mirror of
https://github.com/go-sylixos/elvish.git
synced 2024-12-13 01:47:51 +08:00
pkg/md: Split test utilities from md_test.go into testutils_test.go.
This commit is contained in:
Qi Xiao
parent
b846cfa60b
commit
9d75552d0b
|
|
@ -1,237 +1,28 @@
|
|||
package md_test
|
||||
|
||||
import (
|
||||
_ "embed"
|
||||
"encoding/json"
|
||||
"fmt"
|
||||
"html"
|
||||
"regexp"
|
||||
"strconv"
|
||||
"strings"
|
||||
"testing"
|
||||
|
||||
"github.com/google/go-cmp/cmp"
|
||||
. "src.elv.sh/pkg/md"
|
||||
"src.elv.sh/pkg/must"
|
||||
"src.elv.sh/pkg/testutil"
|
||||
)
|
||||
|
||||
type testCase struct {
|
||||
Markdown string `json:"markdown"`
|
||||
HTML string `json:"html"`
|
||||
Example int `json:"example"`
|
||||
Section string `json:"section"`
|
||||
Name string // supplemental cases only
|
||||
}
|
||||
|
||||
//go:embed spec/spec.json
|
||||
var specJSON []byte
|
||||
|
||||
var testCases []testCase
|
||||
|
||||
// When adding supplemental test cases, check a reference implementation to
|
||||
// determine the expected output. https://spec.commonmark.org/dingus (which uses
|
||||
// https://github.com/commonmark/commonmark.js) is the most convenient.
|
||||
var supplementalCases = []testCase{
|
||||
{
|
||||
Name: "Fenced code blocks supplemental/Empty line in list item",
|
||||
Markdown: "- ```\n a\n\n ```\n",
|
||||
HTML: dedent(`
|
||||
<ul>
|
||||
<li>
|
||||
<pre><code>a
|
||||
|
||||
</code></pre>
|
||||
</li>
|
||||
</ul>
|
||||
`),
|
||||
},
|
||||
{
|
||||
Name: "HTML blocks supplemental/Closed by lack of blockquote marker",
|
||||
Markdown: dedent(`
|
||||
> <pre>
|
||||
|
||||
a
|
||||
`),
|
||||
HTML: dedent(`
|
||||
<blockquote>
|
||||
<pre>
|
||||
</blockquote>
|
||||
<p>a</p>
|
||||
`),
|
||||
},
|
||||
{
|
||||
Name: "HTML blocks supplemental/Closed by insufficient list item indentation",
|
||||
Markdown: dedent(`
|
||||
- <pre>
|
||||
a
|
||||
`),
|
||||
HTML: dedent(`
|
||||
<ul>
|
||||
<li>
|
||||
<pre>
|
||||
</li>
|
||||
</ul>
|
||||
<p>a</p>
|
||||
`),
|
||||
},
|
||||
{
|
||||
Name: "Blockquotes supplemental/Increasing level",
|
||||
Markdown: dedent(`
|
||||
> a
|
||||
>> b
|
||||
`),
|
||||
HTML: dedent(`
|
||||
<blockquote>
|
||||
<p>a</p>
|
||||
<blockquote>
|
||||
<p>b</p>
|
||||
</blockquote>
|
||||
</blockquote>
|
||||
`),
|
||||
},
|
||||
{
|
||||
Name: "Blockquotes supplemental/Reducing level",
|
||||
Markdown: dedent(`
|
||||
>> a
|
||||
>
|
||||
> b
|
||||
`),
|
||||
HTML: dedent(`
|
||||
<blockquote>
|
||||
<blockquote>
|
||||
<p>a</p>
|
||||
</blockquote>
|
||||
<p>b</p>
|
||||
</blockquote>
|
||||
`),
|
||||
},
|
||||
{
|
||||
Name: "List items supplemental/Two leading empty lines with spaces",
|
||||
Markdown: dedent(`
|
||||
-
|
||||
|
||||
a
|
||||
`),
|
||||
HTML: dedent(`
|
||||
<ul>
|
||||
<li></li>
|
||||
</ul>
|
||||
<p>a</p>
|
||||
`),
|
||||
},
|
||||
{
|
||||
Name: "Links supplemental/Backslash and entity in destination",
|
||||
Markdown: `[a](\>)`,
|
||||
HTML: `<p><a href="&gt;">a</a></p>` + "\n",
|
||||
},
|
||||
{
|
||||
Name: "Links supplemental/Backslash and entity in title",
|
||||
Markdown: `[a](b (\>))`,
|
||||
HTML: `<p><a href="b" title="&gt;">a</a></p>` + "\n",
|
||||
},
|
||||
{
|
||||
Name: "Links supplemental/Unmatched ( in destination, with title",
|
||||
Markdown: `[a](http://( "b")`,
|
||||
HTML: "<p>[a](http://( "b")</p>\n",
|
||||
},
|
||||
{
|
||||
Name: "Links supplemental/Unescaped ( in title started with (",
|
||||
Markdown: `[a](b (()))`,
|
||||
HTML: "<p>[a](b (()))</p>\n",
|
||||
},
|
||||
{
|
||||
Name: "Links supplemental/Literal & in destination",
|
||||
Markdown: `[a](http://b?c&d)`,
|
||||
HTML: `<p><a href="http://b?c&d">a</a></p>` + "\n",
|
||||
},
|
||||
{
|
||||
Name: "Image supplemental/Omit hard line break tag in alt",
|
||||
Markdown: dedent(`
|
||||
|
||||
`),
|
||||
HTML: dedent(`
|
||||
<p><img src="c.png" alt="a
|
||||
b" /></p>
|
||||
`),
|
||||
},
|
||||
// This behavior is intentionally under-specified in the spec. The reference
|
||||
// implementations puts the raw HTML in the alt attribute, so we match their
|
||||
// behavior.
