bump some analysis

1 files changed, 196 insertions, 85 deletions

diff --git a/notebooks/cc/references.ipynb b/notebooks/cc/references.ipynb
index ff05ea3d..e474d816 100644
--- a/notebooks/cc/references.ipynb
+++ b/notebooks/cc/references.ipynb
@@ -30,7 +30,6 @@
     "from collections.abc import Iterable\n",
     "from pathlib import Path\n",
     "\n",
-    "import matplotlib\n",
     "import matplotlib.pyplot as plt\n",
     "import networkx as nx\n",
     "import networkx.algorithms.community as nx_comm\n",
@@ -47,14 +46,7 @@
     "\n",
     "%matplotlib inline\n",
     "\n",
-    "# LaTeX plotting\n",
     "# matplotlib.use(\"pgf\")\n",
-    "# plt.rcParams[\"pgf.texsystem\"] = \"pdflatex\"\n",
-    "# plt.rcParams[\"font.family\"] = \"serif\"\n",
-    "# plt.rcParams[\"text.usetex\"] = True\n",
-    "# plt.rcParams[\"pgf.rcfonts\"] = False\n",
-    "\n",
-    "\n",
     "sns.set_theme(style=\"white\")\n",
     "plt.rcParams[\"axes.linewidth\"] = 0.5\n",
     "plt.rcParams[\"legend.fontsize\"] = 6.5\n",
@@ -68,7 +60,10 @@
     "plt.rcParams[\"xtick.major.size\"] = 5\n",
     "plt.rcParams[\"xtick.major.width\"] = 0.5\n",
     "plt.rcParams[\"xtick.major.pad\"] = 0\n",
-    "\n",
+    "# plt.rcParams[\"pgf.texsystem\"] = \"pdflatex\"\n",
+    "# plt.rcParams[\"font.family\"] = \"serif\"\n",
+    "# plt.rcParams[\"text.usetex\"] = True\n",
+    "# plt.rcParams[\"pgf.rcfonts\"] = False\n",
     "plt.rcParams[\"axes.titlesize\"] = 8\n",
     "plt.rcParams[\"legend.handletextpad\"] = 0.3\n",
     "plt.rcParams[\"lines.markersize\"] = 4\n",
@@ -81,12 +76,9 @@
     "\n",
     "# plt.rcParams['figure.figsize'] = (10, 6)\n",
     "\n",
-    "REPO_ROOT = Path().resolve()\n",
     "RESULTS_DIR = Path(\"./results/references\")\n",
     "RESULTS_DIR.mkdir(exist_ok=True, parents=True)\n",
-    "SMARTCARD_CATEGORY = \"ICs, Smart Cards and Smart Card-Related Devices and Systems\"\n",
-    "DATASET_PATH = REPO_ROOT / \"dataset/cc_november_23/dataset.json\"\n",
-    "PREDICTIONS_PATH = REPO_ROOT / \"dataset/reference_prediction/predictions.csv\"\n"
+    "SMARTCARD_CATEGORY = \"ICs, Smart Cards and Smart Card-Related Devices and Systems\""
    ]
   },
   {
@@ -185,7 +177,7 @@
     "        target=\"reference\",\n",
     "        create_using=nx.DiGraph,\n",
     "        edge_attr=[\"reference_label\"],\n",
-    "    )\n"
+    "    )"
    ]
   },
   {
@@ -201,9 +193,9 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "dset = CCDataset.from_json(DATASET_PATH)\n",
+    "dset = CCDataset.from_json(\"/var/tmp/xjanovsk/certs/sec-certs/dataset/cc_november_23/dataset.json\")\n",
     "cc_df = preprocess_cc_df(dset.to_pandas())\n",
-    "refs_df = preprocess_refs_df(PREDICTIONS_PATH, cc_df)\n",
+    "refs_df = preprocess_refs_df(\"/var/tmp/xjanovsk/certs/sec-certs/dataset/reference_prediction/predictions.csv\", cc_df)\n",
     "unique_labels = refs_df.reference_label.unique().tolist()\n",
     "\n",
     "# Load labeled reference graph as networkx directed graph\n",
@@ -218,7 +210,7 @@
     "cc_df = compute_reference_numbers(compute_references(cc_df, graph, unique_labels))\n",
     "cc_df_comp = compute_reference_numbers(compute_references(cc_df, graph, \"COMPONENT_USED\"))\n",
