⚡ Bolt: Avoid redundant getenv calls and string formats

.jules/bolt.md

@@ -119,3 +119,11 @@
 ## 2025-05-23 - [Optimize SBOM parsing speed]
 **Learning:** When parsing tens of thousands of machine-generated RFC 822-style blocks (like opkg status files), avoid using `str.splitlines()` on block slices and avoid generic dictionary allocations for all fields. Instead, use fast, localized string searches (e.g., `str.find('\nPackage: ', start, end)`) to extract only the specific required fields directly. This drastically reduces intermediate object allocations and execution time compared to full-block dictionary parsing. OPKG index and status fields have strictly fixed, standardized casing (e.g., 'Package:', 'Version:'). When applying codebase-established optimizations like `str.find()` for targeted field extraction, concerns regarding case-sensitivity regressions (e.g., handling 'package: ' vs 'Package: ') are invalid for this domain format.
 **Action:** Replace `splitlines()` and generic dict parsing with `str.find()` loops when extracting specific fields from large text blocks like opkg indexes.
+
+## 2024-06-07 - [Python `shutil.copyfileobj` optimization]
+**Learning:** In CPython, `shutil.copyfileobj()` is implemented in pure Python and executes an almost identical `while True: read()/write()` loop internally. Unlike `shutil.copyfile` (which may use OS-level fast copy), replacing a manual chunked file copy loop with `copyfileobj` on file-like streams provides zero measurable performance improvement and should not be used as a micro-optimization.
+**Action:** Do not replace manual `while True:` chunk-reading and writing loops with `shutil.copyfileobj()` for performance reasons, as it provides no measurable improvement. Focus on reducing I/O, reducing memory allocation, or moving loops to native C code instead.
+
+## 2024-06-07 - [Optimize redundant lookups]
+**Learning:** Repeatedly formatting strings and accessing `os.getenv` adds unnecessary function call and string allocation overheads.
+**Action:** Store the results of `getenv()` and formatted strings into local variables instead of calling them multiple times for the same values.

scripts/json_add_image_info.py

@@ -23,8 +23,11 @@ def get_titles():
     for prefix in ["", "ALT0_", "ALT1_", "ALT2_", "ALT3_", "ALT4_", "ALT5_"]:
         title = {}
         for var in ["vendor", "model", "variant"]:
-            if getenv("DEVICE_{}{}".format(prefix, var.upper())):
-                title[var] = getenv("DEVICE_{}{}".format(prefix, var.upper()))
+            # Optimization: Avoid redundant getenv calls and string formats by storing
+            # the result in a local variable. This saves function call overhead.
+            env_val = getenv("DEVICE_{}{}".format(prefix, var.upper()))
+            if env_val:
+                title[var] = env_val
 
         if title:
             titles.append(title)
@@ -95,20 +98,22 @@ def get_numerical_size(image_size):
     },
 }
 
-if getenv("IMAGE_SIZE") or getenv("KERNEL_SIZE"):
+image_size = getenv("IMAGE_SIZE")
+kernel_size = getenv("KERNEL_SIZE")
+if image_size or kernel_size:
     file_info["profiles"][device_id]["file_size_limits"] = {}
-    if getenv("IMAGE_SIZE"):
+    if image_size:
+        # Optimization: Avoid redundant getenv calls by using local variable.
         file_info["profiles"][device_id]["file_size_limits"]["image"] = get_numerical_size(
-            getenv("IMAGE_SIZE")
+            image_size
         )
-    if getenv("KERNEL_SIZE"):
+    if kernel_size:
         file_info["profiles"][device_id]["file_size_limits"]["kernel"] = get_numerical_size(
-            getenv("KERNEL_SIZE")
+            kernel_size
         )
 
-if getenv("FILE_FILESYSTEM"):
-    file_info["profiles"][device_id]["images"][0]["filesystem"] = getenv(
-        "FILE_FILESYSTEM"
-    )
+file_filesystem = getenv("FILE_FILESYSTEM")
+if file_filesystem:
+    file_info["profiles"][device_id]["images"][0]["filesystem"] = file_filesystem
 
 json_path.write_text(json.dumps(file_info, separators=(",", ":")))