sleeper_api 1.5.0 → 1.5.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 0fdd2ef6194210448caf049cdd89a37182bcd13205e45ea9543945c172d36fd4
-  data.tar.gz: 811a0991f5c863956c62c637971571aef98ccd379f9a3bff4b33ad1d28405290
+  metadata.gz: c1ac943353dae30f0fd6efa95f54f3287a85e116abd8408e843e980103557e29
+  data.tar.gz: 11569ff23125b08872ead08956c3395612a0092cf254a93450997fef069b2f19
 SHA512:
-  metadata.gz: 16f575a6d822157d44b7d5aef135626371e0fb03ed8359c70abf82f49bf6d9ae1373ff7a5ee49cbf4c32986593eec23e838b08e70ce5306e29d5fb3df5229be4
-  data.tar.gz: b7b325b17bfd0c9d921ec16dbaa07fdcf6f277d242732db1f414d11f0c5846664ac300c2fc64f2401a3ef54402c112b8a9e89651f986ba3628a5fdcabc38f394
+  metadata.gz: cf5797fea22593f99b428469ea2b99089f9f463f170c86bf7536c9a50398996d6a2edf2f06707b2ae6e3077ccce00d28c1bfe3ee2799943708f65a56090a536d
+  data.tar.gz: '08b1850e3dfb474167c1bc8df50700cab7669c7ac77c8d83ad6d757a92968ac345a82c5ecfc05b22b5238351d448782b48d297372acc33448dce0d1f4e9f1618'

data/CHANGELOG.md

@@ -1,3 +1,21 @@
+## [1.5.1] - 2026-09-25
+
+### Fixed
+
+- **A connection that never got an answer now raises `SleeperApi::Error`.**
+  Before, only non-2xx responses and timeouts did: no DNS (`SocketError`), a
+  refused, reset or unreachable connection (`SystemCallError`), a failed TLS
+  handshake (`OpenSSL::SSL::SSLError`) and a socket closed mid-response
+  (`EOFError`) escaped `make_request` as themselves. A caller rescuing
+  `SleeperApi::Error` for "Sleeper is down" missed the commonest outage of all —
+  a laptop waking before its network — and one consumer recorded 196 failed
+  imports that its retry policy never saw.
+
+  The message is `"Could not reach <path>: <original message>"`, naming the host
+  for a `WEB_HOST` call. **These are not retried** whatever `retries` is set to:
+  they fail at once, and the caller's backoff is the one that can wait long
+  enough to matter. `retries` still governs timeouts only.
+
 ## [1.5.0] - 2026-09-25
 
 ### Added

data/CLAUDE.md

@@ -28,7 +28,7 @@ Four layers, with a deliberate split between HTTP and modeling:
 - **`SleeperApi`** (`lib/sleeper_api.rb`) — module-level config + memoized global `SleeperApi.client`. `Configuration` validates `timeout` (10–60) and `retries` (0–5), raising `SleeperApi::Error` outside those bounds.
 - **`Client`** — the only thing that talks HTTP. `include HTTParty` with `base_uri "https://api.sleeper.app"` — **the bare host; the `/v1` lives in each path**. That is deliberate and load-bearing: not every Sleeper endpoint is versioned. `/schedule/{sport}/{season_type}/{season}` is served from the host root, and while `base_uri` carried the version that endpoint was unreachable at any path a caller could pass in. A new endpoint spells out where it lives; do not move the version back into `base_uri` to shorten the paths.
 
-  Every call funnels through the private `make_request`, which handles retry-on-timeout, logging, and converts non-2xx into `SleeperApi::Error`. **That error quotes the path**, so the path prefix is part of a public string — v1.2.0 changed it from `"Failed to fetch /user/x: 404"` to `"Failed to fetch /v1/user/x: 404"`. `Client` also owns the 24-hour in-memory player cache.
+  Every call funnels through the private `make_request`, which handles retry-on-timeout, logging, and converts non-2xx into `SleeperApi::Error` — and, since 1.5.1, **every connection failure too** (`CONNECTION_ERRORS`: `SocketError`, `SystemCallError`, `OpenSSL::SSL::SSLError`, `EOFError`), unretried, as `"Could not reach <path>: …"`. A consumer can rescue `SleeperApi::Error` alone and see every way a call fails. **That error quotes the path**, so the path prefix is part of a public string — v1.2.0 changed it from `"Failed to fetch /user/x: 404"` to `"Failed to fetch /v1/user/x: 404"`. `Client` also owns the 24-hour in-memory player cache.
 
   ⚠️ **There are two Sleeper hosts and v1.4.0 added the second.** `WEB_HOST`
   (`https://api.sleeper.com`) is the one the web app uses, and it serves

data/lib/sleeper_api/client.rb

@@ -1,4 +1,5 @@
 require "httparty"
+require "openssl"
 require "json"
 require "erb"
 
@@ -28,6 +29,12 @@ module SleeperApi
     # because every other endpoint here lives on `.app`.
     WEB_HOST = "https://api.sleeper.com".freeze
 
+    # The request never got an answer (1.5.1): no DNS, a refused or reset
+    # connection, a failed TLS handshake, a socket closed mid-response. Before,
+    # these escaped as themselves, so a caller rescuing SleeperApi::Error for
+    # an outage missed the commonest one. Timeouts are handled separately.
+    CONNECTION_ERRORS = [SocketError, SystemCallError, OpenSSL::SSL::SSLError, EOFError].freeze
+
     # Not HTTParty's own. Its JSON branch passes `quirks_mode`, which json 3.0
     # removed, so without this every response raises ArgumentError the moment a
     # consumer resolves json 3. See JsonParser.
@@ -516,6 +523,11 @@ module SleeperApi
           @config.logger&.error("Request timed out for #{named} after #{retries} retries")
           raise SleeperApi::Error, "Request timed out after #{retries} retries"
         end
+      rescue *CONNECTION_ERRORS => e
+        # No retry: these fail at once, and the caller's backoff is the one
+        # that can wait long enough to matter.
+        @config.logger&.error("Could not reach #{named}: #{e.message}")
+        raise SleeperApi::Error, "Could not reach #{named}: #{e.message}"
       end
     end
   end

data/lib/sleeper_api/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module SleeperApi
-  VERSION = "1.5.0"
+  VERSION = "1.5.1"
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: sleeper_api
 version: !ruby/object:Gem::Version
-  version: 1.5.0
+  version: 1.5.1
 platform: ruby
 authors:
 - Eruity1