|
| 1 | +using System; |
| 2 | +using System.Collections.Generic; |
| 3 | +using System.Linq; |
| 4 | +using System.Text.RegularExpressions; |
| 5 | +using SIL.Scripture; |
| 6 | + |
| 7 | +/// <summary> |
| 8 | +/// This is a helper class that will assist in parsing various scripture range formats that are accepted by Serval. |
| 9 | +/// </summary> |
| 10 | +/// <remarks> |
| 11 | +/// Further information can be found in the Serval documentation: <see cref="https://github.com/sillsdev/serval/wiki/Filtering-Paratext-Project-Data-with-a-Scripture-Range/" /> |
| 12 | +/// This uses the same regexes as SIL.Machine to verify the type of book range. |
| 13 | +/// Our specific use case for parsing scripture ranges is different from SIL.Machine and therefore requires it's own implementation. |
| 14 | +/// </remarks> |
| 15 | +public class ScriptureRangeParser |
| 16 | +{ |
| 17 | + private HashSet<string> Books = []; |
| 18 | + private HashSet<string> BooksToRemove = []; |
| 19 | + private readonly Regex CommaSeparatedBooks = new Regex( |
| 20 | + @"^([A-Z\d]{3}|OT|NT)(, ?([A-Z\d]{3}|OT|NT))*$", |
| 21 | + RegexOptions.Compiled |
| 22 | + ); |
| 23 | + |
| 24 | + private readonly Regex SemiColonSeparatedBooks = new Regex( |
| 25 | + @"^([A-Z\d]{3}|OT|NT)(; ?([A-Z\d]{3}|OT|NT))*$", |
| 26 | + RegexOptions.Compiled |
| 27 | + ); |
| 28 | + |
| 29 | + private readonly Regex BookRange = new Regex(@"^-?[A-Z\d]{3}-[A-Z\d]{3}$", RegexOptions.Compiled); |
| 30 | + |
| 31 | + public IEnumerable<string> ParseScriptureRange(string scriptureRange) |
| 32 | + { |
| 33 | + if (string.IsNullOrWhiteSpace(scriptureRange)) |
| 34 | + { |
| 35 | + return new HashSet<string>(); |
| 36 | + } |
| 37 | + |
| 38 | + try |
| 39 | + { |
| 40 | + // Allow a single book as a range |
| 41 | + if (scriptureRange.Length == 3) |
| 42 | + { |
| 43 | + Books.Add(scriptureRange); |
| 44 | + } |
| 45 | + // Allow semi-colon separated HashSet |
| 46 | + else if (SemiColonSeparatedBooks.IsMatch(scriptureRange)) |
| 47 | + { |
| 48 | + var booksHashSet = scriptureRange.Split([';'], StringSplitOptions.RemoveEmptyEntries); |
| 49 | + foreach (var book in booksHashSet) |
| 50 | + { |
| 51 | + ProcessBook(book.ToUpperInvariant()); |
| 52 | + } |
| 53 | + } |
| 54 | + // Allow comma separated HashSet |
| 55 | + else if (CommaSeparatedBooks.IsMatch(scriptureRange)) |
| 56 | + { |
| 57 | + var booksHashSet = scriptureRange.Split([','], StringSplitOptions.RemoveEmptyEntries); |
| 58 | + foreach (var book in booksHashSet) |
| 59 | + { |
| 60 | + ProcessBook(book.ToUpperInvariant()); |
| 61 | + } |
| 62 | + } |
| 63 | + } |
| 64 | + catch (Exception ex) |
| 65 | + { |
| 66 | + // Handle any exceptions that may occur during parsing |
| 67 | + Console.WriteLine($"Error parsing scripture range: {ex.Message}"); |
| 68 | + } |
| 69 | + |
| 70 | + // Remove any books that are marked for removal |
| 71 | + foreach (var book in BooksToRemove) |
| 72 | + { |
| 73 | + if (Books.Contains(book)) |
| 74 | + { |
| 75 | + Books.Remove(book); |
| 76 | + } |
| 77 | + } |
| 78 | + return Books; |
| 79 | + } |
| 80 | + |
| 81 | + /// <summary> |
| 82 | + /// |
| 83 | + /// </summary> |
| 84 | + /// <param name="book">The book or range of books to process.</param> |
| 85 | + /// <returns>A HashSet of books that are part of the range.</returns> |
| 86 | + /// <remarks> |
| 87 | + /// A "book" may be a single book (GEN), a range of books (e.g. "GEN-LEV"), an entire testament (OT or NT), |
| 88 | + /// or marked to remove a book (NT;-REV). |
| 89 | + /// </remarks> |
| 90 | + private void ProcessBook(string book) |
| 91 | + { |
| 92 | + switch (book) |
| 93 | + { |
| 94 | + case "OT": |
| 95 | + AddAllBooksForTestament(book); |
| 96 | + break; |
| 97 | + case "NT": |
| 98 | + AddAllBooksForTestament(book); |
| 99 | + break; |
| 100 | + default: |
| 101 | + if (book.Length == 3) |
| 102 | + { |
| 103 | + Books.Add(book); |
| 104 | + } |
| 105 | + else if (BookRange.IsMatch(book)) |
| 106 | + { |
| 107 | + AddAllBooksInRange(book); |
| 108 | + } |
| 109 | + else if (book.StartsWith('-')) |
| 110 | + { |
| 111 | + BooksToRemove.Add(book[..'-']); |
| 112 | + } |
| 113 | + else |
| 114 | + { |
| 115 | + throw new ArgumentException($"Invalid book range: {book}"); |
| 116 | + } |
| 117 | + break; |
| 118 | + } |
| 119 | + } |
| 120 | + |
| 121 | + private void AddAllBooksInRange(string range) |
| 122 | + { |
| 123 | + string[] rangeParts = range.Split('-'); |
| 124 | + // we should expect the first book to come before the second book (e.g. GEN-LEV) not the other way around (e.g. LEV-GEN) |
| 125 | + if (Array.IndexOf(Canon.AllBookIds, rangeParts[1]) > Array.IndexOf(Canon.AllBookIds, rangeParts[0])) |
| 126 | + { |
| 127 | + throw new ArgumentException($"Invalid book range: {range}"); |
| 128 | + } |
| 129 | + |
| 130 | + string endBook = Canon.AllBookIds[Array.IndexOf(Canon.AllBookIds, rangeParts[1]) + 1]; |
| 131 | + foreach ( |
| 132 | + var book in Canon.AllBookIds.SkipWhile(book => book != rangeParts[0]).TakeWhile(book => book != endBook) |
| 133 | + ) |
| 134 | + { |
| 135 | + Books.Add(book); |
| 136 | + } |
| 137 | + } |
| 138 | + |
| 139 | + private void AddAllBooksForTestament(string testment) |
| 140 | + { |
| 141 | + if (testment == "OT") |
| 142 | + { |
| 143 | + foreach (var book in (string[])Canon.AllBookIds.Where(Canon.IsBookOT)) |
| 144 | + { |
| 145 | + Books.Add(book.ToUpperInvariant()); |
| 146 | + } |
| 147 | + } |
| 148 | + else if (testment == "NT") |
| 149 | + { |
| 150 | + foreach (var book in (string[])Canon.AllBookIds.Where(Canon.IsBookNT)) |
| 151 | + { |
| 152 | + Books.Add(book.ToUpperInvariant()); |
| 153 | + } |
| 154 | + } |
| 155 | + else |
| 156 | + { |
| 157 | + throw new ArgumentException($"Invalid testament: {testment}"); |
| 158 | + } |
| 159 | + } |
| 160 | +} |
0 commit comments