Spaces:
Sleeping
Sleeping
File size: 7,421 Bytes
ef78361 | 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 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 | """
Export Utilities - ํตํฉ ๋ด๋ณด๋ด๊ธฐ ์ปดํฌ๋ํธ
๊ณตํต ๋ด๋ณด๋ด๊ธฐ ๊ธฐ๋ฅ์ ์ ๊ณตํฉ๋๋ค:
- CSV/Excel ๋ณํ
- ํํฐ๋ง ์ต์
- ์ ์ฒด ๋ต๋ณ ํฌํจ ์ต์
"""
import pandas as pd
import streamlit as st
from io import BytesIO
from typing import Callable
from .supabase_client import get_sentiment_data_for_export
from .athena_client import fetch_full_answers_batch
# openpyxl ์ค์น ์ฌ๋ถ ํ์ธ (Excel export์ฉ)
try:
import openpyxl
EXCEL_AVAILABLE = True
except ImportError:
EXCEL_AVAILABLE = False
# LLM ๊ฒ์ฆ ์ํ ๋ผ๋ฒจ (์ฉ์ด ํต์ผ)
LLM_STATUS_LABELS = {
"all": "์ ์ฒด",
"verified": "๊ฒ์ฆ์๋ฃ",
"false_positive": "์คํ (๋ถ์ โ๋น๋ถ์ )", # ๋ถ์ ์๋
"true_negative": "์ ํ (๋ถ์ ํ์ )", # ๋ถ์ ํ์
"unverified": "๋ฏธ๊ฒ์ฆ",
}
POLARITY_LABELS = {
"all": "์ ์ฒด",
"negative": "๋ถ์ ",
"positive": "๊ธ์ ",
"neutral": "์ค๋ฆฝ",
}
def prepare_dataframe_for_export(
data: list[dict],
include_full_answers: bool = False,
) -> pd.DataFrame:
"""๋ฐ์ดํฐ๋ฅผ DataFrame์ผ๋ก ๋ณํํ๊ณ ๋ด๋ณด๋ด๊ธฐ์ฉ์ผ๋ก ์ ๋ฆฌํฉ๋๋ค.
Args:
data: ๋ด๋ณด๋ผ ๋ฐ์ดํฐ ๋ฆฌ์คํธ
include_full_answers: ์ ์ฒด ๋ต๋ณ ํฌํจ ์ฌ๋ถ
Returns:
์ ๋ฆฌ๋ DataFrame
"""
if not data:
return pd.DataFrame()
df = pd.DataFrame(data)
# ๋ฆฌ์คํธ ์ปฌ๋ผ์ ๋ฌธ์์ด๋ก ๋ณํ
list_columns = ['in_house_brands', 'mentioned_brands', 'llm_evidence_spans']
for col in list_columns:
if col in df.columns:
df[col] = df[col].apply(
lambda x: ', '.join(x) if isinstance(x, list) else str(x) if x else ''
)
# ์ปฌ๋ผ ์์ ์ ๋ฆฌ - answer_full์ answer_preview ๋ค์์ ๋ฐฐ์น
if 'answer_full' in df.columns and 'answer_preview' in df.columns:
cols = list(df.columns)
cols.remove('answer_full')
idx = cols.index('answer_preview') + 1
cols.insert(idx, 'answer_full')
df = df[cols]
return df
def export_to_csv(df: pd.DataFrame) -> bytes:
"""DataFrame์ CSV ๋ฐ์ดํธ๋ก ๋ณํํฉ๋๋ค."""
return df.to_csv(index=False).encode('utf-8-sig')
def export_to_excel(df: pd.DataFrame) -> bytes | None:
"""DataFrame์ Excel ๋ฐ์ดํธ๋ก ๋ณํํฉ๋๋ค.
Returns:
Excel ๋ฐ์ดํธ ๋ฐ์ดํฐ, ๋๋ openpyxl์ด ์์ผ๋ฉด None
"""
if not EXCEL_AVAILABLE:
return None
output = BytesIO()
with pd.ExcelWriter(output, engine='openpyxl') as writer:
df.to_excel(writer, index=False, sheet_name='Data')
return output.getvalue()
def render_export_component(
campaign_id: int,
key_prefix: str,
title: str = "๐ฅ ๋ฐ์ดํฐ ๋ด๋ณด๋ด๊ธฐ",
show_polarity_filter: bool = True,
show_llm_filter: bool = True,
default_polarity: str = "negative",
default_llm_status: str = "all",
in_house_only: bool = True,
):
"""ํตํฉ ๋ด๋ณด๋ด๊ธฐ ์ปดํฌ๋ํธ๋ฅผ ๋ ๋๋งํฉ๋๋ค.
