process PP maintenances

2 files changed, 47 insertions, 4 deletions

diff --git a/src/sec_certs/sample/protection_profile.py b/src/sec_certs/sample/protection_profile.py
index 079c9319..9f499451 100644
--- a/src/sec_certs/sample/protection_profile.py
+++ b/src/sec_certs/sample/protection_profile.py
@@ -19,7 +19,7 @@ from sec_certs.sample.certificate import Heuristics as BaseHeuristics
 from sec_certs.sample.certificate import PdfData as BasePdfData
 from sec_certs.sample.document_state import DocumentState
 from sec_certs.serialization.json import ComplexSerializableType
-from sec_certs.utils import helpers, sanitization
+from sec_certs.utils import cc_html_parsing, helpers, sanitization
 
 
 class ProtectionProfile(
@@ -55,7 +55,7 @@ class ProtectionProfile(
         report_link: str | None
         pp_link: str | None
         scheme: str | None
-        maintenances: list[tuple[date, str, str]]
+        maintenances: list[tuple[Any]]
 
         @property
         def eal(self) -> str | None:
@@ -88,7 +88,12 @@ class ProtectionProfile(
             if not sanitization.sanitize_cc_link(pp_link):
                 raise ValueError(f"pp_link for PP {pp_name} is empty, cannot create PP record")
 
-            # TODO: Parse maintenance div here. See CC parsing.
+            mu_div = cc_html_parsing.html_row_get_maintenance_div(row)
+            maintenance_updates = cc_html_parsing.parse_maintenance_div(mu_div) if mu_div else []
+            if maintenance_updates:
+                # Drop ST links, not filled in for PPs
+                maintenance_updates = [x[:3] for x in maintenance_updates]
+
             return cls(
                 category,
                 status,
@@ -101,7 +106,7 @@ class ProtectionProfile(
                 cls._html_row_get_link(cells[-1]),
                 pp_link,
                 cls._html_row_get_scheme(cells[-2]),
-                [],
+                maintenance_updates,
             )
 
         @classmethod
diff --git a/src/sec_certs/utils/cc_html_parsing.py b/src/sec_certs/utils/cc_html_parsing.py
new file mode 100644
index 00000000..956353bf
--- /dev/null
+++ b/src/sec_certs/utils/cc_html_parsing.py
@@ -0,0 +1,38 @@
+import logging
+from datetime import datetime
+from typing import Any
+
+from bs4 import Tag
+
+from sec_certs import constants
+
+logger = logging.getLogger(__name__)
+
+
+def html_row_get_maintenance_div(cell: Tag) -> Tag | None:
+    divs = cell.find_all("div")
+    for d in divs:
+        if d.find("div") and d.stripped_strings and list(d.stripped_strings)[0] == "Maintenance Report(s)":
+            return d
+    return None
+
+
+def parse_maintenance_div(main_div: Tag) -> list[tuple[Any, ...]]:
+    possible_updates = list(main_div.find_all("li"))
+    maintenance_updates = set()
+    for u in possible_updates:
+        text = list(u.stripped_strings)[0]
+        main_date = datetime.strptime(text.split(" ")[0], "%Y-%m-%d").date() if text else None
+        main_title = text.split("– ")[1]
+        main_report_link = None
+        main_st_link = None
+        links = u.find_all("a")
+        for link in links:
+            if link.get("title").startswith("Maintenance Report:"):
+                main_report_link = constants.CC_PORTAL_BASE_URL + link.get("href")
+            elif link.get("title").startswith("Maintenance ST"):
+                main_st_link = constants.CC_PORTAL_BASE_URL + link.get("href")
+            else:
+                logger.error("Unknown link in Maintenance part!")
+        maintenance_updates.add((main_date, main_title, main_report_link, main_st_link))
+    return list(maintenance_updates)