From 14b64a0596ae71bdd8e9818cdc3ba7a5424b1aab Mon Sep 17 00:00:00 2001
From: devdudumuniz <82589615+devdudumuniz@users.noreply.github.com>
Date: Sat, 29 Aug 2026 21:30:32 +0000
Subject: [PATCH] =?UTF-8?q?=E2=9A=A1=20Bolt:=20[performance=20improvement]?=
 =?UTF-8?q?=20optimize=20=5Fdecode=20string=20logic=20in=20dbf=5Freader?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 pysus/data/dbf_reader.py | 28 ++++++++++++++++------------
 1 file changed, 16 insertions(+), 12 deletions(-)

diff --git a/pysus/data/dbf_reader.py b/pysus/data/dbf_reader.py
index 4730f48e..7b4b0fff 100644
--- a/pysus/data/dbf_reader.py
+++ b/pysus/data/dbf_reader.py
@@ -158,12 +158,14 @@ def read_dbf_fast(
     data = {}
     for fld in target:
         col: np.ndarray = records[fld.name]
-        decoded = np.empty(n, dtype=object)
-        for i in range(n):
-            val = col[i]
-            b = val if isinstance(val, bytes) else val.tobytes()
-            decoded[i] = _decode(b)
-        data[fld.name] = decoded
+        # Optimize decoding loop with list comprehension
+        data[fld.name] = [
+            (b if isinstance(b, bytes) else b.tobytes())
+            .decode(_ENCODING, errors="replace")
+            .replace("\x00", "")
+            .strip()
+            for b in col
+        ]
 
     return pd.DataFrame(data)
 
@@ -273,12 +275,14 @@ def stream_dbf_fast(
         data = {}
         for fld in schema.fields:
             col: np.ndarray = records[fld.name]
-            decoded = np.empty(chunk_n, dtype=object)
-            for i in range(chunk_n):
-                val = col[i]
-                b = val if isinstance(val, bytes) else val.tobytes()
-                decoded[i] = _decode(b)
-            data[fld.name] = decoded
+            # Optimize decoding loop with list comprehension
+            data[fld.name] = [
+                (b if isinstance(b, bytes) else b.tobytes())
+                .decode(_ENCODING, errors="replace")
+                .replace("\x00", "")
+                .strip()
+                for b in col
+            ]
 
         yield pd.DataFrame(data)