From cbae88fc4e9b98764cfccca2e85f265c4b894573 Mon Sep 17 00:00:00 2001
From: Adam Brangenberg <adambrangenberg@proton.me>
Date: Mon, 31 Jul 2023 04:08:32 +0000
Subject: [PATCH] Removing redundant properties (#356)
---
quartz/path.ts | 303 ++++++++++++++++++++++++++++++++++++++++++++------
1 files changed, 266 insertions(+), 37 deletions(-)
diff --git a/quartz/path.ts b/quartz/path.ts
index 18d74d7..fca2c05 100644
--- a/quartz/path.ts
+++ b/quartz/path.ts
@@ -1,51 +1,280 @@
-import path from 'path'
-import { slug as slugAnchor } from 'github-slugger'
+import { slug } from "github-slugger"
+import { trace } from "./trace"
-function slugSegment(s: string): string {
- return s.replace(/\s/g, '-')
+// Quartz Paths
+// Things in boxes are not actual types but rather sources which these types can be acquired from
+//
+// ┌────────────┐
+// ┌───────────┤ Browser ├────────────┐
+// │ └────────────┘ │
+// │ │
+// ▼ ▼
+// ┌────────┐ ┌─────────────┐
+// ┌───────────────────┤ Window │ │ LinkElement │
+// │ └────┬───┘ └──────┬──────┘
+// │ │ │
+// │ getClientSlug() │ .href │
+// │ ▼ ▼
+// │
+// │ Client Slug ┌───► Relative URL
+// getCanonicalSlug() │ https://test.ca/note/abc#anchor?query=123 │ ../note/def#anchor
+// │ │
+// │ canonicalizeClient() │ │ ▲ ▲
+// │ ▼ │ │ │
+// │ pathToRoot() │ │ │
+// └───────────────► Canonical Slug ────────────────┘ │ │
+// note/abc │ │
+// ──────────────────────────┘ │
+// ▲ resolveRelative() │
+// canonicalizeServer() │ │
+// │
+// HTML File Server Slug │
+// note/abc/index.html ◄───────────── note/abc/index │
+// │
+// ▲ ┌────────┴────────┐
+// slugifyFilePath() │ transformLink() │ │
+// │ │ │
+// ┌─────────┴──────────┐ ┌─────┴─────┐ ┌────────┴──────┐
+// │ File Path │ │ Wikilinks │ │ Markdown Link │
+// │ note/abc/index.md │ └───────────┘ └───────────────┘
+// └────────────────────┘ ▲ ▲
+// ▲ │ │
+// │ ┌─────────┐ │ │
+// └────────────┤ MD File ├─────┴─────────────────┘
+// └─────────┘
+
+const STRICT_TYPE_CHECKS = false
+const HARD_EXIT_ON_FAIL = false
+
+function conditionCheck<T>(name: string, label: "pre" | "post", s: T, chk: (x: any) => x is T) {
+ if (STRICT_TYPE_CHECKS && !chk(s)) {
+ trace(`${name} failed ${label}-condition check: ${s} does not pass ${chk.name}`, new Error())
+ if (HARD_EXIT_ON_FAIL) {
+ process.exit(1)
+ }
+ }
}
-export function trimPathSuffix(fp: string): string {
- let [cleanPath, anchor] = fp.split("#", 2)
- anchor = anchor === undefined ? "" : "#" + anchor
+/// Utility type to simulate nominal types in TypeScript
+type SlugLike<T> = string & { __brand: T }
- return cleanPath + anchor
+/** Client-side slug, usually obtained through `window.location` */
+export type ClientSlug = SlugLike<"client">
+export function isClientSlug(s: string): s is ClientSlug {
+ const res = /^https?:\/\/.+/.test(s)
+ return res
}
-export function slugify(s: string): string {
- const [fp, anchor] = s.split("#", 2)
- const sluggedAnchor = anchor === undefined ? "" : "#" + slugAnchor(anchor)
- const withoutFileExt = fp.replace(new RegExp(path.extname(fp) + '$'), '')
- const rawSlugSegments = withoutFileExt.split(path.sep)
- const slugParts: string = rawSlugSegments
- .map((segment) => slugSegment(segment))
- .join(path.posix.sep)
- .replace(/\/$/, '')
- return path.normalize(slugParts) + sluggedAnchor
+/** Canonical slug, should be used whenever you need to refer to the location of a file/note.
