daftlistings 1.8.1

A library that enables programmatic interaction with daft.ie. Daft.ie has nationwide coverage and contains about 80% of the total available properties in Ireland.

A library that enables programmatic interaction with Daft.ie. Daft.ie has nationwide coverage and contains about 80% of the total available properties in Ireland.

Installation

Daftlistings is available on the Python Package Index (PyPI). You can install daftlistings using pip.

virtualenv env
source env/bin/activate
pip install daftlistings

To install the development version, run:

pip install https://github.com/AnthonyBloomer/daftlistings/archive/dev.zip

Usage

from daftlistings import Daft

daft = Daft()
listings = daft.search()

for listing in listings:
    print(listing.formalised_address)
    print(listing.daft_link)
    print(listing.price)

By default, the Daft search function iterates over each page of results and appends each Listing object to the array that is returned. If you wish to disable this feature, you can set fetch_all to False:

daft.search(fetch_all=False)

Examples

Get apartments to let in Dublin City that are between €1000 and €1500 and contact the advertiser of each listing.

from daftlistings import Daft, RentType

daft = Daft()

daft.set_county("Dublin City")
daft.set_listing_type(RentType.APARTMENTS)
daft.set_min_price(1000)
daft.set_max_price(1500)

listings = daft.search()

for listing in listings:

    contact = listing.contact_advertiser(
        name="Jane Doe",
        contact_number="019202222",
        email="[e-mail]",
        message="Hi, I seen your listing on daft.ie and I would like to schedule a viewing."
    )

    if contact:
        print("Advertiser contacted")

You can sort the listings by price, distance, upcoming viewing or date using the SortType object. The SortOrder object allows you to sort the listings descending or ascending.

from daftlistings import Daft, SortOrder, SortType, RentType

daft = Daft()

daft.set_county("Dublin City")
daft.set_listing_type(RentType.ANY)
daft.set_sort_order(SortOrder.ASCENDING)
daft.set_sort_by(SortType.PRICE)
daft.set_max_price(2500)

listings = daft.search()

for listing in listings:
    print(listing.formalised_address)
    print(listing.daft_link)
    print(listing.price)
    features = listing.features
    if features is not None:
        print('Features: ')
        for feature in features:
            print(feature)
    print("")

Parse listing data from a given search result url.

from daftlistings import Daft

daft = Daft()
daft.set_result_url("https://www.daft.ie/dublin/apartments-for-rent?")
listings = daft.search()

for listing in listings:
    print(listing.formalised_address)
    print(listing.price)
    print(' ')

Find student accommodation near UCD that is between 850 and 1000 per month

from daftlistings import Daft, SortOrder, SortType, RentType, University, StudentAccommodationType

daft = Daft()
daft.set_listing_type(RentType.STUDENT_ACCOMMODATION)
daft.set_university(University.UCD)
daft.set_student_accommodation_type(StudentAccommodationType.ROOMS_TO_SHARE)
daft.set_min_price(850)
daft.set_max_price(1000)
daft.set_sort_by(SortType.PRICE)
daft.set_sort_order(SortOrder.ASCENDING)
daft.set_offset(offset)
listings = daft.search()

for listing in listings:
    print(listing.price)
    print(listing.formalised_address)
    print(listing.daft_link)

Map the 2-bed rental properties in Dublin and color code them wrt to prices. Save the map in a html file.

from daftlistings import Daft, SortOrder, SortType, RentType, MapVisualization
import pandas as pd

daft = Daft()
daft.set_county("Dublin City")
daft.set_listing_type(RentType.ANY)
daft.set_sort_order(SortOrder.ASCENDING)
daft.set_sort_by(SortType.PRICE)
# must sort by price in asending order, MapVisualization class will take care of the weekly/monthly value mess
daft.set_max_price(2400)
daft.set_min_beds(2)
daft.set_max_beds(2)

listings = daft.search()
properties = []
print("Translating {} listing object into json, it will take a few minutes".format(str(len(listings))))
print("Igonre the error message")
for listing in listings:
    try:
        if listing.search_type != 'rental':
            continue
        properties.append(listing.as_dict_for_mapping())
    except:
        continue


df = pd.DataFrame(properties)
print(df)

dublin_map = MapVisualization(df)
dublin_map.add_markers()
dublin_map.add_colorbar()
dublin_map.save("dublin_apartment_to_rent_2_bed_price_map.html")
print("Done, please checkout the html file")

For more examples, check the Examples folder

Parallel as_dict()

lisitng.as_dict() is relatively slow for large volume of listings. Below is an exmple script using threading and joblib library technique to speedup this process

from daftlistings import Daft, RentType
from joblib import Parallel, delayed
import time

def translate_listing_to_json(listing):
    try:
        if listing.search_type != 'rental':
            return None
        return listing.as_dict_for_mapping()
    except:
        return None

daft = Daft()
daft.set_county("Dublin City")
daft.set_listing_type(RentType.ANY)
daft.set_max_price(2000)
daft.set_min_beds(2)
daft.set_max_beds(2)

listings = daft.search()
properties = []
print("Translating {} listing object into json, it will take a few minutes".format(str(len(listings))))
print("Igonre the error message")

# time the translation
start = time.time()
properties = Parallel(n_jobs=6, prefer="threads")(delayed(translate_listing_to_json)(listing) for listing in listings)
properties = [p for p in properties if p is not None] # remove the None
end = time.time()
print("Time for json translations {}s".format(end-start))

Table of perfomance speedup for 501 listings Threads | Time (s) | Speedup ———— | ————- | ————- 1 | 178 | 1.0 2 | 101 | 1.8 3 | 72 | 2.5 4 | 61 | 2.9 6 | 54 | 3.3

Tests

The Python unittest module contains its own test discovery function, which you can run from the command line:

python -m unittest discover tests/

Contributing

• Fork the project and clone locally.

• Create a new branch for what you’re going to work on.

• Push to your origin repository.

• Create a new pull request in GitHub.