feat: add PHP language

.gitmodules

@@ -62,3 +62,6 @@
 [submodule "resources/language-submodules/tree-sitter-toml"]
 	path = resources/language-submodules/tree-sitter-toml
 	url = https://github.com/ikatyang/tree-sitter-toml
+[submodule "resources/language-submodules/tree-sitter-php"]
+	path = resources/language-submodules/tree-sitter-php
+	url = https://github.com/tree-sitter/tree-sitter-php

CONTRIBUTING.md

@@ -53,7 +53,7 @@ Here are the steps for adding a new target language:
 3. Copy the grammar file into `resources/metavariable-grammars`. This alternative grammar is used for parsing `snippets` in GritQL.
 4. Patch the metavariable grammar to include  `$.grit_metavariable` anywhere we want to substitute a metavariable. This is usually at least `$identifier` and `$literal`.
     - For a snippet to match, it also needs to be a field. Often you’ll want to wrap `$thing` like: `field('thing', choice($.grit_metavariable, $thing))`
-5. Add a new language implementation in `crates/core/src/languages`. This involves implementing the `Language` trait and adding a new `Language` enum variant.
+5. Add a new language implementation in `crates/core/languages/src`. This involves implementing the `Language` trait and adding a new `Language` enum variant.
 6. Add `snippet_context_strings` [like this](https://github.com/getgrit/gritql/blob/main/crates/language/src/sql.rs#L52) to provide context for snippets to match in.
 7. Add test cases for the language in `crates/core/src/test.rs`. This is a good time to add a few dozen test cases to ensure that the language is parsed correctly, and that the metavariable grammar is working.
 

crates/core/src/snapshots/marzano_core__test__basic_php.snap

@@ -0,0 +1,57 @@
+---
+source: crates/core/src/test.rs
+expression: results
+---
+- __typename: Rewrite
+  original:
+    messages: []
+    variables:
+      - name: $new_files
+        scopedName: 0_0_$new_files
+        ranges: []
+      - name: $program
+        scopedName: 0_1_$program
+        ranges: []
+      - name: $filename
+        scopedName: 0_2_$filename
+        ranges: []
+      - name: $absolute_filename
+        scopedName: 0_3_$absolute_filename
+        ranges: []
+      - name: $str
+        scopedName: 1_0_$str
+        ranges: []
+      - name: $match
+        scopedName: 1_1_$match
+        ranges:
+          - start:
+              line: 1
+              column: 1
+            end:
+              line: 1
+              column: 19
+            startByte: 0
+            endByte: 18
+    sourceFile: foo.php
+    ranges:
+      - start:
+          line: 1
+          column: 1
+        end:
+          line: 1
+          column: 19
+        startByte: 0
+        endByte: 18
+    debug: "{\n  \"SORT\": \"program\",\n  \"RANGE\": {\n    \"start\": {\n      \"line\": 1,\n      \"column\": 1\n    },\n    \"end\": {\n      \"line\": 1,\n      \"column\": 19\n    }\n  },\n  \"TEXT\": \"echo \\\"hello world\\\"\",\n  \"CHILDREN\": [\n    {\n      \"SORT\": \"text\",\n      \"RANGE\": {\n        \"start\": {\n          \"line\": 1,\n          \"column\": 1\n        },\n        \"end\": {\n          \"line\": 1,\n          \"column\": 19\n        }\n      },\n      \"TEXT\": \"echo \\\"hello world\\\"\"\n    }\n  ]\n}"
+  rewritten:
+    messages: []
+    variables: []
+    sourceFile: foo.php
+    content: "echo \"hello world\";"
+    byteRanges:
+      - start: 0
+        end: 19
+  ansiSummary: ""
+  reason: ~
+- __typename: DoneFile
+  relativeFilePath: foo.php

crates/core/src/test.rs

@@ -8073,6 +8073,29 @@ fn test_basic_md() {
     assert_yaml_snapshot!(results);
 }
 
