Trace infrastructure outlier prefix classes

crates/rrt-runtime/src/smp.rs

@@ -2242,6 +2242,42 @@ pub struct SmpSavePlacedStructureDynamicSideBufferNamePreludeProfileSpanCorrelat
     #[serde(default)]
     pub mode_family_counts:
         Vec<SmpSavePlacedStructureDynamicSideBufferNamePreludeModeFamilyCount>,
+    #[serde(default)]
+    pub compact_prefix_pattern_summaries:
+        Vec<SmpSavePlacedStructureDynamicSideBufferNamePreludeProfileSpanPrefixSummary>,
+    #[serde(default)]
+    pub sample_rows: Vec<SmpSavePlacedStructureDynamicSideBufferNamePreludeProfileSpanSample>,
+}
+
+#[derive(Debug, Clone, PartialEq, Eq, Serialize, Deserialize)]
+pub struct SmpSavePlacedStructureDynamicSideBufferNamePreludeProfileSpanPrefixSummary {
+    pub prefix_leading_dword: u32,
+    pub prefix_leading_dword_hex: String,
+    pub prefix_trailing_word: u16,
+    pub prefix_trailing_word_hex: String,
+    pub prefix_separator_byte: u8,
+    pub prefix_separator_byte_hex: String,
+    pub count: usize,
+}
+
+#[derive(Debug, Clone, PartialEq, Eq, Serialize, Deserialize)]
+pub struct SmpSavePlacedStructureDynamicSideBufferNamePreludeProfileSpanSample {
+    pub sample_index: usize,
+    pub name_tag_relative_offset: usize,
+    #[serde(default)]
+    pub primary_name: Option<String>,
+    #[serde(default)]
+    pub secondary_name: Option<String>,
+    pub prefix_leading_dword: u32,
+    pub prefix_leading_dword_hex: String,
+    pub prefix_trailing_word: u16,
+    pub prefix_trailing_word_hex: String,
+    pub prefix_separator_byte: u8,
+    pub prefix_separator_byte_hex: String,
+    pub child_count_candidate: u16,
+    pub child_count_candidate_hex: String,
+    pub saved_primary_child_byte_candidate: u8,
+    pub saved_primary_child_byte_candidate_hex: String,
 }
 
 #[derive(Debug, Clone, PartialEq, Eq, Serialize, Deserialize)]
@@ -4687,7 +4723,7 @@ fn build_infrastructure_asset_trace_report(
                     })
                     .map(|correlation| {
                         format!(
-                            "current save-side probe now also shows the short 0x03-byte post-profile gaps collapsing cleanly to the next-record prelude: dominant candidate pattern is {}/{} x{} across {} rows",
+                            "current save-side probe now also shows the short 0x03-byte post-profile gaps collapsing cleanly to the next-record prelude: dominant candidate pattern is {}/{} x{} across {} rows, mode counts={:?}, prefix counts={:?}",
                             correlation
                                 .dominant_candidate_pattern
                                 .as_ref()
@@ -4703,12 +4739,74 @@ fn build_infrastructure_asset_trace_report(
                                 .as_ref()
                                 .map(|pattern| pattern.count)
                                 .unwrap_or_default(),
-                            correlation.row_count
+                            correlation.row_count,
+                            correlation
+                                .mode_family_counts
+                                .iter()
+                                .map(|mode| format!("{}:{}", mode.mode_family, mode.count))
+                                .collect::<Vec<_>>(),
+                            correlation
+                                .compact_prefix_pattern_summaries
+                                .iter()
+                                .map(|prefix| format!(
+                                    "{}/{}/{}:{}",
+                                    prefix.prefix_leading_dword_hex,
+                                    prefix.prefix_trailing_word_hex,
+                                    prefix.prefix_separator_byte_hex,
+                                    prefix.count
+                                ))
+                                .collect::<Vec<_>>()
                         )
                     })
                     .unwrap_or_else(|| {
                         "no grounded 0x03-byte post-profile span correlation was available for the prelude candidates".to_string()
                     }),
+                side_buffer
+                    .and_then(|probe| probe.payload_envelope_summary.as_ref())
+                    .and_then(|summary| summary.name_prelude_candidate_summary.as_ref())
+                    .and_then(|summary| {
+                        summary
+                            .profile_span_correlations
+                            .iter()
+                            .find(|row| row.previous_profile_chunk_len_to_next_name_or_end == 0x27)
+                    })
+                    .map(|correlation| {
+                        format!(
+                            "the sparse 0x27 post-profile outlier is now explicit too: mode counts={:?}, prefix counts={:?}, sample rows={:?}",
+                            correlation
+                                .mode_family_counts
+                                .iter()
+                                .map(|mode| format!("{}:{}", mode.mode_family, mode.count))
+                                .collect::<Vec<_>>(),
+                            correlation
