feat(parser): ✨ Refactor HTML table parsing for improved category and severity handling

* Enhanced handling of `data-category` and `data-severity` attributes.
* Simplified header generation logic for tables without explicit headers.
* Improved error handling and code readability in `IsRunningAsRoot` method.

Author: PrzemyslawKlys

Sources/HtmlTinkerX/HtmlParserFromTable.cs

@@ -247,21 +247,25 @@ internal static (HtmlTableMetadata Metadata, IElement[] Rows, int StartIndex) Re
                 dict[string.IsNullOrEmpty(header) ? i.ToString() : header] = rowValues[i];
             }
             // Override Category/Severity from row-level data attributes when present.
-            if (!string.IsNullOrEmpty(row.GetAttribute("data-category"))) {
-                string key = row.GetAttribute("data-category")!;
-                string val = key;
-                if (dataValueLookup != null && dataValueLookup.TryGetValue(key, out var label)) {
-                    val = label;
+            if (categoryIndex >= 0) {
+                string key = (row.GetAttribute("data-category") ?? string.Empty).Trim();
+                if (key.Length > 0) {
+                    string val = key;
+                    if (dataValueLookup != null && dataValueLookup.TryGetValue(key, out var label)) {
+                        val = label;
+                    }
+                    dict["Category"] = val;
                 }
-                dict["Category"] = val;
             }
-            if (!string.IsNullOrEmpty(row.GetAttribute("data-severity"))) {
-                string key = row.GetAttribute("data-severity")!;
-                string val = key;
-                if (dataValueLookup != null && dataValueLookup.TryGetValue(key, out var label)) {
-                    val = label;
+            if (severityIndex >= 0) {
+                string key = (row.GetAttribute("data-severity") ?? string.Empty).Trim();
+                if (key.Length > 0) {
+                    string val = key;
+                    if (dataValueLookup != null && dataValueLookup.TryGetValue(key, out var label)) {
+                        val = label;
+                    }
+                    dict["Severity"] = val;
                 }
-                dict["Severity"] = val;
             }
             // Second chance: fill Category/Severity from data-* even if a non-empty placeholder was present.
             if (dict.TryGetValue("Category", out var catVal) && string.IsNullOrWhiteSpace(catVal)) {
@@ -338,45 +342,46 @@ internal static (HtmlTableMetadata Metadata, IElement[] Rows, int StartIndex) Re
                         headers.Add(header);
                     }
                 }
-            } else {
-                for (int i = 0; i < headerCells.Length; i++) {
-                    headers.Add($"Column{i + 1}");
-                }
-            }
-
+            } else {
+                int columnCount = SumColSpans(headerCells);
+                for (int i = 0; i < columnCount; i++) {
+                    headers.Add($"Column{i + 1}");
+                }
+            }
+
             int startIndex = hasHeader ? headerRowIndex + 1 : 0;
             List<Dictionary<string, string?>> tableRows = new();
             Dictionary<int, (string? Value, int Remaining)> rowSpans = new();
             int categoryIndex = headers.FindIndex(h => h.Equals("Category", StringComparison.OrdinalIgnoreCase));
             int severityIndex = headers.FindIndex(h => h.Equals("Severity", StringComparison.OrdinalIgnoreCase));
             foreach (var row in rows.Skip(startIndex)) {
-                if (row == null) {
-                    continue;
-                }
-                var cells = row.QuerySelectorAll("th,td");
-                string?[] rowValues = new string?[headers.Count];
-                int col = 0;
-                int cellIndex = 0;
-                while (col < headers.Count) {
-                    if (rowSpans.TryGetValue(col, out var span)) {
-                        rowValues[col] = span.Value;
-                        if (--span.Remaining == 0) {
-                            rowSpans.Remove(col);
-                        } else {
-                            rowSpans[col] = span;
-                        }
-                        col++;
-                        continue;
-                    }
-
-                    if (cellIndex < cells.Length) {
+                if (row == null) {
+                    continue;
+                }
+                var cells = row.QuerySelectorAll("th,td");
+                string?[] rowValues = new string?[headers.Count];
+                int col = 0;
+                int cellIndex = 0;
+                while (col < headers.Count) {
+                    if (rowSpans.TryGetValue(col, out var span)) {
+                        rowValues[col] = span.Value;
+                        if (--span.Remaining == 0) {
+                            rowSpans.Remove(col);
+                        } else {
+                            rowSpans[col] = span;
+                        }
+                        col++;
+                        continue;
+                    }
+
+                    if (cellIndex < cells.Length) {
                         var cell = cells[cellIndex++];
                         string value = FormatCellText(cell, HtmlCellTextFormat.Compact);
-                        if (replaceContent != null) {
-                            foreach (var kv in replaceContent) {
