|
| 1 | +/** |
| 2 | + * Bibliography block rule: [[bibliography]] ... [[/bibliography]] |
| 3 | + * |
| 4 | + * Contains definition list items that define bibliography entries. |
| 5 | + * Format inside: |
| 6 | + * : label : Citation text |
| 7 | + * |
| 8 | + * Works with ((bibcite label)) inline elements. |
| 9 | + */ |
| 10 | +import type { Element, DefinitionListItem } from "@wdprlib/ast"; |
| 11 | +import type { BlockRule, ParseContext, RuleResult } from "../types"; |
| 12 | +import { currentToken } from "../types"; |
| 13 | +import { parseBlockName, parseAttributes } from "./utils"; |
| 14 | +import { parseInlineUntil } from "../inline/utils"; |
| 15 | + |
| 16 | +interface BibliographyEntry { |
| 17 | + label: string; |
| 18 | + key: Element[]; |
| 19 | + content: Element[]; |
| 20 | +} |
| 21 | + |
| 22 | +/** |
| 23 | + * Parse a single bibliography entry |
| 24 | + * Format: : label : content |
| 25 | + */ |
| 26 | +function parseBibliographyEntry( |
| 27 | + ctx: ParseContext, |
| 28 | + startPos: number, |
| 29 | +): { entry: BibliographyEntry; consumed: number } | null { |
| 30 | + let pos = startPos; |
| 31 | + let consumed = 0; |
| 32 | + |
| 33 | + // Expect COLON at line start |
| 34 | + const colonToken = ctx.tokens[pos]; |
| 35 | + if (!colonToken || colonToken.type !== "COLON" || !colonToken.lineStart) { |
| 36 | + return null; |
| 37 | + } |
| 38 | + pos++; |
| 39 | + consumed++; |
| 40 | + |
| 41 | + // Wikidot requires whitespace after first colon |
| 42 | + const whitespaceAfterColon = ctx.tokens[pos]; |
| 43 | + if (!whitespaceAfterColon || whitespaceAfterColon.type !== "WHITESPACE") { |
| 44 | + return null; |
| 45 | + } |
| 46 | + |
| 47 | + // Skip whitespace after first colon |
| 48 | + while (ctx.tokens[pos]?.type === "WHITESPACE") { |
| 49 | + pos++; |
| 50 | + consumed++; |
| 51 | + } |
| 52 | + |
| 53 | + // Collect label tokens until second COLON |
| 54 | + let label = ""; |
| 55 | + let foundSecondColon = false; |
| 56 | + const keyNodes: Element[] = []; |
| 57 | + |
| 58 | + while (pos < ctx.tokens.length) { |
| 59 | + const token = ctx.tokens[pos]; |
| 60 | + if (!token || token.type === "NEWLINE" || token.type === "EOF") { |
| 61 | + break; |
| 62 | + } |
| 63 | + if (token.type === "COLON") { |
| 64 | + foundSecondColon = true; |
| 65 | + pos++; |
| 66 | + consumed++; |
| 67 | + break; |
| 68 | + } |
| 69 | + |
| 70 | + // For bibliography, key is just the label (identifier) |
| 71 | + label += token.value; |
| 72 | + keyNodes.push({ element: "text", data: token.value }); |
| 73 | + pos++; |
| 74 | + consumed++; |
| 75 | + } |
| 76 | + |
| 77 | + if (!foundSecondColon) { |
| 78 | + return null; |
| 79 | + } |
| 80 | + |
| 81 | + label = label.trim(); |
| 82 | + |
| 83 | + // Skip whitespace after second colon |
| 84 | + while (ctx.tokens[pos]?.type === "WHITESPACE") { |
| 85 | + pos++; |
| 86 | + consumed++; |
| 87 | + } |
| 88 | + |
| 89 | + // Parse content (rest of line, can continue with line breaks) |
| 90 | + const contentNodes: Element[] = []; |
| 91 | + while (pos < ctx.tokens.length) { |
| 92 | + const token = ctx.tokens[pos]; |