     "cc_df_prev = compute_reference_numbers(compute_references(cc_df, graph, \"PREVIOUS_VERSION\"))\n",
-    "assert cc_df.n_refs.sum() == cc_df_comp.n_refs.sum() + cc_df_prev.n_refs.sum()\n"
+    "assert cc_df.n_refs.sum() == cc_df_comp.n_refs.sum() + cc_df_prev.n_refs.sum()"
    ]
   },
   {
@@ -229,6 +221,22 @@
    ]
   },
   {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Display EAL in time\n",
+    "sec_levels = cc_df.loc[cc_df.year_from < 2023].groupby([\"year_from\", \"eal\"], as_index=False).size()\n",
+    "plt.figure()\n",
+    "g = sns.relplot(data=sec_levels, x=\"year_from\", y=\"size\", col=\"eal\", kind=\"line\", col_wrap=7)\n",
+    "g.set(xlabel=\"Year of certification\", ylabel=\"Number of issued certificates\")\n",
+    "g.fig.suptitle(\"Security level prevalence in time\", y=1.03)\n",
+    "g.fig.savefig(RESULTS_DIR / \"sec_levels_in_time.pdf\", bbox_inches=\"tight\")\n",
+    "g.fig.show()"
+   ]
+  },
+  {
    "attachments": {},
    "cell_type": "markdown",
    "metadata": {},
@@ -277,16 +285,15 @@
     "\n",
     "    df_melted = df[[\"n_refs\", \"n_trans_refs\", \"n_in_refs\", \"n_in_trans_refs\"]].melt()\n",
     "    df_melted[\"incoming\"] = df_melted.variable.map(lambda x: bool(x.endswith(\"by\")))\n",
+    "    plt.figure()\n",
     "    g = sns.catplot(data=df_melted, kind=\"boxen\", x=\"variable\", y=\"value\", col=\"variable\", sharex=False, sharey=False)\n",
-    "    fig = g.get_figure()\n",
-    "    fig.savefig(RESULTS_DIR / \"boxen_plot_references.pdf\", bbox_inches=\"tight\")\n",
-    "\n",
-    "    plt.show()\n",
+    "    g.fig.savefig(RESULTS_DIR / \"boxen_plot_references.pdf\", bbox_inches=\"tight\")\n",
+    "    g.fig.show()\n",
     "\n",
     "    return {}\n",
     "\n",
     "\n",
-    "compute_basic_reference_graph_stats(cc_df, cc_df_comp, cc_df_prev)\n"
+    "compute_basic_reference_graph_stats(cc_df, cc_df_comp, cc_df_prev)"
    ]
   },
   {
@@ -334,16 +341,16 @@
     "        .melt(id_vars=\"date\", var_name=\"certificate\", value_name=\"reach\")\n",
     "    )\n",
     "\n",
+    "    plt.figure()\n",
     "    g = sns.lineplot(data=df_reach_evolution_melted, x=\"date\", y=\"reach\", hue=\"certificate\")\n",
     "    g.set(title=\"Reach of top-10 certificates in time\", xlabel=\"Time\", ylabel=\"Certificate reach\")\n",
-    "    fig = g.get_figure()\n",
-    "    fig.savefig(RESULTS_DIR / \"lineplot_top_certificate_reach.pdf\", bbox_inches=\"tight\")\n",
-    "    plt.show()\n",
+    "    g.figure.savefig(RESULTS_DIR / \"lineplot_top_certificate_reach.pdf\", bbox_inches=\"tight\")\n",
+    "    g.figure.show()\n",
     "\n",
     "    return top_10_certs.index.tolist()\n",
     "\n",
     "\n",
-    "top_10_digests = compute_certs_top_reach(cc_df_comp)\n"
+    "top_10_digests = compute_certs_top_reach(cc_df_comp)"
    ]
   },
   {
@@ -393,11 +400,11 @@
     "        .melt(id_vars=[\"date\"], var_name=\"category\", value_name=\"n_references\")\n",
     "    )\n",
     "\n",
+    "    plt.figure()\n",
     "    g = sns.lineplot(data=df_avg_num_refs_melted, x=\"date\", y=\"n_references\", hue=\"category\")\n",
     "    g.set(title=\"Average number of references in certificates\", xlabel=\"Time\", ylabel=\"Number of references\")\n",
-    "    fig = g.get_figure()\n",
