@borgar/fx - npm Package Compare versions

Comparing version 5.0.1 to 5.0.2

lib/constants.ts

@@ -17,2 +17,5 @@ export const OPERATOR = 'operator';
 export const REF_STRUCT = 'structured';
+// TODO: in future, we should type the difference between A1:B1 (REF_RANGE) and
+//       A1 (REF_CELL) but this will require a major version bump.
+export const REF_CELL = 'cell'; // internal only
 export const FX_PREFIX = 'fx_prefix';
@@ -19,0 +22,0 @@ export const UNKNOWN = 'unknown';

lib/fixRanges.spec.ts

@@ -69,2 +69,16 @@ import { describe, test, expect } from 'vitest';
 
+describe('fixRanges prefixes', () => {
+  test('Quotes prefixes as needed', () => {
+    isFixed('=Sch1!B2', "='Sch1'!B2");
+    isFixed('=[Foo]Ab12x!B2', '=[Foo]Ab12x!B2');
+    isFixed('=[Foo]Ab12!B2', "='[Foo]Ab12'!B2");
+    isFixed('=ABC123!B2', "='ABC123'!B2");
+    isFixed('=abc123!B2', "='abc123'!B2");
+    isFixed('=C!B2', "='C'!B2");
+    isFixed('=R!B2', "='R'!B2");
+    isFixed('=RC!B2', "='RC'!B2");
+    isFixed('=CR!B2', '=CR!B2');
+  });
+});
+
 describe('fixRanges A1', () => {
@@ -71,0 +85,0 @@ const opt = { allowTernary: true };

lib/lexers/lexRangeA1.ts

@@ -102,3 +102,3 @@ import { REF_RANGE, REF_BEAM, REF_TERNARY, MAX_COLS, MAX_ROWS } from '../constants.ts';
     // LT  : this is A1
-    if (top && canEndRange(str, preOp)) {
+    if (top && canEndRange(str, preOp) && str.charCodeAt(preOp) !== 33) { // 33 = "!"
       return { type: REF_RANGE, value: str.slice(pos, preOp) };
@@ -105,0 +105,0 @@ }

lib/lexers/lexRangeR1C1.ts

@@ -14,2 +14,3 @@ import { REF_RANGE, REF_BEAM, REF_TERNARY, MAX_COLS, MAX_ROWS } from '../constants.ts';
 const MINUS = 45;
+const EXCL = 33;
 
@@ -101,3 +102,3 @@ // C
   p += c1;
-  if (c1 || r1) {
+  if ((c1 || r1) && str.charCodeAt(p) !== EXCL) {
     const op = advRangeOp(str, p);
@@ -104,0 +105,0 @@ const preOp = p;

