@youtyan/code-viewer 0.8.8 → 0.9.0

package/README.md

@@ -32,9 +32,9 @@ Requires Node.js 20 or newer when installed from npm. Development uses
 - Switch the viewer UI between English and Japanese from Viewer Settings —
   the language toggle live-updates every screen including the datastore
   viewer.
-- Browse SQLite, PostgreSQL, MySQL, Redis, Elasticsearch, DynamoDB, and
-  S3-compatible object storage (MinIO, LocalStack) with a built-in datastore
-  viewer.
+- Browse SQLite, PostgreSQL, MySQL, Cloudflare D1, Redis, Elasticsearch,
+  DynamoDB, and S3-compatible object storage (MinIO, LocalStack, Cloudflare R2)
+  with a built-in datastore viewer.
   Local Supabase CLI (`supabase start`) Postgres projects are auto-discovered
   too, without needing a `docker-compose.yml`.
   Table descriptions appear inside expanded table entries and in the Schema
@@ -144,7 +144,11 @@ metadata instead of dumping bytes as text.
 Markdown files use a dedicated preview tab. Relative links and images are
 resolved inside the repository, code blocks are highlighted with Shiki, and
 Mermaid diagrams are rendered lazily in the browser (click any diagram to
-open it in a lightbox).
+open it in a lightbox). Relative links lead to the same destinations as they
+do on GitHub: another Markdown file opens its file page, an `#anchor` opens
+the preview and scrolls to that heading, a non-Markdown file opens in the
+Code view, and a link to a directory opens that folder in the repository
+tree.
 
 A file detail page lays out up to four tabs — **Preview**, **Code**,
 **Blame**, **History** — modelled after the GitHub file view. For text files
@@ -152,7 +156,9 @@ A file detail page lays out up to four tabs — **Preview**, **Code**,
 preview. Media files (images, video, audio, PDF) show a **Preview** tab only
 — there is no Code tab for binary media. `Blame` and `History` each have
 their own canonical URL (`view=blame`, `view=history`), so deep links and
-the browser back/forward stay in sync. The Blame tab reuses the source
+the browser back/forward stay in sync. Opening another file from the
+repository tree keeps the active tab (a file that cannot be previewed falls
+back to Code). The Blame tab reuses the source
 view's row component, so line numbers, drag-selection of `line=` ranges,
 syntax highlighting and the Viewer Settings code font size all match the
 Code tab.
@@ -272,7 +278,23 @@ global/local secondary indexes, and non-key attribute types inferred from
 loaded items) alongside scan or query items, `LastEvaluatedKey` pagination,
 and an item detail view with a copyable key. The explorer is read-only.
 
-**S3-compatible object storage** (MinIO, LocalStack): browse
+**Cloudflare D1** support: add a saved connection with a Cloudflare account ID,
+database ID, and API token (needs the `D1:Read` permission). The database is
+browsed over the D1 REST API (`POST /accounts/{account}/d1/database/{db}/raw`)
+and reuses the SQL screens — table list, row grid, query editor, schema, ER
+diagram, snapshots and diffs. D1 is SQLite, so the same `sqlite_master` and
+`PRAGMA table_info` / `index_list` / `foreign_key_list` introspection is used.
+The connection is read-only: the query editor accepts only
+`SELECT` / `PRAGMA` / `EXPLAIN` / `WITH`, and the grid's row editing is
+unavailable.
+
+**Cloudflare R2**: R2 speaks the S3 API, so it is a saved connection of type
+`s3` with the **Cloudflare R2** provider preset. Enter the account ID and an R2
+access key pair; the endpoint (`https://<account-id>.r2.cloudflarestorage.com`)
+and the required `auto` signing region are filled in for you. Everything in the
+object-storage section below applies.
+
+**S3-compatible object storage** (MinIO, LocalStack, Cloudflare R2): browse
 buckets as a folder tree, search by prefix or filename, sort scanned objects by
 update time, and preview images, video, audio, PDFs, Markdown, HTML, and text
 files. Edit text/markdown/JSON object bodies inline, upload new objects to any
@@ -289,27 +311,45 @@ table grid to insert, update, or delete rows inline. Edits queue as pending
 changes (shown with a yellow highlight on the affected row/cell) and are
 applied as a single transaction on commit. Row updates and deletes require the
 table to have a primary key. The whole batch rolls back on the first
-constraint violation.
+constraint violation. Cloudflare D1 is browsed read-only, so Edit mode is not
+offered there.
 
 ### Browser UI
 
 Open Datastores in the global navigation to access:
 
 - **Saved connections** — use the `+` action beside the datastore selector to
-  add PostgreSQL, MySQL, Redis, Elasticsearch, S3-compatible, or DynamoDB
-  endpoints that are not present in the repository's local discovery files.
+  add PostgreSQL, MySQL, Cloudflare D1, Redis, Elasticsearch, S3-compatible
+  (including Cloudflare R2), or DynamoDB endpoints that are not present in the
+  repository's local discovery files.
   Required fields are marked in the connection dialog, and **Test connection**
   verifies the current values before saving. Saved connections can be edited
   or deleted from the adjacent actions.
   PostgreSQL, MySQL, and Redis use Node.js drivers installed automatically with
-  this package, while Elasticsearch, S3, and DynamoDB use built-in HTTP
-  clients. No database CLI or `curl` installation is required for saved
-  connections.
-  Non-secret settings are stored in `.code-viewer/datastore-connections.json`.
-  User names, access key IDs, passwords, secret keys, and session tokens remain
-  in server memory only and are never written to that file or returned by
-  list/edit responses, so they must be entered again after restarting
-  code-viewer.
+  this package, while Cloudflare D1, Elasticsearch, S3, and DynamoDB use
+  built-in HTTP clients. No database CLI or `curl` installation is required for
+  saved connections.
+  Non-secret settings (host, endpoint, account/database id, region, …) are
+  stored in `.code-viewer/datastore-connections.json`. User names, access key
+  IDs, passwords, secret keys, session tokens, and API tokens are never written
+  to that file and are never returned by list/edit responses.
+
+  Those credentials are kept in the macOS Keychain (one generic-password item
+  per connection, service `code-viewer`) and reloaded on demand, so they
+  survive a restart without being stored anywhere in the repository. The
+  keychain item is written through `security -i`, which reads the command from
+  stdin, so the secret never appears in any process's argument list. Deleting a
+  connection also deletes its keychain item; if that fails (a locked keychain,
+  for example) the connection is still removed and the UI says the credentials
+  were left behind, so you can clear them from Keychain Access yourself.
+
+  Credentials are deliberately not encrypted into a file next to the data: a
+  file the app can decrypt unattended needs its key on the same disk, which is
+  no stronger than plaintext. Key protection is left to the OS.
+
+  On platforms without keychain integration (currently everything except
+  macOS), credentials stay in server memory only and must be entered again
+  after restarting code-viewer.
 
 - **Multi-DB tabs** — open multiple databases side by side, with their own
   sidebar, panes, and history. `+` adds an empty tab; `×` or middle-click