https://github.com/mrd0ll4r/urlparse

https://github.com/mrd0ll4r/urlparse

Last synced: 21 days ago
JSON representation

• Host: GitHub
• URL: https://github.com/mrd0ll4r/urlparse
• Owner: mrd0ll4r
• Created: 2015-11-08T19:22:40.000Z (about 9 years ago)
• Default Branch: master
• Last Pushed: 2015-11-08T19:30:35.000Z (about 9 years ago)
• Last Synced: 2024-10-16T06:35:12.821Z (2 months ago)
• Language: Go
• Size: 0 Bytes
• Stars: 1
• Watchers: 3
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
#urlparse

Benchmarks different methods of url query parsing.

Results:

    BenchmarkParseStdlib-8                    200000              8235 ns/op

    BenchmarkParseStdlibNoUrldecode-8         200000              8525 ns/op

    BenchmarkQueryNew-8                       300000              5460 ns/op

    BenchmarkQueryNewSliced-8                 200000              8015 ns/op

    BenchmarkUrlparseStdlib-8                 200000              8950 ns/op

    BenchmarkPathParseStdlib-8                200000              8105 ns/op

    BenchmarkPathParseStdlibNoUrldecode-8     200000              8130 ns/op

    BenchmarkPathQueryNew-8                   300000              5366 ns/op

    BenchmarkPathQueryNewSliced-8             200000              7955 ns/op

    BenchmarkPathUrlparseStdlib-8             200000              8705 ns/op

The winner is `github.com/chihaya/chihaya/http/query.New()` (`QueryNew` and `PathQueryNew`), but the tradeoff is that every key but `info_hash` is present only once. If the key is found again, the previous value will be overwritten.

As an attempt to solve this, I added `QueryNewSliced`, it's still faster than the stdlib, but way slower than the original.

The stdlibs `net/url.Parse().Values()` (`UrlparseStdlib` and `PathUrlparseStdlib`) is the slowest, because it parses the whole URL, and then the query params.

If we only use `net/url.ParseQuery()` (`ParseStdlib` and `PathParseStdlib`) we are a little bit faster, but not much.

Stripping the URLdecoding from the stdlib implementation turned out to not increase performance (?)