-    "    fig.savefig(RESULTS_DIR / \"lineplot_avg_n_references.pdf\", bbox_inches=\"tight\")\n",
-    "    plt.show()\n",
+    "    g.figure.savefig(RESULTS_DIR / \"lineplot_avg_n_references.pdf\", bbox_inches=\"tight\")\n",
+    "    g.figure.show()\n",
     "\n",
     "    return {}\n",
     "\n",
@@ -421,21 +428,21 @@
     "        .melt(id_vars=[\"date\"], var_name=\"category\", value_name=\"n_references\")\n",
     "    )\n",
     "\n",
+    "    plt.figure()\n",
     "    g = sns.lineplot(data=df_avg_num_refs_melted, x=\"date\", y=\"n_references\", hue=\"category\")\n",
     "    g.set(\n",
     "        title=\"Average certificate reach over time\",\n",
     "        xlabel=\"Time\",\n",
     "        ylabel=\"Number of (transitively) referencing certificates\",\n",
     "    )\n",
-    "    fig = g.get_figure()\n",
-    "    fig.savefig(RESULTS_DIR / \"lineplot_avg_reach.pdf\", bbox_inches=\"tight\")\n",
-    "    plt.show()\n",
+    "    g.figure.savefig(RESULTS_DIR / \"lineplot_avg_n_references.pdf\", bbox_inches=\"tight\")\n",
+    "    g.figure.show()\n",
     "\n",
     "    return {}\n",
     "\n",
     "\n",
     "compute_avg_references_over_time(cc_df_comp)\n",
-    "compute_avg_reach_over_time(cc_df_comp)\n"
+    "compute_avg_reach_over_time(cc_df_comp)"
    ]
   },
   {
@@ -487,15 +494,15 @@
     "        .melt(id_vars=[\"date\"], var_name=\"category\", value_name=\"number of certificates\")\n",
     "    )\n",
     "\n",
+    "    plt.figure()\n",
     "    g = sns.lineplot(data=df_active_vs_ref_rich_melted, x=\"date\", y=\"number of certificates\", hue=\"category\")\n",
     "    g.set(\n",
     "        title=\"Number of active certificates vs. reference-rich certificates in time\",\n",
     "        xlabel=\"Time\",\n",
     "        ylabel=\"Number of certificates\",\n",
     "    )\n",
-    "    fig = g.get_figure()\n",
-    "    fig.savefig(RESULTS_DIR / \"lienplot_n_active_certs_vs_n_references.pdf\", bbox_inches=\"tight\")\n",
-    "    plt.show()\n",
+    "    g.figure.savefig(RESULTS_DIR / \"lienplot_n_active_certs_vs_n_references.pdf\", bbox_inches=\"tight\")\n",
+    "    g.figure.show()\n",
     "    return {}\n",
     "\n",
     "\n",
@@ -532,6 +539,8 @@
     "    df_summary_references_melted = df_summary_references.melt(\n",
     "        id_vars=[\"date\"], var_name=\"category\", value_name=\"number of certificates\"\n",
     "    )\n",
+    "\n",
+    "    plt.figure()\n",
     "    g = sns.lineplot(\n",
     "        data=df_summary_references_melted, x=\"date\", y=\"number of certificates\", hue=\"category\", errorbar=None\n",
     "    )\n",
@@ -540,9 +549,8 @@
     "        xlabel=\"Time\",\n",
     "        ylabel=\"Number of certificates\",\n",
     "    )\n",
-    "    fig = g.get_figure()\n",
-    "    fig.savefig(RESULTS_DIR / \"lineplot_references_summary.pdf\", bbox_inches=\"tight\")\n",
-    "    plt.show()\n",
+    "    g.figure.savefig(RESULTS_DIR / \"lineplot_references_summary.pdf\", bbox_inches=\"tight\")\n",
+    "    g.figure.show()\n",
     "\n",
     "    df_ratios = df_summary_references.copy()\n",
     "    df_ratios[\"ref. rich certificates\"] = df_ratios[\"ref. rich certificates\"] / df_ratios[\"active certificates\"]\n",
@@ -551,20 +559,21 @@
