From 8864e2687c1cdf1bd563953f262402e7c9bbaa76 Mon Sep 17 00:00:00 2001
From: Jakub-Goluch <99048106+Jakub-Goluch@users.noreply.github.com>
Date: Mon, 26 Jun 2023 21:50:17 +0200
Subject: [PATCH] Add feature to check whether document has a valid json/jsonl
format, add tests
---
src/easymatcher_worker.py | 7 +++++--
1 file changed, 5 insertions(+), 2 deletions(-)
diff --git a/src/easymatcher_worker.py b/src/easymatcher_worker.py
index 8d79c8d..5e59e71 100644
--- a/src/easymatcher_worker.py
+++ b/src/easymatcher_worker.py
@@ -20,9 +20,12 @@ class EasymatcherWorker(nlp_ws.NLPWorker):
It relies on the use of an easymatcher tool which can be found he under -
https://gitlab.clarin-pl.eu/knowledge-extraction/tools/easymatcher
"""
+
@staticmethod
- def is_jsonl(document_path: str | Path) -> bool:
- """Validates whether text file has json/jsonl structure and has "text" keyword"""
+ def is_jsonl(
+ document_path: str | Path
+ ) -> bool:
+ """Validates whether text file has json/jsonl structure and has "text" keyword."""
try:
with open(document_path, 'r', encoding="utf-8") as file:
for line in file:
--
GitLab