ADR 0290 rewritten on the operator's word: a decision record lives whole in the knowledge base, reviewed and accepted there; and cycle.py sees issues that live only in the tracker (issue 436)

2026-10-11 02:05:02 +02:00
parent 08b1077b51
commit a5d1e0fbf6
9 changed files with 307 additions and 269 deletions
@@ -27,7 +27,7 @@ indistinguishable from one that cannot.
 | `live-citation` | a governing document citing a **superseded** record names its replacement in the same paragraph | `01-to-be/README.md` citing ADR 0022 as live guidance |
 | `supersession` | if A says it was superseded by B, B says it supersedes A | ADR 0012 never declared that it superseded 0011 |
 | `partial-supersession` | a record's `supersedes-in-part:` names records that exist, each carrying a dated *Superseded in part* note naming it under its heading; and every such note is named back by its superseder's field | 11 of 21 partial supersessions had no note on the record they change, and 6 more were said only in prose ([research 045](../../01-RESEARCH/045-a-knowledge-base-the-mesh-writes/00-overview.md)) |
-| `clauses` | an accepted record dated from 2026-10-11 writes its `## Decision` section as numbered clauses with a bold lead ([ADR 0290](../../02-DECISIONS/0290-the-knowledge-base-holds-each-records-structure-and-clauses-and-hq-keeps-its-reasoning.md) rule 3) | 226 of 263 decision records wrote their decision as prose, so the knowledge base must draft their clauses |
+| `clauses` | an accepted record dated from 2026-10-11 writes its `## Decision` section as numbered clauses with a bold lead ([ADR 0290](../../02-DECISIONS/0290-a-decision-record-lives-whole-in-the-knowledge-base-made-of-clauses.md) rule 3) | 226 of 263 decision records wrote their decision as prose, so the knowledge base must draft their clauses |
 | `numbering` | the number in the filename is the number in the heading | — |
 | `topics` | every record names a topic of the reading order in `02-DECISIONS/README.md` | — |
 | *(index.py)* | the reading order is written, every record's topic is one of it, and the README stores no list of records ([ADR 0248](../../02-DECISIONS/0248-the-decision-index-is-generated-when-it-is-read-and-never-stored.md)) | a stored list made every two open decisions conflict ([issue 297](../../04-ISSUES/297-two-open-decisions-always-conflict/00-report.md)); it failed on the README of `main` before the list was removed |
@@ -117,7 +117,14 @@ def covers(insight_section, insight_text, section, text):


 def issue_refs(text, issues):
-    return {n for a, b in ISSUE_REF.findall(text) for n in (a, b) if n and n in issues}
+    """The issues a text names as tracking it: one with a folder here, or one numbered above every folder.
+
+    Since ADR 0275 an issue from 365 on lives only in the issue tracker, and ADR 0291 removes the folders; a
+    checkout alone cannot see those, so a number above the highest folder counts as one the tracker holds
+    (issue 436). The tracker hands out numbers in order, so such a number was issued after the last folder.
+    """
+    top = max((int(k) for k in issues if k.isdigit()), default=0)
+    return {n for a, b in ISSUE_REF.findall(text) for n in (a, b) if n and (n in issues or int(n) > top)}


 def rel(path):