Synopsis : normalisation SÛRE des text ajouté par formulaire

- fins de ligne \r\n/\r → \n,
- blocs de lignes vides → \n\n, è
- espaces/tabulations/NBSP → espace unique,
- trim.
Les retours à la ligne simples intra-paragraphe sont LAISSÉS INTACTS :
une coupure de ligne peut être une vraie frontière de mot (→ espace)
ou une coupure du mot ("dyna\nmiques" → "dynamiques"),
indistinguables sans dictionnaire — automatiser corromprait l'entrée
("poursuivantsur").

Migration batch 045 (backfill des données existantes) RETIRÉE : le
nettoyage automatique fiable des synopsis existants est impossible.
Chaque modification est testée (277 tests PHPUnit).

TODO.md

@@ -1,13 +1,14 @@
 # TODO
 
 > Last updated: 2026-08-26
-> Context: Docs refactor: verify docs/ accuracy against current xamxam codebase; update/refactor stale docs
+> Context: Fix end-user admin report: unexpected line breaks in some synopsis — resolved as undetectable/unfixable for existing data; only safe normalization on new inputs
 
 ## In Progress
 
 ## Pending
 
 ## Completed
+- [x] #normalize-synopsis-whitespace & Passages à la ligne inopinés dans les synopsis : jugés INDÉTECTABLES/non-corrigeables automatiquement (une coupure de ligne interne peut être une vraie frontière de mot → espace, ou une coupure du mot → à réjoindre ; indistinguables sans dictionnaire, toute tentative corrompt l'entrée p.ex. "poursuivantsur"). Migration 045 SUPPRIMÉE ; dites à l'admin qu'il n'y a pas de nettoyage fiable des données existantes. Conservation de la NORMALISATION SÛRE sur les nouvelles saisies : fin de ligne \r\n/\r → \n, pertes de lignes vides multiples → \n\n, espaces/tabulations/NBSP → espace unique, trim — retour à la ligne simple intra-paragraphe laissé intact.
 - [x] #audit-all-docs-and [!high] Audit all docs/ and classify accurate vs stale
 - [x] #rewrite-development-md-to-match [!high] Rewrite development.md to match current just dev / app/ layout / PHPUnit
 - [x] #rewrite-deployment-md-to-match [!high] Rewrite deployment.md to match just deploy / /var/www/xamxam/ / backup

app/src/Controllers/ThesisCreateController.php

@@ -609,12 +609,39 @@ class ThesisCreateController
     // ── Private: input helpers ────────────────────────────────────────────────
 
     /**
-     * Trim and strip HTML tags from a string value.
-     * htmlspecialchars is applied at render time, not here.
+     * Normalise a submitted string value.
+     *
+     * Only the UNAMBIGUOUS cleanups are applied here. We deliberately do NOT
+     * auto-collapse or rejoin single newlines inside a paragraph: a line break
+     * can be either a real word boundary (should become a space) or a word split
+     * across the line ("dyna\nmiques", should be rejoined to "dynamiques"),
+     * and these are indistinguishable without a dictionary. Trying to guess
+     * corrupts valid input (e.g. "se poursuivant\nsur" → "se poursuivantsur"),
+     * so such breaks are left untouched and must be fixed by the author.
+     *
+     * Responsibilities:
+     *  - strip HTML tags (htmlspecialchars is applied at render time, not here)
+     *  - normalise all line endings to \n (handles \r\n and bare \r)
+     *  - collapse runs of horizontal whitespace (spaces, tabs, NBSP) to one space
+     *  - collapse multiple blank lines into a single paragraph separator (\n\n)
+     *  - trim leading/trailing whitespace and blank lines
      */
     private function sanitiseString(string $input): string
     {
-        return strip_tags(trim($input));
+        $input = strip_tags($input);
+
+        // Normalise all line endings to \n (handles \r\n and bare \r).
+        $input = preg_replace('/\r\n|\r|\n/', "\n", $input);
+
+        // Collapse multiple blank lines into a single paragraph separator.
+        $input = preg_replace("/\n{2,}/u", "\n\n", $input);
+
+        // Collapse runs of horizontal whitespace (spaces, tabs, NBSP gaps, …)
+        // into a single space. Single newlines inside a paragraph are NOT
+        // touched, for the reason explained above.
+        $input = preg_replace('/[ \t\x0B\x00\xC2\xA0]+/u', ' ', $input);
+
+        return trim($input, " \t\n\r\0\x0B\xC2\xA0");
     }
 
