Fix: empty query issue. (#7551)

### What problem does this PR solve?

#5214

### Type of change

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

deepdoc/parser/pdf_parser.py

@@ -309,7 +309,7 @@ class RAGFlowPdfParser:
               "bottom": b[-1][1] / ZM,
               "chars": [],
               "page_number": pagenum} for b, t in bxs if b[0][0] <= b[1][0] and b[0][1] <= b[-1][1]],
-            self.mean_height[-1] / 3
+            self.mean_height[pagenum-1] / 3
         )
 
         # merge chars in the same rect
@@ -355,8 +355,8 @@ class RAGFlowPdfParser:
             del boxes_to_reg[i]["box_image"]
         logging.info(f"__ocr recognize {len(bxs)} boxes cost {timer() - start}s")
         bxs = [b for b in bxs if b["text"]]
-        if self.mean_height[-1] == 0:
-            self.mean_height[-1] = np.median([b["bottom"] - b["top"]
+        if self.mean_height[pagenum-1] == 0:
+            self.mean_height[pagenum-1] = np.median([b["bottom"] - b["top"]
                                               for b in bxs])
         self.boxes.append(bxs)
 

rag/nlp/query.py

@@ -77,6 +77,7 @@ class FulltextQueryer:
             " ",
             rag_tokenizer.tradi2simp(rag_tokenizer.strQ2B(txt.lower())),
         ).strip()
+        otxt = txt
         txt = FulltextQueryer.rmWWW(txt)
 
         if not self.isChinese(txt):
@@ -196,6 +197,8 @@ class FulltextQueryer:
 
         if qs:
             query = " OR ".join([f"({t})" for t in qs if t])
+            if not query:
+                query = otxt
             return MatchTextExpr(
                 self.query_fields, query, 100, {"minimum_should_match": min_match}
             ), keywords