koffi 2.5.16 → 2.5.17

package/CHANGELOG.md

@@ -4,6 +4,10 @@
 
 ### Koffi 2.5
 
+#### Koffi 2.5.17 (2023-08-27)
+
+- Fix compatibility with Electron
+
 #### Koffi 2.5.16 (2023-08-25)
 
 - Run Koffi tests through usual index.js entry point

package/doc/benchmarks.md

@@ -23,10 +23,10 @@ The results presented below were measured on my x86_64 Linux machine (Intel® Co
 
 ### rand results
 
-This test is based around repeated calls to a simple standard C function atoi, and has three implementations:
+This test is based around repeated calls to a simple standard C function `rand`, and has three implementations:
 
-- the first one is the reference, it calls atoi through an N-API module, and is close to the theoretical limit of a perfect (no overhead) Node.js > C FFI implementation (pre-compiled static glue code)
-- the second one calls atoi through Koffi
+- the first one is the reference, it calls rand through an N-API module, and is close to the theoretical limit of a perfect (no overhead) Node.js > C FFI implementation (pre-compiled static glue code)
+- the second one calls rand through Koffi
 - the third one uses the official Node.js FFI implementation, node-ffi-napi
 
 Benchmark     | Iteration time | Relative performance | Overhead
@@ -39,7 +39,7 @@ Because rand is a pretty small function, the FFI overhead is clearly visible.
 
 ### atoi results
 
-This test is similar to the rand one, but it is based on atoi, which takes a string parameter. Javascript (V8) to C string conversion is relatively slow and heavy.
+This test is similar to the rand one, but it is based on `atoi`, which takes a string parameter. Javascript (V8) to C string conversion is relatively slow and heavy.
 
 Benchmark     | Iteration time | Relative performance | Overhead
 ------------- | -------------- | -------------------- | --------
@@ -51,7 +51,7 @@ Because atoi is a pretty small function, the FFI overhead is clearly visible.
 
 ### Raylib results
 
-This benchmark uses the CPU-based image drawing functions in Raylib. The calls are much heavier than in the atoi benchmark, thus the FFI overhead is reduced. In this implementation, Koffi is compared to:
+This benchmark uses the CPU-based image drawing functions in Raylib. The calls are much heavier than in previous benchmarks, thus the FFI overhead is reduced. In this implementation, Koffi is compared to:
 
 - Baseline: Full C++ version of the code (no JS)
 - [node-raylib](https://github.com/RobLoach/node-raylib): This is a native wrapper implemented with N-API
@@ -69,10 +69,10 @@ The results presented below were measured on my x86_64 Windows machine (Intel®
 
 ### rand results
 
-This test is based around repeated calls to a simple standard C function atoi, and has three implementations:
+This test is based around repeated calls to a simple standard C function `rand`, and has three implementations:
 
-- the first one is the reference, it calls atoi through an N-API module, and is close to the theoretical limit of a perfect (no overhead) Node.js > C FFI implementation (pre-compiled static glue code)
-- the second one calls atoi through Koffi
+- the first one is the reference, it calls rand through an N-API module, and is close to the theoretical limit of a perfect (no overhead) Node.js > C FFI implementation (pre-compiled static glue code)
+- the second one calls rand through Koffi
 - the third one uses the official Node.js FFI implementation, node-ffi-napi
 
 Benchmark     | Iteration time | Relative performance | Overhead
@@ -85,7 +85,7 @@ Because rand is a pretty small function, the FFI overhead is clearly visible.
 
 ### atoi results
 
-This test is similar to the rand one, but it is based on atoi, which takes a string parameter. Javascript (V8) to C string conversion is relatively slow and heavy.
+This test is similar to the rand one, but it is based on `atoi`, which takes a string parameter. Javascript (V8) to C string conversion is relatively slow and heavy.
 