+                        if (replaceContent != null) {
+                            foreach (var kv in replaceContent) {
                                 value = ReplaceCaseInsensitive(value, kv.Key, kv.Value);
-                            }
-                        }
+                            }
+                        }
                         int colspan = 1;
                         int rowspan = 1;
                         if (int.TryParse(cell.GetAttribute("colspan"), NumberStyles.Integer, CultureInfo.InvariantCulture, out int cs2)) {
@@ -385,61 +390,61 @@ internal static (HtmlTableMetadata Metadata, IElement[] Rows, int StartIndex) Re
                         if (int.TryParse(cell.GetAttribute("rowspan"), NumberStyles.Integer, CultureInfo.InvariantCulture, out int rs2)) {
                             rowspan = rs2;
                         }
-                        for (int c = 0; c < colspan && col < headers.Count; c++, col++) {
-                            rowValues[col] = value;
-                            if (rowspan > 1) {
-                                rowSpans[col] = (value, rowspan - 1);
-                            }
-                        }
-                    } else {
-                        if (allProperties) {
-                            rowValues[col] = null;
-                        }
-                        col++;
-                    }
-                }
-
-            Dictionary<string, string?> dict = new();
-            for (int i = 0; i < headers.Count; i++) {
-                string header = headers[i];
-                dict[string.IsNullOrEmpty(header) ? i.ToString() : header] = rowValues[i];
-            }
-            if (categoryIndex >= 0) {
-                string key = (row.GetAttribute("data-category") ?? string.Empty).Trim();
-                if (key.Length > 0) {
-                    string val = key;
-                    if (dataValueLookup != null && dataValueLookup.TryGetValue(key, out var label)) {
-                        val = label;
+                        for (int c = 0; c < colspan && col < headers.Count; c++, col++) {
+                            rowValues[col] = value;
+                            if (rowspan > 1) {
+                                rowSpans[col] = (value, rowspan - 1);
+                            }
+                        }
+                    } else {
+                        if (allProperties) {
+                            rowValues[col] = null;
+                        }
+                        col++;
                     }
-                    dict["Category"] = val;
                 }
-            }
-            if (severityIndex >= 0) {
-                string key = (row.GetAttribute("data-severity") ?? string.Empty).Trim();
-                if (key.Length > 0) {
-                    string val = key;
-                    if (dataValueLookup != null && dataValueLookup.TryGetValue(key, out var label)) {
-                        val = label;
+
+                Dictionary<string, string?> dict = new();
+                for (int i = 0; i < headers.Count; i++) {
+                    string header = headers[i];
+                    dict[string.IsNullOrEmpty(header) ? i.ToString() : header] = rowValues[i];
+                }
+                if (categoryIndex >= 0) {
+                    string key = (row.GetAttribute("data-category") ?? string.Empty).Trim();
+                    if (key.Length > 0) {
+                        string val = key;
+                        if (dataValueLookup.TryGetValue(key, out var label)) {
+                            val = label;
+                        }
+                        dict["Category"] = val;
                     }
-                    dict["Severity"] = val;
+                }
+                if (severityIndex >= 0) {
+                    string key = (row.GetAttribute("data-severity") ?? string.Empty).Trim();
+                    if (key.Length > 0) {
+                        string val = key;
+                        if (dataValueLookup.TryGetValue(key, out var label)) {
+                            val = label;
+                        }
+                        dict["Severity"] = val;
+                    }
+                }
+                if (dict.TryGetValue("Category", out var catVal) && string.IsNullOrWhiteSpace(catVal)) {
+                    dict["Category"] = FillFromDataAttributes("Category", catVal, row, dataValueLookup);
+                }
+                if (dict.TryGetValue("Severity", out var sevVal) && string.IsNullOrWhiteSpace(sevVal)) {
+                    dict["Severity"] = FillFromDataAttributes("Severity", sevVal, row, dataValueLookup);
+                }
+                if (dict.Count > 0) {
+                    tableRows.Add(dict);
                 }
             }
-            if (dict.TryGetValue("Category", out var catVal) && string.IsNullOrWhiteSpace(catVal)) {
-                dict["Category"] = FillFromDataAttributes("Category", catVal, row, dataValueLookup);
-            }
-            if (dict.TryGetValue("Severity", out var sevVal) && string.IsNullOrWhiteSpace(sevVal)) {
-                dict["Severity"] = FillFromDataAttributes("Severity", sevVal, row, dataValueLookup);
-            }
-            if (dict.Count > 0) {
-                tableRows.Add(dict);
+
+            if (tableRows.Count > 0) {
+                result.Add(tableRows);
             }
         }
 