lib/mergeRefTokens.ts

@@ -1,2 +0,2 @@
-import { CONTEXT, CONTEXT_QUOTE, REF_RANGE, REF_NAMED, REF_BEAM, REF_TERNARY, OPERATOR, REF_STRUCT } from './constants.ts';
+import { CONTEXT, CONTEXT_QUOTE, REF_RANGE, REF_NAMED, REF_BEAM, REF_TERNARY, OPERATOR, REF_STRUCT, REF_CELL } from './constants.ts';
 import type { Token } from './types.ts';
@@ -7,20 +7,22 @@
 const validRunsMerge = [
-  [ REF_RANGE, ':', REF_RANGE ],
-  [ REF_RANGE, '.:', REF_RANGE ],
-  [ REF_RANGE, ':.', REF_RANGE ],
-  [ REF_RANGE, '.:.', REF_RANGE ],
+  [ REF_CELL, ':', REF_CELL ],
+  [ REF_CELL, '.:', REF_CELL ],
+  [ REF_CELL, ':.', REF_CELL ],
+  [ REF_CELL, '.:.', REF_CELL ],
   [ REF_RANGE ],
   [ REF_BEAM ],
   [ REF_TERNARY ],
-  [ CONTEXT, '!', REF_RANGE, ':', REF_RANGE ],
-  [ CONTEXT, '!', REF_RANGE, '.:', REF_RANGE ],
-  [ CONTEXT, '!', REF_RANGE, ':.', REF_RANGE ],
-  [ CONTEXT, '!', REF_RANGE, '.:.', REF_RANGE ],
+  [ CONTEXT, '!', REF_CELL, ':', REF_CELL ],
+  [ CONTEXT, '!', REF_CELL, '.:', REF_CELL ],
+  [ CONTEXT, '!', REF_CELL, ':.', REF_CELL ],
+  [ CONTEXT, '!', REF_CELL, '.:.', REF_CELL ],
+  [ CONTEXT, '!', REF_CELL ],
   [ CONTEXT, '!', REF_RANGE ],
   [ CONTEXT, '!', REF_BEAM ],
   [ CONTEXT, '!', REF_TERNARY ],
-  [ CONTEXT_QUOTE, '!', REF_RANGE, ':', REF_RANGE ],
-  [ CONTEXT_QUOTE, '!', REF_RANGE, '.:', REF_RANGE ],
-  [ CONTEXT_QUOTE, '!', REF_RANGE, ':.', REF_RANGE ],
-  [ CONTEXT_QUOTE, '!', REF_RANGE, '.:.', REF_RANGE ],
+  [ CONTEXT_QUOTE, '!', REF_CELL, ':', REF_CELL ],
+  [ CONTEXT_QUOTE, '!', REF_CELL, '.:', REF_CELL ],
+  [ CONTEXT_QUOTE, '!', REF_CELL, ':.', REF_CELL ],
+  [ CONTEXT_QUOTE, '!', REF_CELL, '.:.', REF_CELL ],
+  [ CONTEXT_QUOTE, '!', REF_CELL ],
   [ CONTEXT_QUOTE, '!', REF_RANGE ],
@@ -66,3 +68,10 @@ [ CONTEXT_QUOTE, '!', REF_BEAM ],
     if (token) {
-      const key = (token.type === OPERATOR) ? token.value : token.type;
+      const value = token.value;
+      let key = (token.type === OPERATOR) ? value : token.type;
+      // we need to prevent merging ["A1:B2" ":" "C3"] as a range is only
+      // allowed to contain a single ":" operator even if "A1:B2:C3" is
+      // valid Excel syntax
+      if (key === REF_RANGE && !value.includes(':')) {
+        key = REF_CELL;
+      }
       if (key in node) {
@@ -69,0 +78,0 @@ node = node[key];

lib/parseA1Ref.spec.ts

@@ -338,1 +338,8 @@ /* eslint-disable @stylistic/object-property-newline */
 });
+
+describe('Sheet name that looks like an A1 ref', () => {
+  test('parse correctly', () => {
+    isA1Equal("'Sch1'!B2", { context: [ 'Sch1' ], range: { top: 1, left: 1, bottom: 1, right: 1 } });
+    isA1Equal('Sch1!B2', { context: [ 'Sch1' ], range: { top: 1, left: 1, bottom: 1, right: 1 } });
+  });
+});

lib/stringifyA1Ref.spec.ts

@@ -33,2 +33,8 @@ import { describe, test, expect } from 'vitest';
   });
+
+  test('should quote prefixes that look like A1 ranges', () => {
+    expect(stringifyA1Ref({ context: [ 'Ab12' ], range: rangeA1 })).toBe("'Ab12'!A1");
+    expect(stringifyA1Ref({ context: [ 'Sch1' ], range: rangeA1 })).toBe("'Sch1'!A1");
+    expect(stringifyA1Ref({ context: [ 'Foo12345' ], range: rangeA1 })).toBe("'Foo12345'!A1");
+  });
 });
@@ -65,2 +71,14 @@
   });
+
+  test('should quote prefixes that look like ranges', () => {
+    expect(stringifyA1RefXlsx({ sheetName: 'C', range: rangeA1 })).toBe("'C'!A1");
+    expect(stringifyA1RefXlsx({ sheetName: 'R', range: rangeA1 })).toBe("'R'!A1");
+    expect(stringifyA1RefXlsx({ sheetName: 'RC', range: rangeA1 })).toBe("'RC'!A1");
+    expect(stringifyA1RefXlsx({ sheetName: 'Ab12', range: rangeA1 })).toBe("'Ab12'!A1");
+    expect(stringifyA1RefXlsx({ sheetName: 'Sch1', range: rangeA1 })).toBe("'Sch1'!A1");
+    expect(stringifyA1RefXlsx({ sheetName: 'Foo12345', range: rangeA1 })).toBe("'Foo12345'!A1");
+    expect(stringifyA1RefXlsx({ workbookName: 'Ab12', range: rangeA1 })).toBe("'[Ab12]'!A1");
+    expect(stringifyA1RefXlsx({ workbookName: 'Sch1', range: rangeA1 })).toBe("'[Sch1]'!A1");
+    expect(stringifyA1RefXlsx({ workbookName: 'Foo12345', range: rangeA1 })).toBe("'[Foo12345]'!A1");
+  });
 });

