Skip to content

Implement HTTP caching for file downloads with If-Modified-Since #140

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
Copilot wants to merge 5 commits into
base: main
Choose a base branch
from
Open

Implement HTTP caching for file downloads with If-Modified-Since #140

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
50da980
Initial plan
Copilot Jan 23, 2026
e8cf080
Implement If-Modified-Since caching for file downloads
Copilot Jan 23, 2026
76ba616
Address PR feedback: fix queryOne, move headers, fix attachment_id, a…
Copilot Jan 24, 2026
066bf14
Use queryOneOrNone consistently in getThreadEmailAttachmentWithTimestamp
Copilot Jan 24, 2026
5f4561f
Refactor: Extract attachment mapping to private helper method
Copilot Jan 25, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
71 changes: 59 additions & 12 deletions organizer/src/class/ThreadStorageManager.php
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -46,21 +46,29 @@ public function getThreadEmailContent($thread_id, $email_id) {

return stream_get_contents($content);
}
public function getThreadEmailAttachment(Thread $thread, $attachment_location) {
// Get attachment from database
$rows = Database::queryOne(
"SELECT tea.name, tea.filename, tea.filetype, tea.location, tea.status_type, tea.status_text,
tea.content
FROM thread_email_attachments tea
JOIN thread_emails te ON tea.email_id = te.id
WHERE te.thread_id = ? AND tea.location = ?",
[$thread->id, $attachment_location]

public function getThreadEmailContentWithTimestamp($thread_id, $email_id) {
// Get email content and timestamp from database
$row = Database::queryOneOrNone(
"SELECT content, timestamp_received FROM thread_emails WHERE thread_id = ? AND id = ?",
[$thread_id, $email_id]
);

if (empty($rows)) {
throw new Exception("Thread Email Attachment not found [thread_id=$thread->id, attachment_id=$attachment_location]");
if (empty($row)) {
return null;
}


return [
'content' => stream_get_contents($row['content']),
'timestamp' => $row['timestamp_received']
];
}
/**
* Maps database row to ThreadEmailAttachment object
* @param array $rows Database row with attachment data
* @return ThreadEmailAttachment
*/
private function mapRowToAttachment(array $rows) {
$attachment = new ThreadEmailAttachment();
$attachment->name = $rows['name'];
$attachment->filename = $rows['filename'];
Expand All @@ -82,4 +90,43 @@ public function getThreadEmailAttachment(Thread $thread, $attachment_location) {

return $attachment;
}

public function getThreadEmailAttachment(Thread $thread, $attachment_location) {
// Get attachment from database
$rows = Database::queryOne(
"SELECT tea.name, tea.filename, tea.filetype, tea.location, tea.status_type, tea.status_text,
tea.content
FROM thread_email_attachments tea
JOIN thread_emails te ON tea.email_id = te.id
WHERE te.thread_id = ? AND tea.location = ?",
[$thread->id, $attachment_location]
);

if (empty($rows)) {
throw new Exception("Thread Email Attachment not found [thread_id=$thread->id, attachment_id=$attachment_location]");
}

return $this->mapRowToAttachment($rows);
}

public function getThreadEmailAttachmentWithTimestamp(Thread $thread, $attachment_location) {
// Get attachment from database with timestamp
$rows = Database::queryOneOrNone(
"SELECT tea.name, tea.filename, tea.filetype, tea.location, tea.status_type, tea.status_text,
tea.content, tea.created_at
FROM thread_email_attachments tea
JOIN thread_emails te ON tea.email_id = te.id
WHERE te.thread_id = ? AND tea.location = ?",
[$thread->id, $attachment_location]
);

if (empty($rows)) {
throw new Exception("Thread Email Attachment not found [thread_id=$thread->id, attachment_id=$attachment_location]");
}

return [
'attachment' => $this->mapRowToAttachment($rows),
'timestamp' => $rows['created_at']
];
}
}
90 changes: 90 additions & 0 deletions organizer/src/e2e-tests/pages/FilePageTest.php
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -17,11 +17,101 @@ public function testFileEmailBodyHappy() {

// :: Assert content type header for file download
$this->assertStringContainsString('Content-Type: text/html; charset=UTF-8', $response->headers);

// :: Assert Last-Modified header is present
$this->assertStringContainsString('Last-Modified:', $response->headers);

