From 4d57f8ea07cedd3b54be4f8c086a0319803c369b Mon Sep 17 00:00:00 2001 From: Yoann Schneider <yschneider@teklia.com> Date: Mon, 19 Feb 2024 12:36:29 +0100 Subject: [PATCH] Ignore empty lines inside the BIO files --- nerval/parse.py | 2 +- tests/fixtures/end_of_file.bio | 1 + 2 files changed, 2 insertions(+), 1 deletion(-) diff --git a/nerval/parse.py b/nerval/parse.py index 65f8d94..04173bf 100644 --- a/nerval/parse.py +++ b/nerval/parse.py @@ -108,7 +108,7 @@ def parse_bio(lines: List[str]) -> dict: and get_type_label(future_label) != last_tag ): index += 1 - if index < len(lines): + if index < len(lines) and lines[index]: future_label = lines[index].split()[1] # Check for continuation of the original entity diff --git a/tests/fixtures/end_of_file.bio b/tests/fixtures/end_of_file.bio index 56f8e52..aa6af75 100644 --- a/tests/fixtures/end_of_file.bio +++ b/tests/fixtures/end_of_file.bio @@ -6,6 +6,7 @@ de I-PER Dieu I-PER roy I-PER de I-PER + France B-LOC et I-PER de I-PER -- GitLab