| 1 | package org.egothor.methodatlas; | |
| 2 | ||
| 3 | import java.io.IOException; | |
| 4 | import java.nio.file.Path; | |
| 5 | import java.util.List; | |
| 6 | import java.util.Map; | |
| 7 | ||
| 8 | import com.fasterxml.jackson.annotation.JsonIgnoreProperties; | |
| 9 | import com.fasterxml.jackson.annotation.JsonProperty; | |
| 10 | import com.fasterxml.jackson.databind.DeserializationFeature; | |
| 11 | import com.fasterxml.jackson.databind.ObjectMapper; | |
| 12 | import com.fasterxml.jackson.dataformat.yaml.YAMLFactory; | |
| 13 | ||
| 14 | /** | |
| 15 | * Loads a YAML configuration file that provides default values for | |
| 16 | * command-line options. | |
| 17 | * | |
| 18 | * <p> | |
| 19 | * When a {@code -config <file>} argument is present, {@link CliArgs} calls | |
| 20 | * {@link #load(Path)} before processing the remaining arguments. The returned | |
| 21 | * {@link YamlConfigFile} seeds the initial values; any matching command-line | |
| 22 | * flag then overrides the YAML value. | |
| 23 | * </p> | |
| 24 | * | |
| 25 | * <h2>Supported fields</h2> | |
| 26 | * | |
| 27 | * <pre> | |
| 28 | * outputMode: csv # csv | plain | sarif | json (default: csv) | |
| 29 | * emitMetadata: false # (default: false) | |
| 30 | * contentHash: false # (default: false) | |
| 31 | * securityOnly: false # (default: false) | |
| 32 | * includeNonSecurity: false # opt-in: include non-security methods in SARIF output (default: false) | |
| 33 | * sarifOmitScores: false # opt-out: omit interaction score / confidence from SARIF message text (default: false) | |
| 34 | * minConfidence: 0.0 # drop AI results below this threshold (requires ai.confidence: true; default: 0.0 = off) | |
| 35 | * driftDetect: false # (default: false) | |
| 36 | * overrideFile: .methodatlas-overrides.yaml # optional | |
| 37 | * fileSuffixes: | |
| 38 | * - Test.java | |
| 39 | * testMarkers: # annotation/attribute names; empty = provider defaults | |
| 40 | * - Test | |
| 41 | * - ParameterizedTest | |
| 42 | * properties: # plugin-specific key/multi-value pairs (optional) | |
| 43 | * functionNames: # example: for a Jest/Mocha/Vitest TypeScript plugin | |
| 44 | * - test | |
| 45 | * - it | |
| 46 | * ai: | |
| 47 | * enabled: true | |
| 48 | * provider: ollama # auto | ollama | openai | openrouter | anthropic | azure_openai | groq | xai | github_models | mistral | |
| 49 | * model: qwen2.5-coder:7b | |
| 50 | * baseUrl: http://localhost:11434 | |
| 51 | * apiKey: sk-... | |
| 52 | * apiKeyEnv: MY_KEY_ENV | |
| 53 | * taxonomyFile: /path/to/taxonomy.txt | |
| 54 | * taxonomyMode: default # default | optimized | |
| 55 | * maxClassChars: 100000 | |
| 56 | * timeoutSec: 30 | |
| 57 | * maxRetries: 3 | |
| 58 | * confidence: false | |
| 59 | * apiVersion: 2024-02-01 # Azure OpenAI REST API version (azure_openai only) | |
| 60 | * </pre> | |
| 61 | * | |
| 62 | * <p> | |
| 63 | * Unknown fields in the YAML file are silently ignored. | |
| 64 | * </p> | |
| 65 | * | |
| 66 | * @see CliArgs | |
| 67 | */ | |
| 68 | final class YamlConfig { | |
| 69 | ||
| 70 | /** | |
| 71 | * Prevents instantiation of this utility class. | |
| 72 | */ | |
| 73 | private YamlConfig() { | |
| 74 | } | |
| 75 | ||
| 76 | /** | |
| 77 | * Loads a YAML configuration file. | |
| 78 | * | |
| 79 | * @param configFile path to the YAML file | |
| 80 | * @return parsed configuration; never {@code null} | |
| 81 | * @throws IOException if the file cannot be read | |
| 82 | * @throws IllegalArgumentException if the file cannot be parsed as valid YAML | |
| 83 | */ | |
| 84 | /* default */ static YamlConfigFile load(Path configFile) throws IOException { | |
