Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

encode file contents as base64 when the result is smaller #13

Open
wants to merge 2 commits into
base: flatcar-master
Choose a base branch
from
Open

encode file contents as base64 when the result is smaller #13

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
c22588b
config/types/files: encode file contents as base64 when the result is…
jepio Nov 5, 2021
324132f
config/types/files: add unit test for makeDataUrl function
jepio Jan 18, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
24 changes: 16 additions & 8 deletions config/types/files.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15,6 +15,7 @@
package types

import (
"encoding/base64"
"errors"
"flag"
"fmt"
Expand Down Expand Up @@ -129,6 +130,19 @@ func (fc FileContents) Validate() report.Report {
return report.Report{}
}

func makeDataUrl(contents []byte) string {
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I'd suggest putting private function below the init() function, which tends to be at the top of the file.

opaque := "," + dataurl.Escape(contents)
b64 := ";base64," + base64.StdEncoding.EncodeToString(contents)
if len(b64) < len(opaque) {
opaque = b64
}
uri := (&url.URL{
Scheme: "data",
Opaque: opaque,
}).String()
return uri
}
pothos marked this conversation as resolved.
Show resolved Hide resolved
Comment on lines +133 to +144
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think formatting can be more compact:

Suggested change
func makeDataUrl(contents []byte) string {
opaque := "," + dataurl.Escape(contents)
b64 := ";base64," + base64.StdEncoding.EncodeToString(contents)
if len(b64) < len(opaque) {
opaque = b64
}
uri := (&url.URL{
Scheme: "data",
Opaque: opaque,
}).String()
return uri
}
func makeDataUrl(contents []byte) string {
opaque := "," + dataurl.Escape(contents)
if b64 := ";base64," + base64.StdEncoding.EncodeToString(contents); len(b64) < len(opaque) {
opaque = b64
}
return (&url.URL{
Scheme: "data",
Opaque: opaque,
}).String()
}

b64 declaration can be moved outside of if block though, no preferences here.


func init() {
register(func(in Config, ast astnode.AstNode, out ignTypes.Config, platform string) (ignTypes.Config, report.Report, astnode.AstNode) {
r := report.Report{}
Expand Down Expand Up @@ -167,10 +181,7 @@ func init() {

if file.Contents.Inline != "" {
newFile.Contents = ignTypes.FileContents{
Source: (&url.URL{
Scheme: "data",
Opaque: "," + dataurl.EscapeString(file.Contents.Inline),
}).String(),
Source: makeDataUrl([]byte(file.Contents.Inline)),
}
}

Expand Down Expand Up @@ -203,10 +214,7 @@ func init() {

// Include the contents of the local file as if it were provided inline.
newFile.Contents = ignTypes.FileContents{
Source: (&url.URL{
Scheme: "data",
Opaque: "," + dataurl.Escape(contents),
}).String(),
Source: makeDataUrl(contents),
}
}

Expand Down
43 changes: 43 additions & 0 deletions config/types/files_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,43 @@
package types

import (
"github.com/vincent-petithory/dataurl"
"strings"
"testing"
Comment on lines +4 to +6
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

3rd party imports are usually separated from stdlib imports.

Suggested change
"github.com/vincent-petithory/dataurl"
"strings"
"testing"
"strings"
"testing"
"github.com/vincent-petithory/dataurl"

)

func TestDataURL(t *testing.T) {
testString1 := "hello world"
dataUrl1 := makeDataUrl([]byte(testString1))
decodedUrl1, err := dataurl.DecodeString(dataUrl1)
if err != nil {
t.Fatalf("DecodeString #1 failed: %v", err)
}
resultStr1 := string(decodedUrl1.Data)
jepio marked this conversation as resolved.
Show resolved Hide resolved
if strings.HasPrefix(dataUrl1, "data:;base64,") {
t.Errorf("base64 encoding used instead of url encoding: %v", dataUrl1)
}
if resultStr1 != testString1 {
t.Errorf("wanted %v, got %v", testString1, resultStr1)
}

testString2 := `
#!/bin/bash
msg="hello"
f() {
( echo "${msg}" ) &
}
f`
dataUrl2 := makeDataUrl([]byte(testString2))
if !strings.HasPrefix(dataUrl2, "data:;base64,") {
t.Errorf("base64 encoding missing: %v", dataUrl2)
}
decodedUrl2, err := dataurl.DecodeString(dataUrl2)
if err != nil {
t.Fatalf("DecodeString #2 failed: %v", err)
}
resultStr2 := string(decodedUrl2.Data)
if resultStr2 != testString2 {
t.Errorf("wanted %v, got %v", testString2, resultStr2)
}
}
Comment on lines +1 to +43
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

How about some property-based blackbox tests like below? I believe they communicate more about the underlying code and they are not dependent on the implementation so much.