+                                .compact_prefix_pattern_summaries
+                                .iter()
+                                .map(|prefix| format!(
+                                    "{}/{}/{}:{}",
+                                    prefix.prefix_leading_dword_hex,
+                                    prefix.prefix_trailing_word_hex,
+                                    prefix.prefix_separator_byte_hex,
+                                    prefix.count
+                                ))
+                                .collect::<Vec<_>>(),
+                            correlation
+                                .sample_rows
+                                .iter()
+                                .map(|sample| format!(
+                                    "{}:{:?}/{:?}@{}/{}/{}",
+                                    sample.name_tag_relative_offset,
+                                    sample.primary_name,
+                                    sample.secondary_name,
+                                    sample.prefix_leading_dword_hex,
+                                    sample.prefix_trailing_word_hex,
+                                    sample.prefix_separator_byte_hex
+                                ))
+                                .collect::<Vec<_>>()
+                        )
+                    })
+                    .unwrap_or_else(|| {
+                        "no grounded 0x27 post-profile outlier correlation was available for the prelude candidates".to_string()
+                    }),
                 side_buffer
                     .and_then(|probe| probe.payload_envelope_summary.as_ref())
                     .and_then(|summary| summary.name_prelude_candidate_summary.as_ref())
@@ -4769,7 +4867,7 @@ fn build_infrastructure_asset_trace_report(
                 "the smaller attach primitive 0x00490a3c no longer looks like the semantic fork by itself: it just allocates one literal Infrastructure child, seeds it through 0x455b70 with caller-provided stem input, attaches it through 0x5395d0, seeds position lanes through 0x539530/0x53a5b0, and optionally caches it as primary child".to_string(),
             ],
             blockers: vec![
-                "how the remaining compact-prefix regimes subdivide the tiny residual outlier spans inside the dominant mixed 0x0001/0xff class, now that the new mode-family correlations already narrow grounded q.gms to bridge:62 / track_cap:21 / tunnel:19 overall and further split the prior 0x55f3 spans as 0x06 -> pure bridge, 0x03 -> mostly tunnel with a small track-cap residue, 0x0f/0x2d/0x33 -> track_cap, 0x36 -> tunnel, leaving only sparse mixed outliers such as 0x27".to_string(),
+                "how the exact residual compact-prefix patterns map back onto constructor semantics now that the mixed 0x0001/0xff class is narrowed to concrete outliers: span 0x03 stays entirely on 0x000055f3/0x0001/0xff but still mixes tunnel with a tiny track-cap residue, while span 0x27 splits cleanly between 0xff0000ff/0x0001/0xff tunnel residue and 0xff0000ff/0x0002/0xff bridge residue".to_string(),
                 "how the payload streams reached through 0x00518380 -> 0x00518140 align with the embedded 0x55f1 name-pair groups and compact-prefix regimes surfaced by the save-side probe".to_string(),
                 "how the observed 0x55f3-to-next-0x55f1 gaps partition between the two 0x52ebd0 flag bytes and the next-record prelude now that 0x0048a6c0 is grounded as the writer for the outer child-count / primary-child prelude".to_string(),
                 "which fields written through the grounded 0x00490960 -> 0x004559d0 -> slot +0x4c -> 0x52ec50 chain retain the 0x38a5 embedded name-pair semantics before route/local-runtime follow-ons take over".to_string(),
@@ -4884,7 +4982,7 @@ fn build_infrastructure_asset_trace_report(
         ),
     ];
     let notes = vec![
-        "Infrastructure asset trace now makes the side-buffer-versus-triplet split explicit: owner seam identity is grounded, the pure bridge-only 0x0002/0xff candidate class is grounded save-side, the upstream chooser above the child attach path is grounded as paired DT/ST siblings at 0x004a2c80 and 0x004a34e0 with decoded Bridge/Tunnel/BallastCap/Overpass families, grounded top-level branch meaning, grounded bridge/tunnel material selector roles, a concrete child-construction/write-side chain through 0x00490960, 0x00491c60, 0x0048a6c0, 0x00455a40, and 0x004559d0, and stable 0x00490960 mode families for BallastCap, TrackCap, Overpass, Tunnel, and Bridge branches. The current save-side name corpus already maps BallastCap, TrackCap, Tunnel, and Bridge rows onto those families directly, the candidate-pattern correlation narrows the dominant mixed 0x0001/0xff class to bridge:62 / track_cap:21 / tunnel:19, and the profile-span correlation further splits most of that mixed class by prior 0x55f3 span, leaving only sparse outlier spans such as 0x27 and the small track-cap residue inside 0x03 as the remaining unknown.".to_string(),
