shlesha 0.5.7

High-performance extensible transliteration library with hub-and-spoke architecture
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257

#!/usr/bin/env python3
"""
Generate a clean, consolidated benchmark report from all benchmark results.
Produces a single markdown file with just data - no commentary.
"""

import os
import csv
import json
from datetime import datetime

def read_csv_results(filename):
    """Read CSV benchmark results"""
    results = []
    if os.path.exists(filename):
        with open(filename, 'r') as f:
            reader = csv.DictReader(f)
            results = list(reader)
    return results

def format_number(num_str):
    """Format number string for display"""
    try:
        return f"{float(num_str):,.0f}"
    except:
        return num_str

def generate_clean_report():
    """Generate consolidated benchmark report"""
    report = []
    report.append(f"# Shlesha Benchmark Results")
    report.append(f"*Generated: {datetime.now().strftime('%Y-%m-%d %H:%M:%S')}*")
    report.append("")
    
    # Rust/Native benchmarks
    rust_results = []
    for category in ["hub", "standard", "extension", "cross_category"]:
        rust_results.extend(read_csv_results(f"target/benchmark_results_{category}.csv"))
    
    if rust_results:
        report.append("## Rust Native Performance")
        report.append("")
        report.append("### Hub Scripts (Devanagari ↔ ISO-15919)")
        report.append("| From | To | Text Size | Throughput (chars/sec) | Latency (ns) |")
        report.append("|------|----|-----------|-----------------------|-------------|")
        
        for r in rust_results:
            if r.get('category') == 'hub':
                report.append(f"| {r['script_from']} | {r['script_to']} | {r['text_size']} | {format_number(r['throughput_chars_per_sec'])} | {format_number(r['latency_ns'])} |")
        
        report.append("")
        report.append("### Standard Indic Scripts")
        report.append("| From | To | Text Size | Throughput (chars/sec) | Latency (ns) |")
        report.append("|------|----|-----------|-----------------------|-------------|")
        
        for r in rust_results:
            if r.get('category') == 'standard':
                report.append(f"| {r['script_from']} | {r['script_to']} | {r['text_size']} | {format_number(r['throughput_chars_per_sec'])} | {format_number(r['latency_ns'])} |")
        
        report.append("")
        report.append("### Extension Scripts (Roman/ASCII)")
        report.append("| From | To | Text Size | Throughput (chars/sec) | Latency (ns) |")
        report.append("|------|----|-----------|-----------------------|-------------|")
        
        for r in rust_results:
            if r.get('category') == 'extension':
                report.append(f"| {r['script_from']} | {r['script_to']} | {r['text_size']} | {format_number(r['throughput_chars_per_sec'])} | {format_number(r['latency_ns'])} |")
    
    # Python benchmarks
    python_results = read_csv_results("target/python_benchmark_results.csv")
    
    if python_results:
        report.append("")
        report.append("## Python API Performance")
        report.append("")
        
        # Group by API type
        api_types = set(r['api_type'] for r in python_results)
        
        for api_type in sorted(api_types):
            report.append(f"### {api_type.replace('_', ' ').title()}")
            report.append("| From | To | Category | Text Size | Throughput (chars/sec) | Latency (ns) |")
            report.append("|------|-----|----------|-----------|------------------------|-------------|")
            
            for r in python_results:
                if r['api_type'] == api_type:
                    report.append(f"| {r['script_from']} | {r['script_to']} | {r['category']} | {r['text_size']} | {format_number(r['throughput_chars_per_sec'])} | {format_number(r['latency_ns'])} |")
            report.append("")
    
    # Comparison benchmarks
    comparison_results = read_csv_results("target/comparison_benchmark_results.csv")
    
    if comparison_results:
        report.append("")
        report.append("## Library Comparison")
        report.append("")
        
        # Group by conversion
        conversions = {}
        for r in comparison_results:
            if r.get('success') == 'True':
                key = f"{r['from_script']}_{r['to_script']}"
                if key not in conversions:
                    conversions[key] = []
                conversions[key].append(r)
        
        for conversion, results in sorted(conversions.items()):
            from_script, to_script = conversion.split('_')
            report.append(f"### {from_script} → {to_script}")
            report.append("| Library | Text Size | Throughput (chars/sec) | Latency (ns) |")
            report.append("|---------|-----------|------------------------|-------------|")
            
            for r in sorted(results, key=lambda x: (x['text_size'], x['library'])):
                report.append(f"| {r['library']} | {r['text_size']} | {format_number(r['throughput_chars_per_sec'])} | {format_number(r['latency_ns'])} |")
            report.append("")
    
