def generate_config():
# TODO ensure this works still
config = qtoml.loads(Path("example.toml").read_text())
botsettings = config["Bot"]
for each in botsettings.keys():
if type(botsettings[each]) is bool:
botsettings[each] = prompt(
f"Would you like to enable {each}? y/N ")
elif type(botsettings[each]) is int:
numerical_option = input(f"Please enter a number for {each}: ")
assert numerical_option.isdigit() is True, "Must be a number!"
botsettings[each] = int(numerical_option)
else:
botsettings[each] = input(f"Please enter your {each}: ")
modules = getmodules()
print(
"Enter the number for each module you'd like to enable, separated by commas"
)
print("Example: 1,5,8")
message = ", ".join([f"{i}) {v}" for i, v in enumerate(modules)])
to_enable = input(f"{message}\n")
for module_index in to_enable.split(","):
if module_index.isdigit() and int(module_index) <= len(modules):
config["Modules"]["enabled"].append(modules[int(module_index)])
return qtoml.dumps(config)
def write_config(config: str, path: str) -> bool:
check = prompt(f"\n{config}\nDoes this look correct? y/N")
if check is False:
return False
else:
path = Path.cwd() / path
with open(path, "w") as tomlfile:
qtoml.dump(config, tomlfile)
return True
def main(args):
# TODO: remove this code, check if there is an issue and unit test
# convert to unicode to avoid windows errors
# args.corpus_path = args.corpus_path
# config corpus_path
# process bibtex files
args.bibtex = args.corpus_path.endswith('.bib')
if args.bibtex:
args.bibtex = args.corpus_path
args.corpus_path = process_bibtex(args.corpus_path)
# set corpus_name
args.corpus_name = os.path.basename(args.corpus_path)
if not args.corpus_name:
args.corpus_name = os.path.basename(os.path.dirname(args.corpus_path))
if not args.corpus_print_name and not args.quiet:
args.corpus_print_name = prompt("Corpus Name", default=args.corpus_name)
# configure model-path
if args.model_path is None:
if os.path.isdir(args.corpus_path):
args.model_path = os.path.join(args.corpus_path, '../models/')
else:
args.model_path = os.path.dirname(args.corpus_path)
if args.model_path and not os.path.exists(args.model_path):
os.makedirs(args.model_path)
args.corpus_filename = get_corpus_filename(
args.corpus_path, args.model_path, stop_freq=args.stop_freq)
if not args.rebuild and os.path.exists(args.corpus_filename):
args.rebuild = bool_prompt("\nCorpus file found. Rebuild? ",
default=False)
else:
args.rebuild = True
if args.htrc:
import vsm.extensions.htrc as htrc
if os.path.isdir(args.corpus_path):
htrc.proc_htrc_coll(args.corpus_path)
ids = listdir_nohidden(args.corpus_path)
args.htrc_metapath = os.path.abspath(args.corpus_path + '/../')
args.htrc_metapath = os.path.join(args.htrc_metapath,
os.path.dirname(args.corpus_path) + '.metadata.json')
else:
import topicexplorer.extensions.htrc_features as htrc_features
with open(args.corpus_path) as idfile:
ids = [row.strip() for row in idfile]
c = htrc_features.create_corpus(ids)
c.save(args.corpus_filename)
args.htrc_metapath = os.path.abspath(args.corpus_path)
args.htrc_metapath = os.path.join(
os.path.dirname(args.htrc_metapath),
os.path.basename(args.htrc_metapath) + '.metadata.json')
import htrc.metadata
print("Downloading metadata to ", args.htrc_metapath)
htrc.metadata.get_metadata(ids, output_file=args.htrc_metapath)
if args.rebuild and (not args.htrc or os.path.isdir(args.corpus_path)):
try:
args.corpus_filename = build_corpus(args.corpus_path, args.model_path,
stop_freq=args.stop_freq, decode=args.decode,
sentences=args.sentences,
simple=args.simple, tokenizer=args.tokenizer)
except IOError:
print("ERROR: invalid path, please specify either:")
print(" * a single plain-text or PDF file,")
print(" * a single bibtex (.bib) file with 'file' fields,")
print(" * a folder of plain-text or PDF files, or")
print(" * a folder of folders of plain-text or PDF files.")
print("\nExiting...")
sys.exit(74)
"""
except LookupError as e:
if 'punkt' in e.message:
print "\nERROR: sentence tokenizer not available, download by running:"
print " python -m nltk.downloader punkt"
elif 'stopwords' in e.message:
print "\nERROR: stopwords not available, download by running:"
print " python -m nltk.downloader stopwords"
else:
raise e
print "\nExiting..."
sys.exit(74)
"""
args.config_file = write_config(args, args.config_file)
args.corpus_desc = args.config_file + '.md'
if not args.quiet and os.path.exists(args.corpus_desc):
while args.corpus_desc not in ['y', 'n', False]:
args.corpus_desc = input("\nExisting corpus description found. Remove? [y/N] ")
args.corpus_desc = args.corpus_desc.lower().strip()
if args.corpus_desc == '':
args.corpus_desc = False
else:
if args.corpus_desc == 'y':
args.corpus_desc = args.config_file + '.md'
if args.corpus_desc:
with open(args.corpus_desc, 'w') as outfile:
outfile.write(
"""This is an instance of the [InPhO Topic Explorer](http://inphodata.cogs.indiana.edu/). If you would like
to add a custom corpus description, either:
- Modify the contents of the file `{}`
- Change the main:corpus_desc path in `{}` to an existing Markdown file.
""".format(os.path.abspath(args.corpus_desc),
os.path.abspath(args.config_file)))
return args.config_file