Args:
campaign_id: ์บ ํ์ธ ID
key_prefix: Streamlit ์์ ฏ ํค ์ ๋์ฌ (์ค๋ณต ๋ฐฉ์ง)
title: ์น์
์ ๋ชฉ
show_polarity_filter: ๊ฐ์ ํํฐ ํ์ ์ฌ๋ถ
show_llm_filter: LLM ์ํ ํํฐ ํ์ ์ฌ๋ถ
default_polarity: ๊ธฐ๋ณธ ๊ฐ์ ํํฐ ๊ฐ
default_llm_status: ๊ธฐ๋ณธ LLM ์ํ ํํฐ ๊ฐ
in_house_only: ์์ฌ ๋ธ๋๋๋ง ํํฐ๋ง
"""
with st.expander(title, expanded=False):
# ํํฐ ์ต์
filter_col1, filter_col2 = st.columns(2)
with filter_col1:
if show_polarity_filter:
polarity_options = list(POLARITY_LABELS.keys())
polarity_labels = list(POLARITY_LABELS.values())
default_idx = polarity_options.index(default_polarity) if default_polarity in polarity_options else 0
selected_polarity = st.selectbox(
"๊ฐ์ ํํฐ",
options=polarity_options,
format_func=lambda x: POLARITY_LABELS[x],
index=default_idx,
key=f"{key_prefix}_polarity"
)
else:
selected_polarity = default_polarity
with filter_col2:
if show_llm_filter:
llm_options = list(LLM_STATUS_LABELS.keys())
default_idx = llm_options.index(default_llm_status) if default_llm_status in llm_options else 0
selected_llm_status = st.selectbox(
"LLM ๊ฒ์ฆ ์ํ",
options=llm_options,
format_func=lambda x: LLM_STATUS_LABELS[x],
index=default_idx,
key=f"{key_prefix}_llm_status"
)
else:
selected_llm_status = default_llm_status
# ๋ด๋ณด๋ด๊ธฐ ์ต์
opt_col1, opt_col2 = st.columns(2)
with opt_col1:
include_full_answers = st.checkbox(
"์ ์ฒด ๋ต๋ณ ํฌํจ",
value=False,
help="Athena์์ ์ ์ฒด ๋ต๋ณ์ ๊ฐ์ ธ์ต๋๋ค (ํ์ผ ํฌ๊ธฐ ์ฆ๊ฐ)",
key=f"{key_prefix}_full_answers"
)
with opt_col2:
include_evidence = st.checkbox(
"LLM ๊ทผ๊ฑฐ ํฌํจ",
value=False,
help="LLM ํ๋จ ๊ทผ๊ฑฐ(reasoning, evidence_spans)๋ฅผ ํฌํจํฉ๋๋ค",
key=f"{key_prefix}_evidence"
)
st.markdown("---")
# ๋ค์ด๋ก๋ ๋ฒํผ
btn_col1, btn_col2, btn_col3 = st.columns([1, 1, 2])
# ๋ฐ์ดํฐ ๊ฐ์ ธ์ค๊ธฐ
data = get_sentiment_data_for_export(
campaign_id=campaign_id,
polarity=selected_polarity if selected_polarity != "all" else None,
llm_status=selected_llm_status if selected_llm_status != "all" else None,
in_house_only=in_house_only,
include_full_answers=include_full_answers,
include_evidence=include_evidence,
)
if data:
df = prepare_dataframe_for_export(data, include_full_answers)
count = len(df)
with btn_col1:
csv_data = export_to_csv(df)
st.download_button(
label=f"๐ฅ CSV ({count}๊ฑด)",
data=csv_data,
file_name=f"campaign_{campaign_id}_export_{count}๊ฑด.csv",
mime="text/csv",
key=f"{key_prefix}_csv_download"
)
with btn_col2:
if EXCEL_AVAILABLE:
excel_data = export_to_excel(df)
st.download_button(
label=f"๐ฅ Excel ({count}๊ฑด)",
data=excel_data,
file_name=f"campaign_{campaign_id}_export_{count}๊ฑด.xlsx",
mime="application/vnd.openxmlformats-officedocument.spreadsheetml.sheet",
key=f"{key_prefix}_excel_download"
)
else:
st.caption("Excel: openpyxl ํ์")
with btn_col3:
st.caption(f"์ด {count}๊ฑด | ํํฐ: {POLARITY_LABELS.get(selected_polarity, '์ ์ฒด')} / {LLM_STATUS_LABELS.get(selected_llm_status, '์ ์ฒด')}")
else:
st.info("๋ด๋ณด๋ผ ๋ฐ์ดํฐ๊ฐ ์์ต๋๋ค.")
|