+#[test]
+fn test_basic_php() {
+    let pattern = r#"
+    |language php
+    |
+    |`echo $str` => `echo $str;`
+    |"#
+    .trim_margin()
+    .unwrap();
+
+    let source = r#"echo "hello world""#.trim_margin().unwrap();
+    let file = "foo.php";
+
+    let context = ExecutionContext::default();
+    let language: TargetLanguage = PatternLanguage::Php.try_into().unwrap();
+    // println!("language: {language} pattern: {pattern}");
+
+    let pattern = src_to_problem(pattern, language).unwrap();
+    let results = pattern.execute_file(&RichFile::new(file.to_owned(), source), &context);
+    // println!("results: {results}");
+    assert_yaml_snapshot!(results);
+}
+
 #[test]
 fn md_link_metavariable() {
     let pattern = r#"
@@ -12725,3 +12748,51 @@ fn go_package_type() {
     })
     .unwrap();
 }
+
+#[test]
+fn call_php() {
+    run_test_no_match({
+        TestArg {
+            pattern: r#"
+                |language php
+                |
+                |`TEST`
+                |"#
+            .trim_margin()
+            .unwrap(),
+            source: r#"
+                |echo "hello world"
+                |"#
+            .trim_margin()
+            .unwrap(),
+        }
+    })
+    .unwrap();
+}
+
+
+#[test]
+fn call_php_2() {
+    run_test_expected({
+        TestArgExpected {
+            pattern: r#"
+                |language php
+                |
+                |`echo $str` => `echo "modified"`
+                |"#
+            .trim_margin()
+            .unwrap(),
+            source: r#"
+                |echo "this is a message"
+                |"#
+            .trim_margin()
+            .unwrap(),
+            expected: r#"
+                |echo "modified"
+                |"#
+            .trim_margin()
+            .unwrap(),
+        }
+    })
+    .unwrap();
+}

crates/gritmodule/src/patterns_directory.rs

@@ -26,6 +26,7 @@ pub struct PatternsDirectory {
     sql: BTreeMap<String, String>,
     vue: BTreeMap<String, String>,
     toml: BTreeMap<String, String>,
+    php: BTreeMap<String, String>,
     universal: BTreeMap<String, String>,
 }
 
@@ -80,6 +81,7 @@ impl PatternsDirectory {
             sql: BTreeMap::new(),
             vue: BTreeMap::new(),
             toml: BTreeMap::new(),
+            php: BTreeMap::new(),
             universal: BTreeMap::new(),
         }
     }
@@ -109,6 +111,7 @@ impl PatternsDirectory {
             PatternLanguage::Sql => &mut self.sql,
             PatternLanguage::Vue => &mut self.vue,
             PatternLanguage::Toml => &mut self.toml,
+            PatternLanguage::Php => &mut self.php,
             PatternLanguage::Universal => &mut self.universal,
         }
     }
@@ -135,6 +138,7 @@ impl PatternsDirectory {
             PatternLanguage::Sql => &self.sql,
             PatternLanguage::Vue => &self.vue,
             PatternLanguage::Toml => &self.toml,
+            PatternLanguage::Php => &self.php,
             PatternLanguage::Universal => &self.universal,
         }
     }

crates/language/Cargo.toml

@@ -27,6 +27,7 @@ tree-sitter-ruby = { path = "../../resources/language-metavariables/tree-sitter-
 tree-sitter-sql = { path = "../../resources/language-metavariables/tree-sitter-sql", optional = true }
 tree-sitter-vue = { path = "../../resources/language-metavariables/tree-sitter-vue", optional = true }
 tree-sitter-toml = { path = "../../resources/language-metavariables/tree-sitter-toml", optional = true }
+tree-sitter-php = { path = "../../resources/language-metavariables/tree-sitter-php", optional = true }
 serde_json = { version = "1.0.91", features = ["preserve_order"] }
 marzano-util = { path = "../util" }
 regex = "1.7.1"
@@ -63,5 +64,6 @@ builtin-parser = [
     "tree-sitter-sql",
     "tree-sitter-vue",
     "tree-sitter-toml",
+    "tree-sitter-php",
 ]
 finder = ["ignore"]

