clustring - npm Package Compare versions

Comparing version 0.0.8 to 0.0.9

index.js

@@ -123,2 +123,4 @@ 'use strict';
           nIterationsBetweenTickChecks = _this$options.nIterationsBetweenTickChecks;
+    const usedStrs = {}; // strs that have been placed in a cluster already
+
     const strs = Object.keys(bucket);
@@ -133,2 +135,8 @@ const nStrs = strs.length;
       const a = strs[ai];
+
+      if (a in usedStrs) {
+        i += nStrs - ai - 1;
+        continue;
+      }
+
       const aCount = bucket[a];
@@ -157,2 +165,3 @@ let bin = null; // set iff any b clusters with a
         const b = strs[bi];
+        if (b in usedStrs) continue;
         const d = distance(a, b);
@@ -181,2 +190,3 @@
           bin.bucket[b] = bCount;
+          usedStrs[b] = null;
         }
@@ -183,0 +193,0 @@ }

package.json

 {
   "name": "clustring",
-  "version": "0.0.8",
+  "version": "0.0.9",
   "description": "Algorithms for clustering strings",
@@ -8,3 +8,4 @@ "main": "index.js",
     "build": "rollup -c",
-    "prepublish": "npm run build",
+    "prepublishOnly": "npm test",
+    "prepare": "npm run build",
     "test": "jest"
@@ -11,0 +12,0 @@ },

src/knn/clusterByKnn.js

@@ -22,2 +22,3 @@ import { tick } from '../util'
 
+    const usedStrs = {} // strs that have been placed in a cluster already
     const strs = Object.keys(bucket)
@@ -33,2 +34,8 @@ const nStrs = strs.length
       const a = strs[ai]
+
+      if (a in usedStrs) {
+        i += (nStrs - ai - 1)
+        continue
+      }
+
       const aCount = bucket[a]
@@ -56,2 +63,5 @@ let bin = null // set iff any b clusters with a
         const b = strs[bi]
+
+        if (b in usedStrs) continue
+
         const d = distance(a, b)
@@ -76,2 +86,3 @@
           bin.bucket[b] = bCount
+          usedStrs[b] = null
         }
@@ -78,0 +89,0 @@ }

src/knn/clusterByKnn.test.js

@@ -32,2 +32,15 @@ import clusterByKnn from './clusterByKnn'
 
+  it('should not put a value into two clusters', async () => {
+    const bucket = { a: 1, b: 1, c: 1, d: 1 }
+    const distance = (a, b) => 1
+    const bins = await clusterByKnn(bucket, distance, 1).cluster()
+    expect(bins).toEqual([
+      {
+        name: 'a',
+        count: 4,
+        bucket: { a: 1, b: 1, c: 1, d: 1 }
+      }
+    ])
+  })
+
   it('should pick highest-count name', async () => {
@@ -34,0 +47,0 @@ const bucket = { a: 3, b: 4 }

src/knn/levenshtein.test.js

@@ -18,3 +18,3 @@ import levenshtein from './levenshtein'
   [ '', 'left empty', 10 ],
-  [ 'accentś', 'accėnts', 2 ],
+  [ 'accentś', 'accėnts', 2 ]
 ]
@@ -21,0 +21,0 @@

New alerts

License Policy Violation

License

This package is not allowed per your license policy. Review the package's license to ensure compliance.

Found 1 instance in 1 package

Fixed alerts

License Policy Violation

License

This package is not allowed per your license policy. Review the package's license to ensure compliance.

Found 1 instance in 1 package

Improved metrics

Total package byte prevSize

95294

increased by1.33%

Lines of code

1167

increased by2.28%

No dependency changes