Spaces:
Running
Running
summary_json = {"summary": str(summary)}
Browse files
app.py
CHANGED
@@ -527,7 +527,7 @@ def process_youtube_link(link):
|
|
527 |
simple_html_content = format_simple_transcript_to_html(formatted_simple_transcript)
|
528 |
first_image = formatted_transcript[0]['screenshot_path']
|
529 |
first_text = formatted_transcript[0]['text']
|
530 |
-
mind_map_json = get_mind_map(video_id, formatted_simple_transcript)
|
531 |
mind_map = mind_map_json["mind_map"]
|
532 |
mind_map_html = get_mind_map_html(mind_map)
|
533 |
|
@@ -596,27 +596,47 @@ def process_web_link(link):
|
|
596 |
soup = BeautifulSoup(response.content, 'html.parser')
|
597 |
return soup.get_text()
|
598 |
|
599 |
-
def get_mind_map(video_id, df_string):
|
600 |
-
|
601 |
-
|
602 |
-
|
603 |
-
|
604 |
-
|
605 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
606 |
|
607 |
-
|
608 |
-
|
609 |
-
|
610 |
-
|
611 |
-
|
612 |
-
|
613 |
-
|
614 |
-
|
615 |
-
|
616 |
-
|
617 |
-
|
618 |
-
|
619 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
620 |
|
621 |
return mind_map_json
|
622 |
|
@@ -712,7 +732,8 @@ def get_video_id_summary(video_id, df_string, source):
|
|
712 |
# 检查 summary_file 是否存在
|
713 |
is_summary_file_exists = gcs_check_file_exists(gcs_client, bucket_name, summary_file_blob_name)
|
714 |
if not is_summary_file_exists:
|
715 |
-
|
|
|
716 |
summary_text = json.dumps(summary_json, ensure_ascii=False, indent=2)
|
717 |
upload_file_to_gcs_with_json_string(gcs_client, bucket_name, summary_file_blob_name, summary_text)
|
718 |
print("summary已上传到GCS")
|
@@ -733,7 +754,6 @@ def get_video_id_summary(video_id, df_string, source):
|
|
733 |
exists, file_id = check_file_exists(service, folder_id, file_name)
|
734 |
if not exists:
|
735 |
summary = generate_summarise(df_string)
|
736 |
-
# processed_summary = processed_video_summary_to_json(summary)
|
737 |
summary_json = {"summary": str(summary)}
|
738 |
summary_text = json.dumps(summary_json, ensure_ascii=False, indent=2)
|
739 |
|
@@ -746,14 +766,6 @@ def get_video_id_summary(video_id, df_string, source):
|
|
746 |
print("===get_video_id_summary error===")
|
747 |
print(error_msg)
|
748 |
print("===get_video_id_summary error===")
|
749 |
-
|
750 |
-
|
751 |
-
|
752 |
-
# 存在 local at OUTPUT_PATH as {video_id}_summary.json
|
753 |
-
# with open(f'{OUTPUT_PATH}/{video_id}_summary.json', 'w') as f:
|
754 |
-
# f.write(summary_text)
|
755 |
-
# print(f"summary已存在 local at {OUTPUT_PATH}/{video_id}_summary.json")
|
756 |
-
# file_id = upload_file_directly(service, file_name, folder_id, f'{OUTPUT_PATH}/{video_id}_summary.json')
|
757 |
else:
|
758 |
# 逐字稿已存在,下载逐字稿内容
|
759 |
print("summary已存在Google Drive中")
|
|
|
527 |
simple_html_content = format_simple_transcript_to_html(formatted_simple_transcript)
|
528 |
first_image = formatted_transcript[0]['screenshot_path']
|
529 |
first_text = formatted_transcript[0]['text']
|
530 |
+
mind_map_json = get_mind_map(video_id, formatted_simple_transcript, source)
|
531 |
mind_map = mind_map_json["mind_map"]
|
532 |
mind_map_html = get_mind_map_html(mind_map)
|
533 |
|
|
|
596 |
soup = BeautifulSoup(response.content, 'html.parser')
