TrainLCD · TinyKitten · Mar 5, 2026 · Mar 5, 2026 · Mar 5, 2026 · Mar 5, 2026
@@ -9,10 +9,51 @@ on:
 
 name: Verify station data integrity
 
+permissions:
+  contents: read
+  pull-requests: write
+  issues: write
+
 jobs:
   verify_migration_data:
     name: Verify pushed migration data
     runs-on: ubuntu-latest
     steps:
       - uses: actions/checkout@v4
-      - run: cargo run --bin data_validator
+
+      - name: Run data validator
+        id: validate
+        run: |
+          if cargo run --bin data_validator; then
+            echo "result=success" >> "$GITHUB_OUTPUT"
+          else
+            echo "result=failure" >> "$GITHUB_OUTPUT"
+          fi
+
+      - name: Find existing comment
+        if: github.event_name == 'pull_request'
+        uses: peter-evans/find-comment@v3
+        id: find_comment
+        with:
+          issue-number: ${{ github.event.pull_request.number }}
+          comment-author: "github-actions[bot]"
+          body-includes: "<!-- data-validator -->"
+
+      - name: Post or update validation failure comment
+        if: github.event_name == 'pull_request' && steps.validate.outputs.result == 'failure'
+        uses: peter-evans/create-or-update-comment@v4
+        with:
+          issue-number: ${{ github.event.pull_request.number }}
+          comment-id: ${{ steps.find_comment.outputs.comment-id }}
+          body-path: /tmp/validation_report.md
+          edit-mode: replace
+
+      - name: Delete comment if validation passed
+        if: github.event_name == 'pull_request' && steps.validate.outputs.result == 'success' && steps.find_comment.outputs.comment-id != ''
+        run: gh api repos/${{ github.repository }}/issues/comments/${{ steps.find_comment.outputs.comment-id }} -X DELETE
+        env:
+          GH_TOKEN: ${{ github.token }}
+
+      - name: Fail job if validation failed
+        if: steps.validate.outputs.result == 'failure'
+        run: exit 1
@@ -1,55 +1,114 @@
 use core::panic;
+use std::collections::HashSet;
 use std::path::Path;
 
 use csv::{ReaderBuilder, StringRecord};
 
 fn main() -> Result<(), Box<dyn std::error::Error>> {
-    let mut has_err = false;
+    let mut invalid_station_ids: Vec<String> = Vec::new();
+    let mut invalid_type_ids: Vec<String> = Vec::new();
 
     let data_path: &Path = Path::new("data");
     let mut rdr = ReaderBuilder::new().from_path(data_path.join("3!stations.csv"))?;
-    let records: Vec<StringRecord> = rdr.records().filter_map(|row| row.ok()).collect();
-    let station_ids: Vec<u32> = records
+    let records: Vec<StringRecord> = rdr.records().collect::<Result<Vec<_>, _>>()?;
+    let station_ids: HashSet<u32> = records
         .iter()
         .map(|row| row.get(0).unwrap().parse::<u32>().unwrap())
         .collect();
 
     let mut rdr = ReaderBuilder::new().from_path(data_path.join("4!types.csv"))?;
-    let records: Vec<StringRecord> = rdr.records().filter_map(|row| row.ok()).collect();
-    let type_ids: Vec<u32> = records
+    let records: Vec<StringRecord> = rdr.records().collect::<Result<Vec<_>, _>>()?;
+    let type_ids: HashSet<u32> = records
         .iter()
         .map(|row| row.get(1).unwrap().parse::<u32>().unwrap())
         .collect();
 
     let mut rdr = ReaderBuilder::new().from_path(data_path.join("5!station_station_types.csv"))?;
-    let records: Vec<StringRecord> = rdr.records().filter_map(|row| row.ok()).collect();
+    let records: Vec<StringRecord> = rdr.records().collect::<Result<Vec<_>, _>>()?;
 