+ * On the client, this is normally stored in `document.body.dataset.slug`
+ */
+export type CanonicalSlug = SlugLike<"canonical">
+export function isCanonicalSlug(s: string): s is CanonicalSlug {
+ const validStart = !(s.startsWith(".") || s.startsWith("/"))
+ const validEnding = !(s.endsWith("/") || s.endsWith("/index") || s === "index")
+ return validStart && !_containsForbiddenCharacters(s) && validEnding && !_hasFileExtension(s)
+}
+
+/** A relative link, can be found on `href`s but can also be constructed for
+ * client-side navigation (e.g. search and graph)
+ */
+export type RelativeURL = SlugLike<"relative">
+export function isRelativeURL(s: string): s is RelativeURL {
+ const validStart = /^\.{1,2}/.test(s)
+ const validEnding = !(s.endsWith("/") || s.endsWith("/index") || s === "index")
+ return validStart && validEnding && !_hasFileExtension(s)
+}
+
+/** A server side slug. This is what Quartz uses to emit files so uses index suffixes */
+export type ServerSlug = SlugLike<"server">
+export function isServerSlug(s: string): s is ServerSlug {
+ const validStart = !(s.startsWith(".") || s.startsWith("/"))
+ const validEnding = !s.endsWith("/")
+ return validStart && validEnding && !_containsForbiddenCharacters(s) && !_hasFileExtension(s)
+}
+
+/** The real file path to a file on disk */
+export type FilePath = SlugLike<"filepath">
+export function isFilePath(s: string): s is FilePath {
+ const validStart = !s.startsWith(".")
+ return validStart && _hasFileExtension(s)
+}
+
+export function getClientSlug(window: Window): ClientSlug {
+ const res = window.location.href as ClientSlug
+ conditionCheck(getClientSlug.name, "post", res, isClientSlug)
+ return res
+}
+
+export function getCanonicalSlug(window: Window): CanonicalSlug {
+ const res = window.document.body.dataset.slug! as CanonicalSlug
+ conditionCheck(getCanonicalSlug.name, "post", res, isCanonicalSlug)
+ return res
+}
+
+export function canonicalizeClient(slug: ClientSlug): CanonicalSlug {
+ conditionCheck(canonicalizeClient.name, "pre", slug, isClientSlug)
+ const { pathname } = new URL(slug)
+ let fp = pathname.slice(1)
+ fp = fp.replace(new RegExp(_getFileExtension(fp) + "$"), "")
+ const res = _canonicalize(fp) as CanonicalSlug
+ conditionCheck(canonicalizeClient.name, "post", res, isCanonicalSlug)
+ return res
+}
+
+export function canonicalizeServer(slug: ServerSlug): CanonicalSlug {
+ conditionCheck(canonicalizeServer.name, "pre", slug, isServerSlug)
+ let fp = slug as string
+ const res = _canonicalize(fp) as CanonicalSlug
+ conditionCheck(canonicalizeServer.name, "post", res, isCanonicalSlug)
+ return res
+}
+
+export function slugifyFilePath(fp: FilePath): ServerSlug {
+ conditionCheck(slugifyFilePath.name, "pre", fp, isFilePath)
+ fp = _stripSlashes(fp) as FilePath
+ const withoutFileExt = fp.replace(new RegExp(_getFileExtension(fp) + "$"), "")
+ let slug = withoutFileExt
+ .split("/")
+ .map((segment) => segment.replace(/\s/g, "-")) // slugify all segments
+ .join("/") // always use / as sep
+ .replace(/\/$/, "") // remove trailing slash
+
+ // treat _index as index
+ if (_endsWith(slug, "_index")) {
+ slug = slug.replace(/_index$/, "index")
+ }
+
+ conditionCheck(slugifyFilePath.name, "post", slug, isServerSlug)
+ return slug as ServerSlug
+}
+
+export function transformInternalLink(link: string): RelativeURL {
+ let [fplike, anchor] = splitAnchor(decodeURI(link))
+ let segments = fplike.split("/").filter((x) => x.length > 0)
+ let prefix = segments.filter(_isRelativeSegment).join("/")
+ let fp = segments.filter((seg) => !_isRelativeSegment(seg)).join("/")
+
+ // implicit markdown
+ if (!_hasFileExtension(fp)) {
+ fp += ".md"
+ }
+
+ fp = canonicalizeServer(slugifyFilePath(fp as FilePath))
+ fp = _trimSuffix(fp, "index")
+
+ let joined = joinSegments(_stripSlashes(prefix), _stripSlashes(fp))
+ const res = (_addRelativeToStart(joined) + anchor) as RelativeURL
+ conditionCheck(transformInternalLink.name, "post", res, isRelativeURL)
+ return res
}
// resolve /a/b/c to ../../
-export function resolveToRoot(slug: string): string {
- let fp = trimPathSuffix(slug)
+export function pathToRoot(slug: CanonicalSlug): RelativeURL {
+ conditionCheck(pathToRoot.name, "pre", slug, isCanonicalSlug)
+ let rootPath = slug
+ .split("/")
+ .filter((x) => x !== "")
+ .map((_) => "..")
