Merge pull request 'Fix the antispam test: pin the rules it scores against, stop live Pyzor' (#24) from fix/antispam-test into main

Reviewed-on: #24

THIRD-PARTY.md

@@ -24,6 +24,7 @@ carry their own license files.
 | `crates/common/src/network/acme/directory.rs`, `crates/common/src/network/acme/jose.rs`, `crates/common/src/network/acme/order.rs` | [rustls-acme](https://github.com/FlorianUekermann/rustls-acme) (MIT or Apache-2.0) | Copyright (c) Florian Uekermann |
 | `crates/types/src/id.rs` | [crockford](https://github.com/archer884/crockford) (MIT or Apache-2.0) | Copyright (c) 2017 J/A <archer884@gmail.com> |
 | `crates/nlp/src/tokenizers/types.rs` | test cases from [linkify](https://github.com/robinst/linkify) (MIT or Apache-2.0) | Copyright (c) 2017 Robin Stocker |
+| `tests/resources/smtp/antispam/spam-filter-rules.json.gz` | the published rules of [spam-filter](https://github.com/stalwartlabs/spam-filter) v3.0.2, unmodified, for the spam filter's tests (MIT or Apache-2.0) | Copyright (C) 2024, Stalwart Labs LLC |
 
 Each notice above applies with this permission notice:
 

crates/spam-filter/src/modules/pyzor.rs

@@ -2,6 +2,8 @@
  * SPDX-FileCopyrightText: 2020 Stalwart Labs LLC <[email protected]>
  *
  * SPDX-License-Identifier: AGPL-3.0-only OR LicenseRef-SEL
+ *
+ * Modified by Coffey Labs in 2026 for INBUXA.
  */
 
 use common::config::mailstore::spamfilter::PyzorConfig;
@@ -43,35 +45,14 @@ pub(crate) async fn pyzor_check(
     // Hash message
     let request = message.pyzor_check_message();
 
+    // Send message to address. inbuxa: in tests, a fixed table answers
+    // instead of a public server (test_response).
+    #[cfg(not(feature = "test_mode"))]
+    let response = pyzor_send_message(config.address, config.timeout, &request).await;
     #[cfg(feature = "test_mode")]
-    {
-        if request.contains("b5b476f0b5ba6e1c038361d3ded5818dd39c90a2") {
-            return Ok(PyzorResponse {
-                code: 200,
-                count: 1000,
-                wl_count: 0,
-            }
-            .into());
-        } else if request.contains("d67d4b8bfc3860449e3418bb6017e2612f3e2a99") {
-            return Ok(PyzorResponse {
-                code: 200,
-                count: 60,
-                wl_count: 10,
-            }
-            .into());
-        } else if request.contains("81763547012b75e57a20d18ce0b93014208cdfdb") {
-            return Ok(PyzorResponse {
-                code: 200,
-                count: 50,
-                wl_count: 20,
-            }
-            .into());
-        }
-    }
+    let response = std::io::Result::Ok(test_response(&request));
 
-    // Send message to address
-    pyzor_send_message(config.address, config.timeout, &request)
-        .await
+    response
         .map(Into::into)
         .map_err(|err| {
             trc::SpamEvent::PyzorError
@@ -82,6 +63,32 @@ pub(crate) async fn pyzor_check(
         })
 }
 
+/// inbuxa: the answers tests get, by digest, instead of a public server's,
+/// whose counts change and which a test may not be able to reach. Upstream
+/// answered the first three here and sent every other digest to the network.
+#[cfg(feature = "test_mode")]
+fn test_response(request: &str) -> PyzorResponse {
+    let (count, wl_count) = if request.contains("b5b476f0b5ba6e1c038361d3ded5818dd39c90a2")
+        // The digest of an empty body, as an HTML-only message with no text
+        // to hash produces; public servers report it widely.
+        || request.contains("da39a3ee5e6b4b0d3255bfef95601890afd80709")
+    {
+        (1000, 0)
+    } else if request.contains("d67d4b8bfc3860449e3418bb6017e2612f3e2a99") {
+        (60, 10)
+    } else if request.contains("81763547012b75e57a20d18ce0b93014208cdfdb") {
+        (50, 20)
+    } else {
+        (0, 0)
+    };
+    PyzorResponse {
+        code: 200,
+        count,
+        wl_count,
+    }
+}
+
+#[cfg_attr(feature = "test_mode", allow(dead_code))]
 async fn pyzor_send_message(
     addr: SocketAddr,
     timeout: Duration,

tests/src/smtp/inbound/antispam.rs

@@ -68,6 +68,10 @@ use std::{
     time::{Duration, Instant},
 };
 
+// inbuxa: its HTTP listener (19048) is shared with the dkim2 and report
+// tests, which are serial; without this it ran beside them and each got the
+// other's server.
+#[serial_test::serial]
 #[tokio::test(flavor = "multi_thread")]
 async fn antispam() {
     let mut test = TestServerBuilder::new("smtp_antispam_test")
@@ -81,9 +85,18 @@ async fn antispam() {
     admin
         .registry_create_object(SpamSettings {
             score_spam: Float::new(5.0),
+            // inbuxa: the rules carry the scores the expectations are written
+            // against, so they're pinned (spam-filter v3.0.2, beside the test
+            // cases) rather than read from a developer's own checkout, which
+            // left every score at zero. SPAM_RULES_URL still overrides.
             spam_filter_rules_url: std::env::var("SPAM_RULES_URL")
                 .unwrap_or_else(|_| {
-                    "file:///Users/me/code/spam-filter/spam-filter-rules.json.gz".to_string()
+                    concat!(
+                        "file://",
+                        env!("CARGO_MANIFEST_DIR"),
+                        "/resources/smtp/antispam/spam-filter-rules.json.gz"
+                    )
+                    .to_string()
                 })
                 .into(),
             ..Default::default()
@@ -156,7 +169,7 @@ async fn antispam() {
         }))
         .await;
     // inbuxa: a rules file that can't be read is retried later; don't wait
-    // for that retry (the path above is a developer's own checkout)
+    // for that retry (SPAM_RULES_URL may name one that isn't there)
     test.wait_for_tasks_skip_not_due().await;
     admin.reload_settings().await;
     admin.reload_lookup_stores().await;