fix note 7

comunica · Dec 11, 2024 · 76b5313 · 76b5313
1 parent 47411d9
commit 76b5313
Show file tree

Hide file tree

Showing 5 changed files with 143 additions and 50 deletions.
diff --git a/src/grammar/builder/parserBuilder.ts b/src/grammar/builder/parserBuilder.ts
@@ -135,7 +135,7 @@ export class Builder<Names extends string, RuleDefs extends RuleDefMap<Names>> {
     const lexer: Lexer = new Lexer(tokenVocabulary, {
       positionTracking: 'onlyStart',
       recoveryEnabled: false,
-      skipValidations: true,
+      // SkipValidations: true,
       ensureOptimizations: true,
       ...lexerConfig,
     });
@@ -150,7 +150,8 @@ export class Builder<Names extends string, RuleDefs extends RuleDefMap<Names>> {
         parser.input = lexResult.tokens;
         const result = parser[rule.name](...args);
         if (parser.errors.length > 0) {
-          throw new Error(`Parse error on line ${parser.errors[0].token.startLine}`);
+          throw new Error(`Parse error on line ${parser.errors.map(x => x.token.startLine).join(', ')}
+${parser.errors.map(x => `${x.token.startLine}: ${x.message}`).join('\n')}`);
         }
         return result;
       };

diff --git a/src/lexer/sparql11/lexer.ts b/src/lexer/sparql11/lexer.ts
@@ -1,11 +1,10 @@
 /* eslint-disable require-unicode-regexp */
 import type { TokenType } from 'chevrotain';
-import { Lexer } from 'chevrotain';
 import { allBuiltInCalls } from './BuildinCalls.js';
 import { allGraphTokens } from './graph.js';
 import { createToken } from './helpers.js';
 import { allSymbols } from './symbols.js';
-import { allTerminals } from './terminals.js';
+import { allTerminals, wsPattern } from './terminals.js';
 
 export const baseDecl = createToken({ name: 'BaseDecl', pattern: /base/i, label: 'BASE' });
 export const prefixDecl = createToken({ name: 'PrefixDecl', pattern: /prefix/i, label: 'PREFIX' });
@@ -36,9 +35,12 @@ export const add = createToken({ name: 'Add', pattern: /add/i, label: 'ADD' });
 export const to = createToken({ name: 'To', pattern: /to/i, label: 'TO' });
 export const move = createToken({ name: 'Move', pattern: /move/i, label: 'MOVE' });
 export const copy = createToken({ name: 'Copy', pattern: /copy/i, label: 'COPY' });
-export const insertData = createToken({ name: 'InsertData', pattern: /insert data/i, label: 'INSERT DATA' });
-export const deleteData = createToken({ name: 'DeleteData', pattern: /delete data/i, label: 'DELETE DATA' });
-export const deleteWhere = createToken({ name: 'DeleteWhere', pattern: /delete where/i, label: 'DELETE WHERE' });
+const insertDataPattern = new RegExp(`insert(?:${wsPattern.source})*data`, 'i');
+const deleteDataPattern = new RegExp(`delete(?:${wsPattern.source})*data`, 'i');
+const deleteWherePattern = new RegExp(`delete(?:${wsPattern.source})*where`, 'i');
+export const insertData = createToken({ name: 'InsertData', pattern: insertDataPattern, label: 'INSERT DATA' });
+export const deleteData = createToken({ name: 'DeleteData', pattern: deleteDataPattern, label: 'DELETE DATA' });
+export const deleteWhere = createToken({ name: 'DeleteWhere', pattern: deleteWherePattern, label: 'DELETE WHERE' });
 export const modifyWith = createToken({ name: 'ModifyWith', pattern: /with/i, label: 'WITH' });
 export const deleteClause = createToken({ name: 'DeleteClause', pattern: /delete/i, label: 'DELETE' });
 export const insertClause = createToken({ name: 'InsertClause', pattern: /insert/i, label: 'INSERT' });
@@ -119,12 +121,3 @@ export const allTokens: TokenType[] = [
   ...allGraphTokens,
   ...allSymbols,
 ];
-
-export const ChevSparqlLexer = new Lexer(allTokens, {
-  // PositionTracking: 'onlyOffset',
-  recoveryEnabled: false,
-  skipValidations: false,
-  safeMode: true,
-  positionTracking: 'full',
-  // EnsureOptimizations: true,
-});
diff --git a/src/lexer/sparql11/terminals.ts b/src/lexer/sparql11/terminals.ts
@@ -3,51 +3,51 @@ import { Lexer } from 'chevrotain';
 import { createToken } from './helpers.js';
 
