From 902e43ea19a16539937cc4a5a60a5343daf22283 Mon Sep 17 00:00:00 2001
From: Levi Neuwirth <ln@levineuwirth.org>
Date: Wed, 10 Jun 2026 09:25:50 -0400
Subject: [PATCH] Add /poetry/ and /fiction/ indexes; widen tag-collision guard
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Nav, the home portal grid, and the library have linked both URLs since
the portals were added, but no rule generated either index — confirmed
404s in production (AUDIT §2.1). Both rules mirror the essays index;
fiction renders an empty list until content exists.

sectionOwnedTopLevelTags now lists every namespace owning a
<name>/index.html route, not just photography — Hakyll silently
overwrites on duplicate routes, so an essay tagged e.g. "music" would
have clobbered a real section landing (AUDIT §2.2).

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
---
 build/Site.hs | 40 ++++++++++++++++++++++++++++++++++++++++
 build/Tags.hs | 24 ++++++++++++------------
 2 files changed, 52 insertions(+), 12 deletions(-)

diff --git a/build/Site.hs b/build/Site.hs
index 8c0c2f0..8316a0d 100644
--- a/build/Site.hs
+++ b/build/Site.hs
@@ -566,6 +566,46 @@ rules = do
                 >>= loadAndApplyTemplate "templates/default.html"      ctx
                 >>= relativizeUrls
 
+    -- ---------------------------------------------------------------------------
+    -- Poetry index
+    -- ---------------------------------------------------------------------------
+    -- Nav, the home portal grid, and the library all link /poetry/; this
+    -- rule is what keeps those links from 404ing. Lists flat poems and
+    -- collection poems alike; collection index pages are excluded by
+    -- 'P.poetryPattern' itself.
+    create ["poetry/index.html"] $ do
+        route idRoute
+        compile $ do
+            poems <- recentFirst =<< loadAll (P.poetryPattern .&&. hasNoVersion)
+            let ctx =
+                    listField "essays" poetryCtx (return poems)
+                    <> constField "title"  "Poetry"
+                    <> constField "portal" "true"
+                    <> siteCtx
+            makeItem ""
+                >>= loadAndApplyTemplate "templates/essay-index.html" ctx
+                >>= loadAndApplyTemplate "templates/default.html"      ctx
+                >>= relativizeUrls
+
+    -- ---------------------------------------------------------------------------
+    -- Fiction index
+    -- ---------------------------------------------------------------------------
+    -- Same rationale as the poetry index. content/fiction/ has no entries
+    -- yet; an empty match list renders an empty index rather than a 404.
+    create ["fiction/index.html"] $ do
+        route idRoute
+        compile $ do
+            stories <- recentFirst =<< loadAll (P.fictionPattern .&&. hasNoVersion)
+            let ctx =
+                    listField "essays" fictionCtx (return stories)
+                    <> constField "title"  "Fiction"
+                    <> constField "portal" "true"
+                    <> siteCtx
+            makeItem ""
+                >>= loadAndApplyTemplate "templates/essay-index.html" ctx
+                >>= loadAndApplyTemplate "templates/default.html"      ctx
+                >>= relativizeUrls
+
     -- ---------------------------------------------------------------------------
     -- New page — all content sorted by creation date, newest first
     -- ---------------------------------------------------------------------------
diff --git a/build/Tags.hs b/build/Tags.hs
index 951e740..7d97baa 100644
--- a/build/Tags.hs
+++ b/build/Tags.hs
@@ -80,23 +80,23 @@ expandTag t =
 
 -- | Top-level tags that own a section URL outside the tag system, and
 --   therefore must NOT be created as tag pages — doing so would
---   collide with a section landing route. The literal @"photography"@
---   is the only one currently affected: every photo's @tags:@ list
---   begins with the bare @"photography"@ portal tag (per the section's
---   convention), and 'tagIdentifier' would route that to
---   @"photography/index.html"@ — already owned by
---   @photographyLandingRules@.
+--   collide with a section landing route. Hakyll does not error on
+--   duplicate routes (one item silently overwrites the other), so an
+--   essay tagged e.g. @music@ would otherwise clobber
+--   @music/index.html@. The set therefore lists every namespace that
+--   owns a @<name>/index.html@ route, not just the tags currently in
+--   use: @photography@ (every photo's @tags:@ list begins with it, per
+--   the section convention) plus the other section landings and
+--   generated index namespaces.
 --
 --   Sub-tags (@photography/landscape@, @photography/film@, …) are
 --   unaffected; they keep their tag pages because no section landing
 --   claims those URLs.
---
---   Other portal tags (@music@, @poetry@, @fiction@, …) don't appear
---   here because their content types don't currently feed
---   'tagIndexable', so the top-level tag never enters the tag system.
---   Add to this set if that ever changes.
 sectionOwnedTopLevelTags :: [String]
-sectionOwnedTopLevelTags = ["photography"]
+sectionOwnedTopLevelTags =
+    [ "photography", "poetry", "fiction", "music", "essays", "blog"
+    , "cv", "archive", "authors", "bibliography"
+    ]
 
 -- | All expanded tags for an item (reads the "tags" metadata field).
 --   Filters out any 'sectionOwnedTopLevelTags' to prevent route