| 93 | + if (!token || token.type === "EOF") { |
| 94 | + break; |
| 95 | + } |
| 96 | + |
| 97 | + // Check for [[/bibliography]] |
| 98 | + if (token.type === "BLOCK_END_OPEN") { |
| 99 | + const closeNameResult = parseBlockName(ctx, pos + 1); |
| 100 | + if (closeNameResult?.name === "bibliography") { |
| 101 | + break; |
| 102 | + } |
| 103 | + } |
| 104 | + |
| 105 | + // Check for end of entry (newline followed by new entry or end) |
| 106 | + if (token.type === "NEWLINE") { |
| 107 | + const nextToken = ctx.tokens[pos + 1]; |
| 108 | + // Look ahead for new entry or block end |
| 109 | + if (nextToken?.type === "COLON" && nextToken.lineStart) { |
| 110 | + // New entry starts |
| 111 | + pos++; |
| 112 | + consumed++; |
| 113 | + break; |
| 114 | + } |
| 115 | + if (nextToken?.type === "BLOCK_END_OPEN") { |
| 116 | + // Block end |
| 117 | + pos++; |
| 118 | + consumed++; |
| 119 | + break; |
| 120 | + } |
| 121 | + if (nextToken?.type === "NEWLINE" || !nextToken || nextToken.type === "EOF") { |
| 122 | + // Double newline or end |
| 123 | + pos++; |
| 124 | + consumed++; |
| 125 | + break; |
| 126 | + } |
| 127 | + // Single newline - continue (becomes line break) |
| 128 | + } |
| 129 | + |
| 130 | + // Parse inline content |
| 131 | + const inlineCtx: ParseContext = { ...ctx, pos }; |
| 132 | + const result = parseInlineUntil(inlineCtx, "NEWLINE"); |
| 133 | + if (result.elements.length > 0) { |
| 134 | + contentNodes.push(...result.elements); |
| 135 | + pos += result.consumed; |
| 136 | + consumed += result.consumed; |
| 137 | + } else { |
| 138 | + pos++; |
| 139 | + consumed++; |
| 140 | + } |
| 141 | + } |
| 142 | + |
| 143 | + // Trim key nodes |
| 144 | + while (keyNodes.length > 0) { |
| 145 | + const lastNode = keyNodes[keyNodes.length - 1]; |
| 146 | + if ( |
| 147 | + lastNode && |
| 148 | + lastNode.element === "text" && |
| 149 | + typeof lastNode.data === "string" && |
| 150 | + lastNode.data.trim() === "" |
| 151 | + ) { |
| 152 | + keyNodes.pop(); |
| 153 | + } else { |
| 154 | + break; |
| 155 | + } |
| 156 | + } |
| 157 | + |
| 158 | + return { |
| 159 | + entry: { |
| 160 | + label, |
| 161 | + key: keyNodes, |
| 162 | + content: contentNodes, |
| 163 | + }, |
| 164 | + consumed, |
| 165 | + }; |
| 166 | +} |
| 167 | + |
| 168 | +export const bibliographyRule: BlockRule = { |
| 169 | + name: "bibliography", |
| 170 | + startTokens: ["BLOCK_OPEN"], |
| 171 | + requiresLineStart: false, |
| 172 | + |
| 173 | + parse(ctx: ParseContext): RuleResult<Element> { |
| 174 | + const openToken = currentToken(ctx); |
| 175 | + if (openToken.type !== "BLOCK_OPEN") { |
| 176 | + return { success: false }; |
| 177 | + } |
| 178 | + |
| 179 | + let pos = ctx.pos + 1; |
| 180 | + let consumed = 1; |
| 181 | + |
| 182 | + // Parse block name |
| 183 | + const nameResult = parseBlockName(ctx, pos); |
| 184 | + if (!nameResult || nameResult.name !== "bibliography") { |
| 185 | + return { success: false }; |
| 186 | + } |
| 187 | + |
| 188 | + pos += nameResult.consumed; |
| 189 | + consumed += nameResult.consumed; |
| 190 | + |