// :: Assert Cache-Control header is present
$this->assertStringContainsString('Cache-Control:', $response->headers);

// Contain "mailHeaders" as part of the output
$this->assertStringContainsString('HEADERS', $response->body);
$this->assertStringContainsString('From: sender@example.com', $response->body);
}

public function testFileEmailBodyIfModifiedSince() {
$testData = E2ETestSetup::createTestThread();
$file = [
'email_id' => $testData['email_id'],
'thread_id' => $testData['thread']->id,
'entity_id' => $testData['entity_id']
];

// First request - get the Last-Modified header
$response = $this->renderPage('/file?entityId=' . $file['entity_id'] . '&threadId=' . $file['thread_id'] . '&body=' . $file['email_id']);

// Extract Last-Modified header from response
$lastModified = null;
$lines = explode("\n", $response->headers);
foreach($lines as $line) {
if (stripos($line, 'Last-Modified:') !== false) {
$lastModified = trim(substr($line, strlen('Last-Modified:')));
break;
}
}

$this->assertNotNull($lastModified, 'Last-Modified header should be present');

// Second request - with If-Modified-Since header set to the same time
$response2 = $this->renderPage(
'/file?entityId=' . $file['entity_id'] . '&threadId=' . $file['thread_id'] . '&body=' . $file['email_id'],
'dev-user-id',
'GET',
'304 Not Modified',
null,
['If-Modified-Since: ' . $lastModified]
);

// Body should be empty for 304 response
$this->assertEmpty($response2->body, '304 response should have empty body');
}

public function testFileAttachmentIfModifiedSince() {
$testData = E2ETestSetup::createTestThread();
$file = [
'attachment_id' => $testData['attachment_id'],
'thread_id' => $testData['thread']->id,
'entity_id' => $testData['entity_id']
];

// First request - get the Last-Modified header
$response = $this->renderPage('/file?entityId=' . $file['entity_id'] . '&threadId=' . $file['thread_id'] . '&attachmentId=' . $file['attachment_id']);

// :: Assert content type header for attachment download
$this->assertStringContainsString('Content-Type: application/pdf', $response->headers);

// :: Assert Last-Modified header is present
$this->assertStringContainsString('Last-Modified:', $response->headers);

// :: Assert Cache-Control header with immutable directive
$this->assertStringContainsString('Cache-Control:', $response->headers);
$this->assertStringContainsString('immutable', $response->headers);

// Extract Last-Modified header from response
$lastModified = null;
$lines = explode("\n", $response->headers);
foreach($lines as $line) {
if (stripos($line, 'Last-Modified:') !== false) {
$lastModified = trim(substr($line, strlen('Last-Modified:')));
break;
}
}

$this->assertNotNull($lastModified, 'Last-Modified header should be present');

// Second request - with If-Modified-Since header set to the same time
$response2 = $this->renderPage(
'/file?entityId=' . $file['entity_id'] . '&threadId=' . $file['thread_id'] . '&attachmentId=' . $file['attachment_id'],
'dev-user-id',
'GET',
'304 Not Modified',
null,
['If-Modified-Since: ' . $lastModified]
);

// Body should be empty for 304 response
$this->assertEmpty($response2->body, '304 response should have empty body');
}
/*
public function testFilePdfHappy() {
// TODO:
Expand Down
13 changes: 9 additions & 4 deletions organizer/src/e2e-tests/pages/common/E2EPageTestCase.php
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7,17 +7,17 @@ class E2EPageTestCase extends TestCase {

private static $session_cookies = array();

protected function renderPage($path, $user = 'dev-user-id', $method = 'GET', $expected_status = '200 OK', $post_data = null) {
protected function renderPage($path, $user = 'dev-user-id', $method = 'GET', $expected_status = '200 OK', $post_data = null, $extra_headers = array()) {
$url = 'http://localhost:25081' . $path;
if ($user !== null) {
if (!isset(self::$session_cookies[$user])) {
$session_cookie = $this->authenticate($user);
self::$session_cookies[$user] = $session_cookie;
}
$session_cookie = self::$session_cookies[$user];
$response = $this->curl($url, $method, $session_cookie, post_data: $post_data);
$response = $this->curl($url, $method, $session_cookie, post_data: $post_data, extra_headers: $extra_headers);
} else {
$response = $this->curl($url, $method, post_data: $post_data);
$response = $this->curl($url, $method, post_data: $post_data, extra_headers: $extra_headers);
}

if ($expected_status != null) {
Expand Down Expand Up @@ -55,7 +55,7 @@ private function authenticate($user) {
return $response->cookies[0];
}

private function curl($url, $method = 'GET', $session_cookie = null, $headers = array(), $post_data = null) {
private function curl($url, $method = 'GET', $session_cookie = null, $headers = array(), $post_data = null, $extra_headers = array()) {
//echo date('Y-m-d H:i:s') . " - $method $url $session_cookie\n";