|
597 |
return soup.get_text()
|
598 |
|
599 |
+
def get_mind_map(video_id, df_string, source):
|
600 |
+
if source == "gcs":
|
601 |
+
print("===get_mind_map on gcs===")
|
602 |
+
gcs_client = init_gcs_client(GCS_KEY)
|
603 |
+
bucket_name = 'video_ai_assistant'
|
604 |
+
file_name = f'{video_id}_mind_map.json'
|
605 |
+
blob_name = f"{video_id}/{file_name}"
|
606 |
+
# 检查檔案是否存在
|
607 |
+
is_file_exists = gcs_check_file_exists(gcs_client, bucket_name, blob_name)
|
608 |
+
if not is_file_exists:
|
609 |
+
mind_map = generate_mind_map(df_string)
|
610 |
+
mind_map_json = {"mind_map": str(mind_map)}
|
611 |
+
mind_map_text = json.dumps(mind_map_json, ensure_ascii=False, indent=2)
|
612 |
+
upload_file_to_gcs_with_json_string(gcs_client, bucket_name, blob_name, mind_map_text)
|
613 |
+
print("mind_map已上傳到GCS")
|
614 |
+
else:
|
615 |
+
# mindmap已存在,下载内容
|
616 |
+
print("mind_map已存在于GCS中")
|
617 |
+
mind_map_text = download_blob_to_string(gcs_client, bucket_name, blob_name)
|
618 |
+
mind_map_json = json.loads(mind_map_text)
|
619 |
|
620 |
+
elif source == "drive":
|
621 |
+
print("===get_mind_map on drive===")
|
622 |
+
service = init_drive_service()
|
623 |
+
parent_folder_id = '1GgI4YVs0KckwStVQkLa1NZ8IpaEMurkL'
|
624 |
+
folder_id = create_folder_if_not_exists(service, video_id, parent_folder_id)
|
625 |
+
file_name = f'{video_id}_mind_map.json'
|
626 |
+
|
627 |
+
# 检查檔案是否存在
|
628 |
+
exists, file_id = check_file_exists(service, folder_id, file_name)
|
629 |
+
if not exists:
|
630 |
+
mind_map = generate_mind_map(df_string)
|
631 |
+
mind_map_json = {"mind_map": str(mind_map)}
|
632 |
+
mind_map_text = json.dumps(mind_map_json, ensure_ascii=False, indent=2)
|
633 |
+
upload_content_directly(service, file_name, folder_id, mind_map_text)
|
634 |
+
print("mind_map已上傳到Google Drive")
|
635 |
+
else:
|
636 |
+
# mindmap已存在,下载内容
|
637 |
+
print("mind_map已存在于Google Drive中")
|
638 |
+
mind_map_text = download_file_as_string(service, file_id)
|
639 |
+
mind_map_json = json.loads(mind_map_text)
|
640 |
|
641 |
return mind_map_json
|
642 |
|
|
|
732 |
# 检查 summary_file 是否存在
|
733 |
is_summary_file_exists = gcs_check_file_exists(gcs_client, bucket_name, summary_file_blob_name)
|
734 |
if not is_summary_file_exists:
|
735 |
+
summary = generate_summarise(df_string)
|
736 |
+
summary_json = {"summary": str(summary)}
|
737 |
summary_text = json.dumps(summary_json, ensure_ascii=False, indent=2)
|
738 |
upload_file_to_gcs_with_json_string(gcs_client, bucket_name, summary_file_blob_name, summary_text)
|
739 |
print("summary已上传到GCS")
|
|
|
754 |
exists, file_id = check_file_exists(service, folder_id, file_name)
|
755 |
if not exists:
|
756 |
summary = generate_summarise(df_string)
|
|
|
757 |
summary_json = {"summary": str(summary)}
|
758 |
summary_text = json.dumps(summary_json, ensure_ascii=False, indent=2)
|
759 |
|
|
|
766 |
print("===get_video_id_summary error===")
|
767 |
print(error_msg)
|
768 |
print("===get_video_id_summary error===")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
769 |
else:
|
770 |
# 逐字稿已存在,下载逐字稿内容
|
771 |
print("summary已存在Google Drive中")
|