing the objectID variable into the URL
    base_url = 'https://collectionapi.metmuseum.org'
    path = '/public/collection/v1/objects/'
    url = f'{base_url}{path}{item}'
    # grabbing our response for that object
    response = requests.get(url)
    # parsing our response with json
    parsed = response.json()
    # appending the response to our new list
    first_ten.append(parsed)

# because we already know the first, let's check the last item

first_ten[-1]

type(first_ten)

for item in first_ten:
    print(item['artistDisplayName'])

for item in first_ten:
    print(item['title'])

dir(item)

for item in first_ten:
    if item.get('artistDisplayName'):
        print(item['artistDisplayName'])

for item in first_ten:
    title = item.get('artistDisplayName')
    print(title)

first_fifty = []
for item in parsed['objectIDs'][:50]:
    url = f'https://collectionapi.metmuseum.org/public/collection/v1/objects/{item}'
    response = requests.get(url)
    parsed = response.json()
    first_fifty.append(parsed)

# combine get() with conditional to get rid of the None's

for item in first_fifty:
    if item.get('artistDisplayName'):
        print(item['artistDisplayName'])

# or, if we want to show the value, even if empty or None

for item in first_hundred:
    gender = item.get('artistGender')
    print(gender)

# why do you think we see only female?

for item in first_fifty:
    if item.get('artistDisplayName'):
        print(f"artist name: {item['artistDisplayName']}")
        print(f"artist gender: {item['artistGender']}")

# look again at the first object
first_fifty[0]

# let's get a bunch of this data into lists

titles = []
names = []
genders = []
depts = []
countries = []
urls = []

for item in first_fifty:
    title = item.get('artistGender')
    titles.append(title)
    name = item.get('artistDisplayName')
    names.append(name)
    gender = item.get('artistGender')
    genders.append(gender)
    dept = item.get('department')
    depts.append(dept)
    country = item.get('country')
    countries.append(country)
    url = item.get('objectURL')
    urls.append(url)

countries

depts

urls

import pandas as pd

df = pd.DataFrame({
    'title': titles,
    'name': names,
    'gender': genders,
    'department': depts,
    'country': countries,
    'link': urls
})

df

df.info()

df.value_counts('department')

df.department.value_counts().plot(kind = 'barh')

df.department.value_counts().plot(kind = 'pie')