| 85 | ObjectMapper mapper = new ObjectMapper(new YAMLFactory()); | |
| 86 | mapper.configure(DeserializationFeature.FAIL_ON_UNKNOWN_PROPERTIES, false); | |
| 87 |
1
1. load : replaced return value with null for org/egothor/methodatlas/YamlConfig::load → KILLED |
return mapper.readValue(configFile.toFile(), YamlConfigFile.class); |
| 88 | } | |
| 89 | ||
| 90 | // ------------------------------------------------------------------------- | |
| 91 | // POJO classes | |
| 92 | // ------------------------------------------------------------------------- | |
| 93 | ||
| 94 | /** | |
| 95 | * Top-level YAML configuration structure. | |
| 96 | */ | |
| 97 | @JsonIgnoreProperties(ignoreUnknown = true) | |
| 98 | /* default */ static final class YamlConfigFile { | |
| 99 | ||
| 100 | /** Output mode: {@code csv}, {@code plain}, or {@code sarif}. */ | |
| 101 | @JsonProperty("outputMode") | |
| 102 | /* default */ String outputMode; | |
| 103 | ||
| 104 | /** Whether to emit {@code # key: value} metadata comment lines. */ | |
| 105 | @JsonProperty("emitMetadata") | |
| 106 | /* default */ boolean emitMetadata; | |
| 107 | ||
| 108 | /** File name suffixes used to select test source files. */ | |
| 109 | @JsonProperty("fileSuffixes") | |
| 110 | /* default */ List<String> fileSuffixes; | |
| 111 | ||
| 112 | /** | |
| 113 | * Language-neutral test-marker identifiers (annotation/attribute simple | |
| 114 | * names for JVM and .NET providers; ignored by TypeScript providers). | |
| 115 | * Empty or absent means "use provider defaults". | |
| 116 | */ | |
| 117 | @JsonProperty("testMarkers") | |
| 118 | /* default */ List<String> testMarkers; | |
| 119 | ||
| 120 | /** | |
| 121 | * Plugin-specific key/multi-value pairs forwarded verbatim to each | |
| 122 | * {@link org.egothor.methodatlas.api.TestDiscovery} provider. | |
| 123 | * Providers ignore keys they do not recognise. | |
| 124 | */ | |
| 125 | @JsonProperty("properties") | |
| 126 | /* default */ Map<String, List<String>> properties; | |
| 127 | ||
| 128 | /** | |
| 129 | * Whether to include a SHA-256 content-hash fingerprint of each class | |
| 130 | * source as a {@code content_hash} column. | |
| 131 | */ | |
| 132 | @JsonProperty("contentHash") | |
| 133 | /* default */ boolean contentHash; | |
| 134 | ||
| 135 | /** | |
| 136 | * Path to a YAML classification override file. When set, human-authored | |
| 137 | * corrections are applied after AI classification on every run. | |
| 138 | */ | |
| 139 | @JsonProperty("overrideFile") | |
| 140 | /* default */ String overrideFile; | |
| 141 | ||
| 142 | /** | |
| 143 | * When {@code true}, only security-relevant methods are emitted; all | |
| 144 | * other methods are silently dropped from the output. | |
| 145 | */ | |
| 146 | @JsonProperty("securityOnly") | |
| 147 | /* default */ boolean securityOnly; | |
| 148 | ||
| 149 | /** | |
| 150 | * When {@code true}, non-security methods are included in SARIF output | |
| 151 | * even though SARIF mode applies the security-only filter by default. | |
| 152 | * Has no effect in CSV or plain-text modes. | |
| 153 | */ | |
| 154 | @JsonProperty("includeNonSecurity") | |
| 155 | /* default */ boolean includeNonSecurity; | |
| 156 | ||
| 157 | /** | |
| 158 | * When {@code true}, a {@code tag_ai_drift} column is added to CSV/plain | |
| 159 | * output comparing the source-level {@code @Tag("security")} annotation | |
| 160 | * against the AI security-relevance classification. | |
| 161 | */ | |
| 162 | @JsonProperty("driftDetect") | |
| 163 | /* default */ boolean driftDetect; | |
| 164 | ||
| 165 | /** | |
