apr-cli 0.4.15

CLI tool for APR model inspection, debugging, and operations
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
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324


// ========================================================================
// NEW: pipeline non-inference paths (pipeline.rs)
// ========================================================================

#[test]
#[cfg(not(feature = "inference"))]
fn test_run_gguf_inference_no_inference_needs_file() {
    let temp_dir = tempfile::tempdir().unwrap();
    let config = ShowcaseConfig {
        model_dir: temp_dir.path().to_path_buf(),
        ..Default::default()
    };

    // Create a fake GGUF file that's large enough
    let gguf_path = temp_dir.path().join(config.tier.gguf_filename());
    let fake_data = vec![0u8; 2_000_000]; // 2MB
    std::fs::write(&gguf_path, &fake_data).unwrap();

    let result = super::pipeline::run_gguf_inference(&config);
    assert!(result.is_ok());
    assert!(result.unwrap());
}

#[test]
#[cfg(not(feature = "inference"))]
fn test_run_gguf_inference_no_inference_too_small() {
    let temp_dir = tempfile::tempdir().unwrap();
    let config = ShowcaseConfig {
        model_dir: temp_dir.path().to_path_buf(),
        ..Default::default()
    };

    // Create a GGUF file that's too small
    let gguf_path = temp_dir.path().join(config.tier.gguf_filename());
    std::fs::write(&gguf_path, b"tiny").unwrap();

    let result = super::pipeline::run_gguf_inference(&config);
    assert!(result.is_err());
}

#[test]
#[cfg(not(feature = "inference"))]
fn test_run_gguf_inference_no_inference_missing_file() {
    let temp_dir = tempfile::tempdir().unwrap();
    let config = ShowcaseConfig {
        model_dir: temp_dir.path().to_path_buf(),
        ..Default::default()
    };

    // No file exists
    let result = super::pipeline::run_gguf_inference(&config);
    assert!(result.is_err());
}

#[test]
#[cfg(not(feature = "inference"))]
fn test_run_convert_no_inference() {
    let temp_dir = tempfile::tempdir().unwrap();
    let config = ShowcaseConfig {
        model_dir: temp_dir.path().to_path_buf(),
        ..Default::default()
    };

    // Create a fake GGUF file for conversion
    let gguf_path = temp_dir.path().join(config.tier.gguf_filename());
    let fake_data = vec![0u8; 1_000];
    std::fs::write(&gguf_path, &fake_data).unwrap();

    let result = super::pipeline::run_convert(&config);
    assert!(result.is_ok());
    assert!(result.unwrap());

    // Should have created a placeholder APR file
    let apr_basename = config.tier.gguf_filename().replace(".gguf", ".apr");
    let apr_path = temp_dir.path().join(apr_basename);
    assert!(apr_path.exists());
    let content = std::fs::read_to_string(&apr_path).unwrap();
    assert!(content.contains("APR-PLACEHOLDER-V2"));
}

#[test]
#[cfg(not(feature = "inference"))]
fn test_run_apr_inference_no_inference_missing_file() {
    let temp_dir = tempfile::tempdir().unwrap();
    let config = ShowcaseConfig {
        model_dir: temp_dir.path().to_path_buf(),
        ..Default::default()
    };

    // No APR file - should return Ok(false)
    let result = super::pipeline::run_apr_inference(&config);
    assert!(result.is_ok());
    assert!(!result.unwrap());
}

#[test]
#[cfg(not(feature = "inference"))]
fn test_run_apr_inference_no_inference_with_file() {
    let temp_dir = tempfile::tempdir().unwrap();
    let config = ShowcaseConfig {
        model_dir: temp_dir.path().to_path_buf(),
        ..Default::default()
    };

    // Create a placeholder APR file
    let apr_basename = config.tier.gguf_filename().replace(".gguf", ".apr");
    let apr_path = temp_dir.path().join(&apr_basename);
    std::fs::write(&apr_path, "APR-PLACEHOLDER").unwrap();

    let result = super::pipeline::run_apr_inference(&config);
    assert!(result.is_ok());
    assert!(result.unwrap());
}

#[test]
#[cfg(not(feature = "inference"))]
fn test_run_apr_inference_no_inference_with_zram_flag() {
    let temp_dir = tempfile::tempdir().unwrap();
    let config = ShowcaseConfig {
        model_dir: temp_dir.path().to_path_buf(),
        zram: true, // Tests the ZRAM flag printing path
        ..Default::default()
    };

    let apr_basename = config.tier.gguf_filename().replace(".gguf", ".apr");
    let apr_path = temp_dir.path().join(&apr_basename);
    std::fs::write(&apr_path, "APR-PLACEHOLDER").unwrap();

    let result = super::pipeline::run_apr_inference(&config);
    assert!(result.is_ok());
}

// ========================================================================
// NEW: pipeline run_import (pipeline.rs lines 11-70)
// ========================================================================

#[test]
fn test_run_import_model_already_exists() {
    let temp_dir = tempfile::tempdir().unwrap();
    let config = ShowcaseConfig {
        model_dir: temp_dir.path().to_path_buf(),
        ..Default::default()
    };

    // Create the model file so import returns early
    let gguf_path = temp_dir.path().join(config.tier.gguf_filename());
    std::fs::write(&gguf_path, b"existing model data").unwrap();

    let result = super::pipeline::run_import(&config);
    assert!(result.is_ok());
    assert!(result.unwrap()); // Should return true for already-exists
}

// ========================================================================
// NEW: validate_falsification with gguf_inference failure
// ========================================================================

