Fix text_id to use date+name instead of PDF filename

Early OFAC years use batch PDFs where one document covers many penalty
cases (e.g. 56 rows sharing the same PDF in 2003). Deriving text_id from
the PDF filename caused all rows sharing a document to overwrite each other
in the DB, reducing 1061 rows to 348.

Fix: text_id = yyyyMMdd_{slugified_name}, which is unique per table row.
Also add ofac-scrape-only command for fast table-only scraping without PDF downloads.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

src/OFACScraper/Application.cs

@@ -47,6 +47,34 @@ public class Application
         return 0;
     }
 
+    /// <summary>
+    /// Scrapes all years into the checkpoint DB without downloading PDFs or syncing to S3.
+    /// Used to produce a spreadsheet export quickly.
+    /// </summary>
+    public async Task<int> RunScrapeOnlyAsync(CancellationToken ct = default)
+    {
+        var currentYear = DateTime.UtcNow.Year;
+        _logger.LogInformation("Starting scrape-only {StartYear}–{EndYear}", _options.StartYear, currentYear);
+
+        var total = 0;
+        for (var year = _options.StartYear; year <= currentYear; year++)
+        {
+            if (ct.IsCancellationRequested) break;
+            var records = await _scraper.GetYearRecordsAsync(year, ct);
+            foreach (var record in records)
+            {
+                if (_checkpoint.HasRecord(record.TextId)) continue;
+                _checkpoint.MarkProcessed(record.TextId, record.Date, record.Name,
+                    record.PenaltyTotalUsd, record.DocumentUrl, record.FileName, record.Year);
+                total++;
+            }
+        }
+
+        _logger.LogInformation("Scrape-only complete. {Total} new records. DB total: {DbTotal}",
+            total, _checkpoint.GetTotalCount());
+        return 0;
+    }
+
     /// <summary>
     /// Daily/incremental run: scrapes current year only, exports any new records.
     /// </summary>

src/OFACScraper/Models/OFACRecord.cs

@@ -11,8 +11,14 @@ public record OFACRecord
     public required string YearPageUrl { get; init; }
 
     /// <summary>
-    /// Stable text identifier derived from PDF filename (without extension).
-    /// E.g. "20260317_tradestation" from "20260317_tradestation.pdf"
+    /// Stable text identifier derived from date + name (unique per row).
+    /// Early years share batch PDFs across many rows, so the PDF filename is NOT unique per row.
     /// </summary>
-    public string TextId => Path.GetFileNameWithoutExtension(FileName);
+    public string TextId => $"{Date:yyyyMMdd}_{Slugify(Name)}";
+
+    private static string Slugify(string text) =>
+        new string(text.ToLowerInvariant()
+            .Select(c => char.IsLetterOrDigit(c) ? c : '_')
+            .ToArray())
+            .Trim('_')[..Math.Min(60, text.Length)];
 }

src/OFACScraper/Program.cs

@@ -51,7 +51,8 @@ try
     {
         "ofac-full"        => await app.RunFullAsync(),
         "ofac-daily"       => await app.RunDailyAsync(),
-        _ => throw new ArgumentException($"Unknown command: {command}. Use ofac-full or ofac-daily.")
+        "ofac-scrape-only" => await app.RunScrapeOnlyAsync(),
+        _ => throw new ArgumentException($"Unknown command: {command}. Use ofac-full, ofac-daily, or ofac-scrape-only.")
     };
 
     return exitCode;

src/OFACScraper/Scraper.cs

@@ -75,8 +75,7 @@ public class OFACScraper
             if (dateLink == null) continue; // "Year to date totals" summary row
 
             var dateText = HtmlEntity.DeEntitize(dateLink.InnerText).Trim();
-            if (!DateTime.TryParseExact(dateText, "MM/dd/yyyy", CultureInfo.InvariantCulture,
-                    DateTimeStyles.None, out var date))
+            if (!TryParseDate(dateText, out var date))
             {
                 _logger.LogWarning("Could not parse date '{Date}' in year {Year}", dateText, year);
                 continue;
@@ -88,7 +87,7 @@ public class OFACScraper
             var docUrl = docHref.StartsWith("http") ? docHref : _options.BaseUrl + docHref;
             var fileName = dateLink.GetAttributeValue("title", "").Trim();
             if (string.IsNullOrEmpty(fileName))
-                fileName = $"{date:yyyyMMdd}_{Slugify(HtmlEntity.DeEntitize(cells[1].InnerText).Trim())}.pdf";
+                fileName = $"{date:yyyyMMdd}.pdf";
 
             var name = HtmlEntity.DeEntitize(cells[1].InnerText).Trim();
             var penaltyText = HtmlEntity.DeEntitize(cells[3].InnerText).Trim();
@@ -114,6 +113,20 @@ public class OFACScraper
         return records;
     }
 
+    private static bool TryParseDate(string raw, out DateTime date)
+    {
+        // Strip suffixes: " a", " b", " c", " (Revised ...)", etc.
+        var clean = raw.Trim();
+        var parenIdx = clean.IndexOf('(');
+        if (parenIdx > 0) clean = clean[..parenIdx].Trim();
+        // Strip trailing letter suffix like "05/06/2014 b"
+        if (clean.Length > 0 && char.IsLetter(clean[^1]))
+            clean = clean[..^1].Trim();
+
+        string[] formats = ["MM/dd/yyyy", "M/d/yyyy", "M/dd/yyyy", "MM/d/yyyy"];
+        return DateTime.TryParseExact(clean, formats, CultureInfo.InvariantCulture, DateTimeStyles.None, out date);
+    }
+
     private static decimal? ParsePenalty(string text)
     {
         if (string.IsNullOrWhiteSpace(text)) return null;
@@ -123,9 +136,4 @@ public class OFACScraper
             : null;
     }
 
-    private static string Slugify(string text) =>
-        new string(text.ToLowerInvariant()
-            .Select(c => char.IsLetterOrDigit(c) ? c : '_')
-            .ToArray())
-            .Trim('_');
 }