-            if (tableRows.Count > 0) {
-                result.Add(tableRows);
-            }
-        }
-
         return result;
     }
 
@@ -802,13 +807,11 @@ public static List<HtmlTableResult> ParseTablesWithHtmlAgilityPackDetailed(
             if (rows == null || rows.Count == 0) {
                 continue;
             }
-            var dataValueLookup = BuildDataValueLookup(table);
-
             if (reverseTable) {
                 Dictionary<string, string?> obj = new();
-                int index = 0;
-                foreach (var row in rows) {
-                    if (row == null) {
+                int index = 0;
+                foreach (var row in rows) {
+                    if (row == null) {
                         continue;
                     }
                     var cells = row.SelectNodes("th|td");
@@ -835,14 +838,16 @@ public static List<HtmlTableResult> ParseTablesWithHtmlAgilityPackDetailed(
 
                 if (obj.Count > 0) {
                     result.Add(new List<Dictionary<string, string?>> { obj });
-                }
-                continue;
-            }
-
-            int headerRowIndex = 0;
-            bool hasHeader = false;
-            for (int i = 0; i < rows.Count; i++) {
-                if (rows[i].SelectNodes("th")?.Count > 0) {
+                }
+                continue;
+            }
+
+            var dataValueLookup = BuildDataValueLookup(table);
+
+            int headerRowIndex = 0;
+            bool hasHeader = false;
+            for (int i = 0; i < rows.Count; i++) {
+                if (rows[i].SelectNodes("th")?.Count > 0) {
                     headerRowIndex = i;
                     hasHeader = true;
                     break;
@@ -871,13 +876,14 @@ public static List<HtmlTableResult> ParseTablesWithHtmlAgilityPackDetailed(
                     }
                     for (int c = 0; c < colspan; c++) {
                         headers.Add(header);
-                    }
-                }
-            } else {
-                for (int i = 0; i < headerCells.Count; i++) {
-                    headers.Add($"Column{i + 1}");
-                }
-            }
+                    }
+                }
+            } else {
+                int columnCount = SumColSpans(headerCells);
+                for (int i = 0; i < columnCount; i++) {
+                    headers.Add($"Column{i + 1}");
+                }
+            }
 
             int startIndex = hasHeader ? headerRowIndex + 1 : 0;
             List<Dictionary<string, string?>> tableRows = new();
@@ -921,15 +927,15 @@ public static List<HtmlTableResult> ParseTablesWithHtmlAgilityPackDetailed(
                         if (int.TryParse(cell.GetAttributeValue("rowspan", "1"), NumberStyles.Integer, CultureInfo.InvariantCulture, out int rs2)) {
                             rowspan = rs2;
                         }
-                        for (int c = 0; c < colspan && col < headers.Count; c++, col++) {
-                            rowValues[col] = value;
-                            if (rowspan > 1) {
-                                rowSpans[col] = (value, rowspan - 1);
-                            }
-                        }
-                    } else {
-                        if (allProperties) {
-                            rowValues[col] = null;
+                        for (int c = 0; c < colspan && col < headers.Count; c++, col++) {
+                            rowValues[col] = FillFromDataAttributes(headers[col], value, row, dataValueLookup);
+                            if (rowspan > 1) {
+                                rowSpans[col] = (rowValues[col], rowspan - 1);
+                            }
+                        }
+                    } else {
+                        if (allProperties) {
+                            rowValues[col] = null;
                         }
                         col++;
                     }
@@ -1123,13 +1129,9 @@ private static string CleanupHeader(string raw) {
     private static IDictionary<string, string> BuildDataValueLookup(IElement table) {
         var dict = new Dictionary<string, string>(StringComparer.OrdinalIgnoreCase);
         foreach (var li in table.QuerySelectorAll("li[data-value]")) {
-            string key = li.GetAttribute("data-value") ?? string.Empty;
-            key = key.Trim();
-            if (key.Length == 0) {
-                continue;
-            }
             string text = li.TextContent.Trim();
-            if (!dict.ContainsKey(key)) {
+            string key = (li.GetAttribute("data-value") ?? string.Empty).Trim();
+            if (key.Length > 0 && !dict.ContainsKey(key)) {
                 dict[key] = text;
             }
         }
@@ -1141,12 +1143,10 @@ private static IDictionary<string, string> BuildDataValueLookup(HtmlNode table)
         var items = table.SelectNodes(".//li[@data-value]");
         if (items != null) {
             foreach (var li in items) {
-                string key = li.GetAttributeValue("data-value", string.Empty);
+                string key = li.GetAttributeValue("data-value", string.Empty).Trim();
                 string text = HtmlEntity.DeEntitize(li.InnerText ?? string.Empty).Trim();
-                if (!string.IsNullOrEmpty(key)) {
-                    if (!dict.ContainsKey(key)) {
-                        dict[key] = text;
-                    }
+                if (key.Length > 0 && !dict.ContainsKey(key)) {
+                    dict[key] = text;
                 }
             }
         }

Sources/HtmlTinkerX/Playwright/HtmlBrowser.Installer.cs

@@ -492,20 +492,12 @@ private static bool ShouldUsePlaywrightWithDepsOnLinux() {
         return null;
     }
 
-    [DllImport("libc")]
-    private static extern uint geteuid();
-
     private static bool IsRunningAsRoot() {
         if (!RuntimeInformation.IsOSPlatform(OSPlatform.Linux)) {
             return false;
         }
 
-        try {
-            return geteuid() == 0;
-        } catch {
-            // ignore
-        }
-
+        // Prefer /proc/self/status (fully managed) to avoid platform-specific P/Invoke.
         try {
             const string procStatus = "/proc/self/status";
             if (File.Exists(procStatus)) {
@@ -521,7 +513,9 @@ private static bool IsRunningAsRoot() {
                     break;
                 }
             }
-        } catch {
+        } catch (IOException) {
+            // ignore
+        } catch (UnauthorizedAccessException) {
             // ignore
         }