liblen.py

import xlrd
import operator
from fuzzywuzzy import fuzz
from fuzzywuzzy import process

'''
lib - dic of releases [created by calling parse_stack]
possible search categories include: (just strings for now)
    artst name
    album title
    label
    genre
    addedby
    donatedby
keyword - value to be searched
datasetSize - amount of items to be returned
intensity - int from 0 to 100 of how narrow a search
'''
def search(lib, categories, keyword, datasetSize, intensity):
    hits = []
    hitRatios = []
    releases = []

    releases = lib
    for release in releases:
        for category in categories:
            ratio = fuzz.token_sort_ratio(release[category], keyword)
            if ratio >= intensity:
                hits.append(release)
                hitRatios.append(ratio)

    hits = [k for k, v in sorted(zip(hits, hitRatios), key=operator.itemgetter(1))]
    hits.reverse()
    return hits[:datasetSize]



def parse_stack(path):
    releases = []
    book = xlrd.open_workbook(path)
    for s in range(book.nsheets):
        sheet = book.sheet_by_index(s)
        # skip first row since it has the headings
        for r in range(1, sheet.nrows):
            row = sheet.row(r)

            year = row[3].value
            if type(year) == float:
                year = int(year)
            elif type(year) != int:
                year = None

            releases.append({
                'artist': str(row[0].value).strip(),
                'title': str(row[1].value).strip(),
                'label': str(row[2].value).strip(),
                'year': year,
                'genres': [genre.strip() for genre in
                           str(row[4].value).split('/')],
                'stack': str(row[5].value).strip(),
                'missing': str(row[7].value).strip() != '',
                'addedby': str(row[8].value).strip(),
                'donatedby': str(row[9].value).strip()
            })

    return releases