+        "Infrastructure asset trace now makes the side-buffer-versus-triplet split explicit: owner seam identity is grounded, the pure bridge-only 0x0002/0xff candidate class is grounded save-side, the upstream chooser above the child attach path is grounded as paired DT/ST siblings at 0x004a2c80 and 0x004a34e0 with decoded Bridge/Tunnel/BallastCap/Overpass families, grounded top-level branch meaning, grounded bridge/tunnel material selector roles, a concrete child-construction/write-side chain through 0x00490960, 0x00491c60, 0x0048a6c0, 0x00455a40, and 0x004559d0, and stable 0x00490960 mode families for BallastCap, TrackCap, Overpass, Tunnel, and Bridge branches. The current save-side name corpus already maps BallastCap, TrackCap, Tunnel, and Bridge rows onto those families directly, the candidate-pattern correlation narrows the dominant mixed 0x0001/0xff class to bridge:62 / track_cap:21 / tunnel:19, and the profile-span/prefix correlation now pins the remaining outliers down to exact residual prefix patterns: 0x03 is a 0x000055f3/0x0001/0xff tunnel-dominant class with a tiny track-cap residue, while 0x27 splits between 0xff0000ff/0x0001/0xff tunnel residue and 0xff0000ff/0x0002/0xff bridge residue.".to_string(),
         if st_only_name_pair_corpus {
             "The current save-side side-buffer corpus is ST-only, so this trace directly exercises the ST chooser sibling while the DT sibling remains grounded statically but unexercised in this save.".to_string()
         } else {
@@ -13537,10 +13635,12 @@ fn parse_save_placed_structure_dynamic_side_buffer_probe(
             )
             .take(8)
             .collect::<Vec<_>>();
-        let mut name_prelude_profile_span_groups =
+        let mut name_prelude_profile_span_groups = BTreeMap::<
-            BTreeMap::<usize, Vec<(u16, u8, Option<String>, Option<String>)>>::new();
+            usize,
+            Vec<(usize, Option<String>, Option<String>, u32, u16, u8, u16, u8)>,
+        >::new();
         for (
-            _,
+            name_tag_relative_offset,
             primary_name,
             secondary_name,
             child_count_candidate,
@@ -13549,14 +13649,26 @@ fn parse_save_placed_structure_dynamic_side_buffer_probe(
         ) in &name_prelude_candidate_rows
         {
             if let Some(previous_span) = previous_span {
+                let Some(row) = embedded_name_rows.get(
+                    payload_envelope_rows
+                        .iter()
+                        .position(|row| row.name_tag_relative_offset == *name_tag_relative_offset)
+                        .unwrap_or(usize::MAX),
+                ) else {
+                    continue;
+                };
                 name_prelude_profile_span_groups
                     .entry(*previous_span)
                     .or_default()
                     .push((
-                        *child_count_candidate,
+                        *name_tag_relative_offset,
-                        *saved_primary_child_byte_candidate,
                         primary_name.clone(),
                         secondary_name.clone(),
+                        row.prefix_leading_dword,
+                        row.prefix_trailing_word,
+                        row.prefix_separator_byte,
+                        *child_count_candidate,
+                        *saved_primary_child_byte_candidate,
                     ));
             }
         }
@@ -13567,11 +13679,16 @@ fn parse_save_placed_structure_dynamic_side_buffer_probe(
                 let mut child_count_counts = BTreeMap::<u16, usize>::new();
                 let mut saved_primary_counts = BTreeMap::<u8, usize>::new();
                 let mut mode_family_counts = BTreeMap::<String, usize>::new();
+                let mut prefix_counts = BTreeMap::<(u32, u16, u8), usize>::new();
                 for (
-                    child_count_candidate,
+                    _name_tag_relative_offset,
-                    saved_primary_child_byte_candidate,
                     primary_name,
                     secondary_name,
+                    prefix_leading_dword,
+                    prefix_trailing_word,
+                    prefix_separator_byte,
+                    child_count_candidate,
+                    saved_primary_child_byte_candidate,
                 ) in &rows
                 {
                     *pattern_counts
@@ -13583,6 +13700,13 @@ fn parse_save_placed_structure_dynamic_side_buffer_probe(
                     *saved_primary_counts
                         .entry(*saved_primary_child_byte_candidate)
                         .or_default() += 1;
+                    *prefix_counts
+                        .entry((
+                            *prefix_leading_dword,
+                            *prefix_trailing_word,
+                            *prefix_separator_byte,
+                        ))
+                        .or_default() += 1;
                     *mode_family_counts
                         .entry(
                             classify_side_buffer_mode_family(
@@ -13668,6 +13792,79 @@ fn parse_save_placed_structure_dynamic_side_buffer_probe(
                             }
                         })
