stcoats
commited on
Commit
·
a0a9509
1
Parent(s):
a47efdc
Add application file
Browse files
app.py
CHANGED
|
@@ -37,6 +37,9 @@ except Exception as e:
|
|
| 37 |
st.error(f"DuckDB connection failed: {e}")
|
| 38 |
st.stop()
|
| 39 |
|
|
|
|
|
|
|
|
|
|
| 40 |
# Search
|
| 41 |
query = st.text_input("Search text (case-insensitive)", "").strip()
|
| 42 |
|
|
@@ -44,7 +47,7 @@ if query:
|
|
| 44 |
sql = """
|
| 45 |
SELECT id, channel, video_id, video_title, speaker, start_time, end_time, text, pos_tags, upload_date, audio
|
| 46 |
FROM data
|
| 47 |
-
WHERE
|
| 48 |
LIMIT 100
|
| 49 |
"""
|
| 50 |
df = con.execute(sql, [query]).df()
|
|
@@ -76,27 +79,26 @@ else:
|
|
| 76 |
return None
|
| 77 |
|
| 78 |
df["audio_file"] = df["audio"].apply(render_audio_cell)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 79 |
|
| 80 |
-
|
| 81 |
-
|
| 82 |
-
|
| 83 |
-
|
| 84 |
-
|
| 85 |
-
|
| 86 |
-
with col1:
|
| 87 |
-
st.write(f"**ID:** {row['id']}")
|
| 88 |
-
st.write(f"**Channel:** {row['channel']}")
|
| 89 |
-
st.write(f"**Video ID:** {row['video_id']}")
|
| 90 |
-
st.write(f"**Video Title:** {row['video_title']}")
|
| 91 |
-
st.write(f"**Speaker:** {row['speaker']}")
|
| 92 |
-
st.write(f"**Start Time:** {row['start_time']}")
|
| 93 |
-
st.write(f"**End Time:** {row['end_time']}")
|
| 94 |
-
st.write(f"**Upload Date:** {row['upload_date']}")
|
| 95 |
-
st.write(f"**POS Tags:** {row['pos_tags']}")
|
| 96 |
-
|
| 97 |
-
with col2:
|
| 98 |
-
st.markdown(f"**Text:** {row['text']}")
|
| 99 |
-
if row['audio_file']:
|
| 100 |
-
st.audio(row['audio_file'], format="audio/mp3")
|
| 101 |
-
else:
|
| 102 |
-
st.warning("Audio not available or invalid format.")
|
|
|
|
| 37 |
st.error(f"DuckDB connection failed: {e}")
|
| 38 |
st.stop()
|
| 39 |
|
| 40 |
+
# Enable full-text search index on first run (one-time setup if not exists)
|
| 41 |
+
con.execute("PRAGMA create_fts_index('data', 'text')")
|
| 42 |
+
|
| 43 |
# Search
|
| 44 |
query = st.text_input("Search text (case-insensitive)", "").strip()
|
| 45 |
|
|
|
|
| 47 |
sql = """
|
| 48 |
SELECT id, channel, video_id, video_title, speaker, start_time, end_time, text, pos_tags, upload_date, audio
|
| 49 |
FROM data
|
| 50 |
+
WHERE text % ?
|
| 51 |
LIMIT 100
|
| 52 |
"""
|
| 53 |
df = con.execute(sql, [query]).df()
|
|
|
|
| 79 |
return None
|
| 80 |
|
| 81 |
df["audio_file"] = df["audio"].apply(render_audio_cell)
|
| 82 |
+
df_display = df.drop(columns=["audio"]).copy()
|
| 83 |
+
|
| 84 |
+
# Add HTML audio tag column
|
| 85 |
+
def audio_html(path):
|
| 86 |
+
if path:
|
| 87 |
+
return f'<audio controls preload="none" style="height:20px;"> <source src="file://{path}" type="audio/mpeg"> </audio>'
|
| 88 |
+
return ""
|
| 89 |
+
|
| 90 |
+
df_display["Audio"] = df["audio_file"].apply(audio_html)
|
| 91 |
+
|
| 92 |
+
# Reorder columns
|
| 93 |
+
column_order = ["id", "channel", "video_id", "video_title", "speaker", "start_time", "end_time", "upload_date", "text", "pos_tags", "Audio"]
|
| 94 |
+
df_display = df_display[column_order]
|
| 95 |
+
|
| 96 |
+
st.markdown("### Full Table View (Sortable)")
|
| 97 |
+
st.write("Note: Audio is embedded using HTML tags; not all browsers allow playback from local temp paths.")
|
| 98 |
+
st.dataframe(df_display.drop(columns=["Audio"]))
|
| 99 |
|
| 100 |
+
st.markdown("### Audio Previews")
|
| 101 |
+
for i, row in df_display.iterrows():
|
| 102 |
+
if row["Audio"]:
|
| 103 |
+
st.markdown(f"**{row['speaker']} | {row['text'][:80]}**", unsafe_allow_html=True)
|
| 104 |
+
st.markdown(row["Audio"], unsafe_allow_html=True)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|