Files
lmxopcua/src/ZB.MOM.WW.OtOpcUa.Admin/Services/EquipmentCsvImporter.cs
Joseph Doherty 560a961cca Phase 6.4 Stream A + B data layer — UnsImpactAnalyzer + EquipmentCsvImporter (parser)
Ships the pure-logic data layer of Phase 6.4. Blazor UI pieces
(UnsTab drag/drop, CSV import modal, preview table, FinaliseImportBatch txn,
staging tables) are deferred to visual-compliance follow-ups (tasks #153,
#155, #157).

Admin.Services additions:

- UnsImpactAnalyzer.Analyze(snapshot, move) — pure-function, no I/O. Three
  move variants: LineMove, AreaRename, LineMerge. Returns UnsImpactPreview
  with AffectedEquipmentCount + AffectedTagCount + CascadeWarnings +
  RevisionToken + HumanReadableSummary the Admin UI shows in the confirm
  modal. Cross-cluster moves rejected with CrossClusterMoveRejectedException
  per decision #82. Missing source/target throws UnsMoveValidationException.
  Surfaces sibling-line same-name ambiguity as a cascade warning.
- DraftRevisionToken — opaque revision fingerprint. Preview captures the
  token; Confirm compares it. The 409-concurrent-edit UX plumbs through on
  the Razor-page follow-up (task #153). Matches(other) is null-safe.
- UnsTreeSnapshot + UnsAreaSummary + UnsLineSummary — snapshot shape the
  caller hands to the analyzer. Tests build them in-memory without a DB.

- EquipmentCsvImporter.Parse(csvText) — RFC 4180 CSV parser per decision #95.
  Version-marker contract: line 1 must be "# OtOpcUaCsv v1" (future shapes
  bump the version). Required columns from decision #117 + optional columns
  from decision #139. Rejects unknown columns, duplicate column names,
  blank required fields, duplicate ZTags within the file. Quoted-field
  handling supports embedded commas + escaped "" quotes. Returns
  EquipmentCsvParseResult { AcceptedRows, RejectedRows } so the preview
  modal renders accept/reject counts without re-parsing.

Tests (22 new, all pass):

- UnsImpactAnalyzerTests (9): line move counts equipment + tags; cross-
  cluster throws; unknown source/target throws validation; ambiguous same-
  name target raises warning; area rename sums across lines; line merge
  cross-area warns; same-area merge no warning; DraftRevisionToken matches
  semantics.
- EquipmentCsvImporterTests (13): empty file throws; missing version marker;
  missing required column; unknown column; duplicate column; valid single
  row round-trips; optional columns populate when present; blank required
  field rejects row; duplicate ZTag rejects second; RFC 4180 quoted fields
  with commas + escaped quotes; mismatched column count rejects; blank
  lines between rows ignored; required + optional column constants match
  decisions #117 + #139 exactly.

Full solution dotnet test: 1159 passing (Phase 6.3 = 1137, Phase 6.4 A+B
data = +22). Pre-existing Client.CLI Subscribe flake unchanged.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-19 10:09:47 -04:00

