Skip to content

release: v0.2.0 #2

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
kolkov merged 6 commits into from
Feb 4, 2026
Merged

release: v0.2.0 #2

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
326c301
bench: add three-way comparison with rivo/uniseg
kolkov Feb 4, 2026
4c7f02b
perf: SWAR optimization for ASCII fast paths
kolkov Feb 4, 2026
1ca4923
perf: replace binary search with O(1) multi-stage lookup table
kolkov Feb 4, 2026
06aa51e
feat: ZWJ emoji sequence and skin tone modifier support
kolkov Feb 4, 2026
e576f96
docs: update public documentation for v0.2.0 release
kolkov Feb 4, 2026
2f20abd
chore: update CI workflows and fix code formatting
kolkov Feb 4, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
296 changes: 296 additions & 0 deletions .github/workflows/benchmarks.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,296 @@
name: Benchmarks

# Two benchmark jobs:
# 1. Regression detection: benchstat compares base vs PR branch (same benchmarks)
# 2. Library comparison: uniwidth vs go-runewidth vs uniseg (three-way table)
#
# Note: GitHub Actions shared runners have ~10-20% variance.
# Results are directional, not absolute. Major regressions (>30%) are reliable.

on:
pull_request:

concurrency:
group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
cancel-in-progress: true

permissions:
contents: read
pull-requests: write

jobs:
# ============================================================================
# Job 1: Regression Detection (benchstat base vs PR)
# ============================================================================
regression:
name: Regression Detection
runs-on: ubuntu-latest
if: github.event.pull_request.draft == false

steps:
- name: Checkout PR branch
uses: actions/checkout@v6
with:
ref: ${{ github.event.pull_request.head.sha }}
path: pr

- name: Checkout base branch
uses: actions/checkout@v6
with:
ref: ${{ github.event.pull_request.base.sha }}
path: base

- name: Set up Go
uses: actions/setup-go@v6
with:
go-version: '1.25'
cache: false

- name: Install benchstat
run: go install golang.org/x/perf/cmd/benchstat@latest

- name: Run base branch benchmarks
working-directory: base
run: |
go test -bench=. -benchmem -count=5 -benchtime=100ms \
-run=^$ ./... 2>/dev/null > ../base-bench.txt || true

- name: Run PR branch benchmarks
working-directory: pr
run: |
go test -bench=. -benchmem -count=5 -benchtime=100ms \
-run=^$ ./... 2>/dev/null > ../pr-bench.txt || true

- name: Compare benchmarks
run: |
benchstat base-bench.txt pr-bench.txt > full-comparison.txt 2>&1 || echo "benchstat comparison failed" > full-comparison.txt

GEOMEAN=$(grep -E "^geomean" full-comparison.txt | head -1 || echo "")
REGRESSIONS=$(grep -E "\+[0-9]+\.[0-9]+%" full-comparison.txt | grep -v "~" | head -10 || echo "")

{
echo "## Regression Detection"
echo ""
echo "Comparing \`${{ github.event.pull_request.base.ref }}\` → PR #${{ github.event.pull_request.number }}"
echo ""

if [ -n "$GEOMEAN" ]; then
echo "**Summary:** \`$GEOMEAN\`"
echo ""
fi

if [ -n "$REGRESSIONS" ]; then
echo "⚠️ **Potential regressions detected:**"
echo "\`\`\`"
echo "$REGRESSIONS"
echo "\`\`\`"
echo ""
else
echo "✅ No significant regressions detected."
echo ""
fi

echo "<details>"
echo "<summary>Full benchstat output</summary>"
echo ""
echo "\`\`\`"
cat full-comparison.txt
echo "\`\`\`"
echo ""
echo "</details>"
} > regression.md

- name: Upload benchmark results
uses: actions/upload-artifact@v4
with:
name: regression-benchmarks
path: |
base-bench.txt
pr-bench.txt
full-comparison.txt
retention-days: 30

- name: Find existing regression comment
uses: peter-evans/find-comment@v3
id: fc-regression
with:
issue-number: ${{ github.event.pull_request.number }}
comment-author: 'github-actions[bot]'
body-includes: '## Regression Detection'

- name: Post regression comment
uses: peter-evans/create-or-update-comment@v4
with:
comment-id: ${{ steps.fc-regression.outputs.comment-id }}
issue-number: ${{ github.event.pull_request.number }}
body-path: regression.md
edit-mode: replace

# ============================================================================
# Job 2: Library Comparison (uniwidth vs go-runewidth vs uniseg)
# ============================================================================
comparison:
name: Library Comparison
runs-on: ubuntu-latest
if: github.event.pull_request.draft == false

steps:
- name: Checkout code
uses: actions/checkout@v6

- name: Set up Go
uses: actions/setup-go@v6
with:
go-version: '1.25'
cache: true

- name: Run three-way benchmarks
working-directory: bench
run: |
go test -bench=. -benchmem -count=3 -benchtime=100ms \
-run=^$ ./... 2>/dev/null > ../bench-raw.txt || true

- name: Generate comparison table
run: |
# Parse benchmark output and build comparison table
# Benchmark names follow: BenchmarkStringWidth_{Category}_{Size}_{Library}-N
# Extract: category, ns/op for each library

echo "## Library Comparison" >> $GITHUB_STEP_SUMMARY
echo "" >> $GITHUB_STEP_SUMMARY
echo "**Platform**: Ubuntu \$(lsb_release -rs), \$(uname -m)" >> $GITHUB_STEP_SUMMARY
echo "**CPU**: \$(grep 'model name' /proc/cpuinfo | head -1 | cut -d: -f2 | xargs)" >> $GITHUB_STEP_SUMMARY
echo "" >> $GITHUB_STEP_SUMMARY

