Spaces:
Running
on
CPU Upgrade
Running
on
CPU Upgrade
File size: 10,610 Bytes
9bf7fc2 c709985 75c8d26 9bf7fc2 75c8d26 9bf7fc2 c709985 9bf7fc2 a1180f7 3279ad9 911afbf a63ba9d 07505ba 75c8d26 7b33394 07505ba 7b33394 9bf7fc2 083e322 9bf7fc2 c709985 9bf7fc2 79b887e 9bf7fc2 083e322 9bf7fc2 75c8d26 c709985 75c8d26 c709985 75c8d26 c709985 75c8d26 c709985 75c8d26 c709985 3279ad9 8ed2dcd 3279ad9 c709985 ef93247 3279ad9 c709985 9bf7fc2 083e322 9bf7fc2 083e322 9320e6c 083e322 9bf7fc2 083e322 79b887e c709985 9bf7fc2 083e322 9bf7fc2 c709985 75c8d26 a00436d 75c8d26 faece1b 75c8d26 faece1b 75c8d26 faece1b 7b33394 faece1b 7b33394 faece1b 7b33394 faece1b 7b33394 faece1b 0f0122d 7608d93 faece1b a63ba9d 0f0122d cb9d3ce a63ba9d 7608d93 faece1b 7608d93 0f0122d a63ba9d 7b33394 a63ba9d 7b33394 a63ba9d 7b33394 a63ba9d 7b33394 a63ba9d c3c8276 a63ba9d 4289ed3 07505ba |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 |
# server.py
import json
import random
import traceback
from typing import Optional
import uuid
from fastapi import APIRouter, Request
from fastapi.responses import JSONResponse
import pycountry
from pydantic import BaseModel
from chat_utils import chat
from config import SanatanConfig
from db import SanatanDatabase
from metadata import MetadataWhereClause
from modules.audio.model import AudioRequest
from modules.audio.service import svc_get_audio_urls
from modules.quiz.answer_validator import validate_answer
from modules.quiz.models import Question
from modules.quiz.quiz_helper import generate_question
import logging
from modules.video.model import VideoRequest
from modules.video.service import svc_get_video_urls
logging.basicConfig()
logger = logging.getLogger(__name__)
logger.setLevel(logging.INFO)
router = APIRouter()
# In-memory mapping from session_id -> thread_id
# For production, you may want Redis or a DB for persistence
thread_map = {}
class Message(BaseModel):
language: str
text: str
session_id: str | None = None # Optional session ID from client
class QuizGeneratePayload(BaseModel):
language: Optional[str] = "English"
scripture: Optional[str] = None
complexity: Optional[str] = None
mode: Optional[str] = None
session_id: Optional[str] = None # Optional session ID from client
class QuizEvalPayload(BaseModel):
language: Optional[str] = "English"
q: Question
answer: str
session_id: Optional[str] = None # Optional session ID from client
LANG_NATIVE_NAMES = {
"en": "English",
"fr": "Français",
"es": "Español",
"hi": "हिन्दी",
"bn": "বাংলা",
"te": "తెలుగు",
"mr": "मराठी",
"ta": "தமிழ்",
"ur": "اردو",
"gu": "ગુજરાતી",
"kn": "ಕನ್ನಡ",
"ml": "മലയാളം",
"pa": "ਪੰਜਾਬੀ",
"as": "অসমীয়া",
"mai": "मैथिली",
"sd": "سنڌي",
"sat": "ᱥᱟᱱᱛᱟᱲᱤ",
}
@router.get("/languages")
async def handle_fetch_languages():
supported_lang_codes = [
"en",
"fr",
"es",
"hi",
"bn",
"te",
"mr",
"ta",
"ur",
"gu",
"kn",
"ml",
"pa",
"as",
"mai",
"sd",
"sat",
]
languages = []
for code in supported_lang_codes:
lang = pycountry.languages.get(alpha_2=code) or pycountry.languages.get(
alpha_3=code
)
if lang is None:
continue # skip unknown codes
english_name = lang.name
native_name = LANG_NATIVE_NAMES.get(code, english_name)
languages.append(
{
"code": code,
"name": english_name,
"native_name": native_name,
}
)
languages.sort(key=lambda x: x["name"])
return languages
@router.post("/greet")
async def handle_greet(msg: Message):
markdown = "Namaskaram 🙏 I am **bhashyam.ai** and I can help you explore the following scriptures:\n---\n"
for scripture in SanatanConfig().scriptures:
num_units = SanatanDatabase().count(
collection_name=scripture["collection_name"]
)
markdown += f"- {scripture['title']} : `{num_units}` {scripture["unit"]}s\n"
session_id = msg.session_id
if not session_id:
session_id = str(uuid.uuid4())
return {"reply": markdown, "session_id": session_id}
@router.post("/chat")
async def handle_chat(msg: Message, request: Request):
try:
# Use existing session_id if provided, else generate new
session_id = msg.session_id
if not session_id:
session_id = str(uuid.uuid4())
print(session_id, ": user sent message : ", msg.text)
# Get or create a persistent thread_id for this session
if session_id not in thread_map:
thread_map[session_id] = str(uuid.uuid4())
thread_id = thread_map[session_id]
# Call your graph/chat function
reply_text = chat(
debug_mode=False,
message=msg.text,
history=None,
thread_id=thread_id,
preferred_language=msg.language or "English",
)
# Return both reply and session_id to the client
return {"reply": reply_text, "session_id": session_id}
except Exception as e:
traceback.print_exc()
return JSONResponse(status_code=500, content={"reply": f"Error: {e}"})
@router.post("/quiz/generate")
async def handle_quiz_generate(payload: QuizGeneratePayload, request: Request):
q = generate_question(
collection=payload.scripture
or random.choice(
[
s["collection_name"]
for s in SanatanConfig.scriptures
if s["collection_name"] != "yt_metadata"
]
),
complexity=payload.complexity
or random.choice(["beginner", "intermediate", "advanced"]),
mode=payload.mode or random.choice(["mcq", "open"]),
preferred_lamguage=payload.language or "English",
)
print(q.model_dump_json(indent=1))
return q.model_dump()
@router.post("/quiz/eval")
async def handle_quiz_eval(payload: QuizEvalPayload, request: Request):
result = validate_answer(
payload.q, payload.answer, preferred_language=payload.language or "English"
)
print(result.model_dump_json(indent=1))
return result
@router.get("/scriptures")
async def handle_get_scriptures():
return_values = {}
for scripture in SanatanConfig().scriptures:
if scripture["collection_name"] != "yt_metadata":
return_values[scripture["collection_name"]] = scripture["title"]
return return_values
class ScriptureRequest(BaseModel):
scripture_name: str
unit_index: int
@router.post("/scripture")
async def get_scripture(req: ScriptureRequest):
"""
Return a scripture unit (page or verse, based on config),
including all metadata fields separately.