                         .collect(),
+                    compact_prefix_pattern_summaries: prefix_counts
+                        .into_iter()
+                        .map(
+                            |(
+                                (prefix_leading_dword, prefix_trailing_word, prefix_separator_byte),
+                                count,
+                            )| {
+                                SmpSavePlacedStructureDynamicSideBufferNamePreludeProfileSpanPrefixSummary {
+                                    prefix_leading_dword,
+                                    prefix_leading_dword_hex: format!(
+                                        "0x{prefix_leading_dword:08x}"
+                                    ),
+                                    prefix_trailing_word,
+                                    prefix_trailing_word_hex: format!(
+                                        "0x{prefix_trailing_word:04x}"
+                                    ),
+                                    prefix_separator_byte,
+                                    prefix_separator_byte_hex: format!(
+                                        "0x{prefix_separator_byte:02x}"
+                                    ),
+                                    count,
+                                }
+                            },
+                        )
+                        .collect(),
+                    sample_rows: rows
+                        .iter()
+                        .take(8)
+                        .enumerate()
+                        .map(
+                            |(
+                                sample_index,
+                                (
+                                    name_tag_relative_offset,
+                                    primary_name,
+                                    secondary_name,
+                                    prefix_leading_dword,
+                                    prefix_trailing_word,
+                                    prefix_separator_byte,
+                                    child_count_candidate,
+                                    saved_primary_child_byte_candidate,
+                                ),
+                            )| {
+                                SmpSavePlacedStructureDynamicSideBufferNamePreludeProfileSpanSample {
+                                    sample_index,
+                                    name_tag_relative_offset: *name_tag_relative_offset,
+                                    primary_name: primary_name.clone(),
+                                    secondary_name: secondary_name.clone(),
+                                    prefix_leading_dword: *prefix_leading_dword,
+                                    prefix_leading_dword_hex: format!(
+                                        "0x{prefix_leading_dword:08x}"
+                                    ),
+                                    prefix_trailing_word: *prefix_trailing_word,
+                                    prefix_trailing_word_hex: format!(
+                                        "0x{prefix_trailing_word:04x}"
+                                    ),
+                                    prefix_separator_byte: *prefix_separator_byte,
+                                    prefix_separator_byte_hex: format!(
+                                        "0x{prefix_separator_byte:02x}"
+                                    ),
+                                    child_count_candidate: *child_count_candidate,
+                                    child_count_candidate_hex: format!(
+                                        "0x{child_count_candidate:04x}"
+                                    ),
+                                    saved_primary_child_byte_candidate:
+                                        *saved_primary_child_byte_candidate,
+                                    saved_primary_child_byte_candidate_hex: format!(
+                                        "0x{saved_primary_child_byte_candidate:02x}"
+                                    ),
+                                }
+                            },
+                        )
+                        .collect(),
                 }
             })
             .take(8)
@@ -24563,6 +24760,35 @@ mod tests {
                                                 count: 15,
                                             },
                                         ],
+                                        compact_prefix_pattern_summaries: vec![
+                                            SmpSavePlacedStructureDynamicSideBufferNamePreludeProfileSpanPrefixSummary {
+                                                prefix_leading_dword: 0x0000_55f3,
+                                                prefix_leading_dword_hex: "0x000055f3".to_string(),
+                                                prefix_trailing_word: 0x0001,
+                                                prefix_trailing_word_hex: "0x0001".to_string(),
+                                                prefix_separator_byte: 0xff,
+                                                prefix_separator_byte_hex: "0xff".to_string(),