+ .join("/")
- if (fp === "") {
- return "."
+ const res = _addRelativeToStart(rootPath) as RelativeURL
+ conditionCheck(pathToRoot.name, "post", res, isRelativeURL)
+ return res
+}
+
+export function resolveRelative(current: CanonicalSlug, target: CanonicalSlug): RelativeURL {
+ conditionCheck(resolveRelative.name, "pre", current, isCanonicalSlug)
+ conditionCheck(resolveRelative.name, "pre", target, isCanonicalSlug)
+ const res = joinSegments(pathToRoot(current), target) as RelativeURL
+ conditionCheck(resolveRelative.name, "post", res, isRelativeURL)
+ return res
+}
+
+export function splitAnchor(link: string): [string, string] {
+ let [fp, anchor] = link.split("#", 2)
+ anchor = anchor === undefined ? "" : "#" + slugAnchor(anchor)
+ return [fp, anchor]
+}
+
+export function slugAnchor(anchor: string) {
+ return slug(anchor)
+}
+
+export function slugTag(tag: string) {
+ return tag
+ .split("/")
+ .map((tagSegment) => slug(tagSegment))
+ .join("/")
+}
+
+export function joinSegments(...args: string[]): string {
+ return args.filter((segment) => segment !== "").join("/")
+}
+
+export function getAllSegmentPrefixes(tags: string): string[] {
+ const segments = tags.split("/")
+ const results: string[] = []
+ for (let i = 0; i < segments.length; i++) {
+ results.push(segments.slice(0, i + 1).join("/"))
}
-
- return "./" + fp
- .split('/')
- .filter(x => x !== '')
- .map(_ => '..')
- .join('/')
-}
-
-export function relativeToRoot(slug: string, fp: string): string {
- return path.join(resolveToRoot(slug), fp)
-}
-
-export function relative(src: string, dest: string): string {
- return "./" + path.relative(src, dest)
-
+ return results
}
export const QUARTZ = "quartz"
+
+function _canonicalize(fp: string): string {
+ fp = _trimSuffix(fp, "index")
+ return _stripSlashes(fp)
+}
+
+function _endsWith(s: string, suffix: string): boolean {
+ return s === suffix || s.endsWith("/" + suffix)
+}
+
+function _trimSuffix(s: string, suffix: string): string {
+ if (_endsWith(s, suffix)) {
+ s = s.slice(0, -suffix.length)
+ }
+ return s
+}
+
+function _containsForbiddenCharacters(s: string): boolean {
+ return s.includes(" ") || s.includes("#") || s.includes("?")
+}
+
+function _hasFileExtension(s: string): boolean {
+ return _getFileExtension(s) !== undefined
+}
+
+function _getFileExtension(s: string): string | undefined {
+ return s.match(/\.[A-Za-z]+$/)?.[0]
+}
+
+function _isRelativeSegment(s: string): boolean {
+ return /^\.{0,2}$/.test(s)
+}
+
+function _stripSlashes(s: string): string {
+ if (s.startsWith("/")) {
+ s = s.substring(1)
+ }
+
+ if (s.endsWith("/")) {
+ s = s.slice(0, -1)
+ }
+
+ return s
+}
+
+function _addRelativeToStart(s: string): string {
+ if (s === "") {
+ s = "."
+ }
+
+ if (!s.startsWith(".")) {
+ s = joinSegments(".", s)
+ }
+
+ return s
+}
--
Gitblit v1.10.0