Labyricorn/liquid
1
0
Fork 0
mirror of https://github.com/Shopify/liquid.git synced 2026-09-16 17:30:43 -07:00
Code Issues Packages Projects Releases Wiki Activity

Add rigid mode to rake benchmark task

Browse Source 
The benchmark results show that rigid mode performs a bit better than both strict
and lax modes across most metrics, including tokenization, parsing, rendering,
and their combined operations. Rigid mode consistently delivers the highest
number of iterations per second, with performance differences staying within
1–2% compared to the other modes

```
================================================================================
/opt/rubies/3.4.1/bin/ruby ./performance/benchmark.rb lax

Running benchmark for 20 seconds (with 10 seconds warmup).

ruby 3.4.1 (2024-12-25 revision 48d4efcb85) +YJIT +PRISM [arm64-darwin23]
Warming up --------------------------------------
           tokenize:   332.000 i/100ms
              parse:    14.000 i/100ms
             render:    61.000 i/100ms
     parse & render:    11.000 i/100ms
Calculating -------------------------------------
           tokenize:      3.325k (± 1.0%) i/s  (300.73 μs/i) -    66.732k in  20.070562s
              parse:    148.166 (± 0.7%) i/s    (6.75 ms/i)  -     2.968k in  20.032971s
             render:    654.428 (± 4.0%) i/s    (1.53 ms/i)  -    13.115k in  20.090452s
     parse & render:    116.108 (± 1.7%) i/s    (8.61 ms/i)  -     2.332k in  20.089221s

================================================================================
/opt/rubies/3.4.1/bin/ruby ./performance/benchmark.rb strict

Running benchmark for 20 seconds (with 10 seconds warmup).

ruby 3.4.1 (2024-12-25 revision 48d4efcb85) +YJIT +PRISM [arm64-darwin23]
Warming up --------------------------------------
           tokenize:   332.000 i/100ms
              parse:    14.000 i/100ms
             render:    61.000 i/100ms
     parse & render:    11.000 i/100ms
Calculating -------------------------------------
           tokenize:      3.332k (± 0.2%) i/s  (300.14 μs/i) -    66.732k in  20.029095s
              parse:    145.674 (± 0.0%) i/s    (6.86 ms/i)  -     2.926k in  20.086104s
             render:    656.711 (± 4.6%) i/s    (1.52 ms/i)  -    13.115k in  20.050810s
     parse & render:    114.705 (± 0.0%) i/s    (8.72 ms/i)  -     2.299k in  20.043028s

================================================================================
/opt/rubies/3.4.1/bin/ruby ./performance/benchmark.rb rigid

Running benchmark for 20 seconds (with 10 seconds warmup).

ruby 3.4.1 (2024-12-25 revision 48d4efcb85) +YJIT +PRISM [arm64-darwin23]
Warming up --------------------------------------
           tokenize:   333.000 i/100ms
              parse:    14.000 i/100ms
             render:    62.000 i/100ms
     parse & render:    11.000 i/100ms
Calculating -------------------------------------
           tokenize:      3.334k (± 0.3%) i/s  (299.93 μs/i) -    66.933k in  20.075484s
              parse:    148.349 (± 2.0%) i/s    (6.74 ms/i)  -     2.968k in  20.019775s
             render:    663.752 (± 2.6%) i/s    (1.51 ms/i)  -    13.268k in  20.010303s
     parse & render:    116.464 (± 2.6%) i/s    (8.59 ms/i)  -     2.332k in  20.037869s
liquid$
```
This commit is contained in:
Guilherme Carreiro
2025-10-27 16:33:31 +01:00
committed by Guilherme Carreiro
parent 6aa4041c0f
commit 15430c0770
1 changed files with 7 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files
Rakefile
+7 -2
View File
@@ -88,8 +88,13 @@ namespace :benchmark do
ruby "./performance/benchmark.rb strict"
end
desc "Run the liquid benchmark with both lax and strict parsing"
task run: [:lax, :strict]
desc "Run the liquid benchmark with rigid parsing"
task :rigid do
ruby "./performance/benchmark.rb rigid"
end
desc "Run the liquid benchmark with lax, strict, and rigid parsing"
task run: [:lax, :strict, :rigid]
desc "Run unit benchmarks"
namespace :unit do