     /**

tests/phpunit/ThesisCreateValidationTest.php

@@ -263,6 +263,96 @@ class ThesisCreateValidationTest extends TestCase
         $this->assertStringContainsString('Bold synopsis', $data['synopsis']);
     }
 
+    // ── Whitespace normalisation (safe, unambiguous subset only) ────────────────
+    // Single newlines inside a paragraph are deliberately LEFT untouched: they can
+    // be a word boundary (→ space) or a word split across a line ("dyna\nmiques"
+    // → "dynamiques") and are indistinguishable, so we never guess and risk
+    // corrupting valid input (e.g. "se poursuivant\nsur" → "poursuivantsur").
+
+    public function testSingleIntraParagraphNewlineIsLeftUntouched(): void
+    {
+        // A mid-word line break is ambiguous (could be a real split or a wrap),
+        // so we do not auto-alter it — the author is responsible for cleaning it.
+        $post = $this->validPost();
+        $post['synopsis'] = "…les dyn\namiques complexes…";
+
+        $data = $this->validate($post);
+        $this->assertSame("…les dyn\namiques complexes…", $data['synopsis']);
+    }
+
+    public function testWordWrapSingleNewlineNotMerged(): void
+    {
+        // We must NOT merge two real words across a line break ("poursuivantsur").
+        $post = $this->validPost();
+        $post['synopsis'] = "se poursuivant\nsur une seconde ligne";
+
+        $data = $this->validate($post);
+        $this->assertSame(
+            "se poursuivant\nsur une seconde ligne",
+            $data['synopsis']
+        );
+    }
+
+    public function testLegitimateParagraphBreakPreserved(): void
+    {
+        $post = $this->validPost();
+        $post['synopsis'] = "Premier paragraphe\n\nSecond paragraphe";
+
+        $data = $this->validate($post);
+        $this->assertSame(
+            "Premier paragraphe\n\nSecond paragraphe",
+            $data['synopsis']
+        );
+    }
+
+    public function testMultipleBlankLinesCollapseToSingleParagraphBreak(): void
+    {
+        $post = $this->validPost();
+        $post['synopsis'] = "A\n\n\n\nB";
+
+        $data = $this->validate($post);
+        $this->assertSame("A\n\nB", $data['synopsis']);
+    }
+
+    public function testMixedWindowsAndUnixLineEndingsNormalised(): void
+    {
+        $post = $this->validPost();
+        $post['synopsis'] = "une pause\r\n\r\ner une future";
+
+        $data = $this->validate($post);
+        // \r\n and \r line-endings all collapse to \n, and blank lines mark
+        // a single paragraph separator.
+        $this->assertSame("une pause\n\ner une future", $data['synopsis']);
+    }
+
+    public function testConsecutiveSpacesAndTabsCollapse(): void
+    {
+        $post = $this->validPost();
+        $post['synopsis'] = "Très    espacé\t\ttab";
+
+        $data = $this->validate($post);
+        $this->assertSame('Très espacé tab', $data['synopsis']);
+    }
+
+    public function testNonBreakingSpacesNormalised(): void
+    {
+        $post = $this->validPost();
+        // Word pasted NBSP (U+00A0) between words must become a regular space.
+        $post['synopsis'] = "Premier\u{00A0}mot second mot";
+
+        $data = $this->validate($post);
+        $this->assertSame('Premier mot second mot', $data['synopsis']);
+    }
+
+    public function testSynopsisSingleLineFieldTrimmed(): void
+    {
+        $post = $this->validPost();
+        $post['synopsis'] = "   texte qui flotte  ";
+
+        $data = $this->validate($post);
+        $this->assertSame('texte qui flotte', $data['synopsis']);
+    }
+
     // ── Author processing ────────────────────────────────────────────────────
 
     public function testMultipleAuthorsAreSorted(): void