чистый pdfplumber (camelot/tabula дают мусор), v1.36

2026-06-17 22:09:26 +04:00
parent a150d3f2a8
commit 44ce0e09ab
2 changed files with 9 additions and 35 deletions
@@ -123,30 +123,6 @@ def _parse_doc(file_bytes: bytes) -> dict:

 def _parse_pdf(file_bytes: bytes) -> dict:
    result = {"elements": [], "errors": []}
-
-    # Попробовать camelot (pdfium, без системных зависимостей)
-    tables_from_camelot = False
-    try:
-        import camelot
-        for flavor in ("lattice", "stream"):
-            try:
-                ctables = camelot.read_pdf(io.BytesIO(file_bytes), pages="all", flavor=flavor)
-                for ct in ctables:
-                    rows = [list(ct.df.columns)] + ct.df.values.tolist()
-                    result["elements"].append({
-                        "type": "table",
-                        "rows": rows,
-                        "page": ct.page,
-                    })
-                    tables_from_camelot = True
-            except Exception:
-                pass
-    except ImportError:
-        result["errors"].append("camelot not installed, using pdfplumber")
-    except Exception as e:
-        result["errors"].append(f"camelot error: {e}")
-
-    # pdfplumber: текст + таблицы (если tabula не сработал)
    try:
        import pdfplumber
        with pdfplumber.open(io.BytesIO(file_bytes)) as pdf:
@@ -163,18 +139,16 @@ def _parse_pdf(file_bytes: bytes) -> dict:
                                "text": line,
                                "page": pn,
                            })
-                if not tables_from_camelot:
-                    tables = page.extract_tables()
-                    for tbl in tables:
-                        if tbl:
-                            result["elements"].append({
-                                "type": "table",
-                                "rows": tbl,
-                                "page": pn,
-                            })
+                tables = page.extract_tables()
+                for tbl in tables:
+                    if tbl:
+                        result["elements"].append({
+                            "type": "table",
+                            "rows": tbl,
+                            "page": pn,
+                        })
    except Exception as e:
        result["errors"].append(f"pdfplumber: {e}")
-
    return result


@@ -59,7 +59,7 @@
 <body>
  <div class="topbar">
    <img src="{{ url_for('static', filename='nubes-logo.svg') }}" alt="Nubes">
-    <span class="title">Сверка договоров <span style="font-weight:400;color:var(--muted);font-size:12px;">v1.35</span></span>
+    <span class="title">Сверка договоров <span style="font-weight:400;color:var(--muted);font-size:12px;">v1.36</span></span>
  </div>

  <div class="content">