Spaces:
Running
Running
Update database/operations.py
Browse files- database/operations.py +91 -39
database/operations.py
CHANGED
|
@@ -4,22 +4,23 @@ from typing import Optional, List, Dict, Any, Tuple
|
|
| 4 |
from datetime import datetime
|
| 5 |
from bson import ObjectId
|
| 6 |
from pymongo import ASCENDING, DESCENDING
|
| 7 |
-
from pymongo import MongoClient
|
| 8 |
-
|
| 9 |
-
|
| 10 |
-
from database.connections import get_video_collection, get_script_collection, get_results_collection
|
| 11 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 12 |
|
| 13 |
log = logging.getLogger(__name__)
|
| 14 |
|
| 15 |
-
|
| 16 |
def _ensure_results_indexes(col):
|
| 17 |
try:
|
| 18 |
col.create_index([("created_at", DESCENDING)])
|
| 19 |
col.create_index([("type", ASCENDING), ("created_at", DESCENDING)])
|
| 20 |
col.create_index([("category", ASCENDING), ("created_at", DESCENDING)])
|
| 21 |
col.create_index([("created_by", ASCENDING), ("created_at", DESCENDING)])
|
| 22 |
-
|
| 23 |
except Exception:
|
| 24 |
pass
|
| 25 |
|
|
@@ -31,7 +32,6 @@ def _ensure_video_indexes(col):
|
|
| 31 |
except Exception:
|
| 32 |
pass
|
| 33 |
|
| 34 |
-
|
| 35 |
_rc = get_results_collection()
|
| 36 |
if _rc is not None:
|
| 37 |
_ensure_results_indexes(_rc)
|
|
@@ -44,9 +44,11 @@ _sg = get_script_collection()
|
|
| 44 |
if _sg is not None:
|
| 45 |
_ensure_video_indexes(_sg)
|
| 46 |
|
|
|
|
|
|
|
|
|
|
| 47 |
|
| 48 |
-
|
| 49 |
-
# ---------- Image jobs (generation / variation) ----------
|
| 50 |
def start_job(
|
| 51 |
col,
|
| 52 |
*,
|
|
@@ -70,14 +72,12 @@ def start_job(
|
|
| 70 |
"created_by": created_by,
|
| 71 |
"created_at": now,
|
| 72 |
}
|
| 73 |
-
|
| 74 |
if "file_name" in (inputs or {}):
|
| 75 |
doc["file_name"] = inputs["file_name"]
|
| 76 |
|
| 77 |
res = col.insert_one(doc)
|
| 78 |
return str(res.inserted_id)
|
| 79 |
|
| 80 |
-
|
| 81 |
def finish_job(
|
| 82 |
col,
|
| 83 |
job_id: str,
|
|
@@ -164,7 +164,85 @@ def find_video_analyses(
|
|
| 164 |
out.append(d)
|
| 165 |
return out
|
| 166 |
|
| 167 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 168 |
def find_generation_jobs(
|
| 169 |
*,
|
| 170 |
category: Optional[str] = None,
|
|
@@ -177,7 +255,6 @@ def find_generation_jobs(
|
|
| 177 |
col = get_results_collection()
|
| 178 |
if col is None:
|
| 179 |
return [], 0
|
| 180 |
-
|
| 181 |
q: Dict[str, Any] = {"type": "generation", "source": "text"}
|
| 182 |
if category:
|
| 183 |
q["category"] = category
|
|
@@ -188,7 +265,6 @@ def find_generation_jobs(
|
|
| 188 |
if start_date: rng["$gte"] = start_date
|
| 189 |
if end_date: rng["$lt"] = end_date
|
| 190 |
q["created_at"] = rng
|
| 191 |
-
|
| 192 |
total = col.count_documents(q)
|
| 193 |
cur = (
|
| 194 |
col.find(q)
|
|
@@ -196,7 +272,6 @@ def find_generation_jobs(
|
|
| 196 |
.skip(page * page_size)
|
| 197 |
.limit(page_size)
|
| 198 |
)
|
| 199 |
-
|
| 200 |
out: List[Dict[str, Any]] = []
|
| 201 |
for d in cur:
|
| 202 |
d["_id"] = str(d.get("_id"))
|
|
@@ -219,8 +294,7 @@ def insert_script_result(
|
|
| 219 |
) -> None:
