feat(inspector): set CSV delimiter, quote, encoding, and line ending with reload (#1943)

Author: datlechin

CHANGELOG.md

@@ -15,6 +15,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - In the CSV editor, insert a column to the left or right, and delete several selected columns at once. Column verbs also appear in the Edit menu, and deleting a column that holds data asks you to confirm first. (#1469)
 - In the CSV editor, split a column into several by a delimiter or regular expression, and merge a column with the one next to it using a separator. Both are single undo steps. (#1469)
 - In the CSV editor, switch the first row between header and data with `Cmd+Shift+H` when auto-detection guesses wrong. Files without a header are no longer saved with a generated header row. (#1469)
+- In the CSV editor, set the delimiter, quote character, encoding, and line ending by hand from Edit > Set CSV Properties…, then Reload to re-read the file with those settings. (#1469)
 - Add llama.cpp and MLX as local AI providers. Each preset points at the server's default local endpoint and needs no API key, alongside the existing Ollama and custom OpenAI-compatible options. (#1777)
 
 ### Fixed

Plugins/CSVInspectorPlugin/CSVDocument.swift

@@ -2,7 +2,7 @@ import AppKit
 import TableProPluginKit
 import os
 
-public final class CSVDocument: NSDocument, InspectorDocument {
+public final class CSVDocument: NSDocument, CSVConfigurableDocument {
     static let logger = Logger(subsystem: "com.TablePro", category: "CSVInspector")
 
     private static let typeInferenceSampleSize = 200
@@ -356,6 +356,18 @@ public final class CSVDocument: NSDocument, InspectorDocument {
         }
     }
 
+    public var csvDialect: CSVDialect { dialect }
+
+    public func reload(with newDialect: CSVDialect) {
+        let data = store.data
+        dialect = newDialect
+        store = CSVRowStore(data: data, dialect: newDialect)
+        recomputeInferredTypes()
+        undoManager?.removeAllActions()
+        updateChangeCount(.changeDone)
+        onChange?()
+    }
+
     private func recomputeInferredTypes() {
         let sample = store.pageRows(offset: 0, limit: Self.typeInferenceSampleSize)
         inferredTypes = CSVTypeInferrer.inferColumns(rows: sample, columnCount: store.columnCount)

Plugins/TableProPluginKit/DocumentInspectorPlugin.swift

@@ -72,3 +72,9 @@ public extension InspectorDocument {
     func mergeColumns(at index: Int, separator: String) {}
     func toggleHeaderRow() {}
 }
+
+@MainActor
+public protocol CSVConfigurableDocument: InspectorDocument {
+    var csvDialect: CSVDialect { get }
+    func reload(with dialect: CSVDialect)
+}

TablePro/TableProApp.swift

@@ -651,6 +651,11 @@ struct AppMenuCommands: Commands {
             .optionalKeyboardShortcut(shortcut(for: .toggleHeaderRow))
             .disabled(!keyWindowIsInspector)
 
+            Button(String(localized: "Set CSV Properties…")) {
+                NSApp.sendAction(#selector(InspectorViewController.inspectorSetCSVProperties(_:)), to: nil, from: nil)
+            }
+            .disabled(!keyWindowIsInspector)
+
             Divider()
 
             // Table operations (work when tables selected in sidebar)

