ref(cli): Parse event view arguments with the shared ID grammar

packages/cli/AGENTS.md

@@ -228,8 +228,8 @@ export const myRoute = buildRouteMap({
 ### Positional Arguments
 
 Required identifiers (trace IDs, span IDs) should be **positional args**, not
-flags. Every command that takes trace, span, replay, or log IDs parses them
-with `parseIdArguments` (or `parseIdArgument` for exactly one) from
+flags. Every command that takes trace, span, replay, log, or event IDs parses
+them with `parseIdArguments` (or `parseIdArgument` for exactly one) from
 `src/lib/id-arguments.ts`, so they all take the same forms:
 
 - `<id>`, `<org>/<id>`, `<org>/<project>/<id>`, `<target> <id>`, a Sentry URL
@@ -248,8 +248,8 @@ target is known. The result is a `target` string for `parseOrgProjectInput`
 and the IDs. Add a kind to `ID_KINDS` rather than validating IDs in a command;
 the `sentry-cli/no-ad-hoc-id-arguments` lint rule rejects hand-written ID
 parsing (`parseSlashSeparatedArg`, `HEX_ID_RE`, ...) in commands and kinds
-defined outside `ID_KINDS`. Issue IDs (and `event view`, which takes them)
-keep their own syntax.
+defined outside `ID_KINDS`. Issue IDs keep their own syntax; `event view`,
+which also takes them, checks for an issue first (`parseEventViewInput`).
 
 **Issue positionals:** issue IDs, short IDs, selectors, and issue URLs are
 parsed by `parseIssueArgument` (`src/lib/issue-arguments.ts`), whose forms
@@ -328,7 +328,7 @@ command's help instead. The same purity rule applies: return a pasted URL's serv
 (`baseUrl`) and any warnings from the parse, and apply them in `resolve`
 (`parseTraceTargetInput` and `parseDualModeInput` in `trace-target.ts` do
 this for trace targets). Commands that recover a malformed ID (`trace view`,
-`span view`, `log view`) parse it unvalidated and validate or recover it in
+`span view`, `log view`, `event view`) parse it unvalidated and validate or recover it in
 `resolve`, which can look the prefix up: `parseTraceArgumentsInput` and
 `resolveTraceArguments` do this for trace IDs. A missing ID or a bad target
 still fails the parse.

packages/cli/grammar/id-arguments.pegjs

@@ -53,11 +53,11 @@ arguments
     }
   // Any other URL is a target
   / sep t:url x:more { return { target: t, ids: x }; }
-  // `<id> <org>/<project>`: the arguments are reversed. When IDs are
-  // optional, only a real ID can come first.
+  // `<id> <org>/<project> [<id>...]`: the arguments are reversed. When IDs
+  // are optional, only a real ID can come first.
   / sep i:segment &{ return !optional || isId(i); } sep !idPath
-    t:$([^/\0]* "/" [^\0]*) end
-    { return { target: t, ids: [{ text: i, id: i }], reversed: true }; }
+    t:$([^/\0]* "/" [^\0]*) x:more
+    { return { target: t, ids: [{ text: i, id: i }, ...x], reversed: true }; }
   // `[<org>/[<project>/]]<id> [<id>...]`
   / sep p:idPath x:more {
       return {

packages/cli/lint-rules/cli-oxlint-plugin.js

@@ -66,6 +66,7 @@ const BOUND_TARGET_TYPES = new Set([
   "OrgProjectTarget",
   "ProjectBoundTarget",
   "BoundArguments",
+  "EventViewArguments",
   "LogListArguments",
   "LogViewArguments",
   "ReplayArguments",
@@ -165,12 +166,6 @@ const AD_HOC_ID_PARSING = new Set([
   "SPAN_ID_RE",
 ]);
 
-/**
- * Commands still allowed to: `event view` also takes issue IDs, which are
- * not hex IDs and have their own syntax.
- */
-const AD_HOC_ID_PARSING_ALLOWED = ["/src/commands/event/view.ts"];
-
 const ID_ARGUMENT_PARSERS = new Set(["parseIdArguments", "parseIdArgument"]);
 
 /** Whether `node` is `ID_KINDS.<kind>`. */
@@ -355,9 +350,7 @@ const rules = {
     (_context, file, report) => {
       if (!file.includes("/src/") || file.endsWith("/lib/id-arguments.ts"))
         return {};
-      const commands =
-        file.includes("/src/commands/") &&
-        !AD_HOC_ID_PARSING_ALLOWED.some((allowed) => file.endsWith(allowed));
+      const commands = file.includes("/src/commands/");
       return {
         Identifier(node) {
           if (