    # Performance Summary Tables
    report.append("")
    report.append("## Performance Summary")
    report.append("")
    
    # Calculate averages
    if rust_results:
        categories = {}
        for r in rust_results:
            cat = r['category']
            if cat not in categories:
                categories[cat] = []
            try:
                categories[cat].append(float(r['throughput_chars_per_sec']))
            except:
                pass
        
        report.append("### Average Throughput by Category (Rust)")
        report.append("| Category | Avg Throughput (chars/sec) |")
        report.append("|----------|----------------------------|")
        
        for cat, throughputs in sorted(categories.items()):
            if throughputs:
                avg = sum(throughputs) / len(throughputs)
                report.append(f"| {cat} | {avg:,.0f} |")
    
    if python_results:
        report.append("")
        report.append("### Average Throughput by API Type (Python)")
        report.append("| API Type | Avg Throughput (chars/sec) |")
        report.append("|----------|----------------------------|")
        
        api_throughputs = {}
        for r in python_results:
            api = r['api_type']
            if api not in api_throughputs:
                api_throughputs[api] = []
            try:
                api_throughputs[api].append(float(r['throughput_chars_per_sec']))
            except:
                pass
        
        for api, throughputs in sorted(api_throughputs.items()):
            if throughputs:
                avg = sum(throughputs) / len(throughputs)
                report.append(f"| {api} | {avg:,.0f} |")
    
    if comparison_results:
        report.append("")
        report.append("### Library Performance Ranking")
        report.append("| Library | Avg Throughput (chars/sec) | Success Rate |")
        report.append("|---------|----------------------------|--------------|")
        
        lib_stats = {}
        for r in comparison_results:
            lib = r['library']
            if lib not in lib_stats:
                lib_stats[lib] = {'throughputs': [], 'total': 0, 'success': 0}
            lib_stats[lib]['total'] += 1
            if r.get('success') == 'True':
                lib_stats[lib]['success'] += 1
                try:
                    lib_stats[lib]['throughputs'].append(float(r['throughput_chars_per_sec']))
                except:
                    pass
        
        # Sort by average throughput
        lib_ranking = []
        for lib, stats in lib_stats.items():
            if stats['throughputs']:
                avg = sum(stats['throughputs']) / len(stats['throughputs'])
                success_rate = (stats['success'] / stats['total']) * 100
                lib_ranking.append((lib, avg, success_rate))
        
        lib_ranking.sort(key=lambda x: x[1], reverse=True)
        
        for lib, avg, success_rate in lib_ranking:
            report.append(f"| {lib} | {avg:,.0f} | {success_rate:.0f}% |")
    
    return '\n'.join(report)

def main():
    """Generate consolidated benchmark report"""
    os.makedirs("target", exist_ok=True)
    
    report = generate_clean_report()
    
    with open("target/BENCHMARK_DATA.md", "w") as f:
        f.write(report)
    
    print("Generated consolidated benchmark report: target/BENCHMARK_DATA.md")
    
    # Also create a minimal summary
    summary = []
    summary.append("# Shlesha Performance Summary")
    summary.append("")
    summary.append("## Quick Reference")
    summary.append("")
    summary.append("| Metric | Value |")
    summary.append("|--------|-------|")
    
    # Add key metrics
    rust_results = []
    for category in ["hub", "standard", "extension"]:
        rust_results.extend(read_csv_results(f"target/benchmark_results_{category}.csv"))
    
    if rust_results:
        throughputs = []
        for r in rust_results:
            try:
                throughputs.append(float(r['throughput_chars_per_sec']))
            except:
                pass
        if throughputs:
            avg_throughput = sum(throughputs) / len(throughputs)
            summary.append(f"| Rust Avg Throughput | {avg_throughput:,.0f} chars/sec |")
            summary.append(f"| Rust Max Throughput | {max(throughputs):,.0f} chars/sec |")
            summary.append(f"| Rust Min Throughput | {min(throughputs):,.0f} chars/sec |")
    
    python_results = read_csv_results("target/python_benchmark_results.csv")
    if python_results:
        throughputs = []
        for r in python_results:
            try:
                throughputs.append(float(r['throughput_chars_per_sec']))
            except:
                pass
        if throughputs:
            avg_throughput = sum(throughputs) / len(throughputs)
            summary.append(f"| Python Avg Throughput | {avg_throughput:,.0f} chars/sec |")
    
    summary.append("")
    summary.append("*See target/BENCHMARK_DATA.md for complete results*")
    
    with open("target/BENCHMARK_SUMMARY.md", "w") as f:
        f.write('\n'.join(summary))
    
    print("Generated benchmark summary: target/BENCHMARK_SUMMARY.md")

if __name__ == "__main__":
    main()