[caching] save/load source hierarchy and extracted log statements

Cargo.toml

@@ -6,19 +6,25 @@ version = "0.1.0"
 edition = "2021"
 
 [dependencies]
-clap = { version = "4.5.36", features = ["derive"] }
+bincode = { version = "2.0.1", features = ["serde", "std"]}
+clap = { version = "4.5.36", features = ["std", "derive"] }
 colored_json = "5.0.0"
+directories = "6.0.0"
 indicatif = "0.18.0"
 itertools = "0.14.0"
 regex = "1.11.1"
 serde = { version = "1.0.219", features = ["derive"] }
 serde_json = "1.0.140"
+serde_regex = "1.1.0"
+sha2 = "0.10.9"
+tempfile = "3.23.0"
 thiserror = "2.0.16"
 tree-sitter = "0.25.3"
 tree-sitter-cpp = "0.23.4"
 tree-sitter-rust-orchard = "0.12.0"
 tree-sitter-java = "0.23.5"
 tree-sitter-python = "0.25.0"
+ignore = "0.4"
 rayon = "1.11.0"
 miette = { version = "7.6.0", features = ["fancy", "serde"] }
 
@@ -34,3 +40,4 @@ log = "0.4.27"
 rand = "0.9.0"
 tempfile = "3.22.0"
 fs_extra = "1.3"
+walkdir = "2.5.0"

docs/Tasks.md

@@ -6,8 +6,8 @@
     and it can find the message.
 - [X] Extract a thread id from log when available and associate with source ref.
 - [X] Generate call stack from exceptions.
-- [ ] Support multiple source roots from CLI.
-- [ ] Serialize state for re-use on subsequent executions
+- [X] Support multiple source roots from CLI.
+- [X] Serialize state for re-use on subsequent executions
 
 ## Extension
 

schemas/cache-header-v1.json

@@ -0,0 +1,34 @@
+{
+    "$schema": "https://json-schema.org/draft-07/schema",
+    "title": "log2src cache entry header v1",
+    "description": "Schema for the header of a log2src cache entry.",
+    "type": "object",
+    "additionalProperties": false,
+    "properties": {
+        "$schema": {
+            "type": "string",
+            "format": "uri"
+        },
+        "revision": {
+            "type": "string"
+        },
+        "format": {
+            "type": "string",
+            "enum": [
+                "Bincode"
+            ]
+        },
+        "path": {
+            "type": "string"
+        },
+        "timestamp": {
+            "type": "integer"
+        }
+    },
+    "required": [
+        "$schema",
+        "format",
+        "path",
+        "timestamp"
+    ]
+}