Generate Confluence Wiki table from NLTK Corpus

gen_table_corpus.py

from nltk.downloader import Downloader

dl = Downloader()

def format(x, fields, type_name, delim):
    x = x.__dict__
    out = delim
    for col in fields.split(delim):
        if col == 'type':
            out += type_name
        else:
            out += x.get(col, "N/A").replace(delim, ";" ).encode("utf8")
        out += delim
    return out
    
headers = 'name languages license author contact type'
delim = "|"
headers = delim.join(headers.split())
out = [delim+headers+delim, ]
out += [format(x, headers, 'models', delim) for x in dl.models()]
out += [format(x, headers, 'packages', delim) for x in dl.packages()]
out += [format(x, headers, 'collections', delim) for x in dl.collections()]
out[0] = out[0].replace("|", "||")
print "\n".join(out)