From 508a09ef25b429456eac152aae98ac20cf893097 Mon Sep 17 00:00:00 2001
From: Fariss <rev.fariss@gmail.com>
Date: Tue, 4 Jun 2024 18:47:41 +0200
Subject: [PATCH] include rule caching in PyInstaller build process (#2097)

* include rule caching in PyInstaller build process

The following commit introduces a new function that caches the capa
rule set, so that users don't have to manually run ./scripts/cache-
ruleset.py, before running pyinstaller.

* ci: omit Cache rule set step from build.yml workflow

* refactor: move cache generation to cache.py

* mkdir cache directory when it does not exist

---------

Co-authored-by: Soufiane Fariss <soufiane.fariss@um5s.net.ma>
Co-authored-by: Moritz <mr-tz@users.noreply.github.com>
---
 .github/pyinstaller/pyinstaller.spec | 15 ++++++++++++---
 .github/workflows/build.yml          |  2 --
 CHANGELOG.md                         |  1 +
 capa/rules/cache.py                  | 22 ++++++++++++++++++++++
 4 files changed, 35 insertions(+), 5 deletions(-)

diff --git a/.github/pyinstaller/pyinstaller.spec b/.github/pyinstaller/pyinstaller.spec
index f103ba16e..021a2b294 100644
--- a/.github/pyinstaller/pyinstaller.spec
+++ b/.github/pyinstaller/pyinstaller.spec
@@ -1,10 +1,19 @@
 # -*- mode: python -*-
 # Copyright (C) 2020 Mandiant, Inc. All Rights Reserved.
-import os.path
-import subprocess
+import sys
 
 import wcwidth
+import capa.rules.cache
 
+from pathlib import Path
+
+# SPECPATH is a global variable which points to .spec file path
+capa_dir = Path(SPECPATH).parent.parent
+rules_dir = capa_dir / 'rules'
+cache_dir = capa_dir / 'cache'
+
+if not capa.rules.cache.generate_rule_cache(rules_dir, cache_dir):
+    sys.exit(-1)
 
 a = Analysis(
     # when invoking pyinstaller from the project root,
@@ -26,7 +35,7 @@ a = Analysis(
         # so we manually embed the wcwidth resources here.
         #
         # ref: https://stackoverflow.com/a/62278462/87207
-        (os.path.dirname(wcwidth.__file__), "wcwidth"),
+        (Path(wcwidth.__file__).parent, "wcwidth"),
     ],
     # when invoking pyinstaller from the project root,
     # this gets run from the project root.
diff --git a/.github/workflows/build.yml b/.github/workflows/build.yml
index bc9c3908c..4a9bf555f 100644
--- a/.github/workflows/build.yml
+++ b/.github/workflows/build.yml
@@ -52,8 +52,6 @@ jobs:
         run: python -m pip install --upgrade pip setuptools
       - name: Install capa with build requirements
         run: pip install -e .[build]
-      - name: Cache the rule set
-        run: python ./scripts/cache-ruleset.py ./rules/ ./cache/
       - name: Build standalone executable
         run: pyinstaller --log-level DEBUG .github/pyinstaller/pyinstaller.spec
       - name: Does it run (PE)?
diff --git a/CHANGELOG.md b/CHANGELOG.md
index 5962ade6a..f27121a8b 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -50,6 +50,7 @@
 - ci: update github workflows to use latest version of actions that were using a deprecated version of node #1967 #2003 capa-rules#883 @sjha2048 @Ana06
 - ci: update binja version to stable 4.0 #2016 @xusheng6
 - ci: update github workflows to reflect the latest ghidrathon installation and bumped up jep, ghidra versions  #2020 @psahithireddy
+- ci: include rule caching in PyInstaller build process #2097 @s-ff
 - add deptry support #1497 @s-ff
 
 ### Raw diffs
diff --git a/capa/rules/cache.py b/capa/rules/cache.py
index 5cc1d1d4a..6f87570ef 100644
--- a/capa/rules/cache.py
+++ b/capa/rules/cache.py
@@ -159,3 +159,25 @@ def load_cached_ruleset(cache_dir: Path, rule_contents: List[bytes]) -> Optional
         return None
     else:
         return cache.ruleset
+
+
+def generate_rule_cache(rules_dir: Path, cache_dir: Path) -> bool:
+    if not rules_dir.is_dir():
+        logger.error("rules directory %s does not exist", rules_dir)
+        return False
+
+    try:
+        cache_dir.mkdir(parents=True, exist_ok=True)
+        rules = capa.rules.get_rules([rules_dir], cache_dir)
+    except (IOError, capa.rules.InvalidRule, capa.rules.InvalidRuleSet) as e:
+        logger.error("%s", str(e))
+        return False
+
+    content = capa.rules.cache.get_ruleset_content(rules)
+    id = capa.rules.cache.compute_cache_identifier(content)
+    path = capa.rules.cache.get_cache_path(cache_dir, id)
+
+    assert path.exists()
+    logger.info("rules cache saved to: %s", path)
+
+    return True