Fix browser-like symbol runs in long words

Broaden the no-space punctuation preprocessing into a word-internal symbol pass so keyboard and Unicode symbols stay with surrounding long-word text while browser-break symbols stay breakable.

Add a compact Chrome/Safari symbol-run oracle and refresh accuracy/benchmark dashboards.

Refs #169

Author: chenglou

CHANGELOG.md

@@ -10,7 +10,7 @@
 ### Fixed
 
 - `{ wordBreak: 'keep-all' }` now handles no-space mixed Latin, numeric, and CJK text more like browsers.
-- No-space punctuation chains now stay together for non-ASCII word-like text too, instead of only ASCII words.
+- No-space punctuation and symbol chains now stay together for non-ASCII word-like text too, instead of only ASCII words (#169).
 - Opening punctuation such as `¡`, `¿`, German low quotes, and `⸘` now stays with the following word instead of dangling at line end (#165).
 - Numeric prefix/postfix symbols like `$`, `%`, `€`, `+`, `−`, and `°` now stay attached to adjacent text the way browser line breaking does (#105).
 - Soft-hyphen breaks now stay at the soft-hyphen insertion point instead of pulling post-hyphen graphemes onto the broken line (#162).

DEVELOPMENT.md

@@ -34,6 +34,7 @@ bun install
 - `bun run benchmark-check:safari`
 - `bun run pre-wrap-check` — compact batched browser oracle for `{ whiteSpace: 'pre-wrap' }`
 - `bun run keep-all-check` — compact batched browser oracle for `{ wordBreak: 'keep-all' }`, including mixed-script no-space canaries
+- `bun run symbol-check` — compact batched Chrome + Safari oracle for no-space symbol runs inside long words
 - `bun run letter-spacing-check` — compact batched browser oracle for `{ letterSpacing }`, using one posted-report probe per browser and covering narrow wraps, combining marks, bidi, CJK, emoji, digits, RTL punctuation, `pre-wrap`, and soft hyphens
 - `bun run letter-spacing-snapshot` — refresh `accuracy/letter-spacing.json` from the Chrome + Safari compact `{ letterSpacing }` oracle
 - `bun run probe-check` — smaller browser probe/diagnostic entrypoint

accuracy/chrome.json

@@ -76825,5 +76825,5 @@
       "diff": 0
     }
   ],
-  "requestId": "1778455134064-dzaeth"
+  "requestId": "1778458451443-udrw2b"
 }

accuracy/firefox.json

@@ -76825,5 +76825,5 @@
       "diff": 0
     }
   ],
-  "requestId": "1778455145984-0ujr0y"
+  "requestId": "1778458462178-8bdtvw"
 }

accuracy/safari.json

@@ -76825,5 +76825,5 @@
       "diff": 0
     }
   ],
-  "requestId": "1778455139711-8jiaqg"
+  "requestId": "1778458457473-esv0s5"
 }

benchmarks/chrome.json

@@ -3,22 +3,22 @@
   "results": [
     {
       "label": "Our library: prepare()",
-      "ms": 17.75,
+      "ms": 17.050000071525574,
       "desc": "One cold 500-text measurement batch"
     },
     {
       "label": "Our library: layout()",
-      "ms": 0.14424999952316286,
+      "ms": 0.1440000003576279,
       "desc": "Normalized hot-path throughput per 500-text batch"
     },
     {
       "label": "DOM batch",
-      "ms": 3.9000000953674316,
+      "ms": 3.8000000715255737,
       "desc": "Single 400→300px batch resize: write all, then read all"
     },
     {
       "label": "DOM interleaved",
-      "ms": 44.59999990463257,
+      "ms": 43.25000011920929,
       "desc": "Single 400→300px batch resize: write + read per div"
     }
   ],
@@ -30,7 +30,7 @@
     },
     {
       "label": "Our library: layoutWithLines()",
-      "ms": 0.0550000011920929,
+      "ms": 0.05250000059604645,
       "desc": "60-text shared-corpus batch across widths 180/220/260px; materializes text lines"
     },
     {
@@ -57,7 +57,7 @@
     },
     {
       "label": "Our library: materializeRichInlineLineRange()",
-      "ms": 0.0449999988079071,
+      "ms": 0.04375,
       "desc": "36 rich-text inline flow shared-corpus texts across widths 180/220/260px; range walker plus per-line materialization"
     },
     {
@@ -69,44 +69,44 @@
   "richPreWrapResults": [
     {
       "label": "Our library: measureLineStats()",
-      "ms": 0.3074999988079071,
+      "ms": 0.3024999976158142,
       "desc": "12 generated pre-wrap texts with 320 hard-break chunks across widths 220/260/320px; stats only"
     },
     {
       "label": "Our library: layoutWithLines()",
-      "ms": 0.42000000476837157,
+      "ms": 0.4149999976158142,
       "desc": "12 generated pre-wrap texts with 320 hard-break chunks across widths 220/260/320px; materializes text lines"
     },
     {
       "label": "Our library: walkLineRanges()",
-      "ms": 0.24499999284744262,
+      "ms": 0.2399999976158142,
       "desc": "12 generated pre-wrap texts with 320 hard-break chunks across widths 220/260/320px; line ranges only, no line text strings"
     },
     {
       "label": "Our library: layoutNextLine()",
-      "ms": 0.9724999964237213,
+      "ms": 0.9475000023841857,
       "desc": "12 generated pre-wrap texts with 320 hard-break chunks across widths 220/260/320px; streaming line-by-line layout"
     }
   ],
   "richLongResults": [
     {
       "label": "Our library: measureLineStats()",
-      "ms": 1.6787499994039536,
+      "ms": 1.6599999994039536,
       "desc": "8 Arabic long-form texts across widths 240/300/360px; stats only"
     },
     {
       "label": "Our library: layoutWithLines()",
-      "ms": 5.534999996423721,
+      "ms": 5.548750001192093,
       "desc": "8 Arabic long-form texts across widths 240/300/360px; materializes text lines"
     },
     {
       "label": "Our library: walkLineRanges()",
-      "ms": 2.23125,
+      "ms": 2.2125,
       "desc": "8 Arabic long-form texts across widths 240/300/360px; line ranges only, no line text strings"
     },
     {
       "label": "Our library: layoutNextLine()",
-      "ms": 5.313749998807907,
+      "ms": 5.21875,
       "desc": "8 Arabic long-form texts across widths 240/300/360px; streaming line-by-line layout"
     }
   ],
@@ -122,9 +122,9 @@
       "width": 300,
       "lineCount": 193,
       "analysisMs": 2.5999999046325684,
-      "measureMs": 4.299999713897705,
-      "prepareMs": 6.800000190734863,
-      "layoutMs": 0.02649999976158142
+      "measureMs": 4.099999904632568,
+      "prepareMs": 6.700000047683716,
+      "layoutMs": 0.025499999523162842
     },
     {
       "id": "ja-rashomon",
@@ -136,10 +136,10 @@
       "breakableSegments": 1,