They also capture the case when base64 encoding results in the same data as url encoding, which in this case I guess we should choose simpler url encoding? We could benchmark here, but that seems like an overkill 😄

Suggested change
package types
import (
"github.com/vincent-petithory/dataurl"
"strings"
"testing"
)
func TestDataURL(t *testing.T) {
testString1 := "hello world"
dataUrl1 := makeDataUrl([]byte(testString1))
decodedUrl1, err := dataurl.DecodeString(dataUrl1)
if err != nil {
t.Fatalf("DecodeString #1 failed: %v", err)
}
resultStr1 := string(decodedUrl1.Data)
if strings.HasPrefix(dataUrl1, "data:;base64,") {
t.Errorf("base64 encoding used instead of url encoding: %v", dataUrl1)
}
if resultStr1 != testString1 {
t.Errorf("wanted %v, got %v", testString1, resultStr1)
}
testString2 := `
#!/bin/bash
msg="hello"
f() {
( echo "${msg}" ) &
}
f`
dataUrl2 := makeDataUrl([]byte(testString2))
if !strings.HasPrefix(dataUrl2, "data:;base64,") {
t.Errorf("base64 encoding missing: %v", dataUrl2)
}
decodedUrl2, err := dataurl.DecodeString(dataUrl2)
if err != nil {
t.Fatalf("DecodeString #2 failed: %v", err)
}
resultStr2 := string(decodedUrl2.Data)
if resultStr2 != testString2 {
t.Errorf("wanted %v, got %v", testString2, resultStr2)
}
}
package types_test
import (
"encoding/base64"
"strings"
"testing"
"github.com/vincent-petithory/dataurl"
"github.com/flatcar-linux/container-linux-config-transpiler/config/types"
)
const (
base64overhead = len(";base64")
baseOverhead = len("data:,")
)
func Test_When_converting_files_inline_content_gets(t *testing.T) {
t.Parallel()
t.Run("url_encoded_when_base64_encoding", func(t *testing.T) {
t.Parallel()
for name, testInput := range map[string]string{
"increase_contents_size": "hello world",
"does_not_change_contents_size": "1234567 aa",
} {
testInput := testInput
t.Run(name, func(t *testing.T) {
t.Parallel()
c := types.Config{
Storage: types.Storage{
Files: []types.File{
{
Path: "/foo",
Contents: types.FileContents{
Inline: testInput,
},
},
},
},
}
f, r := types.Convert(c, "", nil)
if r.IsFatal() {
t.Fatalf("Unexpected conversion error: %v", r)
}
if strings.Contains(f.Storage.Files[0].Contents.Source, "base64") {
t.Fatalf("Expected data to be URl-encoded, got: %q", f.Storage.Files[0].Contents.Source)
}
b64encodedData := base64.StdEncoding.EncodeToString([]byte(testInput))
b64encodedDataLengthWithOverhead := base64overhead + baseOverhead + len(b64encodedData)
if b64encodedDataLengthWithOverhead < len(f.Storage.Files[0].Contents.Source) {
t.Fatalf("Expected encoded file contents to be smaller than base64 encoded data with overhead")
}
})
}
})
t.Run("base64_encoded_when_it_reduces_content_size", func(t *testing.T) {
base64reducibleData := `
#!/bin/bash
msg="hello"
f() {
( echo "${msg}" ) &
}
f`
c := types.Config{
Storage: types.Storage{
Files: []types.File{
{
Path: "/foo",
Contents: types.FileContents{
Inline: base64reducibleData,
},
},
},
},
}
f, r := types.Convert(c, "", nil)
if r.IsFatal() {
t.Fatalf("Unexpected conversion error: %v", r)
}
if !strings.Contains(f.Storage.Files[0].Contents.Source, "base64") {
t.Fatalf("Expected data to be base64-encoded, got: %q", f.Storage.Files[0].Contents.Source)
}
urlEncodedDataLengthWithOverhead := len(dataurl.Escape([]byte(base64reducibleData))) + baseOverhead
if urlEncodedDataLengthWithOverhead <= len(f.Storage.Files[0].Contents.Source) {
t.Fatalf("Expected encoded file contents to be smaller than base64 encoded data with overhead.")
}
})
}