 // eslint-disable-next-line max-len
-const pnCharsBasePattern = /[A-Za-z\u00C0-\u00D6\u00D8-\u00F6\u00F8-\u02FF\u0370-\u037D\u037F-\u1FFF\u200C-\u200D\u2070-\u218F\u2C00-\u2FEF\u3001-\uD7FF\uF900-\uFDCF\uFDF0-\uFFFD]|[\uD800-\uDB7F][\uDC00-\uDFFF]/;
-const pnCharsUPattern = new RegExp(`${pnCharsBasePattern.source}|_`);
+export const pnCharsBasePattern = /[A-Za-z\u00C0-\u00D6\u00D8-\u00F6\u00F8-\u02FF\u0370-\u037D\u037F-\u1FFF\u200C-\u200D\u2070-\u218F\u2C00-\u2FEF\u3001-\uD7FF\uF900-\uFDCF\uFDF0-\uFFFD]|[\uD800-\uDB7F][\uDC00-\uDFFF]/;
+export const pnCharsUPattern = new RegExp(`${pnCharsBasePattern.source}|_`);
 // eslint-disable-next-line no-misleading-character-class
-const varNamePattern = new RegExp(`((${pnCharsUPattern.source})|[0-9])((${pnCharsUPattern.source})|[0-9]|[\u00B7\u0300-\u036F\u203F-\u2040])*`);
+export const varNamePattern = new RegExp(`((${pnCharsUPattern.source})|[0-9])((${pnCharsUPattern.source})|[0-9]|[\u00B7\u0300-\u036F\u203F-\u2040])*`);
 // eslint-disable-next-line no-control-regex
-const iriRefPattern = /<([^\\<>"{}|^`\u0000-\u0020])*>/;
+export const iriRefPattern = /<([^\\<>"{}|^`\u0000-\u0020])*>/;
 // eslint-disable-next-line no-misleading-character-class
-const pnCharsPattern = new RegExp(`(${pnCharsUPattern.source})|[\\-0-9\u00B7\u0300-\u036F\u203F-\u2040]`);
+export const pnCharsPattern = new RegExp(`(${pnCharsUPattern.source})|[\\-0-9\u00B7\u0300-\u036F\u203F-\u2040]`);
 // eslint-disable-next-line no-misleading-character-class
-const pnPrefixPattern = new RegExp(`(${pnCharsBasePattern.source})(((${pnCharsPattern.source})|\\.)*(${pnCharsPattern.source}))?`);
+export const pnPrefixPattern = new RegExp(`(${pnCharsBasePattern.source})(((${pnCharsPattern.source})|\\.)*(${pnCharsPattern.source}))?`);
 // eslint-disable-next-line no-misleading-character-class
-const pNameNsPattern = new RegExp(`(${pnPrefixPattern.source})?:`);
-const percentPattern = /%[0-9A-Fa-f][0-9A-Fa-f]/;
-const pnLocalEscPattern = /\\[_~.\-!$&'()*+,;=\\/?#@%]/;
-const plxPattern = new RegExp(`(${percentPattern.source})|(${pnLocalEscPattern.source})`);
+export const pNameNsPattern = new RegExp(`(${pnPrefixPattern.source})?:`);
+export const percentPattern = /%[0-9A-Fa-f][0-9A-Fa-f]/;
+export const pnLocalEscPattern = /\\[_~.\-!$&'()*+,;=\\/?#@%]/;
+export const plxPattern = new RegExp(`(${percentPattern.source})|(${pnLocalEscPattern.source})`);
 // eslint-disable-next-line no-misleading-character-class
-const pnLocalPattern = new RegExp(`((${pnCharsUPattern.source})|:|[0-9]|(${plxPattern.source}))(((${pnCharsPattern.source})|\\.|:|(${plxPattern.source}))*((${pnCharsPattern.source})|:|(${plxPattern.source})))?`);
+export const pnLocalPattern = new RegExp(`((${pnCharsUPattern.source})|:|[0-9]|(${plxPattern.source}))(((${pnCharsPattern.source})|\\.|:|(${plxPattern.source}))*((${pnCharsPattern.source})|:|(${plxPattern.source})))?`);
 // eslint-disable-next-line no-misleading-character-class
-const pNameLnPattern = new RegExp(`(${pNameNsPattern.source})(${pnLocalPattern.source})`);
+export const pNameLnPattern = new RegExp(`(${pNameNsPattern.source})(${pnLocalPattern.source})`);
 // eslint-disable-next-line no-misleading-character-class
-const blankNodeLabelPattern = new RegExp(`_:((${pnCharsUPattern.source})|[0-9])(((${pnCharsPattern.source})\\.)*(${pnCharsPattern.source}))?`);
+export const blankNodeLabelPattern = new RegExp(`_:((${pnCharsUPattern.source})|[0-9])(((${pnCharsPattern.source})\\.)*(${pnCharsPattern.source}))?`);