260 lines
11 KiB
C#
Raw Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
using System.Globalization;
using System.Text;
namespace ZB.MOM.WW.OtOpcUa.Admin.Services;
/// <summary>
/// RFC 4180 CSV parser for equipment import per decision #95 and Phase 6.4 Stream B.1.
/// Produces a validated <see cref="EquipmentCsvParseResult"/> the caller (CSV import
/// modal + staging tables) consumes. Pure-parser concern — no DB access, no staging
/// writes; those live in the follow-up Stream B.2 work.
/// </summary>
/// <remarks>
/// <para><b>Header contract</b>: line 1 must be exactly <c># OtOpcUaCsv v1</c> (version
/// marker). Line 2 is the column header row. Unknown columns are rejected; required
/// columns must all be present. The version bump handshake lets future shapes parse
/// without ambiguity — v2 files go through a different parser variant.</para>
///
/// <para><b>Required columns</b> per decision #117: ZTag, MachineCode, SAPID,
/// EquipmentId, EquipmentUuid, Name, UnsAreaName, UnsLineName.</para>
///
/// <para><b>Optional columns</b> per decision #139: Manufacturer, Model, SerialNumber,
/// HardwareRevision, SoftwareRevision, YearOfConstruction, AssetLocation,
/// ManufacturerUri, DeviceManualUri.</para>
///
/// <para><b>Row validation</b>: blank required field → rejected; duplicate ZTag within
/// the same file → rejected. Duplicate against the DB isn't detected here — the
/// staged-import finalize step (Stream B.4) catches that.</para>
/// </remarks>
public static class EquipmentCsvImporter
{
public const string VersionMarker = "# OtOpcUaCsv v1";
public static IReadOnlyList<string> RequiredColumns { get; } = new[]
{
"ZTag", "MachineCode", "SAPID", "EquipmentId", "EquipmentUuid",
"Name", "UnsAreaName", "UnsLineName",
};
public static IReadOnlyList<string> OptionalColumns { get; } = new[]
{
"Manufacturer", "Model", "SerialNumber", "HardwareRevision", "SoftwareRevision",
"YearOfConstruction", "AssetLocation", "ManufacturerUri", "DeviceManualUri",
};
public static EquipmentCsvParseResult Parse(string csvText)
{
ArgumentNullException.ThrowIfNull(csvText);
var rows = SplitLines(csvText);
if (rows.Count == 0)
throw new InvalidCsvFormatException("CSV is empty.");
if (!string.Equals(rows[0].Trim(), VersionMarker, StringComparison.Ordinal))
throw new InvalidCsvFormatException(
$"CSV header line 1 must be exactly '{VersionMarker}' — got '{rows[0]}'. " +
"Files without the version marker are rejected so future-format files don't parse ambiguously.");
if (rows.Count < 2)
throw new InvalidCsvFormatException("CSV has no column header row (line 2) or data rows.");
var headerCells = SplitCsvRow(rows[1]);
ValidateHeader(headerCells);
var accepted = new List<EquipmentCsvRow>();
var rejected = new List<EquipmentCsvRowError>();
var ztagsSeen = new HashSet<string>(StringComparer.OrdinalIgnoreCase);
var colIndex = headerCells
.Select((name, idx) => (name, idx))
.ToDictionary(t => t.name, t => t.idx, StringComparer.OrdinalIgnoreCase);
for (var i = 2; i < rows.Count; i++)
{
if (string.IsNullOrWhiteSpace(rows[i])) continue;
try
{
var cells = SplitCsvRow(rows[i]);
if (cells.Length != headerCells.Length)
{
rejected.Add(new EquipmentCsvRowError(
LineNumber: i + 1,
Reason: $"Column count {cells.Length} != header count {headerCells.Length}."));
continue;
}
var row = BuildRow(cells, colIndex);
var missing = RequiredColumns.Where(c => string.IsNullOrWhiteSpace(GetCell(row, c))).ToList();
if (missing.Count > 0)
{
rejected.Add(new EquipmentCsvRowError(i + 1, $"Blank required column(s): {string.Join(", ", missing)}"));
continue;
}
if (!ztagsSeen.Add(row.ZTag))
{
rejected.Add(new EquipmentCsvRowError(i + 1, $"Duplicate ZTag '{row.ZTag}' within file."));
continue;
}
accepted.Add(row);
}
catch (InvalidCsvFormatException ex)
{
rejected.Add(new EquipmentCsvRowError(i + 1, ex.Message));
}
}
return new EquipmentCsvParseResult(accepted, rejected);
}
private static void ValidateHeader(string[] headerCells)
{
var seen = new HashSet<string>(headerCells, StringComparer.OrdinalIgnoreCase);
// Missing required
var missingRequired = RequiredColumns.Where(r => !seen.Contains(r)).ToList();
if (missingRequired.Count > 0)
throw new InvalidCsvFormatException($"Header is missing required column(s): {string.Join(", ", missingRequired)}");
// Unknown columns (not in required optional)
var known = new HashSet<string>(RequiredColumns.Concat(OptionalColumns), StringComparer.OrdinalIgnoreCase);
var unknown = headerCells.Where(c => !known.Contains(c)).ToList();
