This project does not seem to handle request data directly as such no vulnerable execution paths were found.
include
, or for example
via PHP's auto-loading mechanism.
1 | <?php |
||||
2 | |||||
3 | declare(strict_types = 1); |
||||
4 | |||||
5 | namespace App\Model; |
||||
6 | |||||
7 | use App\Repository\BlameRepository; |
||||
8 | |||||
9 | /** |
||||
10 | * A Blame will search the given page for the given text and return the relevant revisions and authors. |
||||
11 | */ |
||||
12 | class Blame extends Authorship |
||||
13 | { |
||||
14 | /** @var string Text to search for. */ |
||||
15 | protected string $query; |
||||
16 | |||||
17 | /** @var array|null Matches, keyed by revision ID, each with keys 'edit' <Edit> and 'tokens' <string[]>. */ |
||||
18 | protected ?array $matches; |
||||
19 | |||||
20 | /** @var Edit|null Target revision that is being blamed. */ |
||||
21 | protected ?Edit $asOf; |
||||
22 | |||||
23 | /** |
||||
24 | * Blame constructor. |
||||
25 | * @param BlameRepository $repository |
||||
26 | * @param Page $page The page to process. |
||||
27 | * @param string $query Text to search for. |
||||
28 | * @param string|null $target Either a revision ID or date in YYYY-MM-DD format. Null to use latest revision. |
||||
29 | */ |
||||
30 | public function __construct( |
||||
31 | BlameRepository $repository, |
||||
32 | Page $page, |
||||
33 | string $query, |
||||
34 | ?string $target = null |
||||
35 | ) { |
||||
36 | parent::__construct($repository, $page, $target); |
||||
37 | $this->query = $query; |
||||
38 | } |
||||
39 | |||||
40 | /** |
||||
41 | * Get the search query. |
||||
42 | * @return string |
||||
43 | */ |
||||
44 | public function getQuery(): string |
||||
45 | { |
||||
46 | return $this->query; |
||||
47 | } |
||||
48 | |||||
49 | /** |
||||
50 | * Matches, keyed by revision ID, each with keys 'edit' <Edit> and 'tokens' <string[]>. |
||||
51 | * @return array|null |
||||
52 | */ |
||||
53 | public function getMatches(): ?array |
||||
54 | { |
||||
55 | return $this->matches; |
||||
56 | } |
||||
57 | |||||
58 | /** |
||||
59 | * Get all the matches as Edits. |
||||
60 | * @return Edit[]|null |
||||
61 | */ |
||||
62 | public function getEdits(): ?array |
||||
63 | { |
||||
64 | return array_column($this->matches, 'edit'); |
||||
0 ignored issues
–
show
Bug
introduced
by
![]() |
|||||
65 | } |
||||
66 | |||||
67 | /** |
||||
68 | * Strip out spaces, since they are not accounted for in the WikiWho API. |
||||
69 | * @return string |
||||
70 | */ |
||||
71 | public function getTokenizedQuery(): string |
||||
72 | { |
||||
73 | return strtolower(preg_replace('/\s*/m', '', $this->query)); |
||||
74 | } |
||||
75 | |||||
76 | /** |
||||
77 | * Get the first "token" of the search query. A "token" in this case is a word or group of syntax, |
||||
78 | * roughly correlating to the token structure returned by the WikiWho API. |
||||
79 | * @return string |
||||
80 | */ |
||||
81 | public function getFirstQueryToken(): string |
||||
82 | { |
||||
83 | return strtolower(preg_split('/[\n\s]/', $this->query)[0]); |
||||
84 | } |
||||
85 | |||||
86 | /** |
||||
87 | * Get the target revision that is being blamed. |
||||
88 | * @return Edit|null |
||||
89 | */ |
||||
90 | public function getAsOf(): ?Edit |
||||
91 | { |
||||
92 | if (isset($this->asOf)) { |
||||
93 | return $this->asOf; |
||||
94 | } |
||||
95 | |||||
96 | $this->asOf = $this->target |
||||
97 | ? $this->repository->getEditFromRevId($this->page, $this->target) |
||||
0 ignored issues
–
show
The method
getEditFromRevId() does not exist on App\Repository\Repository . It seems like you code against a sub-type of App\Repository\Repository such as App\Repository\BlameRepository .
