Skip to content

wasilibs/go-libinjection

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

21 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

go-libinjection

go-libinjection is a library for libinjection which wraps the C libinjection library. It provides the same API as libinjection-go and is a drop-in replacement with full API and behavior compatibility. By default, libinjection is packaged as a WebAssembly module and accessed with the pure Go runtime, wazero. This means that it is compatible with any Go application, regardless of availability of cgo.

For TinyGo applications being built for WASM, this library will perform significantly better. For Go applications, it seems to be slower for SQL injection and faster for XSS. The API is a drop-in replacement, so it is best to try it and benchmark to see the effect. It is likely that in the future, libinjection-go will improve to have competitive XSS detection and this library will be relevant only to TinyGo or case studies of wrapping vs rewriting libraries.

Usage

go-libinjection is a standard Go library package and can be added to a go.mod file. It will work fine in Go or TinyGo projects.

go get github.com/wasilibs/go-libinjection

Because the library is a drop-in replacement for petar-dambovaliev/libinjection, an import rewrite can make migrating code to use it simple.

import "github.com/corazawaf/libinjection-go"

can be changed to

import "github.com/wasilibs/go-libinjection"

cgo

This library also supports opting into using cgo to wrap libinjection instead of using WebAssembly. This requires having a built version of the library available - pkg-config will be used to locate the library. The build tag aho_corasick_cgo can be used to enable cgo support.

Performance

Benchmarks are run against every commit in the bench workflow. GitHub action runners are highly virtualized and do not have stable performance across runs, but the relative numbers within a run should still be somewhat, though not precisely, informative.

Microbenchmarks

Microbenchmarks are the same as included in libinjection-go. Full results can be viewed in the workflow, a sample of results for one run looks like this

SQLiDriver/tests/test-sqli-001.txt-2                         1.27µs ± 0%      5.40µs ± 0%          2.13µs ± 2%
SQLiDriver/tests/test-sqli-002.txt-2                          838ns ± 0%      3409ns ± 1%          1441ns ± 1%
SQLiDriver/tests/test-sqli-012.txt-2                          508ns ± 1%      2039ns ± 0%           885ns ± 0%
SQLiDriver/tests/test-sqli-013.txt-2                          450ns ± 1%      1340ns ± 0%           681ns ± 1%
SQLiDriver/tests/test-sqli-014.txt-2                          412ns ± 2%      1165ns ± 0%           576ns ± 1%
SQLiDriver/tests/test-sqli-015.txt-2                         1.11µs ± 0%      4.74µs ± 0%          1.93µs ± 1%
SQLiDriver/tests/test-sqli-016.txt-2                         1.05µs ± 1%      4.57µs ± 0%          1.88µs ± 1%
SQLiDriver/tests/test-sqli-017.txt-2                         1.03µs ± 0%      4.43µs ± 0%          1.82µs ± 0%
SQLiDriver/tests/test-sqli-018.txt-2                         1.11µs ± 3%      4.98µs ± 0%          1.98µs ± 0%
SQLiDriver/tests/test-sqli-033.txt-2                          441ns ± 0%      1058ns ± 1%           566ns ± 0%
SQLiDriver/tests/test-sqli-034.txt-2                          434ns ± 1%      1482ns ± 0%           687ns ± 1%
SQLiDriver/tests/test-sqli-035.txt-2                          222ns ± 1%         3ns ± 0%             3ns ± 0%
SQLiDriver/tests/test-sqli-036.txt-2                         1.15µs ± 0%      4.48µs ± 0%          2.17µs ± 2%
SQLiDriver/tests/test-sqli-037.txt-2                         1.15µs ± 1%      4.49µs ± 0%          2.16µs ± 1%
SQLiDriver/tests/test-sqli-038.txt-2                          580ns ± 1%      2430ns ± 1%           968ns ± 1%
SQLiDriver/tests/test-sqli-049.txt-2                          883ns ± 1%      2587ns ± 0%          1427ns ± 1%
SQLiDriver/tests/test-sqli-050.txt-2                          443ns ± 2%       930ns ± 0%           586ns ± 0%
IsXSS/<script>alert(1);</script>-2                           1.12µs ± 0%      0.46µs ± 1%          0.31µs ± 1%
IsXSS/><script>alert(1);</script>-2                          1.13µs ± 0%      0.49µs ± 1%          0.32µs ± 2%
IsXSS/x_><script>alert(1);</script>-2                        1.13µs ± 0%      0.49µs ± 1%          0.32µs ± 2%
IsXSS/'_><script>alert(1);</script>-2                        1.13µs ± 0%      0.49µs ± 1%          0.32µs ± 1%
IsXSS/"><script>alert(1);</script>-2                         1.13µs ± 0%      0.49µs ± 2%          0.32µs ± 1%
IsXSS/<a_href='javascript:alert(1)'>-2                       2.08µs ± 1%      0.79µs ± 3%          0.38µs ± 1%
IsXSS/<a_href=javascript:alert(1)>-2                         2.11µs ± 1%      0.77µs ± 3%          0.51µs ± 0%
IsXSS/<a_href__=___javascript:alert(1);_>-2                  2.18µs ± 1%      0.81µs ± 2%          0.54µs ± 1%
IsXSS/<a_href="__javascript:alert(1);"_>-2                   2.14µs ± 0%      0.82µs ± 1%          0.38µs ± 0%
IsXSS/<a_href="JAVASCRIPT:alert(1);"_>-2                     2.09µs ± 0%      0.78µs ± 1%          0.37µs ± 0%

The library seems to consistently perform 4x slower for SQL injection, with one odd case, and 2x faster for XSS.

wafbench

wafbench tests the performance of replacing the detect operators of the OWASP CoreRuleSet and Coraza implementation with this library. This benchmark is considered a real world performance test, though the bulk of processing is in regex, not detection.\

WAF/FTW-2                          35.6s ± 0%          36.5s ± 1%              35.3s ± 0%
WAF/POST/1-2                      4.13ms ± 2%         4.22ms ± 1%             4.10ms ± 1%
WAF/POST/1000-2                   21.6ms ± 1%         22.1ms ± 3%             21.2ms ± 1%
WAF/POST/10000-2                   189ms ± 1%          195ms ± 0%              186ms ± 1%
WAF/POST/100000-2                  1.86s ± 1%          1.93s ± 1%              1.83s ± 1%

Using the WebAssembly wrapped version of this library, performance is a little slower, especially at larger payload sizes. With cgo, performance is a little faster.