Denpa Finder

Searches several denpa / jcore dumps (called "data sources") for albums matching a certain query.

Current Data Sources

RTL
Silence The Discord (all pages, even /requests): http://135.181.29.38
Audio For You: https://audioforyou.top/?p=184
Denpa Gist: https://gist.github.com/dnpcllctns/f79394cd283ee30834ee6e4bb484b502

Installation

Install from pip:

pip install denpa-finder

Usage

Basic usage:

import pprint
from denpa_finder import DenpaFinder

df = DenpaFinder()
matches = df.matches("ななひら")
pprint.pprint(matches, indent=4)

Fuzzy Matching

Denpa Finder performs fuzzy matching on the search string by default. This means that an album with a title one letter off from your search query will still be returned in the results. This is convenient for typos or weird characters in an artist's name which some data sources write differently (P*light vs P＊light, for instance).

If you'd like to adust the ratio at which fuzzy matches are made, pass a float between 0 and 1 to the ratio parameter of df.matches (the default is 0.8). If you'd like to disable fuzzy matching completely, pass ratio=1.

# more fuzzy matches will be returned (less exact ones)
matches = df.matches("ななひら", ratio=0.7)
# less fuzzy matches will be returned
matches = df.matches("ななひら", ratio=0.9)
# no fuzzy matches will be returned, only exact ones
matches = df.matches("ななひら", ratio=1)

Complex Queries

I often find myself needing to express a more complex query than just "search for this string". For that, denpa finder provides the Q (short for "query") object, which can be AND'd, OR'd, and NOT'd:

from denpa_finder import DenpaFinder, Q

df = DenpaFinder()
# anything that matches either "camellia" or "かめりあ"
q1 = Q("camellia") | Q("かめりあ")
# anything that matches "haru" and "ama" and "nia".
# used this recently when I wanted to find albums by
# haru＊ama＊nia but wanted to avoid getting tripped up
# by some people using weird asterisks for the artist name
q2 = Q("haru") & Q("ama") & Q("nia")
# albums by camellia, but not with nanahira
q3 = (Q("camellia") | Q("かめりあ")) & (~(Q("nanahira") | Q("ななひら")))

matches1 = df.matches(q1)
matches2 = df.matches(q2)
matches3 = df.matches(q3)

Caching

DenpaFinder caches scraped albums from each data source, since loading and parsing the html is the majority of the cost of matching a query. If you'd like to refresh the caches, call df.refresh().