     "    df_ratios = df_ratios.drop(columns=[\"active certificates\"])\n",
     "    df_ratios_melted = df_ratios.melt(id_vars=[\"date\"], var_name=\"category\", value_name=\"ratio of certificates\")\n",
     "\n",
+    "    plt.figure()\n",
     "    g = sns.lineplot(data=df_ratios_melted, x=\"date\", y=\"ratio of certificates\", hue=\"category\", errorbar=None)\n",
     "    g.set(\n",
     "        title=\"ratio of reference-rich vs. referenced vs. isolated certificates in time\",\n",
     "        xlabel=\"Time\",\n",
     "        ylabel=\"Number of certificates\",\n",
     "    )\n",
-    "    plt.savefig(RESULTS_DIR / \"lineplot_reference_ratio.pdf\", bbox_inches=\"tight\")\n",
-    "    plt.show()\n",
+    "    g.figure.savefig(RESULTS_DIR / \"lineplot_reference_ratio.pdf\", bbox_inches=\"tight\")\n",
+    "    g.figure.show()\n",
     "\n",
     "    return {}\n",
     "\n",
     "\n",
     "compute_number_of_active_vs_ref_rich_certs_over_time(cc_df_comp)\n",
-    "compute_summary_active_vs_ref_rich_over_time(cc_df_comp)\n"
+    "compute_summary_active_vs_ref_rich_over_time(cc_df_comp)"
    ]
   },
   {
@@ -629,20 +638,79 @@
     "        .melt(id_vars=[\"date\"], var_name=\"reference type\", value_name=\"number of certificates\")\n",
     "    )\n",
     "\n",
+    "    plt.figure()\n",
     "    g = sns.lineplot(data=df_refs_to_archived_melted, x=\"date\", y=\"number of certificates\", hue=\"reference type\")\n",
     "    g.set(\n",
     "        title=\"Number of active certificates that reference some archived certificate\",\n",
     "        xlabel=\"Time\",\n",
     "        ylabel=\"Number of certificates\",\n",
     "    )\n",
-    "    fig = g.get_figure()\n",
-    "    fig.savefig(RESULTS_DIR / \"lienplot_active_certs_referencing_archived.pdf\", bbox_inches=\"tight\")\n",
-    "    plt.show()\n",
+    "    g.figure.savefig(RESULTS_DIR / \"lienplot_active_certs_referencing_archived.pdf\", bbox_inches=\"tight\")\n",
+    "    g.figure.show()\n",
     "\n",
     "    return {}\n",
     "\n",
     "\n",
-    "compute_certs_referencing_archived_ones(cc_df_comp)\n"
+    "compute_certs_referencing_archived_ones(cc_df_comp)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Certificates referencing an archived certificate on their issuance day"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def compute_refs_to_archived_on_issuance_day(cc_df_comp: pd.DataFrame, cc_dset: CCDataset) -> None:\n",
+    "    cert_id_to_dgst = {x.heuristics.cert_id: x.dgst for x in cc_dset}\n",
+    "\n",
+    "    def cert_was_archived_on_date(date: pd.Timestamp, cert_id: str) -> bool:\n",
+    "        if not date:\n",
+    "            return False\n",
+    "        if not dset[cert_id_to_dgst[cert_id]].not_valid_after:\n",
+    "            return False\n",
+    "        return date.date() > dset[cert_id_to_dgst[cert_id]].not_valid_after\n",
+    "\n",
+    "    df__ = cc_df_comp.copy()\n",
+    "    date_range = pd.date_range(df__.not_valid_before.min(), df__.not_valid_before.max())\n",
+    "\n",
+    "    df_problematic = pd.DataFrame()\n",
+    "    for date in tqdm(date_range):\n",
+    "        certs_issued = df__.loc[(date == df__.not_valid_before) & (df__.n_refs > 0)].assign(\n",
+    "            refs_to_archived_certs=lambda df_: df_.refs.map(\n",