# Build the table
{
echo "| Scenario | uniwidth | go-runewidth | uniseg | vs runewidth | vs uniseg | Winner |"
echo "|----------|----------|-------------|--------|-------------|----------|--------|"
} > table.md

# Get unique benchmark scenarios (strip library suffix and CPU count)
grep "^Benchmark" bench-raw.txt | sed 's/_Uniwidth-[0-9]*//' | sed 's/_GoRunewidth-[0-9]*//' | sed 's/_Uniseg-[0-9]*//' | awk '{print $1}' | sort -u | while read scenario; do
# Pretty name: strip "Benchmark" prefix, replace _ with spaces
pretty=$(echo "$scenario" | sed 's/^Benchmark//' | sed 's/_/ /g')

# Get median ns/op for each library (take middle value of count=3)
uni_ns=$(grep "^${scenario}_Uniwidth-" bench-raw.txt | awk '{print $3}' | sort -n | head -2 | tail -1)
rw_ns=$(grep "^${scenario}_GoRunewidth-" bench-raw.txt | awk '{print $3}' | sort -n | head -2 | tail -1)
seg_ns=$(grep "^${scenario}_Uniseg-" bench-raw.txt | awk '{print $3}' | sort -n | head -2 | tail -1)

# Skip if uniwidth result is missing
[ -z "$uni_ns" ] && continue

# Format times with units
uni_fmt="${uni_ns} ns"
rw_fmt="${rw_ns:-—} ns"
seg_fmt="${seg_ns:-—} ns"

# Calculate vs go-runewidth speedup
vs_rw="—"
if [ -n "$rw_ns" ] && [ "$uni_ns" != "0" ]; then
speedup=$(echo "scale=1; $rw_ns / $uni_ns" | bc 2>/dev/null || echo "")
if [ -n "$speedup" ]; then
# Check if speedup >= 2
is_fast=$(echo "$speedup >= 2" | bc 2>/dev/null || echo "0")
if [ "$is_fast" = "1" ]; then
vs_rw="**${speedup}x**"
else
vs_rw="${speedup}x"
fi
fi
fi

# Calculate vs uniseg speedup
vs_seg="—"
if [ -n "$seg_ns" ] && [ "$uni_ns" != "0" ]; then
speedup=$(echo "scale=1; $seg_ns / $uni_ns" | bc 2>/dev/null || echo "")
if [ -n "$speedup" ]; then
is_fast=$(echo "$speedup >= 2" | bc 2>/dev/null || echo "0")
if [ "$is_fast" = "1" ]; then
vs_seg="**${speedup}x**"
else
vs_seg="${speedup}x"
fi
fi
fi

# Determine winner
winner="uniwidth"
min_ns="$uni_ns"
if [ -n "$rw_ns" ]; then
rw_faster=$(echo "$rw_ns < $min_ns" | bc 2>/dev/null || echo "0")
[ "$rw_faster" = "1" ] && winner="go-runewidth" && min_ns="$rw_ns"
fi
if [ -n "$seg_ns" ]; then
seg_faster=$(echo "$seg_ns < $min_ns" | bc 2>/dev/null || echo "0")
[ "$seg_faster" = "1" ] && winner="uniseg"
fi

# Bold the winner's time
if [ "$winner" = "uniwidth" ]; then
uni_fmt="**${uni_ns} ns**"
elif [ "$winner" = "go-runewidth" ]; then
rw_fmt="**${rw_ns} ns**"
elif [ "$winner" = "uniseg" ]; then
seg_fmt="**${seg_ns} ns**"
fi

echo "| ${pretty} | ${uni_fmt} | ${rw_fmt} | ${seg_fmt} | ${vs_rw} | ${vs_seg} | ${winner} |" >> table.md
done

# Append table to summary
cat table.md >> $GITHUB_STEP_SUMMARY
echo "" >> $GITHUB_STEP_SUMMARY

# Raw output in details
echo "<details>" >> $GITHUB_STEP_SUMMARY
echo "<summary>Raw benchmark output</summary>" >> $GITHUB_STEP_SUMMARY
echo "" >> $GITHUB_STEP_SUMMARY
echo '```' >> $GITHUB_STEP_SUMMARY
cat bench-raw.txt >> $GITHUB_STEP_SUMMARY
echo '```' >> $GITHUB_STEP_SUMMARY
echo "</details>" >> $GITHUB_STEP_SUMMARY

# Build PR comment (table + note)
{
echo "## Library Comparison"
echo ""
cat table.md
echo ""
echo "<details>"
echo "<summary>Raw benchmark output</summary>"
echo ""
echo "\`\`\`"
cat bench-raw.txt
echo "\`\`\`"
echo ""
echo "</details>"
echo ""
echo "> CI runners have ~10-20% variance. For accurate results, run locally: \`cd bench && go test -bench=. -benchmem -count=10\`"
} > comparison.md

- name: Upload comparison results
uses: actions/upload-artifact@v4
with:
name: library-comparison
path: |
bench-raw.txt
table.md
retention-days: 30

- name: Find existing comparison comment
uses: peter-evans/find-comment@v3
id: fc-comparison
with:
issue-number: ${{ github.event.pull_request.number }}
comment-author: 'github-actions[bot]'
body-includes: '## Library Comparison'

- name: Post comparison comment
uses: peter-evans/create-or-update-comment@v4
with:
comment-id: ${{ steps.fc-comparison.outputs.comment-id }}
issue-number: ${{ github.event.pull_request.number }}
body-path: comparison.md
edit-mode: replace
Loading
Loading