From 45e7465f7bf16cbbea74be59b291d589ebebd5e2 Mon Sep 17 00:00:00 2001
From: Harwinder Sekhon <hs.sekhon3@gmail.com>
Date: Sat, 11 Jul 2026 11:38:41 -0400
Subject: [PATCH] Probe video durations at scan time

Parse the mp4 mvhd box in pure PHP (ffprobe fallback) for videos with no
stored duration, so folder listings show durations before first play.
---
 README.md              |  6 ++-
 includes/functions.php | 97 ++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 101 insertions(+), 2 deletions(-)

diff --git a/README.md b/README.md
index 5a59d00..fa54848 100644
--- a/README.md
+++ b/README.md
@@ -106,8 +106,10 @@ attempt, and quiz scores count toward folder progress alongside videos.
   Completion is sticky — rewatching doesn't un-complete it.
 - Folder progress is computed over the folder's entire subtree using a path
   prefix match, with partial credit for videos in progress.
-- Video duration is captured from the browser on first play (no ffprobe
-  dependency) and cached on the `videos` row.
+- Video duration is read at scan time by parsing the mp4's `mvhd` header in
+  pure PHP (ffprobe fallback if the parse fails), so listings show durations
+  before anyone plays the video. The browser still reports the duration on
+  first play as a last-resort fallback.
 
 ## Production notes
 
diff --git a/includes/functions.php b/includes/functions.php
index 4c8d60d..842af98 100644
--- a/includes/functions.php
+++ b/includes/functions.php
@@ -6,6 +6,86 @@ function h(?string $s): string
     return htmlspecialchars($s ?? '', ENT_QUOTES, 'UTF-8');
 }
 
+/* ---------------------------------------------------------------------------
+ * Video duration probing (used by the scanner so listings can show durations
+ * before anyone plays the video). Pure-PHP mp4 parse first, ffprobe fallback.
+ * ------------------------------------------------------------------------- */
+function probeDuration(string $file): ?float
+{
+    $d = mp4Duration($file);
+    if ($d !== null && $d > 0) return $d;
+
+    if (function_exists('shell_exec')) {
+        $out = @shell_exec('ffprobe -v error -show_entries format=duration -of csv=p=0 '
+                           . escapeshellarg($file) . ' 2>/dev/null');
+        if ($out !== null && is_numeric(trim($out))) {
+            return (float)trim($out);
+        }
+    }
+    return null;
+}
+
+/** Read the duration from an mp4's moov/mvhd box. Returns seconds or null. */
+function mp4Duration(string $path): ?float
+{
+    $fp = @fopen($path, 'rb');
+    if (!$fp) return null;
+    $fileSize = filesize($path) ?: 0;
+    $duration = null;
+
+    $readBox = function (int $pos) use ($fp): ?array {
+        fseek($fp, $pos);
+        $hdr = fread($fp, 8);
+        if (strlen($hdr) < 8) return null;
+        $size   = unpack('N', substr($hdr, 0, 4))[1];
+        $type   = substr($hdr, 4, 4);
+        $hdrLen = 8;
+        if ($size === 1) {                    // 64-bit box size
+            $ext = fread($fp, 8);
+            if (strlen($ext) < 8) return null;
+            $size   = unpack('J', $ext)[1];
+            $hdrLen = 16;
+        }
+        return ($size >= $hdrLen) ? [$size, $type, $hdrLen] : null;
+    };
+
+    // Walk top-level boxes to find moov, then its mvhd child.
+    for ($pos = 0; $pos < $fileSize;) {
+        $box = $readBox($pos);
+        if (!$box) break;
+        [$size, $type, $hdrLen] = $box;
+        if ($size === 0) $size = $fileSize - $pos;   // box extends to EOF
+
+        if ($type === 'moov') {
+            for ($p = $pos + $hdrLen, $end = $pos + $size; $p < $end;) {
+                $child = $readBox($p);
+                if (!$child) break;
+                [$cSize, $cType, $cHdrLen] = $child;
+                if ($cType === 'mvhd') {
+                    fseek($fp, $p + $cHdrLen);
+                    $body = fread($fp, 32);
+                    if (strlen($body) >= 20) {
+                        if (ord($body[0]) === 1 && strlen($body) >= 32) {   // version 1
+                            $timescale = unpack('N', substr($body, 20, 4))[1];
+                            $dur       = unpack('J', substr($body, 24, 8))[1];
+                        } else {                                            // version 0
+                            $timescale = unpack('N', substr($body, 12, 4))[1];
+                            $dur       = unpack('N', substr($body, 16, 4))[1];
+                        }
+                        if ($timescale > 0) $duration = $dur / $timescale;
+                    }
+                    break 2;
+                }
+                $p += $cSize;
+            }
+            break;
+        }
+        $pos += $size;
+    }
+    fclose($fp);
+    return $duration;
+}
+
 /* ---------------------------------------------------------------------------
  * Quiz file parser
  * Format ("*_quiz_solved.txt"):
@@ -160,6 +240,23 @@ function scanLibrary(): array
         $db->rollBack();
         throw $e;
     }
+
+    // Fill in durations for videos that don't have one yet (new files, or
+    // indexed before duration probing existed). Only missing ones are probed,
+    // so rescans stay fast.
+    $stats['durations_probed'] = 0;
+    $rows = $db->query(
+        'SELECT id, path FROM videos WHERE duration_seconds IS NULL AND is_missing = 0'
+    )->fetchAll();
+    $upd = $db->prepare('UPDATE videos SET duration_seconds = ? WHERE id = ?');
+    foreach ($rows as $r) {
+        $d = probeDuration($root . '/' . $r['path']);
+        if ($d !== null && $d > 0) {
+            $upd->execute([$d, (int)$r['id']]);
+            $stats['durations_probed']++;
+        }
+    }
+
     return $stats;
 }