[FOU-470] ANTLR Code Generation

python_modules/dagster/.gitattributes

@@ -1 +1,2 @@
 dagster_tests/**/snapshots/* linguist-generated=true
+dagster/_core/definitions/antlr_asset_selection/generated/* linguist-generated=true

python_modules/dagster/dagster/_core/definitions/antlr_asset_selection/AssetSelection.g4

@@ -0,0 +1,85 @@
+grammar AssetSelection;
+
+start: expr EOF;
+
+// Root rule for parsing expressions
+expr
+    : assetExpr                                    # AssetExpression
+    | attributeExpr                                # AttributeExpression
+    | traversal expr                               # UpTraversalExpression
+    | traversal expr traversal                     # UpAndDownTraversalExpression
+    | expr traversal                               # DownTraversalExpression
+    | NOT expr                                     # NotExpression
+    | expr AND expr                                # AndExpression
+    | expr OR expr                                 # OrExpression
+    | functionName LPAREN expr RPAREN              # FunctionCallExpression
+    | LPAREN expr RPAREN                           # ParenthesizedExpression
+    ;
+
+// Traversal operators
+traversal
+    : STAR
+    | PLUS+
+    ;
+
+// Function names as tokens
+functionName
+    : SINKS
+    | ROOTS
+    ;
+
+// Attribute expressions for specific attributes
+attributeExpr
+    : TAG COLON value (EQUAL value)?               # TagAttributeExpr
+    | OWNER COLON value                            # OwnerAttributeExpr
+    | GROUP COLON value                            # GroupAttributeExpr
+    | KIND COLON value                             # KindAttributeExpr
+    | CODE_LOCATION COLON value                    # CodeLocationAttributeExpr
+    ;
+
+// Define the EQUAL token for tag:value=value syntax
+EQUAL : '=';
+
+// Value can be a quoted or unquoted string
+value
+    : QUOTED_STRING
+    | UNQUOTED_STRING
+    ;
+
+// Asset expressions
+assetExpr
+    : QUOTED_STRING                                # ExactMatchAsset
+    | UNQUOTED_STRING                              # PrefixMatchAsset
+    ;
+
+// Tokens for operators and keywords
+AND : 'and';
+OR : 'or';
+NOT : 'not';
+
+STAR : '*';
+PLUS : '+';
+
+COLON : ':';
+
+LPAREN : '(';
+RPAREN : ')';
+COMMA : ',';
+
+// Tokens for keys
+OWNER : 'owner';
+GROUP : 'group';
+TAG : 'tag';
+KIND : 'kind';
+CODE_LOCATION : 'code_location';
+
+// Tokens for function names
+SINKS : 'sinks';
+ROOTS : 'roots';
+
+// Tokens for strings
+QUOTED_STRING : '"' (~["\\\r\n])* '"' ;
+UNQUOTED_STRING : [a-zA-Z_][a-zA-Z0-9_]*;
+
+// Whitespace
+WS : [ \t\r\n]+ -> skip ;

python_modules/dagster/dagster/_core/definitions/antlr_asset_selection/README.md

@@ -0,0 +1,80 @@
+# How to set up ANTLR
+
+## Install Java
+
+Using M1 MacBook and Homebrew.
+
+Install java:
+
+```bash
+$ brew install openjdk
+```
+
+Verify it's installed:
+
+```bash
+$ $(brew --prefix openjdk)/bin/java --version
+```
+
+Verify it's for the arm64 hardware:
+
+```bash
+$ file $(brew --prefix openjdk)/bin/java
+```
+
+Create symlink:
+
+```bash
+$ sudo ln -sfn /opt/homebrew/opt/openjdk/libexec/openjdk.jdk /Library/Java/JavaVirtualMachines/openjdk.jdk
+$ java -version
+```
+
+## Install ANTLR
+
+https://github.com/antlr/antlr4/blob/master/doc/getting-started.md
+
+1. Download:
+
+```bash
+$ cd /usr/local/lib
+$ curl -O https://www.antlr.org/download/antlr-4.13.2-complete.jar
+```
+
+2. Add antlr-4.13.2-complete.jar to your `CLASSPATH`:
+
+```bash
+$ export CLASSPATH=".:/usr/local/lib/antlr-4.13.2-complete.jar:$CLASSPATH"
+```
+
+Also add to `.bash_profile` or whatever your startup script is.
+
+3. Create alias for ANTLR:
+
+```bash
+$ alias antlr4='java -Xmx500M -cp "/usr/local/lib/antlr-4.13.2-complete.jar:$CLASSPATH" org.antlr.v4.Tool'
+```
+
+4. Install python runtime:
+
+```bash
+pip install antlr4-python3-runtime
+```
+
+## Generate Python Parser
+
+Under the `python_modules/dagster/dagster/_core/definitions/asset_selection` directory, run
+
+```bash
+$ antlr4 -Dlanguage=Python3 -visitor AssetSelection.g4 -o generated
+```
+
+This will generate the following files from the grammar file `AssetSelection.g4` in a `generated` folder:
+
+- `AssetSelection.interp`
+- `AssetSelection.tokens`
+- `AssetSelectionLexer.interp`
+- `AssetSelectionLexer.py`
+- `AssetSelectionLexer.tokens`
+- `AssetSelectionListener.py`
+- `AssetSelectionParser.py`
+- `AssetSelectionVisitor.py`