 // eslint-disable-next-line no-misleading-character-class
-const var1Pattern = new RegExp(`\\?(${varNamePattern.source})`);
+export const var1Pattern = new RegExp(`\\?(${varNamePattern.source})`);
 // eslint-disable-next-line no-misleading-character-class
-const var2Pattern = new RegExp(`\\$(${varNamePattern.source})`);
-const langTagPattern = /@[a-zA-Z]+(-[a-zA-Z0-9]+)*/;
-const integerPattern = /[0-9]+/;
-const decimalPattern = /[0-9]+\.[0-9]+/;
-const exponentPattern = /[eE][+-]?[0-9]+/;
-const doublePattern = new RegExp(`([0-9]+\\.[0-9]*(${exponentPattern.source}))|(\\.[0-9]+(${exponentPattern.source}))|([0-9]+(${exponentPattern.source}))`);
-const interferePositivePattern = new RegExp(`\\+${integerPattern.source}`);
-const decimalPositivePattern = new RegExp(`\\+${decimalPattern.source}`);
-const doublePositivePattern = new RegExp(`\\+${doublePattern.source}`);
-const integerNegativePattern = new RegExp(`-${integerPattern.source}`);
-const decimalNegativePattern = new RegExp(`-${decimalPattern.source}`);
-const doubleNegativePattern = new RegExp(`-${doublePattern.source}`);
-const echarPattern = /\\[\\"'bfnrt]/u;
-const stringLiteral1Pattern = new RegExp(`'(?:([^\\u0027\\u005C\\u000A\u000D])|(?:${echarPattern.source}))*'`);
-const stringLiteral2Pattern = new RegExp(`"(?:([^\\u0022\\u005C\\u000A\\u000D])|(?:${echarPattern.source}))*"`);
-const stringLiteralLong1Pattern = new RegExp(`'''(('|(''))?([^'\\\\]|(${echarPattern.source})))*'''`);
-const stringLiteralLong2Pattern = new RegExp(`"""(("|(""))?([^"\\\\]|(${echarPattern.source})))*"""`);
+export const var2Pattern = new RegExp(`\\$(${varNamePattern.source})`);
+export const langTagPattern = /@[a-zA-Z]+(-[a-zA-Z0-9]+)*/;
+export const integerPattern = /[0-9]+/;
+export const decimalPattern = /[0-9]+\.[0-9]+/;
+export const exponentPattern = /[eE][+-]?[0-9]+/;
+export const doublePattern = new RegExp(`([0-9]+\\.[0-9]*(${exponentPattern.source}))|(\\.[0-9]+(${exponentPattern.source}))|([0-9]+(${exponentPattern.source}))`);
+export const interferePositivePattern = new RegExp(`\\+${integerPattern.source}`);
+export const decimalPositivePattern = new RegExp(`\\+${decimalPattern.source}`);
+export const doublePositivePattern = new RegExp(`\\+${doublePattern.source}`);
+export const integerNegativePattern = new RegExp(`-${integerPattern.source}`);
+export const decimalNegativePattern = new RegExp(`-${decimalPattern.source}`);
+export const doubleNegativePattern = new RegExp(`-${doublePattern.source}`);
+export const echarPattern = /\\[\\"'bfnrt]/u;
+export const stringLiteral1Pattern = new RegExp(`'(?:([^\\u0027\\u005C\\u000A\u000D])|(?:${echarPattern.source}))*'`);
+export const stringLiteral2Pattern = new RegExp(`"(?:([^\\u0022\\u005C\\u000A\\u000D])|(?:${echarPattern.source}))*"`);
+export const stringLiteralLong1Pattern = new RegExp(`'''(('|(''))?([^'\\\\]|(${echarPattern.source})))*'''`);
+export const stringLiteralLong2Pattern = new RegExp(`"""(("|(""))?([^"\\\\]|(${echarPattern.source})))*"""`);
 // eslint-disable-next-line no-control-regex
-const wsPattern = /[\u0020\u0009\u000D\u000A]/;
-const nilPattern = new RegExp(`\\((${wsPattern.source})*\\)`);
-const anonPattern = new RegExp(`\\[(${wsPattern.source})*\\]`);
+export const wsPattern = /[\u0020\u0009\u000D\u000A]/;
+export const nilPattern = new RegExp(`\\((${wsPattern.source})*\\)`);
+export const anonPattern = new RegExp(`\\[(${wsPattern.source})*\\]`);
 