       "width": 300,
       "lineCount": 380,
-      "analysisMs": 4.900000095367432,
-      "measureMs": 8.5,
-      "prepareMs": 13.600000143051147,
-      "layoutMs": 0.049000000953674315
+      "analysisMs": 5.199999809265137,
+      "measureMs": 8.200000286102295,
+      "prepareMs": 13.599999904632568,
+      "layoutMs": 0.04850000023841858
     },
     {
       "id": "ko-unsu-joh-eun-nal",
@@ -151,10 +151,10 @@
       "breakableSegments": 0,
       "width": 300,
       "lineCount": 428,
-      "analysisMs": 3.4000000953674316,
-      "measureMs": 8.499999761581421,
-      "prepareMs": 12.200000047683716,
-      "layoutMs": 0.07850000023841858
+      "analysisMs": 3,
+      "measureMs": 9,
+      "prepareMs": 11.799999952316284,
+      "layoutMs": 0.07699999928474427
     },
     {
       "id": "ko-sonagi",
@@ -166,25 +166,25 @@
       "breakableSegments": 2,
       "width": 300,
       "lineCount": 332,
-      "analysisMs": 2.5999999046325684,
-      "measureMs": 6.599999904632568,
-      "prepareMs": 9.099999904632568,
-      "layoutMs": 0.05950000047683716
+      "analysisMs": 2.299999952316284,
+      "measureMs": 6.299999952316284,
+      "prepareMs": 8.800000190734863,
+      "layoutMs": 0.05850000023841858
     },
     {
       "id": "zh-zhufu",
       "label": "Chinese prose",
       "font": "20px \"Songti SC\", \"PingFang SC\", \"Noto Serif CJK SC\", serif",
       "chars": 9428,
-      "analysisSegments": 5433,
-      "segments": 7949,
-      "breakableSegments": 4,
+      "analysisSegments": 5428,
+      "segments": 7944,
+      "breakableSegments": 2,
       "width": 300,
       "lineCount": 626,
-      "analysisMs": 7.900000095367432,
-      "measureMs": 14.399999856948853,
-      "prepareMs": 22.100000143051147,
-      "layoutMs": 0.08
+      "analysisMs": 8,
+      "measureMs": 14.400000095367432,
+      "prepareMs": 22.200000047683716,
+      "layoutMs": 0.07800000071525574
     },
     {
       "id": "zh-guxiang",
@@ -196,85 +196,85 @@
       "breakableSegments": 14,
       "width": 300,
       "lineCount": 375,
-      "analysisMs": 4.700000047683716,
-      "measureMs": 8.900000095367432,
-      "prepareMs": 13.799999952316284,
-      "layoutMs": 0.04850000023841858
+      "analysisMs": 4.799999952316284,
+      "measureMs": 7.8999998569488525,
+      "prepareMs": 12.700000047683716,
+      "layoutMs": 0.04700000047683716
     },
     {
       "id": "th-nithan-vetal-story-1",
       "label": "Thai prose",
       "font": "20px \"Thonburi\", \"Noto Sans Thai\", sans-serif",
       "chars": 34033,
-      "analysisSegments": 10281,
-      "segments": 10281,
+      "analysisSegments": 10279,
+      "segments": 10279,
       "breakableSegments": 8087,
       "width": 300,
       "lineCount": 1024,
-      "analysisMs": 10.399999856948853,
-      "measureMs": 7.699999809265137,
-      "prepareMs": 18.299999952316284,
-      "layoutMs": 0.08950000047683716
+      "analysisMs": 10.899999856948853,
+      "measureMs": 7.3999998569488525,
+      "prepareMs": 18.799999952316284,
+      "layoutMs": 0.08649999976158142
     },
     {
       "id": "my-cunning-heron-teacher",
       "label": "Myanmar prose",
       "font": "20px \"Myanmar MN\", \"Myanmar Sangam MN\", \"Noto Sans Myanmar\", serif",
       "chars": 2707,
-      "analysisSegments": 797,
-      "segments": 797,
-      "breakableSegments": 422,
+      "analysisSegments": 786,
+      "segments": 786,
+      "breakableSegments": 420,
       "width": 300,
       "lineCount": 81,
       "analysisMs": 0.7999999523162842,
-      "measureMs": 0.9999997615814209,
-      "prepareMs": 1.7999999523162842,
+      "measureMs": 0.7000002861022949,
+      "prepareMs": 1.6000001430511475,
       "layoutMs": 0.0070000004768371584
     },
     {
       "id": "my-bad-deeds-return-to-you-teacher",