#[test]
fn test_falsification_gguf_inference_failure_full_run() {
    let results = ShowcaseResults {
        import: true,
        gguf_inference: false, // Failed
        convert: true,
        apr_inference: true,
        benchmark: Some(BenchmarkComparison {
            apr_tps: 44.0,
            llama_cpp_tps: Some(35.0),
            ollama_tps: None,
            apr_ttft_ms: 78.0,
            llama_cpp_ttft_ms: Some(120.0),
            ollama_ttft_ms: None,
            speedup_vs_llama: Some(25.7),
            speedup_vs_ollama: None,
            apr_tps_stddev: 1.5,
            runs: 30,
        }),
        ..Default::default()
    };
    assert!(validate_falsification(&results, &full_run_config()).is_err());
}

#[test]
fn test_falsification_apr_inference_failure_full_run() {
    let results = ShowcaseResults {
        import: true,
        gguf_inference: true,
        convert: true,
        apr_inference: false, // Failed
        benchmark: Some(BenchmarkComparison {
            apr_tps: 44.0,
            llama_cpp_tps: Some(35.0),
            ollama_tps: None,
            apr_ttft_ms: 78.0,
            llama_cpp_ttft_ms: Some(120.0),
            ollama_ttft_ms: None,
            speedup_vs_llama: Some(25.7),
            speedup_vs_ollama: None,
            apr_tps_stddev: 1.5,
            runs: 30,
        }),
        ..Default::default()
    };
    assert!(validate_falsification(&results, &full_run_config()).is_err());
}

// ========================================================================
// NEW: ModelTier specific size/filename tests
// ========================================================================

#[test]
fn test_model_tier_tiny_specific_values() {
    let tier = ModelTier::Tiny;
    assert!(tier.size_gb() < 1.0);
    assert!(tier.gguf_filename().contains("0.5b"));
    assert!(tier.model_path().contains("0.5B"));
}

#[test]
fn test_model_tier_large_specific_values() {
    let tier = ModelTier::Large;
    assert!(tier.size_gb() > 10.0);
    assert!(tier.gguf_filename().contains("32B"));
    assert!(tier.model_path().contains("32B"));
}

// ========================================================================
// NEW: ShowcaseConfig model field matches tier
// ========================================================================

#[test]
fn test_showcase_config_default_model_matches_small_tier() {
    let config = ShowcaseConfig::default();
    assert_eq!(config.model, ModelTier::Small.model_path());
}

#[test]
fn test_showcase_config_with_tier_model_field_matches() {
    for tier in &[
        ModelTier::Tiny,
        ModelTier::Small,
        ModelTier::Medium,
        ModelTier::Large,
    ] {
        let config = ShowcaseConfig::with_tier(*tier);
        assert_eq!(config.model, tier.model_path());
    }
}

// ========================================================================
// NEW: BenchmarkComparison serde round-trip with all None optionals
// ========================================================================

#[test]
fn test_benchmark_comparison_serde_all_none() {
    let bench = BenchmarkComparison {
        apr_tps: 50.0,
        llama_cpp_tps: None,
        ollama_tps: None,
        apr_ttft_ms: 60.0,
        llama_cpp_ttft_ms: None,
        ollama_ttft_ms: None,
        speedup_vs_llama: None,
        speedup_vs_ollama: None,
        apr_tps_stddev: 2.0,
        runs: 30,
    };
    let json = serde_json::to_string(&bench).unwrap();
    let parsed: BenchmarkComparison = serde_json::from_str(&json).unwrap();
    assert!((parsed.apr_tps - 50.0).abs() < 0.001);
    assert!(parsed.llama_cpp_tps.is_none());
    assert!(parsed.ollama_tps.is_none());
    assert!(parsed.speedup_vs_llama.is_none());
    assert!(parsed.speedup_vs_ollama.is_none());
}

#[test]
fn test_benchmark_comparison_serde_pretty_print() {
    let bench = BenchmarkComparison {
        apr_tps: 44.0,
        llama_cpp_tps: Some(35.0),
        ollama_tps: None,
        apr_ttft_ms: 78.0,
        llama_cpp_ttft_ms: Some(120.0),
        ollama_ttft_ms: None,
        speedup_vs_llama: Some(25.7),
        speedup_vs_ollama: None,
        apr_tps_stddev: 1.5,
        runs: 30,
    };
    let pretty = serde_json::to_string_pretty(&bench).unwrap();
    assert!(pretty.contains('\n'));
    assert!(pretty.contains("apr_tps"));
    // Verify round-trip with pretty-printed JSON
    let parsed: BenchmarkComparison = serde_json::from_str(&pretty).unwrap();
    assert_eq!(parsed.runs, 30);
}

// ========================================================================
// NEW: run_convert with existing APR file (early return)
// ========================================================================

#[test]
#[cfg(not(feature = "inference"))]
fn test_run_convert_apr_already_exists() {
    let temp_dir = tempfile::tempdir().unwrap();
    let config = ShowcaseConfig {
        model_dir: temp_dir.path().to_path_buf(),
        ..Default::default()
    };

    // Create both GGUF and APR files
    let gguf_path = temp_dir.path().join(config.tier.gguf_filename());
    std::fs::write(&gguf_path, b"fake gguf").unwrap();

    let apr_basename = config.tier.gguf_filename().replace(".gguf", ".apr");
    let apr_path = temp_dir.path().join(&apr_basename);
    std::fs::write(&apr_path, b"existing apr").unwrap();

    // The non-inference convert path still writes a placeholder.
    // But the inference path would short-circuit. Let's verify the function runs:
    let result = super::pipeline::run_convert(&config);
    assert!(result.is_ok());
}