From 8fc4440150ce84c9cc3f5ec13d50beb58a5b65bc Mon Sep 17 00:00:00 2001
From: srdusr <99972264+srdusr@users.noreply.github.com>
Date: Sun, 7 Dec 2025 21:58:00 +0200
Subject: Rebuild the generic packs, add shell and sysadmin, fix short race
 passages

The packs were not proper. An audit found 86 of 253 items (33%) carrying an
attribution that just restated the category - prose *about* a topic with an
invented source, which is the same fault the movies pack had. Six of thirteen
packs were mostly that: technology had 15 items and one distinct attribution.

- science, technology, history, nature and business are now sourced quotes
  with real attributions: Feynman, Curie, Hopper, Dijkstra, Lincoln, Carson,
  Drucker, Goodhart. 82 items, all attributed to a person or a work.
- general is original factual prose, so it now carries no attribution at all
  rather than claiming "General knowledge" as a source. merge_packs no longer
  invents one from the pack's filename.
- Four explanatory passages in philosophy lost their "Philosophy" attribution
  for the same reason.
- One duplicated passage removed.

Generic attributions: 86/253 before, 0/349 now.

New technical packs
- shell: 24 awk, sed and pipeline drills, each explaining what the line does
  - field splitting, associative arrays, !seen[$0]++, process substitution,
  xargs -0, strict mode.
- sysadmin: 24 operational one-liners across systemd, disk, processes,
  network, permissions, SSH, backup and containers.
- programming grew to 37 and hacking to 30, with git bisect, window
  functions, EXPLAIN ANALYZE, certificate transparency and capability audits.
- All 115 technical drills carry an explanation.

Race passage length
- Multiplayer drew from the same pool as single player, so a race could land
  on a 22-character quote and be over before anyone had their hands in
  position. Races now require 120 characters; the filter is applied when the
  pool is loaded, not to the packs, since a short quote is fine to type alone.

For reference, TypeRacer organises by difficulty and language rather than
topic: one default English pool of ~11,900 texts plus per-language universes
and specials (accuracy, repeat, easytexts, anime). Their scale comes from user
submission with moderation, which is still the feature this does not have.
---
 crates/server/src/main.rs | 39 +++++++++++++++++++++++++++++++--------
 1 file changed, 31 insertions(+), 8 deletions(-)

(limited to 'crates/server')

diff --git a/crates/server/src/main.rs b/crates/server/src/main.rs
index e9133b9..0ce169d 100644
--- a/crates/server/src/main.rs
+++ b/crates/server/src/main.rs
@@ -36,19 +36,42 @@ struct TextEntry {
 // falls back to a couple of plain sentences if the file isn't reachable
 // (e.g. the server binary run from somewhere other than the repo root),
 // so a room can still start rather than erroring on an empty pool.
+/// Shortest passage a multiplayer race will use.
+const MIN_RACE_TEXT_CHARS: usize = 120;
+
+fn fallback_race_texts() -> Vec<RaceText> {
+    vec![
+        RaceText {
+            text: "The quick brown fox jumps over the lazy dog, and the dog, being lazy, does not mind at all. Pack my box with five dozen liquor jugs.".to_string(),
+            attribution: None,
+        },
+    ]
+}
+
 fn load_race_texts() -> Vec<RaceText> {
     let path = std::env::var("TEXTS_JSON_PATH").unwrap_or_else(|_| "texts.json".to_string());
     match std::fs::read_to_string(&path).ok().and_then(|raw| serde_json::from_str::<Vec<TextEntry>>(&raw).ok()) {
-        Some(entries) if !entries.is_empty() => entries
-            .into_iter()
-            .map(|e| RaceText { text: e.content, attribution: e.attribution })
-            .collect(),
+        Some(entries) if !entries.is_empty() => {
+            let pool: Vec<RaceText> = entries
+                .into_iter()
+                // A race on a 22-character quote is over before anyone has
+                // their hands in position. The dataset is shared with single
+                // player, where a short quote is fine, so the floor is applied
+                // here rather than to the pack itself.
+                .filter(|e| e.content.chars().count() >= MIN_RACE_TEXT_CHARS)
+                .map(|e| RaceText { text: e.content, attribution: e.attribution })
+                .collect();
+            if pool.is_empty() {
+                tracing::warn!("no passage reached {MIN_RACE_TEXT_CHARS} characters - races will use the fallback pool");
+                fallback_race_texts()
+            } else {
+                tracing::info!("{} passages available for races", pool.len());
+                pool
+            }
+        }
         _ => {
             tracing::warn!("could not load race texts from {path} - using a small built-in fallback pool");
-            vec![
-                RaceText { text: "The quick brown fox jumps over the lazy dog.".to_string(), attribution: None },
-                RaceText { text: "Pack my box with five dozen liquor jugs.".to_string(), attribution: None },
-            ]
+            fallback_race_texts()
         }
     }
 }
-- 
cgit v1.2.3