(
Ignorable by Annotation
)
If this is a false-positive, you can also ignore this issue in your code via the
![]() |
|||||
98 | : null; |
||||
99 | |||||
100 | return $this->asOf; |
||||
101 | } |
||||
102 | |||||
103 | /** |
||||
104 | * Get authorship attribution from the WikiWho API. |
||||
105 | * @see https://www.f-squared.org/wikiwho/ |
||||
106 | */ |
||||
107 | public function prepareData(): void |
||||
108 | { |
||||
109 | if (isset($this->matches)) { |
||||
110 | return; |
||||
111 | } |
||||
112 | |||||
113 | // Set revision data. self::setRevisionData() returns null if there are errors. |
||||
114 | $revisionData = $this->getRevisionData(true); |
||||
115 | if (null === $revisionData) { |
||||
116 | return; |
||||
117 | } |
||||
118 | |||||
119 | $matches = $this->searchTokens($revisionData['tokens']); |
||||
120 | |||||
121 | // We want the results grouped by editor and revision ID. |
||||
122 | $this->matches = []; |
||||
123 | foreach ($matches as $match) { |
||||
124 | if (isset($this->matches[$match['id']])) { |
||||
125 | $this->matches[$match['id']]['tokens'][] = $match['token']; |
||||
126 | continue; |
||||
127 | } |
||||
128 | |||||
129 | $edit = $this->repository->getEditFromRevId($this->page, $match['id']); |
||||
130 | if ($edit) { |
||||
131 | $this->matches[$match['id']] = [ |
||||
132 | 'edit' => $edit, |
||||
133 | 'tokens' => [$match['token']], |
||||
134 | ]; |
||||
135 | } |
||||
136 | } |
||||
137 | } |
||||
138 | |||||
139 | /** |
||||
140 | * Find matches of search query in the given list of tokens. |
||||
141 | * @param array $tokens |
||||
142 | * @return array |
||||
143 | */ |
||||
144 | private function searchTokens(array $tokens): array |
||||
145 | { |
||||
146 | $matchData = []; |
||||
147 | $matchDataSoFar = []; |
||||
148 | $matchSoFar = ''; |
||||
149 | $firstQueryToken = $this->getFirstQueryToken(); |
||||
150 | $tokenizedQuery = $this->getTokenizedQuery(); |
||||
151 | |||||
152 | foreach ($tokens as $token) { |
||||
153 | // The previous matches plus the new token. This is basically a candidate for what may become $matchSoFar. |
||||
154 | $newMatchSoFar = $matchSoFar.$token['str']; |
||||
155 | |||||
156 | // We first check if the first token of the query matches, because we want to allow for partial matches |
||||
157 | // (e.g. for query "barbaz", the tokens ["foobar","baz"] should match). |
||||
158 | if (false !== strpos($newMatchSoFar, $firstQueryToken)) { |
||||
159 | // If the full query is in the new match, use it, otherwise use just the first token. This is because |
||||
160 | // the full match may exist across multiple tokens, but the first match is only a partial match. |
||||
161 | $newMatchSoFar = false !== strpos($newMatchSoFar, $tokenizedQuery) |
||||
162 | ? $newMatchSoFar |
||||
163 | : $firstQueryToken; |
||||
164 | } |
||||
165 | |||||
166 | // Keep track of tokens that match. To allow partial matches, |
||||
167 | // we check the query against $newMatchSoFar and vice versa. |
||||
168 | if (false !== strpos($tokenizedQuery, $newMatchSoFar) || |
||||
169 | false !== strpos($newMatchSoFar, $tokenizedQuery) |
||||
170 | ) { |
||||
171 | $matchSoFar = $newMatchSoFar; |
||||
172 | $matchDataSoFar[] = [ |
||||
173 | 'id' => $token['o_rev_id'], |
||||
174 | 'editor' => $token['editor'], |
||||
175 | 'token' => $token['str'], |
||||
176 | ]; |
||||
177 | } elseif (!empty($matchSoFar)) { |
||||
178 | // We hit a token that isn't in the query string, so start over. |
||||
179 | $matchDataSoFar = []; |
||||
180 | $matchSoFar = ''; |
||||
181 | } |
||||
182 | |||||
183 | // A full match was found, so merge $matchDataSoFar into $matchData, |
||||
184 | // and start over to see if there are more matches in the article. |
||||
185 | if (false !== strpos($matchSoFar, $tokenizedQuery)) { |
||||
186 | $matchData = array_merge($matchData, $matchDataSoFar); |
||||
187 | $matchDataSoFar = []; |
||||
188 | $matchSoFar = ''; |
||||
189 | } |
||||
190 | } |
||||
191 | |||||
192 | // Full matches usually come last, but are the most relevant. |
||||
193 | return array_reverse($matchData); |
||||
194 | } |
||||
195 | } |
||||
196 |