 The results below were measured on my x86_64 Windows machine (Intel® Core™ i5-4460):
 

package/doc/start.md

@@ -20,7 +20,7 @@ import koffi from 'koffi';
 
 ## Simple examples
 
-Below you can find three examples:
+Below you can find two examples:
 
 * The first one runs on Linux. The functions are declared with the C-like prototype language.
 * The second one runs on Windows, and uses the node-ffi like syntax to declare functions.

package/package.json

@@ -1,7 +1,7 @@
 {
     "name": "koffi",
-    "version": "2.5.16",
-    "stable": "2.5.16",
+    "version": "2.5.17",
+    "stable": "2.5.17",
     "description": "Fast and simple C FFI (foreign function interface) for Node.js",
     "keywords": [
         "foreign",

package/src/cnoke/assets/win_delay_hook.c

@@ -29,41 +29,24 @@
 #include <windows.h>
 #include <delayimp.h>
 
-static HMODULE node_dll;
-static HMODULE nw_dll;
-
 static FARPROC WINAPI self_exe_hook(unsigned int event, DelayLoadInfo *info)
 {
-    if (event == dliStartProcessing) {
-        node_dll = GetModuleHandleA("node.dll");
-        nw_dll = GetModuleHandleA("nw.dll");
-
-        return NULL;
-    }
-
-    if (event == dliNotePreGetProcAddress) {
-        if (node_dll) {
-            FARPROC ret = GetProcAddress(node_dll, info->dlp.szProcName);
-            if (ret)
-                return ret;
-        }
-
-        if (nw_dll) {
-            FARPROC ret = GetProcAddress(nw_dll, info->dlp.szProcName);
-            if (ret)
-                return ret;
-        }
-    }
-
-    if (event == dliNotePreLoadLibrary && _stricmp(info->szDll, "node.exe") != 0) {
-        if (!node_dll) {
-            node_dll = GetModuleHandleA(NULL);
+    static const wchar_t *const NodeLibraries[] = {
+        L"node.dll",
+        NULL
+    };
+
+    if (event == dliNotePreLoadLibrary && !stricmp(info->szDll, "node.exe")) {
+        for (int i = 0; i < sizeof(NodeLibraries) / sizeof(*NodeLibraries); i++) {
+            const wchar_t *name = NodeLibraries[i];
+            HMODULE h = GetModuleHandleW(name);
+
+            if (h)
+                return (FARPROC)h;
         }
-        return (FARPROC)node_dll;
     }
 
     return NULL;
 }
 
 const PfnDliHook __pfnDliNotifyHook2 = self_exe_hook;
-const PfnDliHook __pfnDliFailureHook2 = self_exe_hook;

package/src/index.js

@@ -378,8 +378,8 @@ var require_package = __commonJS({
   "build/dist/src/koffi/package.json"(exports2, module2) {
     module2.exports = {
       name: "koffi",
-      version: "2.5.16",
-      stable: "2.5.16",
+      version: "2.5.17",
+      stable: "2.5.17",
       description: "Fast and simple C FFI (foreign function interface) for Node.js",
       keywords: [
         "foreign",

package/src/koffi/examples/electron-builder/package.json

@@ -16,6 +16,6 @@
     "esbuild": "^0.18.17"
   },
   "dependencies": {
-    "koffi": "^2.5.9"
+    "koffi": "2.5.17"
   }
 }

package/src/koffi/examples/electron-forge/package.json

@@ -32,6 +32,6 @@
   },
   "dependencies": {
     "electron-squirrel-startup": "^1.0.0",
-    "koffi": "^2.5.9"
+    "koffi": "^2.5.17"
   }
 }

package/src/koffi/examples/node-esbuild/package.json

@@ -11,6 +11,6 @@
   "license": "ISC",
   "dependencies": {
     "esbuild": "^0.18.17",
-    "koffi": "^2.5.9"
+    "koffi": "^2.5.17"
   }
 }

package/src/koffi/examples/nwjs/src/package.json

@@ -5,6 +5,6 @@
   "author": "Niels Martignène <niels.martignene@protonmail.com>",
   "license": "MIT",
   "dependencies": {
-    "koffi": "^2.5.14"
+    "koffi": "^2.5.17"
   }
 }