-    if let Some(invalid_record) = records
-        .iter()
-        .find(|row| !station_ids.contains(&row.get(1).unwrap().parse::<u32>().unwrap()))
-    {
-        println!(
-            "[INVALID] Unrecognized Station ID {:?} Found!",
-            invalid_record.get(1).unwrap()
-        );
-        has_err = true;
-    }
+    for record in &records {
+        let line = || record.iter().collect::<Vec<&str>>().join(",");
 
-    if let Some(invalid_record) = records
-        .iter()
-        .find(|row| !type_ids.contains(&row.get(2).unwrap().parse::<u32>().unwrap()))
-    {
-        println!(
-            "[INVALID] Unrecognized Type ID {:?} Found!",
-            invalid_record.get(2).unwrap()
-        );
-        has_err = true;
+        let station_cd: u32 = match record.get(1).and_then(|v| v.parse().ok()) {
+            Some(id) => id,
+            None => {
+                println!("[INVALID] Failed to parse station_cd from row: {}", line());
+                invalid_station_ids.push(line());
+                continue;
+            }
+        };
+        let type_cd: u32 = match record.get(2).and_then(|v| v.parse().ok()) {
+            Some(id) => id,
+            None => {
+                println!("[INVALID] Failed to parse type_cd from row: {}", line());
+                invalid_type_ids.push(line());
+                continue;
+            }
+        };
+
+        if !station_ids.contains(&station_cd) {
+            println!("[INVALID] Unrecognized Station ID {:?} Found!", station_cd);
+            invalid_station_ids.push(line());
+        }
+        if !type_ids.contains(&type_cd) {
+            println!("[INVALID] Unrecognized Type ID {:?} Found!", type_cd);
+            invalid_type_ids.push(line());
+        }
     }
 
+    let has_err = !invalid_station_ids.is_empty() || !invalid_type_ids.is_empty();
+
     if has_err {
+        let report = build_markdown_report(&invalid_station_ids, &invalid_type_ids);
+        let report_path =
+            std::env::var("VALIDATION_REPORT_PATH").unwrap_or("/tmp/validation_report.md".into());
+        std::fs::write(&report_path, &report)?;
+        eprintln!("Validation report written to {}", report_path);
         panic!("[FATAL] Verification hasn't been passed!");
     }
 
     println!("[VALID] No errors reported.");
     Ok(())
 }
+
+fn build_markdown_report(invalid_station_ids: &[String], invalid_type_ids: &[String]) -> String {
+    let mut md = String::new();
+
+    md.push_str("<!-- data-validator -->\n");
+    md.push_str("## :x: データ整合性チェックに失敗しました\n\n");
+    md.push_str("`5!station_station_types.csv` に存在しない外部キーへの参照が含まれています。\n\n");
+
+    if !invalid_station_ids.is_empty() {
+        md.push_str(&format!(
+            "### 不正な Station ID ({} 件)\n\n",
+            invalid_station_ids.len()
+        ));
+        md.push_str("`3!stations.csv` に存在しない `station_cd` が参照されています。\n\n");
+        md.push_str("<details>\n<summary>該当レコード一覧</summary>\n\n");
+        md.push_str("| 行データ |\n|---|\n");
+        for line in invalid_station_ids {
+            md.push_str(&format!("| `{}` |\n", escape_markdown_cell(line)));
+        }
+        md.push_str("\n</details>\n\n");
+    }
+
+    if !invalid_type_ids.is_empty() {
+        md.push_str(&format!(
+            "### 不正な Type ID ({} 件)\n\n",
+            invalid_type_ids.len()
+        ));
+        md.push_str("`4!types.csv` に存在しない `type_cd` が参照されています。\n\n");
+        md.push_str("<details>\n<summary>該当レコード一覧</summary>\n\n");
+        md.push_str("| 行データ |\n|---|\n");
+        for line in invalid_type_ids {
+            md.push_str(&format!("| `{}` |\n", escape_markdown_cell(line)));
+        }
+        md.push_str("\n</details>\n\n");
+    }
+
+    md
+}
+
+fn escape_markdown_cell(s: &str) -> String {
+    s.replace('`', "&#96;").replace('|', "&#124;")
+}