if (unknown.Count > 0)
throw new InvalidCsvFormatException(
$"Header has unknown column(s): {string.Join(", ", unknown)}. " +
"Bump the version marker to define a new shape before adding columns.");
// Duplicates
var dupe = headerCells.GroupBy(c => c, StringComparer.OrdinalIgnoreCase).FirstOrDefault(g => g.Count() > 1);
if (dupe is not null)
throw new InvalidCsvFormatException($"Header has duplicate column '{dupe.Key}'.");
}
private static EquipmentCsvRow BuildRow(string[] cells, Dictionary<string, int> colIndex) => new()
{
ZTag = cells[colIndex["ZTag"]],
MachineCode = cells[colIndex["MachineCode"]],
SAPID = cells[colIndex["SAPID"]],
EquipmentId = cells[colIndex["EquipmentId"]],
EquipmentUuid = cells[colIndex["EquipmentUuid"]],
Name = cells[colIndex["Name"]],
UnsAreaName = cells[colIndex["UnsAreaName"]],
UnsLineName = cells[colIndex["UnsLineName"]],
Manufacturer = colIndex.TryGetValue("Manufacturer", out var mi) ? cells[mi] : null,
Model = colIndex.TryGetValue("Model", out var moi) ? cells[moi] : null,
SerialNumber = colIndex.TryGetValue("SerialNumber", out var si) ? cells[si] : null,
HardwareRevision = colIndex.TryGetValue("HardwareRevision", out var hi) ? cells[hi] : null,
SoftwareRevision = colIndex.TryGetValue("SoftwareRevision", out var swi) ? cells[swi] : null,
YearOfConstruction = colIndex.TryGetValue("YearOfConstruction", out var yi) ? cells[yi] : null,
AssetLocation = colIndex.TryGetValue("AssetLocation", out var ai) ? cells[ai] : null,
ManufacturerUri = colIndex.TryGetValue("ManufacturerUri", out var mui) ? cells[mui] : null,
DeviceManualUri = colIndex.TryGetValue("DeviceManualUri", out var dui) ? cells[dui] : null,
};
private static string GetCell(EquipmentCsvRow row, string colName) => colName switch
{
"ZTag" => row.ZTag,
"MachineCode" => row.MachineCode,
"SAPID" => row.SAPID,
"EquipmentId" => row.EquipmentId,
"EquipmentUuid" => row.EquipmentUuid,
"Name" => row.Name,
"UnsAreaName" => row.UnsAreaName,
"UnsLineName" => row.UnsLineName,
_ => string.Empty,
};
/// <summary>Split the raw text on line boundaries. Handles \r\n + \n + \r.</summary>
private static List<string> SplitLines(string csv) =>
csv.Split(["\r\n", "\n", "\r"], StringSplitOptions.None).ToList();
/// <summary>Split one CSV row with RFC 4180 quoted-field handling.</summary>
private static string[] SplitCsvRow(string row)
{
var cells = new List<string>();
var sb = new StringBuilder();
var inQuotes = false;
for (var i = 0; i < row.Length; i++)
{
var ch = row[i];
if (inQuotes)
{
if (ch == '"')
{
// Escaped quote "" inside quoted field.
if (i + 1 < row.Length && row[i + 1] == '"')
{
sb.Append('"');
i++;
}
else
{
inQuotes = false;
}
}
else
{
sb.Append(ch);
}
}
else
{
if (ch == ',')
{
cells.Add(sb.ToString());
sb.Clear();
}
else if (ch == '"' && sb.Length == 0)
{
inQuotes = true;
}
else
{
sb.Append(ch);
}
}
}
cells.Add(sb.ToString());
return cells.ToArray();
}
}
/// <summary>One parsed equipment row with required + optional fields.</summary>
public sealed class EquipmentCsvRow
{
// Required (decision #117)
public required string ZTag { get; init; }
public required string MachineCode { get; init; }
public required string SAPID { get; init; }
public required string EquipmentId { get; init; }
public required string EquipmentUuid { get; init; }
public required string Name { get; init; }
public required string UnsAreaName { get; init; }
public required string UnsLineName { get; init; }
// Optional (decision #139 — OPC 40010 Identification fields)
public string? Manufacturer { get; init; }
public string? Model { get; init; }
public string? SerialNumber { get; init; }
public string? HardwareRevision { get; init; }
public string? SoftwareRevision { get; init; }
public string? YearOfConstruction { get; init; }
public string? AssetLocation { get; init; }
public string? ManufacturerUri { get; init; }
public string? DeviceManualUri { get; init; }
}
/// <summary>One row-level rejection captured by the parser. Line-number is 1-based in the source file.</summary>
public sealed record EquipmentCsvRowError(int LineNumber, string Reason);
/// <summary>Parser output — accepted rows land in staging; rejected rows surface in the preview modal.</summary>
public sealed record EquipmentCsvParseResult(
IReadOnlyList<EquipmentCsvRow> AcceptedRows,
IReadOnlyList<EquipmentCsvRowError> RejectedRows);
/// <summary>Thrown for file-level format problems (missing version marker, bad header, etc.).</summary>
public sealed class InvalidCsvFormatException(string message) : Exception(message);