From 27f86edfda7629642dcc2cccf9c54ae9735468b5 Mon Sep 17 00:00:00 2001
From: Andrew Leech <andrew.leech@planetinnovation.com.au>
Date: Tue, 4 Jul 2023 10:42:02 +1000
Subject: [PATCH] tugger-wix: Sanitise characters in Directory/ComponentGroup
 Id's.

---
 Cargo.lock               | 28 ++++++++++++++++++++++------
 tugger-wix/Cargo.toml    |  1 +
 tugger-wix/src/common.rs | 26 +++++++++-----------------
 3 files changed, 32 insertions(+), 23 deletions(-)

diff --git a/Cargo.lock b/Cargo.lock
index aaa719499..7aaff824b 100644
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -28,6 +28,15 @@ dependencies = [
  "memchr",
 ]
 
+[[package]]
+name = "aho-corasick"
+version = "1.0.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "43f6cb1bf222025340178f382c426f13757b2960e89779dfcb319c32542a5a41"
+dependencies = [
+ "memchr",
+]
+
 [[package]]
 name = "android_system_properties"
 version = "0.1.5"
@@ -2079,7 +2088,7 @@ version = "0.4.10"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "029d74589adefde59de1a0c4f4732695c32805624aec7b68d91503d4dba79afc"
 dependencies = [
- "aho-corasick",
+ "aho-corasick 0.7.20",
  "bstr",
  "fnv",
  "log",
@@ -2561,7 +2570,7 @@ dependencies = [
  "petgraph",
  "pico-args",
  "regex",
- "regex-syntax",
+ "regex-syntax 0.6.28",
  "string_cache",
  "term",
  "tiny-keccak",
@@ -3843,13 +3852,13 @@ dependencies = [
 
 [[package]]
 name = "regex"
-version = "1.7.1"
+version = "1.8.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "48aaa5748ba571fb95cd2c85c09f629215d3a6ece942baa100950af03a34f733"
+checksum = "d0ab3ca65655bb1e41f2a8c8cd662eb4fb035e67c3f78da1d61dffe89d07300f"
 dependencies = [
- "aho-corasick",
+ "aho-corasick 1.0.2",
  "memchr",
- "regex-syntax",
+ "regex-syntax 0.7.2",
 ]
 
 [[package]]
@@ -3864,6 +3873,12 @@ version = "0.6.28"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "456c603be3e8d448b072f410900c09faf164fbce2d480456f50eea6e25f9c848"
 
+[[package]]
+name = "regex-syntax"
+version = "0.7.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "436b050e76ed2903236f032a59761c1eb99e1b0aead2c257922771dab1fc8c78"
+
 [[package]]
 name = "release"
 version = "0.1.0"
@@ -5118,6 +5133,7 @@ dependencies = [
  "log",
  "msi",
  "once_cell",
+ "regex",
  "simple-file-manifest",
  "tempfile",
  "tugger-common",
diff --git a/tugger-wix/Cargo.toml b/tugger-wix/Cargo.toml
index 49d210057..c97b66c72 100644
--- a/tugger-wix/Cargo.toml
+++ b/tugger-wix/Cargo.toml
@@ -16,6 +16,7 @@ anyhow = "1.0.68"
 duct = "0.13.6"
 log = "0.4.17"
 once_cell = "1.17.0"
+regex = "1.8.4"
 simple-file-manifest = "0.11.0"
 url = "2.3.1"
 uuid = { version = "1.2.2", features = ["v4", "v5"] }
diff --git a/tugger-wix/src/common.rs b/tugger-wix/src/common.rs
index 0a57221c8..ab73b06ae 100644
--- a/tugger-wix/src/common.rs
+++ b/tugger-wix/src/common.rs
@@ -7,6 +7,7 @@ use {
     duct::cmd,
     log::warn,
     once_cell::sync::Lazy,
+    regex::Regex,
     simple_file_manifest::FileManifest,
     std::{
         io::{BufRead, BufReader, Write},
@@ -30,16 +31,15 @@ static WIX_TOOLSET: Lazy<RemoteContent> = Lazy::new(|| RemoteContent {
     sha256: "2c1888d5d1dba377fc7fa14444cf556963747ff9a0a289a3599cf09da03b9e2e".to_string(),
 });
 
+pub fn path_id(path: &Path) -> String {
+    // Identifiers may contain ASCII characters A-Z, a-z, digits, underscores (_), or periods (.)
+    let re = Regex::new(r"[^A-Za-z0-9_\.]").unwrap();
+    re.replace_all(&path.to_string_lossy().replace('\\', ".").replace('/', "."), "_").to_string()
+}
+
 /// Compute the `Id` of a directory.
 pub fn directory_to_id(prefix: &str, path: &Path) -> String {
-    format!(
-        "{}.dir.{}",
-        prefix,
-        path.to_string_lossy()
-            .replace('\\', "/")
-            .replace('/', ".")
-            .replace('-', "_")
-    )
+    format!("{}.dir.{}", prefix, path_id(path))
 }
 
 const GUID_NAMESPACE: &str = "https://github.com/indygreg/PyOxidizer/tugger/wix";
@@ -88,15 +88,7 @@ pub fn file_id(prefix: &str, path: &Path) -> String {
 }
 
 pub fn component_group_id(prefix: &str, path: &Path) -> String {
-    format!(
-        "{}.group.{}",
-        prefix,
-        path.display()
-            .to_string()
-            .replace('\\', "/")
-            .replace('/', ".")
-            .replace('-', "_")
-    )
+    format!("{}.group.{}", prefix, path_id(path))
 }
 
 /// Convert a `FileManifest` to WiX XML defining those files.