 export const iriRef = createToken({ name: 'IriRef', pattern: iriRefPattern });
 export const pNameLn = createToken({ name: 'PNameLn', pattern: pNameLnPattern });

diff --git a/test/statics/sparql/spaced-data-update.json b/test/statics/sparql/spaced-data-update.json
@@ -0,0 +1,88 @@
+{
+  "prefixes": {},
+  "type": "update",
+  "updates": [
+    {
+      "insert": [
+        {
+          "triples": [
+            {
+              "object": {
+                "datatype": {
+                  "termType": "NamedNode",
+                  "value": "http://www.w3.org/2001/XMLSchema#string"
+                },
+                "language": "",
+                "termType": "Literal",
+                "value": "object"
+              },
+              "predicate": {
+                "termType": "NamedNode",
+                "value": "http://example.org/predicate"
+              },
+              "subject": {
+                "termType": "NamedNode",
+                "value": "http://example.org/subject"
+              }
+            }
+          ],
+          "type": "bgp"
+        }
+      ],
+      "updateType": "insert"
+    },
+    {
+      "delete": [
+        {
+          "triples": [
+            {
+              "object": {
+                "datatype": {
+                  "termType": "NamedNode",
+                  "value": "http://www.w3.org/2001/XMLSchema#string"
+                },
+                "language": "",
+                "termType": "Literal",
+                "value": "object"
+              },
+              "predicate": {
+                "termType": "NamedNode",
+                "value": "http://example.org/predicate"
+              },
+              "subject": {
+                "termType": "NamedNode",
+                "value": "http://example.org/subject"
+              }
+            }
+          ],
+          "type": "bgp"
+        }
+      ],
+      "updateType": "delete"
+    },
+    {
+      "delete": [
+        {
+          "triples": [
+            {
+              "object": {
+                "termType": "Variable",
+                "value": "o"
+              },
+              "predicate": {
+                "termType": "NamedNode",
+                "value": "http://example.org/predicate"
+              },
+              "subject": {
+                "termType": "NamedNode",
+                "value": "http://example.org/subject"
+              }
+            }
+          ],
+          "type": "bgp"
+        }
+      ],
+      "updateType": "deletewhere"
+    }
+  ]
+}
diff --git a/test/statics/sparql/spaced-data-update.sparql b/test/statics/sparql/spaced-data-update.sparql
@@ -0,0 +1,11 @@
+# Basically tests Note 7 of the SPARQL 1.1 spec
+INsert     Data {
+  <http://example.org/subject> <http://example.org/predicate> "object" .
+};
+Delete
+Data {
+  <http://example.org/subject> <http://example.org/predicate> "object" .
+};
+DeLete WHERE {
+  <http://example.org/subject> <http://example.org/predicate> ?o .
+};