|
| 220 |
col = get_script_collection()
|
| 221 |
if col is None:
|
| 222 |
-
raise ValueError("
|
| 223 |
-
|
| 224 |
doc: Dict[str, Any] = {
|
| 225 |
"type": "script",
|
| 226 |
"source": "script_generator",
|
|
@@ -236,13 +310,11 @@ def insert_script_result(
|
|
| 236 |
"num_scripts": num_scripts,
|
| 237 |
"created_at": datetime.utcnow(),
|
| 238 |
}
|
| 239 |
-
|
| 240 |
try:
|
| 241 |
col.insert_one(doc)
|
| 242 |
except Exception as e:
|
| 243 |
raise ValueError(f"Failed to insert script result: {e}")
|
| 244 |
|
| 245 |
-
|
| 246 |
def find_script_results(
|
| 247 |
*,
|
| 248 |
start_date: Optional[datetime] = None,
|
|
@@ -255,20 +327,16 @@ def find_script_results(
|
|
| 255 |
col = get_script_collection()
|
| 256 |
if col is None:
|
| 257 |
return [], 0
|
| 258 |
-
|
| 259 |
q: Dict[str, Any] = {"type": "script", "source": "script_generator"}
|
| 260 |
if start_date or end_date:
|
| 261 |
rng: Dict[str, Any] = {}
|
| 262 |
if start_date: rng["$gte"] = start_date
|
| 263 |
if end_date: rng["$lt"] = end_date
|
| 264 |
q["created_at"] = rng
|
| 265 |
-
|
| 266 |
if created_by:
|
| 267 |
q["created_by"] = created_by
|
| 268 |
-
|
| 269 |
if video_name_query:
|
| 270 |
q["video_name"] = {"$regex": video_name_query, "$options": "i"}
|
| 271 |
-
|
| 272 |
total = col.count_documents(q)
|
| 273 |
cur = (
|
| 274 |
col.find(q)
|
|
@@ -276,25 +344,9 @@ def find_script_results(
|
|
| 276 |
.skip(page * page_size)
|
| 277 |
.limit(page_size)
|
| 278 |
)
|
| 279 |
-
|
| 280 |
out: List[Dict[str, Any]] = []
|
| 281 |
for d in cur:
|
| 282 |
d["_id"] = str(d.get("_id"))
|
| 283 |
out.append(d)
|
| 284 |
return out, int(total)
|
| 285 |
|
| 286 |
-
|
| 287 |
-
def get_all_scripts(start_date: Optional[datetime] = None, end_date: Optional[datetime] = None, limit: int = 20) -> \
|
| 288 |
-
List[Dict[str, Any]]:
|
| 289 |
-
col = get_script_collection()
|
| 290 |
-
if col is None:
|
| 291 |
-
return []
|
| 292 |
-
query = {"type": "generation", "source": "text"}
|
| 293 |
-
|
| 294 |
-
if start_date:
|
| 295 |
-
query["created_at"] = {"$gte": start_date}
|
| 296 |
-
if end_date:
|
| 297 |
-
query["created_at"]["$lt"] = end_date
|
| 298 |
-
|
| 299 |
-
cursor = col.find(query).sort("created_at", DESCENDING).limit(limit)
|
| 300 |
-
return list(cursor)
|
|
|
|
| 4 |
from datetime import datetime
|
| 5 |
from bson import ObjectId
|
| 6 |
from pymongo import ASCENDING, DESCENDING
|
|
|
|
|
|
|
|
|
|
|
|
|
| 7 |
|
| 8 |
+
from database.connections import (
|
| 9 |
+
get_video_collection,
|
| 10 |
+
get_script_collection,
|
| 11 |
+
get_results_collection,
|
| 12 |
+
get_image_collection,
|
| 13 |
+
)
|
| 14 |
|
| 15 |
log = logging.getLogger(__name__)
|
| 16 |
|
| 17 |
+
# ------------------- Index Setup -------------------
|
| 18 |
def _ensure_results_indexes(col):
|
| 19 |
try:
|
| 20 |
col.create_index([("created_at", DESCENDING)])
|
| 21 |
col.create_index([("type", ASCENDING), ("created_at", DESCENDING)])
|
| 22 |
col.create_index([("category", ASCENDING), ("created_at", DESCENDING)])
|
| 23 |