TablePro/Views/Inspector/CSVPropertiesSheet.swift

@@ -0,0 +1,84 @@
+//
+//  CSVPropertiesSheet.swift
+//  TablePro
+//
+
+import SwiftUI
+import TableProPluginKit
+
+struct CSVPropertiesSheet: View {
+    private let baseDialect: CSVDialect
+    private let onReload: (CSVDialect) -> Void
+    private let onCancel: () -> Void
+
+    @State private var delimiterIndex: Int
+    @State private var quoteIndex: Int
+    @State private var encodingIndex: Int
+    @State private var lineEndingIndex: Int
+
+    init(
+        dialect: CSVDialect,
+        onReload: @escaping (CSVDialect) -> Void,
+        onCancel: @escaping () -> Void
+    ) {
+        self.baseDialect = dialect
+        self.onReload = onReload
+        self.onCancel = onCancel
+        _delimiterIndex = State(initialValue: CSVPropertyOptions.delimiterIndex(for: dialect.delimiter))
+        _quoteIndex = State(initialValue: CSVPropertyOptions.quoteIndex(for: dialect.quoteChar))
+        _encodingIndex = State(initialValue: CSVPropertyOptions.encodingIndex(for: dialect.encoding))
+        _lineEndingIndex = State(initialValue: CSVPropertyOptions.lineEndingIndex(for: dialect.lineEnding))
+    }
+
+    var body: some View {
+        VStack(alignment: .leading, spacing: 16) {
+            Text("CSV Properties").font(.headline)
+            Text("Re-read the file with these settings. This discards unsaved changes.")
+                .font(.subheadline)
+                .foregroundStyle(.secondary)
+
+            Form {
+                Picker("Delimiter", selection: $delimiterIndex) {
+                    ForEach(CSVPropertyOptions.delimiters.indices, id: \.self) { index in
+                        Text(CSVPropertyOptions.delimiters[index].label).tag(index)
+                    }
+                }
+                Picker("Quote character", selection: $quoteIndex) {
+                    ForEach(CSVPropertyOptions.quotes.indices, id: \.self) { index in
+                        Text(CSVPropertyOptions.quotes[index].label).tag(index)
+                    }
+                }
+                Picker("Encoding", selection: $encodingIndex) {
+                    ForEach(CSVPropertyOptions.encodings.indices, id: \.self) { index in
+                        Text(CSVPropertyOptions.encodings[index].label).tag(index)
+                    }
+                }
+                Picker("Line ending", selection: $lineEndingIndex) {
+                    ForEach(CSVPropertyOptions.lineEndings.indices, id: \.self) { index in
+                        Text(CSVPropertyOptions.lineEndings[index].label).tag(index)
+                    }
+                }
+            }
+
+            HStack {
+                Spacer()
+                Button("Cancel", role: .cancel, action: onCancel)
+                    .keyboardShortcut(.cancelAction)
+                Button("Reload") { onReload(selectedDialect) }
+                    .keyboardShortcut(.defaultAction)
+            }
+        }
+        .padding(20)
+        .frame(width: 340)
+    }
+
+    private var selectedDialect: CSVDialect {
+        CSVPropertyOptions.dialect(
+            base: baseDialect,
+            delimiterIndex: delimiterIndex,
+            quoteIndex: quoteIndex,
+            encodingIndex: encodingIndex,
+            lineEndingIndex: lineEndingIndex
+        )
+    }
+}