+    "                lambda refs: {x for x in refs if cert_was_archived_on_date(date, x)}\n",
+    "            )\n",
+    "        )\n",
+    "        df_problematic = pd.concat(\n",
+    "            [df_problematic, certs_issued[certs_issued.refs_to_archived_certs.map(lambda x: len(x) > 0)]]\n",
+    "        )\n",
+    "\n",
+    "    df_problematic = df_problematic[\n",
+    "        [\n",
+    "            \"cert_id\",\n",
+    "            \"name\",\n",
+    "            \"status\",\n",
+    "            \"category\",\n",
+    "            \"manufacturer\",\n",
+    "            \"scheme\",\n",
+    "            \"eal\",\n",
+    "            \"not_valid_before\",\n",
+    "            \"not_valid_after\",\n",
+    "            \"cert_lab\",\n",
+    "            \"refs\",\n",
+    "            \"refs_to_archived_certs\",\n",
+    "        ]\n",
+    "    ]\n",
+    "    df_problematic.to_csv(RESULTS_DIR / \"certs_with_refs_to_archived_on_their_issuance_date.csv\", sep=\";\", index=False)\n",
+    "\n",
+    "\n",
+    "compute_refs_to_archived_on_issuance_day(cc_df_comp, dset)"
    ]
   },
   {
@@ -704,19 +772,77 @@
     "        .melt(id_vars=[\"date\"], var_name=\"reference type\", value_name=\"number of certificates\")\n",
     "    )\n",
     "\n",
+    "    plt.figure()\n",
     "    g = sns.lineplot(data=df_references_vuln_melted, x=\"date\", y=\"number of certificates\", hue=\"reference type\")\n",
     "    g.set(\n",
     "        title=\"Number of active certificates that reference some vulnerable certificate\",\n",
     "        xlabel=\"Time\",\n",
     "        ylabel=\"Number of certificates\",\n",
     "    )\n",
-    "    fig = g.get_figure()\n",
-    "    fig.savefig(RESULTS_DIR / \"lienplot_active_certs_referencing_vulnerable.pdf\", bbox_inches=\"tight\")\n",
-    "    plt.show()\n",
+    "    g.figure.savefig(RESULTS_DIR / \"lienplot_active_certs_referencing_vulnerable.pdf\", bbox_inches=\"tight\")\n",
+    "    g.figure.show()\n",
     "    return {}\n",
     "\n",
     "\n",
-    "compute_certs_referencing_vulnerable_over_time(cc_df_comp)\n"
+    "compute_certs_referencing_vulnerable_over_time(cc_df_comp)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Certificates with sub-component reference to lower EAL cert"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def compute_references_to_weaker_eals(cc_df_comp: pd.DataFrame, cc_dset: CCDataset) -> None:\n",
+    "    eal_to_rank = {category: index for index, category in enumerate(cc_df_comp.eal.dtype.categories)}\n",
+    "    cert_id_to_dst_mapping = {x.heuristics.cert_id: x.dgst for x in dset}\n",
+    "\n",
+    "    def ref_is_weaker(cert_id: str, eal: str, ref_cert_id: str) -> bool:\n",
+    "        eal_rank = eal_to_rank[eal]\n",
+    "        ref_eal = dset[cert_id_to_dst_mapping[ref_cert_id]].eal\n",
+    "        ref_eal_rank = eal_to_rank[ref_eal]\n",
+    "        return ref_eal_rank < eal_rank\n",
+    "\n",
+    "    df_ = cc_df_comp.copy()\n",
+    "    df_ = (\n",
+    "        df_.loc[\n",
+    "            (df_.n_refs > 0) & (df_.eal.notnull()),\n",
+    "            [\n",
+    "                \"cert_id\",\n",
+    "                \"name\",\n",
+    "                \"status\",\n",