col.create_index([("created_by", ASCENDING), ("created_at", DESCENDING)])
|
|
|
|
| 24 |
except Exception:
|
| 25 |
pass
|
| 26 |
|
|
|
|
| 32 |
except Exception:
|
| 33 |
pass
|
| 34 |
|
|
|
|
| 35 |
_rc = get_results_collection()
|
| 36 |
if _rc is not None:
|
| 37 |
_ensure_results_indexes(_rc)
|
|
|
|
| 44 |
if _sg is not None:
|
| 45 |
_ensure_video_indexes(_sg)
|
| 46 |
|
| 47 |
+
_ia = get_image_collection()
|
| 48 |
+
if _ia is not None:
|
| 49 |
+
_ensure_results_indexes(_ia)
|
| 50 |
|
| 51 |
+
# ------------------- Image Jobs -------------------
|
|
|
|
| 52 |
def start_job(
|
| 53 |
col,
|
| 54 |
*,
|
|
|
|
| 72 |
"created_by": created_by,
|
| 73 |
"created_at": now,
|
| 74 |
}
|
|
|
|
| 75 |
if "file_name" in (inputs or {}):
|
| 76 |
doc["file_name"] = inputs["file_name"]
|
| 77 |
|
| 78 |
res = col.insert_one(doc)
|
| 79 |
return str(res.inserted_id)
|
| 80 |
|
|
|
|
| 81 |
def finish_job(
|
| 82 |
col,
|
| 83 |
job_id: str,
|
|
|
|
| 164 |
out.append(d)
|
| 165 |
return out
|
| 166 |
|
| 167 |
+
|
| 168 |
+
def insert_image_analysis(
|
| 169 |
+
*,
|
| 170 |
+
image_name: str,
|
| 171 |
+
response: Dict[str, Any],
|
| 172 |
+
category: Optional[str] = None,
|
| 173 |
+
created_by: Optional[str] = None,
|
| 174 |
+
analyzer_model: Optional[str] = None,
|
| 175 |
+
image_meta: Optional[Dict[str, Any]] = None,
|
| 176 |
+
thumbnail: str = "",
|
| 177 |
+
) -> Optional[str]:
|
| 178 |
+
col = get_image_collection()
|
| 179 |
+
if col is None:
|
| 180 |
+
raise ValueError("Image collection not available")
|
| 181 |
+
|
| 182 |
+
doc: Dict[str, Any] = {
|
| 183 |
+
"type": "image_analysis",
|
| 184 |
+
"source": "image_analyzer",
|
| 185 |
+
"category": category or "general",
|
| 186 |
+
"image": {"name": image_name, **(image_meta or {})},
|
| 187 |
+
"analyzer_model": analyzer_model or "gpt-4o",
|
| 188 |
+
"results": response or {},
|
| 189 |
+
"created_by": created_by,
|
| 190 |
+
"created_at": datetime.utcnow(),
|
| 191 |
+
"thumbnail": thumbnail or "",
|
| 192 |
+
}
|
| 193 |
+
|
| 194 |
+
res = col.insert_one(doc)
|
| 195 |
+
return str(res.inserted_id)
|
| 196 |
+
|
| 197 |
+
|
| 198 |
+
|
| 199 |
+
def list_image_categories(created_by: Optional[str] = None) -> List[str]:
|
| 200 |
+
col = get_image_collection()
|
| 201 |
+
if col is None:
|
| 202 |
+
return []
|
| 203 |
+
try:
|
| 204 |
+
q = {"type": "image_analysis"}
|
| 205 |
+
if created_by:
|
| 206 |
+
q["created_by"] = created_by
|
| 207 |
+
vals = col.distinct("category", q)
|
| 208 |
+
return sorted({v for v in vals if v not in (None, "", [])})
|
| 209 |
+
except Exception:
|
| 210 |
+
return []
|
| 211 |
+
|
| 212 |
+
|
| 213 |
+
def find_image_analyses(
|
| 214 |
+
*,
|
| 215 |
+
category: Optional[str] = None,
|
| 216 |
+
start_date: Optional[datetime] = None,
|
| 217 |
+
end_date: Optional[datetime] = None,
|
| 218 |
+
limit: int = 200,
|
| 219 |
+
created_by: Optional[str] = None,
|
| 220 |
+
) -> List[Dict[str, Any]]:
|
| 221 |
+
col = get_image_collection()
|
| 222 |
+
if col is None:
|
| 223 |
+
return []
|
| 224 |
+