lib/stringifyPrefix.ts

@@ -13,3 +13,24 @@ import type {
 const reBannedChars = /[^0-9A-Za-z._¡¤§¨ª\u00ad¯-\uffff]/;
+// A1-XFD1048575 | R | C | RC
+const reIsRangelike = /^(R|C|RC|[A-Z]{1,3}\d{1,7})$/i;
 
+export function needQuotes (scope: string, yesItDoes = 0): number {
+  if (yesItDoes) {
+    return 1;
+  }
+  if (scope) {
+    if (reBannedChars.test(scope)) {
+      return 1;
+    }
+    if (reIsRangelike.test(scope)) {
+      return 1;
+    }
+  }
+  return 0;
+}
+
+export function quotePrefix (prefix) {
+  return "'" + prefix.replace(/'/g, "''") + "'";
+}
+
 export function stringifyPrefix (
@@ -27,3 +48,3 @@ ref: ReferenceA1 | ReferenceName | ReferenceStruct | ReferenceR1C1
       pre = part + pre;
-      quote += +reBannedChars.test(scope);
+      quote += needQuotes(scope, quote);
       nth++;
@@ -33,3 +54,3 @@ }
   if (quote) {
-    pre = "'" + pre.replace(/'/g, "''") + "'";
+    pre = quotePrefix(pre);
   }
@@ -47,12 +68,12 @@ return pre ? pre + '!' : pre;
     pre += '[' + workbookName + ']';
-    quote += +reBannedChars.test(workbookName);
+    quote += needQuotes(workbookName);
   }
   if (sheetName) {
     pre += sheetName;
-    quote += +reBannedChars.test(sheetName);
+    quote += needQuotes(sheetName);
   }
   if (quote) {
-    pre = "'" + pre.replace(/'/g, "''") + "'";
+    pre = quotePrefix(pre);
   }
   return pre ? pre + '!' : pre;
 }

lib/translateToR1C1.ts

@@ -103,4 +103,3 @@ import { stringifyR1C1RefXlsx } from './stringifyR1C1Ref.ts';
       const tokenValue = token.value;
-      // We can get away with using the xlsx ref-parser here because it is more permissive
-      // and we will end up with the same prefix after serialization anyway:
+      // We can get away with using the xlsx ref-parser here because it is more permissive:
       const ref = quickParseA1(tokenValue);
@@ -107,0 +106,0 @@ if (ref) {

package.json

 {
   "name": "@borgar/fx",
-  "version": "5.0.1",
+  "version": "5.0.2",
   "description": "Utilities for working with Excel formulas",
@@ -71,9 +71,9 @@ "type": "module",
     "typescript": "~5.9.3",
-    "typescript-eslint": "~8.46.2",
-    "vitest": "~4.0.6",
-    "globals": "~16.5.0",
-    "typedoc": "~0.28.14",
-    "typedoc-plugin-markdown": "~4.9.0",
-    "tsup": "~8.5.0"
+    "typescript-eslint": "~8.55.0",
+    "vitest": "~4.0.18",
+    "globals": "~17.3.0",
+    "typedoc": "~0.28.17",
+    "typedoc-plugin-markdown": "~4.10.0",
+    "tsup": "~8.5.1"
   }
 }

New alerts

Long strings

Supply chain risk

Contains long string literals, which may be a sign of obfuscated or packed code.

Found 1 instance in 1 package

Fixed alerts

Long strings

Supply chain risk

Contains long string literals, which may be a sign of obfuscated or packed code.

Found 1 instance in 1 package

Improved metrics

Total package byte prevSize

1823107

0.66%

Lines of code

25969

0.75%

Worsened metrics

Number of low supply chain risk alerts

76

5.56%

No dependency changes