TablePro/Views/Inspector/CSVPropertyOptions.swift

@@ -0,0 +1,69 @@
+//
+//  CSVPropertyOptions.swift
+//  TablePro
+//
+
+import Foundation
+import TableProPluginKit
+
+enum CSVPropertyOptions {
+    static let delimiters: [(label: String, byte: UInt8)] = [
+        (String(localized: "Comma  ,"), 0x2C),
+        (String(localized: "Semicolon  ;"), 0x3B),
+        (String(localized: "Tab"), 0x09),
+        (String(localized: "Pipe  |"), 0x7C),
+        (String(localized: "Colon  :"), 0x3A),
+        (String(localized: "Space"), 0x20),
+    ]
+
+    static let quotes: [(label: String, byte: UInt8)] = [
+        (String(localized: "Double Quote  \""), 0x22),
+        (String(localized: "Single Quote  '"), 0x27),
+    ]
+
+    static let encodings: [(label: String, encoding: String.Encoding)] = [
+        ("UTF-8", .utf8),
+        ("UTF-16 LE", .utf16LittleEndian),
+        ("UTF-16 BE", .utf16BigEndian),
+        ("Latin-1", .isoLatin1),
+        ("Windows-1252", .windowsCP1252),
+    ]
+
+    static let lineEndings: [(label: String, value: CSVDialect.LineEnding)] = [
+        ("LF", .lf),
+        ("CRLF", .crlf),
+        ("CR", .cr),
+    ]
+
+    static func delimiterIndex(for byte: UInt8) -> Int {
+        delimiters.firstIndex { $0.byte == byte } ?? 0
+    }
+
+    static func quoteIndex(for byte: UInt8) -> Int {
+        quotes.firstIndex { $0.byte == byte } ?? 0
+    }
+
+    static func encodingIndex(for encoding: String.Encoding) -> Int {
+        encodings.firstIndex { $0.encoding == encoding } ?? 0
+    }
+
+    static func lineEndingIndex(for value: CSVDialect.LineEnding) -> Int {
+        lineEndings.firstIndex { $0.value == value } ?? 0
+    }
+
+    static func dialect(
+        base: CSVDialect,
+        delimiterIndex: Int,
+        quoteIndex: Int,
+        encodingIndex: Int,
+        lineEndingIndex: Int
+    ) -> CSVDialect {
+        CSVDialect(
+            delimiter: delimiters.indices.contains(delimiterIndex) ? delimiters[delimiterIndex].byte : base.delimiter,
+            quoteChar: quotes.indices.contains(quoteIndex) ? quotes[quoteIndex].byte : base.quoteChar,
+            encoding: encodings.indices.contains(encodingIndex) ? encodings[encodingIndex].encoding : base.encoding,
+            lineEnding: lineEndings.indices.contains(lineEndingIndex) ? lineEndings[lineEndingIndex].value : base.lineEnding,
+            hasBom: base.hasBom
+        )
+    }
+}

TablePro/Views/Inspector/InspectorViewController.swift

@@ -25,6 +25,7 @@ final class InspectorViewController: NSViewController, NSUserInterfaceValidation
     private var lastFilterClauses: [FilterClause] = []
     private var lastSortSpecs: [SortSpec] = []
     private var pendingPostRefresh: PostRefreshAction?
+    private var propertiesSheetController: NSViewController?
 
     private enum PostRefreshAction {
         case selectClamped(displayRow: Int)
@@ -221,6 +222,51 @@ final class InspectorViewController: NSViewController, NSUserInterfaceValidation
         inspectorDocument?.toggleHeaderRow()
     }
 
+    @objc func inspectorSetCSVProperties(_ sender: Any?) {
+        guard let configurable = inspectorDocument as? CSVConfigurableDocument else { return }
+        presentCSVProperties(configurable)
+    }
+
+    private func presentCSVProperties(_ document: CSVConfigurableDocument) {
+        guard propertiesSheetController == nil else { return }
+        let sheet = CSVPropertiesSheet(
+            dialect: document.csvDialect,
+            onReload: { [weak self, weak document] dialect in
+                self?.dismissPropertiesSheet()
+                guard let document else { return }
+                DispatchQueue.main.async { self?.reloadCSV(document, with: dialect) }
+            },
+            onCancel: { [weak self] in self?.dismissPropertiesSheet() }
+        )
+        let hosting = NSHostingController(rootView: sheet)
+        propertiesSheetController = hosting
+        presentAsSheet(hosting)
+    }
+
+    private func dismissPropertiesSheet() {
+        guard let controller = propertiesSheetController else { return }
+        dismiss(controller)
+        propertiesSheetController = nil
+    }
+
+    private func reloadCSV(_ document: CSVConfigurableDocument, with dialect: CSVDialect) {
+        guard nsDocument?.isDocumentEdited == true, let window = view.window else {
+            document.reload(with: dialect)
+            return
+        }
+        let alert = NSAlert()
+        alert.messageText = String(localized: "Reload with new properties?")
+        alert.informativeText = String(localized: "This discards your unsaved changes and re-reads the file with the chosen settings.")
+        alert.alertStyle = .warning
+        let reloadButton = alert.addButton(withTitle: String(localized: "Reload"))
+        reloadButton.hasDestructiveAction = true
+        alert.addButton(withTitle: String(localized: "Cancel"))
+        alert.beginSheetModal(for: window) { [weak document] response in
+            guard response == .alertFirstButtonReturn, let document else { return }
+            document.reload(with: dialect)
+        }
+    }
+
     @objc func inspectorInsertRowAbove(_ sender: Any?) {
         performInsertRow(anchoredBy: sender, below: false)
     }
@@ -595,6 +641,8 @@ final class InspectorViewController: NSViewController, NSUserInterfaceValidation
              #selector(inspectorSplitColumn(_:)), #selector(inspectorMergeColumns(_:)),
              #selector(inspectorToggleHeaderRow(_:)):
             return nsDocument != nil
+        case #selector(inspectorSetCSVProperties(_:)):
+            return inspectorDocument is CSVConfigurableDocument
         case #selector(inspectorDeleteColumn(_:)):
             guard nsDocument != nil else { return false }
             if let menuItem = item as? NSMenuItem, menuItem.representedObject is [Int] { return true }