used for page view to fetch by global index.
"""
logger.info("get_scripture: received request to fetch scripture: %s", req)
# find config entry for the scripture
config = next(
(s for s in SanatanConfig().scriptures if s["name"] == req.scripture_name), None
)
if not config:
return {"error": f"Scripture '{req.scripture_name}' not found"}
# fetch the raw document from DB
raw_doc = SanatanDatabase().fetch_document_by_index(
collection_name=config["collection_name"],
index=req.unit_index,
# unit_name=config.get("unit_field", config.get("unit")),
)
if not raw_doc or isinstance(raw_doc, str) or "error" in raw_doc:
return {"error": f"No data available for unit {req.unit_index}"}
# canonicalize it
canonical_doc = SanatanConfig().canonicalize_document(
scripture_name=req.scripture_name,
document_text=raw_doc.get("document", ""),
metadata_doc=raw_doc,
)
# add unit index & total units (so Flutter can paginate)
canonical_doc["total"] = SanatanDatabase().count(config["collection_name"])
print("canonical_doc = ", canonical_doc)
return canonical_doc
@router.get("/scripture_configs")
async def get_scripture_configs():
scriptures = []
for s in SanatanConfig().scriptures:
num_units = SanatanDatabase().count(collection_name=s["collection_name"])
# Deep copy metadata_fields so we don’t mutate the original config
metadata_fields = []
for f in s.get("metadata_fields", []):
f_copy = dict(f)
lov = f_copy.get("lov")
if callable(lov): # evaluate the function
try:
f_copy["lov"] = lov()
except Exception as e:
f_copy["lov"] = []
metadata_fields.append(f_copy)
scriptures.append(
{
"name": s["name"], # e.g. "bhagavad_gita"
"title": s["title"], # e.g. "Bhagavad Gita"
"unit": s["unit"], # e.g. "verse" or "page"
"unit_field": s.get("unit_field", s.get("unit")),
"total": num_units,
"enabled": "field_mapping" in s,
"source": s.get("source", ""),
"credits": s.get("credits", f"{s.get('source','')}"),
"metadata_fields": metadata_fields,
}
)
return {"scriptures": sorted(scriptures, key=lambda s: s["title"])}
@router.post("/scripture/{scripture_name}/search")
async def search_scripture(
scripture_name: str,
filter_obj: Optional[MetadataWhereClause] = None,
):
"""
Search scripture collection with optional filters.
- `scripture_name`: Name of the collection
- `filter_obj`: MetadataWhereClause (filters, groups, operator)
- `n_results`: number of random results to return
"""
try:
db = SanatanDatabase()
config = next(
(s for s in SanatanConfig().scriptures if s["name"] == scripture_name), None
)
results = db.fetch_first_match(
collection_name=config["collection_name"],
metadata_where_clause=filter_obj,
)
print("results = ", results)
# Flatten + canonicalize results
formatted_results = []
for i in range(len(results["metadatas"])):
id = results["ids"][i]
metadata_doc = results["metadatas"][i]
metadata_doc["id"] = id
print("metadata_doc = ",metadata_doc)
document_text = (
results["documents"][i] if results.get("documents") else None
)
canonical_doc = SanatanConfig().canonicalize_document(
scripture_name, document_text, metadata_doc
)
formatted_results.append(canonical_doc)
print("formatted_results = ", formatted_results)
return {"results": formatted_results}
except Exception as e:
logger.error("Error while searching %s", e, exc_info=True)
return {"error": str(e)}
@router.post("/audio")
async def generate_audio_urls(req: AudioRequest):
logger.info("generate_audio_urls: %s", req)
audio_urls = await svc_get_audio_urls(req)
return audio_urls
@router.post("/video")
async def generate_audio_urls(req: VideoRequest):
logger.info("generate_audio_urls: %s", req)
video_urls = await svc_get_video_urls(req)
return video_urls |