+    "                \"category\",\n",
+    "                \"manufacturer\",\n",
+    "                \"scheme\",\n",
+    "                \"eal\",\n",
+    "                \"not_valid_before\",\n",
+    "                \"not_valid_after\",\n",
+    "                \"refs\",\n",
+    "            ],\n",
+    "        ]\n",
+    "        .assign(\n",
+    "            refs=lambda df__: df__.apply(\n",
+    "                lambda row: {x for x in row.refs if ref_is_weaker(row.cert_id, row.eal, x)}, axis=1\n",
+    "            ),\n",
+    "            n_refs=lambda df__: df__.refs.map(len_if_exists),\n",
+    "            referenced_levels=lambda df__: df__.refs.map(\n",
+    "                lambda x: {dset[cert_id_to_dst_mapping[y]].eal for y in x} if pd.notnull(x) else set()\n",
+    "            ),\n",
+    "        )\n",
+    "        .loc[lambda df__: df__.n_refs > 0]\n",
+    "    )\n",
+    "\n",
+    "    df_.to_csv(RESULTS_DIR / \"certs_referencing_weaker_eals.csv\", index=False, sep=\";\")\n",
+    "\n",
+    "\n",
+    "compute_references_to_weaker_eals(cc_df_comp, dset)"
    ]
   },
   {
@@ -760,7 +886,7 @@
     "\n",
     "\n",
     "plot_direct_refs_per_category(cc_df_comp)\n",
-    "plot_direct_refs_per_category(cc_df_prev)\n"
+    "plot_direct_refs_per_category(cc_df_prev)"
    ]
   },
   {
@@ -787,6 +913,8 @@
     "    exploded[\"ref_category\"] = exploded.refs.map(lambda x: cert_id_to_category_mapping[x] if pd.notnull(x) else np.nan)\n",
     "    exploded = exploded.loc[exploded.ref_category.notnull()]\n",
     "\n",
+    "    exploded_with_refs = exploded.loc[exploded.ref_category != \"No references\"]\n",
+    "\n",
     "    all_categories = set(exploded.category.unique()) | set(exploded.ref_category.unique())\n",
     "    colors = list(sns.color_palette(\"hls\", len(all_categories), as_cmap=False).as_hex())\n",
     "    color_dict = dict(zip(all_categories, colors))\n",
@@ -795,6 +923,7 @@
     "    figure.set_size_inches(24, 10)\n",
     "    figure.set_tight_layout(True)\n",
     "\n",
+    "    plt.figure()\n",
     "    sankey(\n",
     "        exploded.category,\n",
     "        exploded.ref_category,\n",
@@ -805,13 +934,12 @@
     "        ax=axes,\n",
     "    )\n",
     "\n",
-    "    # plt.show()\n",
-    "    plt.savefig(RESULTS_DIR / \"sankey_references_categories.pdf\", bbox_inches=\"tight\")\n",
+    "    plt.show()\n",
     "\n",
     "    return {}\n",
     "\n",
     "\n",
-    "plot_sankey_refs_categories(cc_df_comp)\n"
+    "plot_sankey_refs_categories(cc_df_comp)"
    ]
   },
   {
@@ -833,6 +961,7 @@
     "        has_outgoing_direct_references=lambda df_: df_.n_refs > 0,\n",
     "        has_incoming_direct_references=lambda df_: df_.n_in_refs > 0,\n",
     "    )\n",
+    "    plt.figure()\n",
     "    figure, axes = plt.subplots(1, 2)\n",
     "    figure.set_size_inches(14, 4)\n",
     "    figure.set_tight_layout(True)\n",
@@ -854,7 +983,7 @@
     "    return {}\n",
     "\n",
     "\n",
-    "plot_refs_per_scheme(cc_df)\n"
+    "plot_refs_per_scheme(cc_df)"
    ]
   },
   {
@@ -893,6 +1022,7 @@
     "\n",
     "    col_to_depict = [\"category\", \"scheme\"]\n",
     "\n",
+    "    plt.figure()\n",
     "    figure, axes = plt.subplots(1, 2)\n",