| 166 | * When {@code true}, the interaction score and confidence percentage are | |
| 167 | * omitted from SARIF result message text. Use this when the consuming | |
| 168 | * system already renders the {@code properties} bag and the extra text is | |
| 169 | * unwanted. Default: {@code false} (scores are embedded in messages). | |
| 170 | */ | |
| 171 | @JsonProperty("sarifOmitScores") | |
| 172 | /* default */ boolean sarifOmitScores; | |
| 173 | ||
| 174 | /** | |
| 175 | * Minimum AI confidence score (inclusive) required for a method to be | |
| 176 | * emitted. Methods whose {@code ai_confidence} is below this threshold | |
| 177 | * are silently dropped. Only meaningful when {@code ai.confidence: true} | |
| 178 | * is also set. Default: {@code 0.0} (no filtering). | |
| 179 | */ | |
| 180 | @JsonProperty("minConfidence") | |
| 181 | /* default */ Double minConfidence; | |
| 182 | ||
| 183 | /** AI enrichment settings. */ | |
| 184 | @JsonProperty("ai") | |
| 185 | /* default */ YamlAiConfig ai; | |
| 186 | } | |
| 187 | ||
| 188 | /** | |
| 189 | * AI subsystem configuration within the YAML file. | |
| 190 | */ | |
| 191 | @JsonIgnoreProperties(ignoreUnknown = true) | |
| 192 | /* default */ static final class YamlAiConfig { | |
| 193 | ||
| 194 | /** Whether AI enrichment is enabled. */ | |
| 195 | @JsonProperty("enabled") | |
| 196 | /* default */ Boolean enabled; | |
| 197 | ||
| 198 | /** | |
| 199 | * AI provider: {@code auto}, {@code ollama}, {@code openai}, | |
| 200 | * {@code openrouter}, {@code anthropic}, {@code azure_openai}, | |
| 201 | * {@code groq}, {@code xai}, {@code github_models}, or {@code mistral}. | |
| 202 | */ | |
| 203 | @JsonProperty("provider") | |
| 204 | /* default */ String provider; | |
| 205 | ||
| 206 | /** Provider-specific model name. */ | |
| 207 | @JsonProperty("model") | |
| 208 | /* default */ String model; | |
| 209 | ||
| 210 | /** Provider base URL override. */ | |
| 211 | @JsonProperty("baseUrl") | |
| 212 | /* default */ String baseUrl; | |
| 213 | ||
| 214 | /** API key supplied directly. */ | |
| 215 | @JsonProperty("apiKey") | |
| 216 | /* default */ String apiKey; | |
| 217 | ||
| 218 | /** Name of the environment variable that holds the API key. */ | |
| 219 | @JsonProperty("apiKeyEnv") | |
| 220 | /* default */ String apiKeyEnv; | |
| 221 | ||
| 222 | /** Path to an external taxonomy file. */ | |
| 223 | @JsonProperty("taxonomyFile") | |
| 224 | /* default */ String taxonomyFile; | |
| 225 | ||
| 226 | /** Built-in taxonomy variant: {@code default} or {@code optimized}. */ | |
| 227 | @JsonProperty("taxonomyMode") | |
| 228 | /* default */ String taxonomyMode; | |
| 229 | ||
| 230 | /** Maximum number of characters of class source sent to the AI. */ | |
| 231 | @JsonProperty("maxClassChars") | |
| 232 | /* default */ Integer maxClassChars; | |
| 233 | ||
| 234 | /** AI request timeout in seconds. */ | |
| 235 | @JsonProperty("timeoutSec") | |
| 236 | /* default */ Long timeoutSec; | |
| 237 | ||
| 238 | /** Maximum number of retries for AI requests. */ | |
| 239 | @JsonProperty("maxRetries") | |
| 240 | /* default */ Integer maxRetries; | |
| 241 | ||
| 242 | /** Whether to request a confidence score for each classification. */ | |
| 243 | @JsonProperty("confidence") | |
| 244 | /* default */ Boolean confidence; | |
| 245 | ||
| 246 | /** | |
| 247 | * Azure OpenAI REST API version appended as the {@code api-version} query | |
| 248 | * parameter; only used when {@code provider: azure_openai} is set. | |
| 249 | */ | |
| 250 | @JsonProperty("apiVersion") | |
| 251 | /* default */ String apiVersion; | |
| 252 | } | |
| 253 | } | |
Mutations | ||
| 87 |
1.1 |