       "label": "Myanmar prose (story 2)",
       "font": "20px \"Myanmar MN\", \"Myanmar Sangam MN\", \"Noto Sans Myanmar\", serif",
       "chars": 1666,
-      "analysisSegments": 498,
-      "segments": 498,
-      "breakableSegments": 274,
+      "analysisSegments": 491,
+      "segments": 491,
+      "breakableSegments": 271,
       "width": 300,
       "lineCount": 54,
       "analysisMs": 0.5,
-      "measureMs": 0.6000001430511475,
-      "prepareMs": 1.1999998092651367,
-      "layoutMs": 0.004500000476837158
+      "measureMs": 0.5999999046325684,
+      "prepareMs": 1,
+      "layoutMs": 0.004499999284744263
     },
     {
       "id": "ur-chughd",
       "label": "Urdu prose",
       "font": "20px \"Noto Nastaliq Urdu\", \"DecoType Nastaleeq Urdu UI\", \"Geeza Pro\", serif",
       "chars": 13391,
-      "analysisSegments": 6051,
-      "segments": 6051,
-      "breakableSegments": 2922,
+      "analysisSegments": 5841,
+      "segments": 5841,
+      "breakableSegments": 2916,
       "width": 300,
-      "lineCount": 351,
-      "analysisMs": 3.700000047683716,
-      "measureMs": 4.1000001430511475,
-      "prepareMs": 8,
-      "layoutMs": 0.04350000023841858
+      "lineCount": 352,
+      "analysisMs": 3.4000000953674316,
+      "measureMs": 4.299999952316284,
+      "prepareMs": 7.799999952316284,
+      "layoutMs": 0.04
     },
     {
       "id": "km-prachum-reuang-preng-khmer-volume-7-stories-1-10",
       "label": "Khmer prose",
       "font": "20px \"Khmer Sangam MN\", \"Khmer MN\", \"Noto Sans Khmer\", serif",
       "chars": 24240,
-      "analysisSegments": 11109,
-      "segments": 11109,
-      "breakableSegments": 4010,
+      "analysisSegments": 11107,
+      "segments": 11107,
+      "breakableSegments": 4011,
       "width": 300,
       "lineCount": 591,
-      "analysisMs": 8.599999904632568,
-      "measureMs": 5,
-      "prepareMs": 13.799999952316284,
-      "layoutMs": 0.08600000023841858
+      "analysisMs": 6.900000095367432,
+      "measureMs": 5.6000001430511475,
+      "prepareMs": 12.5,
+      "layoutMs": 0.08400000095367431
     },
     {
       "id": "hi-eidgah",
@@ -286,10 +286,10 @@
       "breakableSegments": 4089,
       "width": 300,
       "lineCount": 653,
-      "analysisMs": 6.099999904632568,
-      "measureMs": 8.000000238418579,
-      "prepareMs": 14.599999904632568,
-      "layoutMs": 0.07049999952316284
+      "analysisMs": 5.1000001430511475,
+      "measureMs": 7.999999761581421,
+      "prepareMs": 13.200000047683716,
+      "layoutMs": 0.06949999928474426
     },
     {
       "id": "synthetic-long-breakable-runs",
@@ -301,25 +301,25 @@
       "breakableSegments": 1540,
       "width": 300,
       "lineCount": 2860,
-      "analysisMs": 8.599999904632568,
-      "measureMs": 15.5,
-      "prepareMs": 24,
-      "layoutMs": 0.39799999952316284
+      "analysisMs": 8.400000095367432,
+      "measureMs": 14.900000095367432,
+      "prepareMs": 23.200000047683716,
+      "layoutMs": 0.3960000002384186
     },
     {
       "id": "ar-risalat-al-ghufran-part-1",
       "label": "Arabic prose",
       "font": "20px \"Geeza Pro\", \"Noto Naskh Arabic\", \"Arial\", serif",
       "chars": 106857,
-      "analysisSegments": 37603,
-      "segments": 37603,
-      "breakableSegments": 18745,
+      "analysisSegments": 37601,
+      "segments": 37601,
+      "breakableSegments": 18743,
       "width": 300,
       "lineCount": 2643,
-      "analysisMs": 29.100000143051147,
-      "measureMs": 45,
-      "prepareMs": 73.10000014305115,
-      "layoutMs": 0.27700000047683715
+      "analysisMs": 23.59999990463257,
+      "measureMs": 45.39999985694885,
+      "prepareMs": 69.39999985694885,
+      "layoutMs": 0.2689999997615814
     }
   ]
 }