Clone Tools
  • last updated a few seconds ago
Constraints
Constraints: committers
 
Constraints: files
Constraints: dates
Perform simplification only if specs contain expected types

Perform simplification only if specs contain expected types

Perform simplification only if specs contain expected types

Use hash set estimate size to avoid resizes

  1. … 1 more file in changeset.
Use hash set estimate size to avoid resizes

  1. … 1 more file in changeset.
Use hash set estimate size to avoid resizes

  1. … 1 more file in changeset.
Give up a bit on immutability for the sake of performance

Creating an immutable list with Guava has a slight performance

overhead. Given we have to mitigate a performance regression,

choice is to give up here, knowing we have enough guarantees

(the structure is internal).

    • -10
    • +12
    ./NormalizingExcludeFactory.java
Give up a bit on immutability for the sake of performance

Creating an immutable list with Guava has a slight performance

overhead. Given we have to mitigate a performance regression,

choice is to give up here, knowing we have enough guarantees

(the structure is internal).

    • -10
    • +12
    ./NormalizingExcludeFactory.java
Give up a bit on immutability for the sake of performance

Creating an immutable list with Guava has a slight performance

overhead. Given we have to mitigate a performance regression,

choice is to give up here, knowing we have enough guarantees

(the structure is internal).

    • -10
    • +12
    ./NormalizingExcludeFactory.java
Add union/intersection simplification

Add union/intersection simplification

Add union/intersection simplification

Use coarce grained locking

Measurements show it's significantly faster than using a read/write lock.

  1. … 1 more file in changeset.
Use coarce grained locking

Measurements show it's significantly faster than using a read/write lock.

Use coarce grained locking

Measurements show it's significantly faster than using a read/write lock.

  1. … 1 more file in changeset.
Use coarce grained locking

Measurements show it's significantly faster than using a read/write lock.

  1. … 1 more file in changeset.
Use coarce grained locking

Measurements show it's significantly faster than using a read/write lock.

  1. … 1 more file in changeset.
Optimize flattening

This commit optimizes flattening by avoiding the creation of intermediate

data structures. In particular using lists we were converting from and to

sets unnecessarily.

    • -0
    • +61
    ./ExcludeFactory.java
    • -29
    • +67
    ./NormalizingExcludeFactory.java
  1. … 8 more files in changeset.
Optimize flattening

This commit optimizes flattening by avoiding the creation of intermediate

data structures. In particular using lists we were converting from and to

sets unnecessarily.

    • -29
    • +64
    ./NormalizingExcludeFactory.java
  1. … 8 more files in changeset.
Optimize flattening

This commit optimizes flattening by avoiding the creation of intermediate

data structures. In particular using lists we were converting from and to

sets unnecessarily.

    • -0
    • +61
    ./ExcludeFactory.java
    • -29
    • +64
    ./NormalizingExcludeFactory.java
  1. … 8 more files in changeset.
Optimize flattening

This commit optimizes flattening by avoiding the creation of intermediate

data structures. In particular using lists we were converting from and to

sets unnecessarily.

    • -29
    • +64
    ./NormalizingExcludeFactory.java
  1. … 8 more files in changeset.
Reuse caching at lower level

The merge cache can be reused at different stages. In particular, when

the normalizing factory normalizes unions/intersections, it is possible

that the normalized union is found in cache.

  1. … 1 more file in changeset.
Reuse caching at lower level

The merge cache can be reused at different stages. In particular, when

the normalizing factory normalizes unions/intersections, it is possible

that the normalized union is found in cache.

  1. … 1 more file in changeset.
Reuse caching at lower level

The merge cache can be reused at different stages. In particular, when

the normalizing factory normalizes unions/intersections, it is possible

that the normalized union is found in cache.

  1. … 1 more file in changeset.
Optimize `ExcludePair` for comparisons

Optimize `ExcludePair` for comparisons

Optimize `ExcludePair` for comparisons

Optimize `ExcludePair` for comparisons

Remove indexed exclude factory

It didn't prove as fast as it was intended to be. Instead, we performed

the same optimization for single groups/modules as we did for module sets.

    • -37
    • +91
    ./NormalizingExcludeFactory.java
  1. … 13 more files in changeset.
Remove indexed exclude factory

It didn't prove as fast as it was intended to be. Instead, we performed

the same optimization for single groups/modules as we did for module sets.

    • -37
    • +91
    ./NormalizingExcludeFactory.java
  1. … 13 more files in changeset.