| 191 | + // Parse optional attributes (title, hide) |
| 192 | + const attrResult = parseAttributes(ctx, pos); |
| 193 | + pos += attrResult.consumed; |
| 194 | + consumed += attrResult.consumed; |
| 195 | + |
| 196 | + // Expect ]] |
| 197 | + if (ctx.tokens[pos]?.type !== "BLOCK_CLOSE") { |
| 198 | + return { success: false }; |
| 199 | + } |
| 200 | + pos++; |
| 201 | + consumed++; |
| 202 | + |
| 203 | + // Skip newline after opening tag |
| 204 | + if (ctx.tokens[pos]?.type === "NEWLINE") { |
| 205 | + pos++; |
| 206 | + consumed++; |
| 207 | + } |
| 208 | + |
| 209 | + // Parse bibliography entries |
| 210 | + const entries: BibliographyEntry[] = []; |
| 211 | + |
| 212 | + while (pos < ctx.tokens.length) { |
| 213 | + const token = ctx.tokens[pos]; |
| 214 | + if (!token || token.type === "EOF") { |
| 215 | + break; |
| 216 | + } |
| 217 | + |
| 218 | + // Check for [[/bibliography]] |
| 219 | + if (token.type === "BLOCK_END_OPEN") { |
| 220 | + const closeNameResult = parseBlockName(ctx, pos + 1); |
| 221 | + if (closeNameResult?.name === "bibliography") { |
| 222 | + // Consume [[/bibliography]] |
| 223 | + pos++; |
| 224 | + consumed++; |
| 225 | + pos += closeNameResult.consumed; |
| 226 | + consumed += closeNameResult.consumed; |
| 227 | + // Skip whitespace |
| 228 | + while (ctx.tokens[pos]?.type === "WHITESPACE") { |
| 229 | + pos++; |
| 230 | + consumed++; |
| 231 | + } |
| 232 | + // Expect ]] |
| 233 | + if (ctx.tokens[pos]?.type === "BLOCK_CLOSE") { |
| 234 | + pos++; |
| 235 | + consumed++; |
| 236 | + } |
| 237 | + break; |
| 238 | + } |
| 239 | + } |
| 240 | + |
| 241 | + // Skip whitespace and newlines |
| 242 | + if (token.type === "WHITESPACE" || token.type === "NEWLINE") { |
| 243 | + pos++; |
| 244 | + consumed++; |
| 245 | + continue; |
| 246 | + } |
| 247 | + |
| 248 | + // Parse entry |
| 249 | + if (token.type === "COLON" && token.lineStart) { |
| 250 | + const result = parseBibliographyEntry(ctx, pos); |
| 251 | + if (result) { |
| 252 | + entries.push(result.entry); |
| 253 | + pos += result.consumed; |
| 254 | + consumed += result.consumed; |
| 255 | + continue; |
| 256 | + } |
| 257 | + } |
| 258 | + |
| 259 | + // Skip unknown tokens |
| 260 | + pos++; |
| 261 | + consumed++; |
| 262 | + } |
| 263 | + |
| 264 | + // Convert to definition list format for AST storage |
| 265 | + const definitionItems: DefinitionListItem[] = entries.map((entry) => ({ |
| 266 | + key_string: entry.label, |
| 267 | + key: entry.key, |
| 268 | + value: entry.content, |
| 269 | + })); |
| 270 | + |
| 271 | + // Get attributes |
| 272 | + const title = attrResult.attrs.title ?? null; |
| 273 | + const hide = attrResult.attrs.hide === "true" || attrResult.attrs.hide === ""; |
| 274 | + |
| 275 | + return { |
| 276 | + success: true, |
| 277 | + elements: [ |
| 278 | + { |
| 279 | + element: "bibliography-block", |
| 280 | + data: { |
| 281 | + entries: definitionItems, |
| 282 | + title: typeof title === "string" ? title : null, |
| 283 | + hide, |
| 284 | + }, |
| 285 | + }, |
| 286 | + ], |
| 287 | + consumed, |
| 288 | + }; |
| 289 | + }, |
| 290 | +}; |
0 commit comments