Fix: presentation parsing & Embedding encode exception handling (#11933)

### What problem does this PR solve?

Fix: presentation parsing #11920
Fix: Embeddin encode exception handling
### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

Author: Magicbook1108

rag/app/presentation.py

@@ -227,8 +227,9 @@ def chunk(filename, binary=None, from_page=0, to_page=100000,
         for pn, (txt, img) in enumerate(sections):
             d = copy.deepcopy(doc)
             pn += from_page
-            if img:
-                d["image"] = img
+            if not isinstance(img, Image.Image):
+                img = None
+            d["image"] = img
             d["page_num_int"] = [pn + 1]
             d["top_int"] = [0]
             d["position_int"] = [(pn + 1, 0, img.size[0] if img else 0, 0,

rag/llm/embedding_model.py

@@ -121,11 +121,16 @@ def encode(self, texts: list):
                 total_tokens += self.total_token_count(res)
             except Exception as _e:
                 log_exception(_e, res)
+                raise Exception(f"Error: {res}")
         return np.array(ress), total_tokens
 
     def encode_queries(self, text):
         res = self.client.embeddings.create(input=[truncate(text, 8191)], model=self.model_name, encoding_format="float",extra_body={"drop_params": True})
-        return np.array(res.data[0].embedding), self.total_token_count(res)
+        try:
+            return np.array(res.data[0].embedding), self.total_token_count(res)
+        except Exception as _e:
+            log_exception(_e, res)
+            raise Exception(f"Error: {res}")
 
 
 class LocalAIEmbed(Base):
@@ -147,6 +152,7 @@ def encode(self, texts: list):
                 ress.extend([d.embedding for d in res.data])
             except Exception as _e:
                 log_exception(_e, res)
+                raise Exception(f"Error: {res}")
         # local embedding for LmStudio donot count tokens
         return np.array(ress), 1024
 
@@ -222,6 +228,7 @@ def encode_queries(self, text):
             return np.array(resp["output"]["embeddings"][0]["embedding"]), self.total_token_count(resp)
         except Exception as _e:
             log_exception(_e, resp)
+            raise Exception(f"Error: {resp}")
 
 
 class ZhipuEmbed(Base):
@@ -249,6 +256,7 @@ def encode(self, texts: list):
                 tks_num += self.total_token_count(res)
             except Exception as _e:
                 log_exception(_e, res)
+                raise Exception(f"Error: {res}")
         return np.array(arr), tks_num
 
     def encode_queries(self, text):
@@ -257,6 +265,7 @@ def encode_queries(self, text):
             return np.array(res.data[0].embedding), self.total_token_count(res)
         except Exception as _e:
             log_exception(_e, res)
+            raise Exception(f"Error: {res}")
 
 
 class OllamaEmbed(Base):
@@ -281,6 +290,7 @@ def encode(self, texts: list):
                 arr.append(res["embedding"])
             except Exception as _e:
                 log_exception(_e, res)
+                raise Exception(f"Error: {res}")
             tks_num += 128
         return np.array(arr), tks_num
 
@@ -293,6 +303,7 @@ def encode_queries(self, text):
             return np.array(res["embedding"]), 128
         except Exception as _e:
             log_exception(_e, res)
+            raise Exception(f"Error: {res}")
 
 
 class XinferenceEmbed(Base):
@@ -315,6 +326,7 @@ def encode(self, texts: list):
                 total_tokens += self.total_token_count(res)
             except Exception as _e:
                 log_exception(_e, res)
+                raise Exception(f"Error: {res}")
         return np.array(ress), total_tokens
 
     def encode_queries(self, text):
@@ -324,6 +336,7 @@ def encode_queries(self, text):
             return np.array(res.data[0].embedding), self.total_token_count(res)
         except Exception as _e:
             log_exception(_e, res)
+            raise Exception(f"Error: {res}")
 
 
 class YoudaoEmbed(Base):
@@ -399,6 +412,7 @@ def encode(self, texts: list[str|bytes], task="retrieval.passage"):
                 token_count += self.total_token_count(res)
             except Exception as _e:
                 log_exception(_e, response)
+                raise Exception(f"Error: {response}")
         return np.array(ress), token_count
 
     def encode_queries(self, text):
@@ -531,6 +545,7 @@ def encode(self, texts: list):
                 ress.extend(result["embedding"])
             except Exception as _e:
                 log_exception(_e, result)
+                raise Exception(f"Error: {result}")
         return np.array(ress), token_count
 
     def encode_queries(self, text):
@@ -541,6 +556,7 @@ def encode_queries(self, text):
             return np.array(result["embedding"]), token_count
         except Exception as _e:
             log_exception(_e, result)
+            raise Exception(f"Error: {result}")
 
 
 class NvidiaEmbed(Base):
@@ -578,10 +594,11 @@ def encode(self, texts: list):
             response = requests.post(self.base_url, headers=self.headers, json=payload)
             try:
                 res = response.json()
+                ress.extend([d["embedding"] for d in res["data"]])
+                token_count += self.total_token_count(res)
             except Exception as _e:
                 log_exception(_e, response)
-            ress.extend([d["embedding"] for d in res["data"]])
-            token_count += self.total_token_count(res)
+                raise Exception(f"Error: {response}")
         return np.array(ress), token_count
 
     def encode_queries(self, text):
@@ -636,6 +653,7 @@ def encode(self, texts: list):
                 token_count += res.meta.billed_units.input_tokens
             except Exception as _e:
                 log_exception(_e, res)
+                raise Exception(f"Error: {res}")
         return np.array(ress), token_count
 
     def encode_queries(self, text):
@@ -649,6 +667,7 @@ def encode_queries(self, text):
             return np.array(res.embeddings.float[0]), int(res.meta.billed_units.input_tokens)
         except Exception as _e:
             log_exception(_e, res)
+            raise Exception(f"Error: {res}")
 
 
 class TogetherAIEmbed(OpenAIEmbed):
@@ -716,6 +735,7 @@ def encode(self, texts: list):
                 token_count += self.total_token_count(res)
             except Exception as _e:
                 log_exception(_e, response)
+                raise Exception(f"Error: {response}")
 
         return np.array(ress), token_count
 
@@ -731,6 +751,7 @@ def encode_queries(self, text):
             return np.array(res["data"][0]["embedding"]), self.total_token_count(res)
         except Exception as _e:
             log_exception(_e, response)
+            raise Exception(f"Error: {response}")
 
 
 class ReplicateEmbed(Base):
@@ -777,6 +798,7 @@ def encode(self, texts: list, batch_size=16):
             )
         except Exception as _e:
             log_exception(_e, res)
+            raise Exception(f"Error: {res}")
 
     def encode_queries(self, text):
         res = self.client.do(model=self.model_name, texts=[text]).body
@@ -787,6 +809,7 @@ def encode_queries(self, text):
             )
         except Exception as _e:
             log_exception(_e, res)
+            raise Exception(f"Error: {res}")
 
 
 class VoyageEmbed(Base):
@@ -809,6 +832,7 @@ def encode(self, texts: list):
                 token_count += res.total_tokens
             except Exception as _e:
                 log_exception(_e, res)
+                raise Exception(f"Error: {res}")
         return np.array(ress), token_count
 
     def encode_queries(self, text):
@@ -817,6 +841,7 @@ def encode_queries(self, text):
             return np.array(res.embeddings)[0], res.total_tokens
         except Exception as _e:
             log_exception(_e, res)
+            raise Exception(f"Error: {res}")
 
 
 class HuggingFaceEmbed(Base):