mongodb-rag-core/build/contentStore/PageFormat.js

Common elements used by MongoDB Chatbot Framework components.

"use strict";
Object.defineProperty(exports, "__esModule", { value: true });
exports.pageFormat = exports.isPageFormat = exports.asPageFormat = exports.pageFormats = void 0;
/**
  This is the definition of the canonical file formats that we support
  for pages. If something is not in this list, we treat it as a "txt"
  file.

  If a format may be referenced by multiple terms (e.g. a file
  extension), we list it as an array where the first element is the
  canonical name and the rest are synonyms.
 */
const pageFormatsWithSynonyms = [
    // Text file formats
    ["txt", "text"],
    ["md", "markdown"],
    "mdx",
    ["restructuredtext", "rst"],
    // Data file formats
    "csv",
    "json",
    ["yaml", "yml"],
    "toml",
    "xml",
    "openapi-yaml",
    "openapi-json",
    "graphql",
    // Code file formats
    ["c", "h"],
    ["cpp", "hpp", "cxx", "hxx"],
    ["csharp", "cs"],
    ["go", "golang"],
    "html",
    "java",
    ["javascript", "js", "cjs", "mjs"],
    ["kotlin", "kt"],
    ["latex", "tex"],
    ["objective-c", "m"],
    "php",
    ["python", "py"],
    ["ruby", "rb"],
    ["rust", "rs"],
    ["scala", "sc"],
    ["shell", "sh"],
    "swift",
    ["typescript", "ts"],
];
/**
  The list of canonical file formats that we support for pages.
 */
exports.pageFormats = pageFormatsWithSynonyms.map((type) => typeof type === "string" ? type : type[0]);
// Helper that makes every element of pageFormatsWithSynonyms a tuple.
// We define this outside of the function because it's derived from a
// const so does not need to be redefined every time the function is
// called.
const pageFormatsAndSynonymsAsTuples = pageFormatsWithSynonyms.map((t) => typeof t === "string" ? [t] : t);
/**
  Maps a string to the canonical page format it represents.

  @returns The canonical page format, or undefined if the string is not
  a recognized page format.
 */
const asPageFormat = (str) => {
    for (const pageFormatAndSynonyms of pageFormatsAndSynonymsAsTuples) {
        if (pageFormatAndSynonyms.includes(str)) {
            return pageFormatAndSynonyms[0];
        }
    }
};
exports.asPageFormat = asPageFormat;
/**
  Type guard to check if a string is a canonical page format.
 */
function isPageFormat(str) {
    return exports.pageFormats.includes(str);
}
exports.isPageFormat = isPageFormat;
/**
  Converts a string to a canonical page format. If the string is not a
  recognized page format or a synonym for one, this returns a default
  page format.
 */
function pageFormat(str, defaultPageFormat = "txt") {
    return (0, exports.asPageFormat)(str) ?? defaultPageFormat;
}
exports.pageFormat = pageFormat;
//# sourceMappingURL=PageFormat.js.map