+                                                count: 17,
+                                            },
+                                        ],
+                                        sample_rows: vec![
+                                            SmpSavePlacedStructureDynamicSideBufferNamePreludeProfileSpanSample {
+                                                sample_index: 0,
+                                                name_tag_relative_offset: 1200,
+                                                primary_name: Some("TunnelSTBrick_Section.3dp".to_string()),
+                                                secondary_name: Some("Infrastructure".to_string()),
+                                                prefix_leading_dword: 0x0000_55f3,
+                                                prefix_leading_dword_hex: "0x000055f3".to_string(),
+                                                prefix_trailing_word: 0x0001,
+                                                prefix_trailing_word_hex: "0x0001".to_string(),
+                                                prefix_separator_byte: 0xff,
+                                                prefix_separator_byte_hex: "0xff".to_string(),
+                                                child_count_candidate: 1,
+                                                child_count_candidate_hex: "0x0001".to_string(),
+                                                saved_primary_child_byte_candidate: 0xff,
+                                                saved_primary_child_byte_candidate_hex: "0xff".to_string(),
+                                            },
+                                        ],
                                     },
                                     SmpSavePlacedStructureDynamicSideBufferNamePreludeProfileSpanCorrelation {
                                         previous_profile_chunk_len_to_next_name_or_end: 6,
@@ -24592,6 +24818,18 @@ mod tests {
                                                 count: 72,
                                             },
                                         ],
+                                        compact_prefix_pattern_summaries: vec![
+                                            SmpSavePlacedStructureDynamicSideBufferNamePreludeProfileSpanPrefixSummary {
+                                                prefix_leading_dword: 0xff00_0000,
+                                                prefix_leading_dword_hex: "0xff000000".to_string(),
+                                                prefix_trailing_word: 0x0001,
+                                                prefix_trailing_word_hex: "0x0001".to_string(),
+                                                prefix_separator_byte: 0xff,
+                                                prefix_separator_byte_hex: "0xff".to_string(),
+                                                count: 72,
+                                            },
+                                        ],
+                                        sample_rows: vec![],
                                     },
                                     SmpSavePlacedStructureDynamicSideBufferNamePreludeProfileSpanCorrelation {
                                         previous_profile_chunk_len_to_next_name_or_end: 0x27,
@@ -24625,6 +24863,60 @@ mod tests {
                                                 count: 1,
                                             },
                                         ],
+                                        compact_prefix_pattern_summaries: vec![
+                                            SmpSavePlacedStructureDynamicSideBufferNamePreludeProfileSpanPrefixSummary {
+                                                prefix_leading_dword: 0xff00_00ff,
+                                                prefix_leading_dword_hex: "0xff0000ff".to_string(),
+                                                prefix_trailing_word: 0x0001,
+                                                prefix_trailing_word_hex: "0x0001".to_string(),
+                                                prefix_separator_byte: 0xff,
+                                                prefix_separator_byte_hex: "0xff".to_string(),
+                                                count: 1,
+                                            },
+                                            SmpSavePlacedStructureDynamicSideBufferNamePreludeProfileSpanPrefixSummary {
+                                                prefix_leading_dword: 0xff00_00ff,
+                                                prefix_leading_dword_hex: "0xff0000ff".to_string(),
+                                                prefix_trailing_word: 0x0002,
+                                                prefix_trailing_word_hex: "0x0002".to_string(),
+                                                prefix_separator_byte: 0xff,
+                                                prefix_separator_byte_hex: "0xff".to_string(),
+                                                count: 2,
+                                            },
