TEMP refactor plan for LLMs

src/lib/components/publications/REFACTOR_PLAN.md

@@ -0,0 +1,35 @@
+# Component Refactoring Plan for Path-Based Routing
+
+This document outlines the necessary changes to Svelte components to support the new path-based routing for publications.
+
+## 1. `PublicationHeader.svelte`
+
+This component generates links to publications and needs to be updated to the new URL format.
+
+### Actions:
+
+1.  **Locate `href` derivation:** Find the `$derived.by` block that computes the `href` constant.
+2.  **Update URL structure:** Modify the logic to generate URLs in the format `/publication/[type]/[identifier]`.
+    -   If the event has a `d` tag and is a replaceable event (e.g., kind 30040), encode it as an `naddr` and use the URL `/publication/naddr/[naddr]`.
+    -   If the event is not replaceable but has an ID (like a kind 30041), encode it as an `nevent` and use the URL `/publication/nevent/[nevent]`.
+    -   Use the existing `naddrEncode` and `neventEncode` utilities from `src/lib/utils.ts` to encode identifiers.
+    -   If needed, add new `naddrDecode` and `neventDecode` utilities to `src/lib/utils.ts`, leveraging functions from the `nip19` module in the `nostr-tools` package.
+
+## 2. `Publication.svelte`
+
+This component is responsible for rendering the publication content. The primary changes will be in how data is passed to it, rather than in its internal logic.
+
+### Actions:
+
+1.  **Review props:** The component accepts `rootAddress`, `publicationType`, and `indexEvent`. This is good.
+2.  **Update parent component:** The new `src/routes/publication/[type]/[identifier]/+page.svelte` will be responsible for providing these props from the data loaded on the server. No direct changes to `Publication.svelte` should be needed unless the data shape from the `load` function requires it. It is expected that the `load` function will provide the `indexEvent` directly.
+3.  **Add identifierType prop:** If the rendering logic needs to know the original identifier type (e.g., `id`, `d`, `naddr`, `nevent`), introduce a new `identifierType` prop to `Publication.svelte`.
+
+## 3. General Codebase Audit
+
+Other parts of the application might contain hardcoded links to publications using the old query parameter format.
+
+### Actions:
+
+1.  **Perform a codebase search:** Search for the strings `"publication?id="` and `"publication?d="` to identify any other places where links are constructed.
+2.  **Update any found links:** Refactor any discovered instances to use the new `/publication/[type]/[identifier]` format. 

src/routes/publication/REFACTOR_PLAN.md

@@ -0,0 +1,60 @@
+# Publication Route Refactoring Plan
+
+This document outlines the plan to refactor the publication routes to improve SSR, add server-side metadata, and switch to path-based routing.
+
+## 1. New Route Structure
+
+The current query-based routing (`/publication?id=...`) will be replaced with a path-based structure: `/publication/[type]/[identifier]`.
+
+### Supported Identifier Types:
+-   `id`: A raw hex event ID.
+-   `d`: A `d` tag identifier from a replaceable event.
+-   `naddr`: A bech32-encoded `naddr` string for a replaceable event.
+-   `nevent`: A bech32-encoded `nevent` string.
+
+### Actions:
+
+1.  **Create new route directory:** `src/routes/publication/[type]/[identifier]`.
+2.  **Move `+page.svelte`:** Relocate the content of the current `src/routes/publication/+page.svelte` to `src/routes/publication/[type]/[identifier]/+page.svelte`.
+3.  **Preserve old query-based route:** Instead of deleting old files, create `src/routes/publication/+page.server.ts` at the root of `src/routes/publication` to parse `?id=` and `?d=` query parameters and delegate to the new path-based routes.
+4.  **Review base route:** Ensure `/publication` either renders the main feed (via `PublicationFeed.svelte`) or redirects to `/start`; keep the existing `+page.svelte` in place for backward compatibility.
+
+## 2. Server-Side Rendering (SSR) and Data Loading
+
+We will use SvelteKit's `load` functions to fetch data on the server.
+
+### Actions:
+
+1.  **Create `+page.server.ts`:** Inside `src/routes/publication/[type]/[identifier]/`, create a `+page.server.ts` file.
+2.  **Implement `load` function:**
+    -   The `load` function will receive `params` containing `type` and `identifier`.
+    -   It will use these params to fetch the publication's root event. The logic will need to handle the different identifier types:
+        -   If `type` is `id`, use the `identifier` as a hex event ID.
+        -   If `type` is `d`, use the `identifier` to search for an event with a matching `d` tag; when multiple events share the same tag, select the event with the latest `created_at` timestamp. // AI-NOTE: choose latest for now; future logic may change.
+        -   If `type` is `naddr` or `nevent`, decode the `identifier` using `nip19.decode()` (from `nostr-tools`) and construct the appropriate filter. Add corresponding `naddrDecode` and `neventDecode` functions to `src/lib/utils.ts` to centralize NIP-19 logic.
+    -   The fetched event will be returned as `data` to the `+page.svelte` component.
+    -   Handle cases where the event is not found by throwing a 404 error using `@sveltejs/kit/error`.
+
+## 3. Server-Side Metadata
+
+Publication-specific metadata will be rendered on the server for better link previews.
+
+### Actions:
+
+1.  **Create `+layout.server.ts`:** Inside `src/routes/publication/[type]/[identifier]/`, create a `+layout.server.ts`. Its `load` function will be very similar to the one in `+page.server.ts`. It will fetch the root event and return the necessary data for metadata (title, summary, image URL).
+2.  **Create `+layout.svelte`:** Inside `src/routes/publication/[type]/[identifier]/`, create a `+layout.svelte`.
+3.  **Implement metadata:**
+    -   The layout will receive `data` from its `load` function.
+    -   It will contain a `<svelte:head>` block.
+    -   Inside `<svelte:head>`, render `<title>` and `<meta>` tags (OpenGraph, Twitter Cards) using properties from the loaded `data`.
+    -   Use `{@render children()}` in `+layout.svelte` to display the page content.
+    -   Refer to https://web.dev/learn/html/metadata/#officially_defined_meta_tags for a compilation of recommended meta tags.
+
+## 4. Handling Authentication
+
+For publications requiring authentication, we need to avoid full SSR of content while still providing a good user experience.
+
+### Actions:
+
+-   Skip authentication/authorization handling in this refactor; it will be addressed separately.
+-   If the `indexEvent` cannot be fetched, display a user-friendly error message in `+page.svelte` indicating the publication cannot be loaded.