m105
4c0cc5a7f5
The CDX languages field (ISO 639 codes, e.g. "spa,eng") has been captured on every page since the first ingestion command, but never surfaced anywhere. Filters via FIND_IN_SET for an exact code match rather than a LIKE substring match, since the column holds a comma-separated list. Verified against real data: lang=spa returns exactly the expected 47 sites (25+22 across two pages), a bogus code renders the empty state, and lang=spa combined with tld=es correctly narrows to 20.
108 lines
3.8 KiB
PHP
108 lines
3.8 KiB
PHP
<?php
|
|
|
|
namespace App\Livewire\Discover;
|
|
|
|
use App\Models\Site;
|
|
use Illuminate\Contracts\View\View;
|
|
use Illuminate\Pagination\LengthAwarePaginator;
|
|
use Livewire\Attributes\Layout;
|
|
use Livewire\Attributes\Url;
|
|
use Livewire\Component;
|
|
use Livewire\WithPagination;
|
|
|
|
/**
|
|
* The competitor-discovery search: "who sells honey with a .es domain?" — keyword
|
|
* matches a site's pages via MySQL fulltext (natural language mode, so free-text input
|
|
* never trips a boolean-mode syntax error), TLD filters the site's domain suffix,
|
|
* platform filters its detected server/powered-by/generator signature, language
|
|
* filters by the CDX-reported ISO 639 code of a page's content (e.g. "spa", "eng").
|
|
*/
|
|
#[Layout('components.layouts.app')]
|
|
class Search extends Component
|
|
{
|
|
use WithPagination;
|
|
|
|
#[Url(as: 'q', history: true)]
|
|
public string $keyword = '';
|
|
|
|
#[Url(as: 'tld', history: true)]
|
|
public string $tld = '';
|
|
|
|
#[Url(as: 'platform', history: true)]
|
|
public string $platform = '';
|
|
|
|
#[Url(as: 'lang', history: true)]
|
|
public string $language = '';
|
|
|
|
public function updatingKeyword(): void
|
|
{
|
|
$this->resetPage();
|
|
}
|
|
|
|
public function updatingTld(): void
|
|
{
|
|
$this->resetPage();
|
|
}
|
|
|
|
public function updatingPlatform(): void
|
|
{
|
|
$this->resetPage();
|
|
}
|
|
|
|
public function updatingLanguage(): void
|
|
{
|
|
$this->resetPage();
|
|
}
|
|
|
|
public function render(): View
|
|
{
|
|
$keyword = trim($this->keyword);
|
|
$tld = strtolower(trim(ltrim($this->tld, '.')));
|
|
$platform = trim($this->platform);
|
|
$language = strtolower(trim($this->language));
|
|
$hasQuery = $keyword !== '' || $tld !== '' || $platform !== '' || $language !== '';
|
|
|
|
// Without any filter this would just be the full site list — that's what the
|
|
// Sites page is for. Discover only returns results once you've asked it something.
|
|
if (! $hasQuery) {
|
|
$sites = new LengthAwarePaginator([], 0, 25, 1, ['path' => request()->url()]);
|
|
} else {
|
|
$sites = Site::query()
|
|
->when(
|
|
$keyword !== '',
|
|
fn ($query) => $query->whereHas(
|
|
'pages',
|
|
fn ($pageQuery) => $pageQuery->whereFullText(['title', 'description', 'keywords', 'content'], $keyword)
|
|
)
|
|
)
|
|
->when($tld !== '', fn ($query) => $query->where('domain', 'like', '%.'.$tld))
|
|
->when($platform !== '', function ($query) use ($platform) {
|
|
$query->where(function ($query) use ($platform) {
|
|
$query->where('server', 'like', '%'.$platform.'%')
|
|
->orWhere('powered_by', 'like', '%'.$platform.'%')
|
|
->orWhere('generator', 'like', '%'.$platform.'%');
|
|
});
|
|
})
|
|
->when(
|
|
$language !== '',
|
|
// languages is a comma-separated list of ISO 639 codes (e.g. "spa,eng") —
|
|
// FIND_IN_SET matches an exact code, not a LIKE substring match.
|
|
fn ($query) => $query->whereHas(
|
|
'pages',
|
|
fn ($pageQuery) => $pageQuery->whereRaw('FIND_IN_SET(?, languages)', [$language])
|
|
)
|
|
)
|
|
->withCount(['pages as matching_pages_count' => function ($pageQuery) use ($keyword) {
|
|
if ($keyword !== '') {
|
|
$pageQuery->whereFullText(['title', 'description', 'keywords', 'content'], $keyword);
|
|
}
|
|
}])
|
|
->orderByDesc('matching_pages_count')
|
|
->orderBy('domain')
|
|
->paginate(25);
|
|
}
|
|
|
|
return view('livewire.discover.search', ['sites' => $sites, 'hasQuery' => $hasQuery]);
|
|
}
|
|
}
|