+                                        ],
+                                        sample_rows: vec![
+                                            SmpSavePlacedStructureDynamicSideBufferNamePreludeProfileSpanSample {
+                                                sample_index: 0,
+                                                name_tag_relative_offset: 2805,
+                                                primary_name: Some("TunnelSTBrick_Section.3dp".to_string()),
+                                                secondary_name: Some("Infrastructure".to_string()),
+                                                prefix_leading_dword: 0xff00_00ff,
+                                                prefix_leading_dword_hex: "0xff0000ff".to_string(),
+                                                prefix_trailing_word: 0x0001,
+                                                prefix_trailing_word_hex: "0x0001".to_string(),
+                                                prefix_separator_byte: 0xff,
+                                                prefix_separator_byte_hex: "0xff".to_string(),
+                                                child_count_candidate: 1,
+                                                child_count_candidate_hex: "0x0001".to_string(),
+                                                saved_primary_child_byte_candidate: 0xff,
+                                                saved_primary_child_byte_candidate_hex: "0xff".to_string(),
+                                            },
+                                            SmpSavePlacedStructureDynamicSideBufferNamePreludeProfileSpanSample {
+                                                sample_index: 1,
+                                                name_tag_relative_offset: 3764,
+                                                primary_name: Some("BridgeSTWood_Section.3dp".to_string()),
+                                                secondary_name: Some("Infrastructure".to_string()),
+                                                prefix_leading_dword: 0xff00_00ff,
+                                                prefix_leading_dword_hex: "0xff0000ff".to_string(),
+                                                prefix_trailing_word: 0x0002,
+                                                prefix_trailing_word_hex: "0x0002".to_string(),
+                                                prefix_separator_byte: 0xff,
+                                                prefix_separator_byte_hex: "0xff".to_string(),
+                                                child_count_candidate: 1,
+                                                child_count_candidate_hex: "0x0001".to_string(),
+                                                saved_primary_child_byte_candidate: 0xff,
+                                                saved_primary_child_byte_candidate_hex: "0xff".to_string(),
+                                            },
+                                        ],
                                     },
                                 ],
                                 sample_rows: Vec::new(),
@@ -24788,11 +25080,11 @@ mod tests {
             trace.candidate_consumer_hypotheses[0]
                 .blockers
                 .iter()
-                .any(|line| line.contains("compact-prefix regimes subdivide")
+                .any(|line| line.contains("exact residual compact-prefix patterns")
-                    && line.contains("bridge:62 / track_cap:21 / tunnel:19")
+                    && line.contains("span 0x03")
-                    && line.contains("0x06 -> pure bridge")
+                    && line.contains("0x000055f3/0x0001/0xff")
-                    && line.contains("0x03 -> mostly tunnel")
+                    && line.contains("span 0x27")
-                    && line.contains("0x27"))
+                    && line.contains("0xff0000ff/0x0002/0xff"))
         );
         assert!(
             trace.candidate_consumer_hypotheses[0]
@@ -24804,6 +25096,25 @@ mod tests {
                     && line.contains("span=0x3 rows=17")
                     && line.contains("span=0x27 rows=3"))
         );
+        assert!(
+            trace.candidate_consumer_hypotheses[0]
+                .evidence
+                .iter()
+                .any(|line| line.contains("short 0x03-byte post-profile gaps")
+                    && line.contains("track_cap:2")
+                    && line.contains("tunnel:15")
+                    && line.contains("0x000055f3/0x0001/0xff:17"))
+        );
+        assert!(
+            trace.candidate_consumer_hypotheses[0]
+                .evidence
+                .iter()
+                .any(|line| line.contains("sparse 0x27 post-profile outlier")
+                    && line.contains("0xff0000ff/0x0001/0xff:1")
+                    && line.contains("0xff0000ff/0x0002/0xff:2")
+                    && line.contains("TunnelSTBrick_Section.3dp")
+                    && line.contains("BridgeSTWood_Section.3dp"))
+        );
         assert!(
             trace.candidate_consumer_hypotheses[0]
                 .evidence