Spaces:
Running
on
Zero
Running
on
Zero
File size: 1,881 Bytes
08fac87 03db0de 08fac87 100d2c7 08fac87 03db0de 3b06efd 03db0de 08fac87 100d2c7 08fac87 03db0de 08fac87 03db0de 08fac87 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 |
from calendar import month_name
from retriever import BuildRetriever, db_dir
import json
import os
import re
def get_collection(compute_mode):
"""
Returns the vectorstore collection.
Usage Examples:
# Number of child documents
collection = get_collection("remote")
len(collection["ids"])
# Number of parent documents (unique doc_ids)
len(set([m["doc_id"] for m in collection["metadatas"]]))
"""
retriever = BuildRetriever(compute_mode, "dense")
return retriever.vectorstore.get()
def get_sources():
"""
Return the source files indexed in the database, e.g. 'R-help/2024-April.txt'.
"""
# Path to your JSON Lines file
file_path = os.path.join(db_dir, "bm25", "corpus.jsonl")
# Reading the JSON Lines file
with open(file_path, "r", encoding="utf-8") as file:
# Parse each line as a JSON object
sources = [json.loads(line.strip())["metadata"]["source"] for line in file]
return sources
def get_start_end_months(sources):
"""
Given a set of filenames like 'R-help/2024-January.txt', return the earliest and latest month in 'Month YYYY' format.
"""
pattern = re.compile(r"R-help/(\d{4})-([A-Za-z]+)\.txt")
months = []
# Start with the unique sources
unique_sources = set(sources)
for src in unique_sources:
m = pattern.match(src)
if m:
year = int(m.group(1))
month_str = m.group(2)
try:
month_num = list(month_name).index(month_str)
except ValueError:
continue
if month_num == 0:
continue
months.append((year, month_num, month_str))
if not months:
return None, None
months.sort()
start = months[0]
end = months[-1]
return f"{start[2]} {start[0]}", f"{end[2]} {end[0]}"
|