From 7f353e925b7dbd6d0c16962e0e822aea0093308b Mon Sep 17 00:00:00 2001
From: Craig Jennings <c@cjennings.net>
Date: Mon, 11 May 2026 17:17:54 -0500
Subject: perf(lorem-optimum): speed up the Markov generation path

`cj/markov-join-tokens' collects tokens in a list and `mapconcat's once instead of repeated string concatenation. `cj/markov-generate' uses `push'/`nreverse' instead of repeated `append'. The Markov keys are cached as a vector so random key selection is O(1). Re-enabled the benchmark tests (the `:slow' tags were stale) and added a `cj/lipsum-title' test after byte-compilation flagged a malformed form there. `assets/liber-primus.txt' is left as-is (36 KB / 5,374 words, small enough not to need trimming). 100K-word learning now measures about 196 ms.
---
 tests/test-lorem-optimum-benchmark.el | 15 ++++-----------
 tests/test-lorem-optimum.el           |  9 +++++++++
 2 files changed, 13 insertions(+), 11 deletions(-)

(limited to 'tests')

diff --git a/tests/test-lorem-optimum-benchmark.el b/tests/test-lorem-optimum-benchmark.el
index 57d5ae5f..6b2f0163 100644
--- a/tests/test-lorem-optimum-benchmark.el
+++ b/tests/test-lorem-optimum-benchmark.el
@@ -63,10 +63,7 @@
     (should (< time 50.0))))  ; Should be < 50ms
 
 (ert-deftest benchmark-learn-10k-words ()
-  "Benchmark learning 10,000 words.
-DISABLED: Takes too long (minutes instead of seconds).
-Needs lorem-optimum performance optimization before re-enabling."
-  :tags '(:slow)
+  "Benchmark learning 10,000 words."
   (let* ((text (generate-test-text 10000))
          (chain (cj/markov-chain-create))
          (time (benchmark-time
@@ -76,7 +73,6 @@ Needs lorem-optimum performance optimization before re-enabling."
 
 (ert-deftest benchmark-learn-100k-words ()
   "Benchmark learning 100,000 words (stress test)."
-  :tags '(:slow)
   (let* ((text (generate-test-text 100000))
          (chain (cj/markov-chain-create))
          (time (benchmark-time
@@ -84,7 +80,8 @@ Needs lorem-optimum performance optimization before re-enabling."
     (benchmark-report "Learn 100K words" time)
     ;; This may be slow due to key rebuild
     (message "Hash table size: %d bigrams"
-             (hash-table-count (cj/markov-chain-map chain)))))
+             (hash-table-count (cj/markov-chain-map chain)))
+    (should (< time 5000.0))))
 
 ;;; Multiple Learning Operations (Exposes Quadratic Behavior)
 
@@ -109,7 +106,6 @@ Needs lorem-optimum performance optimization before re-enabling."
 
 (ert-deftest benchmark-multiple-learns-100x100 ()
   "Benchmark 100 learn operations of 100 words each (key rebuild overhead)."
-  :tags '(:slow)
   (let ((chain (cj/markov-chain-create))
         (times '())
         (measurements '()))
@@ -154,10 +150,7 @@ Needs lorem-optimum performance optimization before re-enabling."
 ;;; Tokenization Performance Tests
 
 (ert-deftest benchmark-tokenize-10k-words ()
-  "Benchmark tokenizing 10,000 words.
-DISABLED: Takes too long (minutes instead of seconds).
-Needs lorem-optimum performance optimization before re-enabling."
-  :tags '(:slow)
+  "Benchmark tokenizing 10,000 words."
   (let* ((text (generate-test-text 10000))
          (time (benchmark-time
                 (lambda () (cj/markov-tokenize text)))))
diff --git a/tests/test-lorem-optimum.el b/tests/test-lorem-optimum.el
index ca2e52f4..40bdc684 100644
--- a/tests/test-lorem-optimum.el
+++ b/tests/test-lorem-optimum.el
@@ -238,5 +238,14 @@
     (let ((result (cj/markov-generate chain 2)))
       (should (stringp result)))))
 
+(ert-deftest test-title-generation-produces-title ()
+  "Should generate a non-empty title from the global chain."
+  (let ((cj/lipsum-chain
+         (test-learn "lorem ipsum dolor sit amet consectetur adipiscing elit")))
+    (let ((result (cj/lipsum-title)))
+      (should (stringp result))
+      (should (> (length result) 0))
+      (should (string-match-p "^[[:upper:]]" result)))))
+
 (provide 'test-lorem-optimum)
 ;;; test-lorem-optimum.el ends here
-- 
cgit v1.2.3