crates/language/src/language.rs

@@ -493,7 +493,7 @@ pub fn fields_for_nodes(language: &TSLanguage, types: &str) -> Vec<Vec<Field>> {
 #[cfg(test)]
 mod tests {
     use super::nodes_from_indices;
-    use crate::{language::Language, tsx::Tsx};
+    use crate::{language::Language, php::Php, tsx::Tsx};
     use tree_sitter::Parser;
     use trim_margin::MarginTrimmable;
 
@@ -536,6 +536,14 @@ mod tests {
         assert_eq!(subbed, "µfoo$('µbar')");
     }
 
+    #[test]
+    fn test_php_substitute_variable() {
+        let snippet = "$foo$('$bar')";
+        let lang = Php::new(None);
+        let subbed = lang.substitute_metavariable_prefix(snippet);
+        assert_eq!(subbed, "µ[foo]$('µ[bar]')");
+    }
+
     #[test]
     fn test_substitute_variable_template() {
         let snippet = r#"

crates/language/src/lib.rs

@@ -12,6 +12,7 @@ pub mod language;
 pub mod markdown_block;
 pub mod markdown_inline;
 pub mod parent_traverse;
+pub mod php;
 pub mod python;
 pub mod ruby;
 pub mod rust;

crates/language/src/php.rs

@@ -0,0 +1,76 @@
+use std::sync::OnceLock;
+
+use crate::language::{fields_for_nodes, Field, Language, SortId, TSLanguage};
+
+static NODE_TYPES_STRING: &str = include_str!("../../../resources/node-types/php-node-types.json");
+
+static NODE_TYPES: OnceLock<Vec<Vec<Field>>> = OnceLock::new();
+static LANGUAGE: OnceLock<TSLanguage> = OnceLock::new();
+
+#[cfg(not(feature = "builtin-parser"))]
+fn language() -> TSLanguage {
+    unimplemented!(
+        "tree-sitter parser must be initialized before use when [builtin-parser] is off."
+    )
+}
+#[cfg(feature = "builtin-parser")]
+fn language() -> TSLanguage {
+    tree_sitter_php::language_php().into()
+}
+
+#[derive(Debug, Clone)]
+pub struct Php {
+    node_types: &'static [Vec<Field>],
+    metavariable_sort: SortId,
+    language: &'static TSLanguage,
+}
+
+impl Php {
+    pub(crate) fn new(lang: Option<TSLanguage>) -> Self {
+        let language = LANGUAGE.get_or_init(|| lang.unwrap_or_else(language));
+        let node_types = NODE_TYPES.get_or_init(|| fields_for_nodes(language, NODE_TYPES_STRING));
+        let metavariable_sort = language.id_for_node_kind("grit_metavariable", true);
+        Self {
+            node_types,
+            metavariable_sort,
+            language,
+        }
+    }
+    pub(crate) fn is_initialized() -> bool {
+        LANGUAGE.get().is_some()
+    }
+}
+
+impl Language for Php {
+    fn get_ts_language(&self) -> &TSLanguage {
+        self.language
+    }
+
+    fn comment_prefix(&self) -> &'static str {
+        "//"
+    }
+
+    fn language_name(&self) -> &'static str {
+        "Php"
+    }
+    fn snippet_context_strings(&self) -> &[(&'static str, &'static str)] {
+        &[("", "")]
+    }
+
+    fn node_types(&self) -> &[Vec<Field>] {
+        self.node_types
+    }
+
+    fn metavariable_sort(&self) -> SortId {
+        self.metavariable_sort
+    }
+
+    fn substitute_metavariable_prefix(&self, src: &str) -> String {
+        self.metavariable_regex()
+            .replace_all(
+                src,
+                format!("{}[$1]", self.metavariable_prefix_substitute()).as_str(),
+            )
+            .to_string()
+    }
+}