$ch = curl_init();
Expand All @@ -68,6 +68,11 @@ private function curl($url, $method = 'GET', $session_cookie = null, $headers =
$headers[] = 'Cookie: ' . $session_cookie;
}
$headers[] = 'User-Agent: Offpost E2E Test';

// Add any extra headers
foreach ($extra_headers as $header) {
$headers[] = $header;
}

if (!empty($headers)) {
curl_setopt($ch, CURLOPT_HTTPHEADER, $headers);
Expand Down
47 changes: 44 additions & 3 deletions organizer/src/file.php
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -44,15 +44,38 @@

foreach ($thread->emails as $email) {
if (isset($_GET['body']) && $_GET['body'] == $email->id) {
$eml = ThreadStorageManager::getInstance()->getThreadEmailContent($thread->id, $email->id);
$emailData = ThreadStorageManager::getInstance()->getThreadEmailContentWithTimestamp($thread->id, $email->id);
if (!$emailData) {
http_response_code(404);
header('Content-Type: text/plain');
die("Email content not found");
}

$eml = $emailData['content'];
$lastModified = strtotime($emailData['timestamp']);

// Check If-Modified-Since header early for 304 response
if (isset($_SERVER['HTTP_IF_MODIFIED_SINCE'])) {
$ifModifiedSince = strtotime($_SERVER['HTTP_IF_MODIFIED_SINCE']);
if ($ifModifiedSince >= $lastModified) {
// Set headers before 304 response
header('Last-Modified: ' . gmdate('D, d M Y H:i:s', $lastModified) . ' GMT');
header('Cache-Control: private, max-age=3600');
http_response_code(304);
exit;
}
}

header('Content-Type: text/html; charset=UTF-8');

// Set Content-Security-Policy header to prevent XSS somewhat
header("Content-Security-Policy: default-src 'none'; script-src 'self'; style-src 'self'; img-src 'self' data:; frame-src 'none'; object-src 'none'; base-uri 'none'; form-action 'none';");

$email_content = ThreadEmailExtractorEmailBody::extractContentFromEmail($eml);

// Set caching headers after successful extraction
header('Last-Modified: ' . gmdate('D, d M Y H:i:s', $lastModified) . ' GMT');
header('Cache-Control: private, max-age=3600');

$subject = Imap\ImapEmail::getEmailSubject($eml);
echo '<h1 id="email-subject">Subject: ' . htmlescape($subject) . '</h1>' . chr(10);
Expand Down Expand Up @@ -109,9 +132,27 @@
// New format of location
$filename = $att->location;
}
$att = ThreadStorageManager::getInstance()->getThreadEmailAttachment($thread, $att->location);
$attachmentData = ThreadStorageManager::getInstance()->getThreadEmailAttachmentWithTimestamp($thread, $att->location);
$att = $attachmentData['attachment'];
$lastModified = strtotime($attachmentData['timestamp']);

if (empty($att->content)) {
throw new Exception("Attachment content empty: threadId={$threadId}, attachmentId={$att->attachment_id}", 404);
throw new Exception("Attachment content empty: threadId={$threadId}, attachmentId={$att->id}", 404);
}

// Set Last-Modified header
header('Last-Modified: ' . gmdate('D, d M Y H:i:s', $lastModified) . ' GMT');

// Set Cache-Control header
header('Cache-Control: private, max-age=31536000, immutable'); // 1 year for attachments

// Check If-Modified-Since header
if (isset($_SERVER['HTTP_IF_MODIFIED_SINCE'])) {
$ifModifiedSince = strtotime($_SERVER['HTTP_IF_MODIFIED_SINCE']);
if ($ifModifiedSince >= $lastModified) {
http_response_code(304);
exit;
}
}

if ($att->filetype == 'pdf') {
Expand Down