TableProTests/Views/CSVPropertyOptionsTests.swift

@@ -0,0 +1,41 @@
+//
+//  CSVPropertyOptionsTests.swift
+//  TableProTests
+//
+
+@testable import TablePro
+import TableProPluginKit
+import Testing
+
+@Suite("CSVPropertyOptions")
+struct CSVPropertyOptionsTests {
+    @Test("Indices map back from a dialect's bytes and values")
+    func indicesFromDialect() {
+        #expect(CSVPropertyOptions.delimiterIndex(for: 0x3B) == 1)
+        #expect(CSVPropertyOptions.quoteIndex(for: 0x27) == 1)
+        #expect(CSVPropertyOptions.encodingIndex(for: .windowsCP1252) == 4)
+        #expect(CSVPropertyOptions.lineEndingIndex(for: .crlf) == 1)
+    }
+
+    @Test("An unknown delimiter falls back to the first option")
+    func unknownFallsBack() {
+        #expect(CSVPropertyOptions.delimiterIndex(for: 0x5E) == 0)
+    }
+
+    @Test("Building a dialect from indices sets the four properties and keeps the BOM")
+    func dialectRoundTrip() {
+        let base = CSVDialect(delimiter: 0x2C, hasBom: true)
+        let dialect = CSVPropertyOptions.dialect(
+            base: base,
+            delimiterIndex: CSVPropertyOptions.delimiterIndex(for: 0x09),
+            quoteIndex: CSVPropertyOptions.quoteIndex(for: 0x27),
+            encodingIndex: CSVPropertyOptions.encodingIndex(for: .utf16LittleEndian),
+            lineEndingIndex: CSVPropertyOptions.lineEndingIndex(for: .cr)
+        )
+        #expect(dialect.delimiter == 0x09)
+        #expect(dialect.quoteChar == 0x27)
+        #expect(dialect.encoding == .utf16LittleEndian)
+        #expect(dialect.lineEnding == .cr)
+        #expect(dialect.hasBom)
+    }
+}

docs/features/csv-inspector.mdx

@@ -26,6 +26,8 @@ When a file opens, TablePro detects:
 - **Line ending**: CRLF, LF, or CR, whichever appears first in the first 64 KB.
 - **Header row**: row one is treated as headers when at least half of its cells are non-empty and not numbers. Otherwise TablePro generates `Column 1`, `Column 2`, ... and treats every row as data.
 
+When a guess is wrong, choose **Edit > Set CSV Properties…** to pick the delimiter, quote character, encoding, and line ending by hand, then **Reload** to re-read the file with those settings. Reload discards unsaved edits and asks first if you have any.
+
 ## Pagination
 
 Rows load in pages. The page size comes from the Default page size setting in [Settings > Data Grid](/customization/settings), 1,000 rows by default. The status bar shows row and column counts, and previous/next page controls when the file spans more than one page.