Include checker name in cache key and tidy up some code.

Author: alitheg

atr/attestable.py

@@ -24,7 +24,7 @@
 import aiofiles.os
 import pydantic
 
-import atr.hashing as hashing
+import atr.hashes as hashes
 import atr.log as log
 import atr.models.attestable as models
 import atr.util as util
@@ -130,7 +130,7 @@ async def paths_to_hashes_and_sizes(directory: pathlib.Path) -> tuple[dict[str,
         if "\\" in path_key:
             # TODO: We should centralise this, and forbid some other characters too
             raise ValueError(f"Backslash in path is forbidden: {path_key}")
-        path_to_hash[path_key] = await hashing.compute_file_hash(full_path)
+        path_to_hash[path_key] = await hashes.compute_file_hash(full_path)
         path_to_size[path_key] = (await aiofiles.os.stat(full_path)).st_size
     return path_to_hash, path_to_size
 

atr/docs/tasks.md

@@ -41,7 +41,7 @@ In `atr/tasks/checks` you will find several modules that perform these check tas
 In `atr/tasks/__init__.py` you will see imports for existing modules where you can add an import for new check task, for example:
 
 ```python
-import atr.tasks.checks.file_hash as file_hash
+import atr.tasks.checks.hashing as file_hash
 import atr.tasks.checks.license as license
 ```
 

atr/hashing.py atr/hashes.pyatr/hashing.py renamed to atr/hashes.py

@@ -24,6 +24,7 @@
 import aiofiles.os
 
 import atr.attestable as attestable
+import atr.hashes as hashes
 import atr.util as util
 
 if TYPE_CHECKING:
@@ -131,7 +132,7 @@ async def _add_from_prior(
     if (prior_hashes is not None) and (path in prior_hashes):
         n_hashes[path] = prior_hashes[path]
     else:
-        n_hashes[path] = await attestable.compute_file_hash(target)
+        n_hashes[path] = await hashes.compute_file_hash(target)
     stat_result = await aiofiles.os.stat(target)
     n_sizes[path] = stat_result.st_size
     return prior_hashes
@@ -211,7 +212,7 @@ async def _merge_all_present(
         if (prior_hashes is not None) and (path in prior_hashes):
             p_hash = prior_hashes[path]
         else:
-            p_hash = await attestable.compute_file_hash(prior_dir / path)
+            p_hash = await hashes.compute_file_hash(prior_dir / path)
         if p_hash != b_hash:
             # Case 11 via hash: base and new have the same content but prior differs
             return await _replace_with_prior(
@@ -250,7 +251,7 @@ async def _replace_with_prior(
     if (prior_hashes is not None) and (path in prior_hashes):
         n_hashes[path] = prior_hashes[path]
     else:
-        n_hashes[path] = await attestable.compute_file_hash(file_path)
+        n_hashes[path] = await hashes.compute_file_hash(file_path)
     stat_result = await aiofiles.os.stat(file_path)
     n_sizes[path] = stat_result.st_size
     return prior_hashes

atr/merge.py

@@ -22,7 +22,7 @@
 from typing import TYPE_CHECKING
 
 import atr.db as db
-import atr.hashing as hashing
+import atr.hashes as hashes
 import atr.models.sql as sql
 import atr.storage as storage
 import atr.storage.types as types
@@ -55,9 +55,9 @@ async def _filter_check_results_by_hash(
                 extra_arg_names = []
             extra_args = checks.resolve_extra_args(extra_arg_names, release)
             cache_key = await checks.resolve_cache_key(
-                policy_keys, release, release.latest_revision_number, extra_args, file=rel_path.name
+                cr.checker, policy_keys, release, release.latest_revision_number, extra_args, file=rel_path.name
             )
-            input_hash_by_module[module_path] = hashing.compute_dict_hash(cache_key) if cache_key else None
+            input_hash_by_module[module_path] = hashes.compute_dict_hash(cache_key) if cache_key else None
 
         if cr.inputs_hash == input_hash_by_module[module_path]:
             filtered_check_results.append(cr)