q: Dict[str, Any] = {"type": "image_analysis"}
|
| 225 |
+
if category:
|
| 226 |
+
q["category"] = category
|
| 227 |
+
if created_by:
|
| 228 |
+
q["created_by"] = created_by
|
| 229 |
+
if start_date or end_date:
|
| 230 |
+
rng: Dict[str, Any] = {}
|
| 231 |
+
if start_date:
|
| 232 |
+
rng["$gte"] = start_date
|
| 233 |
+
if end_date:
|
| 234 |
+
rng["$lt"] = end_date
|
| 235 |
+
q["created_at"] = rng
|
| 236 |
+
|
| 237 |
+
cur = col.find(q).sort("created_at", DESCENDING).limit(max(1, int(limit)))
|
| 238 |
+
out: List[Dict[str, Any]] = []
|
| 239 |
+
for d in cur:
|
| 240 |
+
d["_id"] = str(d.get("_id"))
|
| 241 |
+
out.append(d)
|
| 242 |
+
return out
|
| 243 |
+
|
| 244 |
+
|
| 245 |
+
# ------------------- Generation Jobs -------------------
|
| 246 |
def find_generation_jobs(
|
| 247 |
*,
|
| 248 |
category: Optional[str] = None,
|
|
|
|
| 255 |
col = get_results_collection()
|
| 256 |
if col is None:
|
| 257 |
return [], 0
|
|
|
|
| 258 |
q: Dict[str, Any] = {"type": "generation", "source": "text"}
|
| 259 |
if category:
|
| 260 |
q["category"] = category
|
|
|
|
| 265 |
if start_date: rng["$gte"] = start_date
|
| 266 |
if end_date: rng["$lt"] = end_date
|
| 267 |
q["created_at"] = rng
|
|
|
|
| 268 |
total = col.count_documents(q)
|
| 269 |
cur = (
|
| 270 |
col.find(q)
|
|
|
|
| 272 |
.skip(page * page_size)
|
| 273 |
.limit(page_size)
|
| 274 |
)
|
|
|
|
| 275 |
out: List[Dict[str, Any]] = []
|
| 276 |
for d in cur:
|
| 277 |
d["_id"] = str(d.get("_id"))
|
|
|
|
| 294 |
) -> None:
|
| 295 |
col = get_script_collection()
|
| 296 |
if col is None:
|
| 297 |
+
raise ValueError("Script collection not available.")
|
|
|
|
| 298 |
doc: Dict[str, Any] = {
|
| 299 |
"type": "script",
|
| 300 |
"source": "script_generator",
|
|
|
|
| 310 |
"num_scripts": num_scripts,
|
| 311 |
"created_at": datetime.utcnow(),
|
| 312 |
}
|
|
|
|
| 313 |
try:
|
| 314 |
col.insert_one(doc)
|
| 315 |
except Exception as e:
|
| 316 |
raise ValueError(f"Failed to insert script result: {e}")
|
| 317 |
|
|
|
|
| 318 |
def find_script_results(
|
| 319 |
*,
|
| 320 |
start_date: Optional[datetime] = None,
|
|
|
|
| 327 |
col = get_script_collection()
|
| 328 |
if col is None:
|
| 329 |
return [], 0
|
|
|
|
| 330 |
q: Dict[str, Any] = {"type": "script", "source": "script_generator"}
|
| 331 |
if start_date or end_date:
|
| 332 |
rng: Dict[str, Any] = {}
|
| 333 |
if start_date: rng["$gte"] = start_date
|
| 334 |
if end_date: rng["$lt"] = end_date
|
| 335 |
q["created_at"] = rng
|
|
|
|
| 336 |
if created_by:
|
| 337 |
q["created_by"] = created_by
|
|
|
|
| 338 |
if video_name_query:
|
| 339 |
q["video_name"] = {"$regex": video_name_query, "$options": "i"}
|
|
|
|
| 340 |
total = col.count_documents(q)
|
| 341 |
cur = (
|
| 342 |
col.find(q)
|
|
|
|
| 344 |
.skip(page * page_size)
|
| 345 |
.limit(page_size)
|
| 346 |
)
|
|
|
|
| 347 |
out: List[Dict[str, Any]] = []
|
| 348 |
for d in cur:
|
| 349 |
d["_id"] = str(d.get("_id"))
|
| 350 |
out.append(d)
|
| 351 |
return out, int(total)
|
| 352 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|