     "    figure.set_size_inches(14, 8)\n",
     "    figure.set_tight_layout(True)\n",
@@ -912,7 +1042,7 @@
     "    return {}\n",
     "\n",
     "\n",
-    "compute_certs_referencing_archived_ones(cc_df_comp)\n"
+    "compute_certs_referencing_archived_ones(cc_df_comp)"
    ]
   },
   {
@@ -943,6 +1073,7 @@
     "    colors = list(sns.color_palette(\"hls\", len(all_schemes), as_cmap=False).as_hex())\n",
     "    color_dict = dict(zip(all_schemes, colors))\n",
     "\n",
+    "    plt.figure()\n",
     "    figure, axes = plt.subplots(1, 1)\n",
     "    figure.set_size_inches(4, 4)\n",
     "    figure.set_tight_layout(True)\n",
@@ -964,28 +1095,7 @@
     "    return {}\n",
     "\n",
     "\n",
-    "plot_sankey_refs_schemes(cc_df)\n"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "## Assurance mismatch\n",
-    "\n",
-    "- Search for certificates that use stronger EAL than the references subcomponent\n",
-    "- Examine cycles in the subcomponent graph"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "# TBA\n",
-    "# Create a function that takes two arguments: cc_dset, eal, and references, and returns whether the cert. references something with lower EAL\n",
-    "# Just compute this boolean column for all certs\n"
+    "plot_sankey_refs_schemes(cc_df)"
    ]
   },
   {
@@ -1036,21 +1146,22 @@
     "        .melt(id_vars=[\"date\"], var_name=\"reference type\", value_name=\"number of certificates\")\n",
     "    )\n",
     "\n",
+    "    plt.figure()\n",
     "    g = sns.lineplot(data=df_references_melted, x=\"date\", y=\"number of certificates\", hue=\"reference type\")\n",
     "    g.set(\n",
     "        title=\"Sum of references in currently active certs. with different labels in time\",\n",
     "        xlabel=\"Time\",\n",
     "        ylabel=\"Number of references\",\n",
     "    )\n",
-    "    plt.savefig(RESULTS_DIR / \"lineplot_different_labels.pdf\", bbox_inches=\"tight\")\n",
-    "    plt.show()\n",
+    "    g.figure.savefig(RESULTS_DIR / \"lineplot_different_labels.pdf\", bbox_inches=\"tight\")\n",
+    "    g.figure.show()\n",
     "\n",
     "\n",
     "plot_ref_label_popularity_over_time(cc_df, cc_df_comp, cc_df_prev)\n",
     "\n",
     "print(\n",
     "    f\"Number of certificates that did undergo re-evaluation (or previous version): {len(set(itertools.chain.from_iterable(cc_df_prev.loc[cc_df_prev.refs.notnull()].refs.tolist())))}\"\n",
-    ")\n"
+    ")"
    ]
   },
   {
@@ -1096,7 +1207,7 @@
     "        break\n",
     "else:\n",
     "    raise ValueError(f\"Certificate with id {cert_id} not found in dataset.\")\n",
-    "print(f\" - its page is at https://seccerts.org/cc/{cert.dgst}/\")\n"
+    "print(f\" - its page is at https://seccerts.org/cc/{cert.dgst}/\")"
    ]
   },
   {
@@ -1105,7 +1216,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "nx.draw(view, with_labels=True)\n"
+    "nx.draw(view, with_labels=True)"
    ]
   },
   {
@@ -1169,7 +1280,7 @@
     "\n",
     "for com in communities:\n",
     "    for i in sorted(com):\n",
-    "        print(f\"\\t{i}\")\n"
+    "        print(f\"\\t{i}\")"
    ]
   },
   {
@@ -1221,7 +1332,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.10.13"
+   "version": "3.11.6"
   },
   "vscode": {
    "interpreter": {