Add Status Report to Info page

internal/wisski/ingredient/php/extras/requirements.go

@@ -0,0 +1,66 @@
+package extras
+
+import (
+	"context"
+
+	"github.com/FAU-CDI/wisski-distillery/internal/phpx"
+	"github.com/FAU-CDI/wisski-distillery/internal/status"
+	"github.com/FAU-CDI/wisski-distillery/internal/wisski/ingredient"
+	"github.com/FAU-CDI/wisski-distillery/internal/wisski/ingredient/php"
+	"golang.org/x/exp/slices"
+
+	_ "embed"
+)
+
+type Requirements struct {
+	ingredient.Base
+	Dependencies struct {
+		PHP *php.PHP
+	}
+}
+
+var (
+	_ ingredient.WissKIFetcher = (*Requirements)(nil)
+)
+
+//go:embed requirements.php
+var requirementsPHP string
+
+// Create creates a new block with the given title and html content
+func (requirements *Requirements) Get(ctx context.Context, server *phpx.Server) (data []status.Requirement, err error) {
+	err = requirements.Dependencies.PHP.ExecScript(ctx, server, &data, requirementsPHP, "get_requirements", requirements.URL().String())
+	if err == nil {
+		// sort first by weight, then by id!
+		slices.SortFunc(data, func(a, b status.Requirement) bool {
+			// compare first by weight
+			if a.Weight < b.Weight {
+				return true
+			}
+			if a.Weight > b.Weight {
+				return false
+			}
+
+			// then by severity
+			if a.Severity < b.Severity {
+				return true
+			}
+			if a.Severity > b.Severity {
+				return false
+			}
+
+			// and finally by id
+			return a.ID < b.ID
+		})
+	}
+	return
+}
+
+// Fetch fetches information
+func (requirements *Requirements) Fetch(flags ingredient.FetcherFlags, target *status.WissKI) error {
+	if flags.Quick {
+		return nil
+	}
+
+	target.Requirements, _ = requirements.Get(flags.Context, flags.Server)
+	return nil
+}

internal/wisski/ingredient/php/extras/requirements.php

@@ -0,0 +1,125 @@
+<?php
+
+/**
+ * Returns a well-typed array of all requirements.
+ * Relative URLs will be placed in the given domain.
+ */
+function get_requirements(string $public = ""): array {
+    $results = [];
+
+    $managerService = \Drupal::service('system.manager');
+    $rendererService = \Drupal::service('renderer');
+
+    $requirements = $managerService->listRequirements();
+    foreach($requirements as $id => $req) {
+        $title = $req['title'] ?? NULL;
+        $weight = $req['weight'] ?? NULL;
+        $severity = $req['severity'] ?? NULL;
+        $value = $req['value'] ?? NULL;
+        $description = $req['description'] ?? NULL;
+
+        $results[] = array(
+            "id" => $id,
+            "title" => strip_tags(ensure_html($rendererService, $title)),
+            "weight" => is_numeric($weight) ? $weight : 0,
+            "severity" => is_numeric($severity) ? $severity : 0, 
+            "value" => clean_html(ensure_html($rendererService, $value), $public),
+            "description" => clean_html(ensure_html($rendererService, $description), $public),
+        );
+    }
+
+    return $results;
+}
+
+
+/**
+ * Ensures that the passed data is drupal html code.
+ *
+ * @param mixed $data
+ * @return string
+ */
+function ensure_html(mixed $renderer, mixed $data): string {
+  // already a string => return it!
+  if (is_string($data)) {
+	  return $data;
+  }
+
+  // it is null => return the empty string
+  if ($data === NULL) {
+	  return "";
+  }
+
+  // create a render array and render it!
+  $rary = is_array($data) ? $data : ["#markup" => $data];
+  return $renderer->renderPlain($rary);
+}
+
+/**
+ * Parses source as an html fragment.
+ * It then iterates over all '<a>' elements and performs the following modifications:
+ * - Setup the distillery redirector (/next/?next=...) on relative links according to $public
+ * - Make all links target="_blank" rel="noopener"
+  */
+function clean_html(string $source, string $public): string|bool {
+    // fast path: we don't have a source
+    if ($source === "" ) {
+        return $source;
+    }
+
+    // trim the trailing '/' from the public URL.
+    // this technically trims the beginning as well, but that should not be a problem
+    $public = trim($public, '/');
+
+    // attempt to parse a new document
+    $doc = new DOMDocument();
+    $ok = $doc->loadHTML('<?xml encoding="UTF-8">' . $source);
+    if ($ok === FALSE) {
+        return FALSE;
+    }
+
+    // replace all the <a href="..."> which are relative
+    $as = $doc->getElementsByTagName('a');
+    $modified = $as->count() > 0;
+    foreach($as as $a) {
+        // setup a link to open in a new tab
+        $a->setAttribute('rel', 'noopener noreferer');
+        $a->setAttribute('target', '_blank');
+
+        // if we don't have a domain don't even bother replacing hrefs.
+        if ($domain === "") {
+            continue;
+        }
+
+        // take only href="/" relative to the current domain.
+        $href = $a->getAttribute('href');
+        if (!is_string($href) || !str_starts_with($href, '/')) {
+            continue;
+        }
+
+        $a->setAttribute('href', $public . $href);
+    }
+
+    // we didn't modify the document => return as is
+    // (no need to re-serialize)
+    if (!$modified) {
+        return $source;
+    }
+
+    // find the body
+    $body = $doc->getElementsByTagName('body')->item(0);
+    if ($body === NULL) {
+        return FALSE;
+    }
+
+    // re-build the source document
+    $source = "";
+
+    // get the inner html§
+    $nodes = $body->childNodes;
+    foreach($nodes as $node) {
+        $source .= $doc->saveHTML($node);
+    }
+
+    // and turn it back into html
+    return $source;
+}