Path Patterns#

Simuhook supports three ways to match the request path. A stub may set exactly one of them.

Field	Semantics
`path`	Exact match on the raw URL path: `/api/users` matches `/api/users` only — never `/api/users/1` or `/api/users/1/orders`.
`path_glob`	Segment-wise glob with `*`, ``, and `?` wildcards.
`path_regex`	Go stdlib `regexp` (RE2). Not implicitly anchored — write `^…$` yourself for a full-path match.

Setting more than one is a load-time parse error naming the stub ID and the conflicting fields.

Exact Path — `path`#

id: health
path: /health
response:
  status_code: 200

Globs are matched segment-wise:

Wildcard	Matches
`**`	Zero or more path segments
`*`	Exactly one path segment
`?`	Exactly one character within a segment

id: users-by-id
method: GET
path_glob: /users/**
response:
  status_code: 200

More examples:

# Exactly one segment after /users
path_glob: /users/*

# Exactly four digits: /orders/1234
path_glob: /orders/????

Glob rules:

Trailing/leading slashes are normalized (/users/**/ ≡ /users/**).
An in-segment * (e.g. /files/*.pdf) is a load-time parse error — only whole-segment * and **, and in-segment ?, are supported.
A glob with no wildcards behaves exactly like path.

The pattern is Go RE2 and must be anchored by you — ^…$ for a full-path match:

id: orders-filter
method: POST
path_regex: ^/orders/[0-9]{5}$
response:
  status_code: 200

Request	Matches
`POST /orders/12345`	✅
`POST /orders/1234`	❌ (four digits — regex requires five)
`POST /orders/123456`	❌
`GET /orders/12345`	❌ (method constraint)

An invalid regex is a load-time parse error naming the stub ID and the pattern.

When several stubs could match the same request, specificity decides (see Matching):

exact path  >  glob  >  regex

Among globs, fewer wildcards win (/users/*/orders beats /users/**). Identical specificity falls back to load order.