Merge pull request #447 from crocs-muni/fix/notebooks-aux-dsets

Avoid double serialization in CVE and CPE dataset load.

4 files changed, 11 insertions, 13 deletions

diff --git a/src/sec_certs/dataset/cc.py b/src/sec_certs/dataset/cc.py
index 4c1114f3..ba2ffb58 100644
--- a/src/sec_certs/dataset/cc.py
+++ b/src/sec_certs/dataset/cc.py
@@ -269,9 +269,8 @@ class CCDataset(Dataset[CCCertificate, CCAuxiliaryDatasets], ComplexSerializable
 
         Optionally stores it at the given path (a directory) and also downloads auxiliary datasets and artifacts (PDFs).
 
-        :::{note}
-        Note that including the auxiliary datasets adds several gigabytes and including artifacts adds tens of gigabytes.
-        :::
+        .. note::
+            Note that including the auxiliary datasets adds several gigabytes and including artifacts adds tens of gigabytes.
 
         :param path: Path to a directory where to store the dataset, or `None` if it should not be stored.
         :param auxiliary_datasets: Whether to also download auxiliary datasets (CVE, CPE, CPEMatch datasets).
diff --git a/src/sec_certs/dataset/dataset.py b/src/sec_certs/dataset/dataset.py
index 2da2bdc4..c906fd59 100644
--- a/src/sec_certs/dataset/dataset.py
+++ b/src/sec_certs/dataset/dataset.py
@@ -193,9 +193,8 @@ class Dataset(Generic[CertSubType, AuxiliaryDatasetsSubType], ComplexSerializabl
 
         Optionally stores it at the given path (a directory) and also downloads auxiliary datasets and artifacts (PDFs).
 
-        :::{note}
-        Note that including the auxiliary datasets adds several gigabytes and including artifacts adds tens of gigabytes.
-        :::
+        .. note::
+            Note that including the auxiliary datasets adds several gigabytes and including artifacts adds tens of gigabytes.
 
         :param archive_url: The URL of the full dataset archive.
         :param snapshot_url: The URL of the full dataset snapshot.
@@ -437,10 +436,10 @@ class Dataset(Generic[CertSubType, AuxiliaryDatasetsSubType], ComplexSerializabl
                 logger.info("Fetching new CPE records from NVD API.")
                 with CpeNvdDatasetBuilder(api_key=config.nvd_api_key) as builder:
                     cpe_dataset = builder.build_dataset(cpe_dataset)
+                cpe_dataset.to_json()
             else:
                 logger.info("Preparing CPEDataset from sec-certs.org.")
                 cpe_dataset = CPEDataset.from_web(self.cpe_dataset_path)
-            cpe_dataset.to_json()
 
         return cpe_dataset
 
@@ -462,10 +461,10 @@ class Dataset(Generic[CertSubType, AuxiliaryDatasetsSubType], ComplexSerializabl
                 logger.info("Fetching new CVE records from NVD API.")
                 with CveNvdDatasetBuilder(api_key=config.nvd_api_key) as builder:
                     cve_dataset = builder.build_dataset(cve_dataset)
+                cve_dataset.to_json()
             else:
                 logger.info("Preparing CVEDataset from sec-certs.org")
                 cve_dataset = CVEDataset.from_web(self.cve_dataset_path)
-            cve_dataset.to_json()
 
         return cve_dataset
 
diff --git a/src/sec_certs/dataset/fips.py b/src/sec_certs/dataset/fips.py
index eeaec0a0..0feb920f 100644
--- a/src/sec_certs/dataset/fips.py
+++ b/src/sec_certs/dataset/fips.py
@@ -228,9 +228,8 @@ class FIPSDataset(Dataset[FIPSCertificate, FIPSAuxiliaryDatasets], ComplexSerial
 
         Optionally stores it at the given path (a directory) and also downloads auxiliary datasets and artifacts (PDFs).
 
-        :::{note}
-        Note that including the auxiliary datasets adds several gigabytes and including artifacts adds tens of gigabytes.
-        :::
+        .. note::
+            Note that including the auxiliary datasets adds several gigabytes and including artifacts adds tens of gigabytes.
 
         :param path: Path to a directory where to store the dataset, or `None` if it should not be stored.
         :param auxiliary_datasets: Whether to also download auxiliary datasets (CVE, CPE, CPEMatch datasets).
diff --git a/src/sec_certs/sample/cc.py b/src/sec_certs/sample/cc.py
index aa145706..63c2cca5 100644
--- a/src/sec_certs/sample/cc.py
+++ b/src/sec_certs/sample/cc.py
@@ -484,7 +484,8 @@ class CCCertificate(
     @property
     def actual_sars(self) -> set[SAR] | None:
         """
-        Computes actual SARs. First, SARs implied by EAL are computed. Then, these are augmented with heuristically extracted SARs
+        Computes actual SARs. First, SARs implied by EAL are computed. Then, these are augmented with heuristically extracted SARs.
+
         :return Optional[Set[SAR]]: Set of actual SARs of a certificate, None if empty
         """
         sars = {}
@@ -543,7 +544,7 @@ class CCCertificate(
     def merge(self, other: CCCertificate, other_source: str | None = None) -> None:
         """
         Merges with other CC sample. Assuming they come from different sources, e.g., csv and html.
-        Assuming that html source has better protection profiles, they overwrite CSV info
+        Assuming that html source has better protection profiles, they overwrite CSV info.
         On other values the sanity checks are made.
         """
         if self != other: