feat(pipeline): extract Ofsted report-card judgements (rc_* columns)

Wires the tap TODO in stg_ofsted_inspections.sql: maps the 7 confirmed
report-card MI columns (Safeguarding standards, Inclusion, Curriculum
and teaching, Achievement, Attendance and behaviour, Personal
development and wellbeing, Leadership and governance) into rc_*
fields, parsed via the new parse_report_card_grade macro against
real sampled grade values (Exceptional/Strong standard/Expected
standard/Needs attention/Urgent improvement). rc_safeguarding_met
becomes boolean from Met/Not met. rc_early_years/rc_sixth_form have
no MI column yet and are intentionally omitted from COLUMN_PRIORITY,
staying NULL.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_0146VHeLAWjDVE2B5uU67jCB

pipeline/plugins/extractors/tap-uk-ofsted/tap_uk_ofsted/tap.py

@@ -68,6 +68,19 @@ COLUMN_PRIORITY = {
     "ungraded_inspection_date": [
         "Date of latest ungraded inspection",
     ],
+    # Report Card fields (post-Nov 2025 framework). Confirmed verbatim MI
+    # headers per diagnose_compare_gaps.py's Task 1(c) findings. No MI column
+    # currently exists for early-years or sixth-form report-card grades, so
+    # those two fields are deliberately omitted here (see schema below) --
+    # they stay absent from every record, same as the existing `report_url`
+    # pattern for fields with no COLUMN_PRIORITY entry.
+    "rc_safeguarding_met": ["Safeguarding standards"],
+    "rc_inclusion": ["Inclusion"],
+    "rc_curriculum_teaching": ["Curriculum and teaching"],
+    "rc_achievement": ["Achievement"],
+    "rc_attendance_behaviour": ["Attendance and behaviour"],
+    "rc_personal_development": ["Personal development and wellbeing"],
+    "rc_leadership_governance": ["Leadership and governance"],
 }
 
 
@@ -111,6 +124,17 @@ class OfstedInspectionsStream(Stream):
         th.Property("sixth_form_provision", th.StringType),
         th.Property("ungraded_outcome", th.StringType),
         th.Property("ungraded_inspection_date", th.StringType),
+        th.Property("rc_safeguarding_met", th.StringType),
+        th.Property("rc_inclusion", th.StringType),
+        th.Property("rc_curriculum_teaching", th.StringType),
+        th.Property("rc_achievement", th.StringType),
+        th.Property("rc_attendance_behaviour", th.StringType),
+        th.Property("rc_personal_development", th.StringType),
+        th.Property("rc_leadership_governance", th.StringType),
+        # No MI column exists for these yet; declared for forward
+        # compatibility with the mart schema, always emitted as absent/NULL.
+        th.Property("rc_early_years", th.StringType),
+        th.Property("rc_sixth_form", th.StringType),
         th.Property("report_url", th.StringType),
     ).to_dict()
 

pipeline/transform/macros/parse_report_card_grade.sql

@@ -0,0 +1,17 @@
+-- Macro: Parse Ofsted Report Card grade (post-Nov 2025 framework) from text
+-- into the 5-point scale. Real values confirmed via a live sample of the MI
+-- CSV (see pipeline/scripts/diagnose_compare_gaps.py's
+-- "TASK 7 VALUE SAMPLE 2026-07-12" note) -- unrecognised text (including the
+-- 'NULL' sentinel used by the source CSV for blanks) parses to NULL, never
+-- errors.
+
+{% macro parse_report_card_grade(column_name) %}
+    case lower(trim(nullif({{ column_name }}, 'NULL')))
+        when 'exceptional' then 1
+        when 'strong standard' then 2
+        when 'expected standard' then 3
+        when 'needs attention' then 4
+        when 'urgent improvement' then 5
+        else null
+    end
+{% endmacro %}

pipeline/transform/models/staging/stg_ofsted_inspections.sql

@@ -33,17 +33,20 @@ renamed as (
         nullif(trim(ungraded_outcome), 'NULL')                                  as ungraded_outcome,
         {{ parse_ungraded_outcome('ungraded_outcome') }}::integer               as ungraded_grade,
 
-        -- Report Card fields (post-Nov 2025 framework)
+        -- Report Card fields (post-Nov 2025 framework), 5-point scale:
-        -- TODO: add rc_* columns to tap-uk-ofsted schema once CSV column names are confirmed
+        -- 1 Exceptional · 2 Strong standard · 3 Expected standard
-        null::text as rc_safeguarding_met,
+        -- · 4 Needs attention · 5 Urgent improvement
-        null::text as rc_inclusion,
+        (lower(trim(nullif(rc_safeguarding_met, 'NULL'))) = 'met')        as rc_safeguarding_met,
-        null::text as rc_curriculum_teaching,
+        {{ parse_report_card_grade('rc_inclusion') }}::integer            as rc_inclusion,
-        null::text as rc_achievement,
+        {{ parse_report_card_grade('rc_curriculum_teaching') }}::integer  as rc_curriculum_teaching,
-        null::text as rc_attendance_behaviour,
+        {{ parse_report_card_grade('rc_achievement') }}::integer          as rc_achievement,
-        null::text as rc_personal_development,
+        {{ parse_report_card_grade('rc_attendance_behaviour') }}::integer as rc_attendance_behaviour,
-        null::text as rc_leadership_governance,
+        {{ parse_report_card_grade('rc_personal_development') }}::integer as rc_personal_development,
-        null::text as rc_early_years,
+        {{ parse_report_card_grade('rc_leadership_governance') }}::integer as rc_leadership_governance,
-        null::text as rc_sixth_form,
+        -- No MI column exists for these yet (see tap.py); the tap never
+        -- emits rc_early_years/rc_sixth_form, so these stay NULL.
+        null::integer as rc_early_years,
+        null::integer as rc_sixth_form,
 
         report_url
     from source