|
||||
//
|
||||
// CommonMark.js is inconsistent here and does not escape the < and > in the
|
||||
// alt attribute: https://github.com/commonmark/commonmark.js/issues/264
|
||||
{
|
||||
Name: "Image supplemental/Keep raw HTML in alt",
|
||||
Markdown: "",
|
||||
HTML: `<p><img src="b.png" alt="a <a></a>" /></p>` + "\n",
|
||||
},
|
||||
// CommonMark.js has a bug and will not generate the expected output:
|
||||
// https://github.com/commonmark/commonmark.js/issues/263
|
||||
{
|
||||
Name: "Autolinks supplemental/Entity",
|
||||
Markdown: `<http://>>`,
|
||||
HTML: `<p><a href="http://%3E">http://></a></p>` + "\n",
|
||||
},
|
||||
{
|
||||
Name: "Raw HTML supplemental/unclosed <",
|
||||
Markdown: `a<`,
|
||||
HTML: "<p>a<</p>\n",
|
||||
},
|
||||
{
|
||||
Name: "Raw HTML supplemental/unclosed <!--",
|
||||
Markdown: `a<!--`,
|
||||
HTML: "<p>a<!--</p>\n",
|
||||
},
|
||||
{
|
||||
Name: "Soft line breaks supplemental/trailing spaces in last line",
|
||||
Markdown: "a \n",
|
||||
HTML: "<p>a</p>\n",
|
||||
},
|
||||
}
|
||||
|
||||
func init() {
|
||||
must.OK(json.Unmarshal(specJSON, &testCases))
|
||||
testCases = append(testCases, supplementalCases...)
|
||||
}
|
||||
|
||||
func TestRender(t *testing.T) {
|
||||
testutil.Set(t, &UnescapeEntities, html.UnescapeString)
|
||||
for _, tc := range testCases {
|
||||
name := tc.Name
|
||||
if name == "" {
|
||||
name = fmt.Sprintf("%s/Example %d", tc.Section, tc.Example)
|
||||
}
|
||||
t.Run(name, func(t *testing.T) {
|
||||
switch tc.Section {
|
||||
case "Tabs",
|
||||
"Setext headings",
|
||||
"Link reference definitions":
|
||||
t.Skip("section not supported")
|
||||
}
|
||||
switch tc.Example {
|
||||
case 59, 115, 141, 300:
|
||||
t.Skip("setext heading not supported")
|
||||
case 23, 33, 317,
|
||||
// Link
|
||||
526, 527, 528, 529, 530, 531, 532, 533, 534, 535, 536, 537, 538, 539, 540, 541, 542, 543, 544, 548, 549, 552, 553, 554, 555, 556, 557, 558, 559, 560, 561, 562, 563, 564, 565, 566, 567, 568, 569, 570, 572, 575, 576,
|
||||
// Image
|
||||
581, 582, 583, 584, 585, 586, 587, 588, 590, 591, 592:
|
||||
t.Skip("link reference definitions not supported")
|
||||
case 294, 296, 307, 318, 319, 320, 321, 323:
|
||||
t.Skip("tight list not supported")
|
||||
}
|
||||
|
||||
var codec htmlCodec
|
||||
Render(tc.Markdown, &codec)
|
||||
got := codec.String()
|
||||
t.Run(tc.testName(), func(t *testing.T) {
|
||||
tc.skipIfNotSupported(t)
|
||||
got := render(tc.Markdown, &htmlCodec{})
|
||||
// Try to hide the difference between tight and loose lists by
|
||||
// "loosifying" the output. This only works for tight lists whose
|
||||
// items consist of single lines, so more complex cases are still
|
||||
// skipped.
|
||||
want := loosifyLists(tc.HTML)
|
||||
hr := strings.Repeat("═", 40)
|
||||
if diff := cmp.Diff(want, got); diff != "" {
|
||||
t.Errorf("input:\n%s\ndiff (-want +got):\n%s",
|
||||
hr+"\n"+tc.Markdown+hr, diff)
|
||||
|
|
@ -320,24 +111,3 @@ func (c *htmlCodec) Do(op Op) {
|
|||
type attrBuilder struct{ strings.Builder }
|
||||
|
||||
func (a *attrBuilder) set(k, v string) { fmt.Fprintf(a, ` %s="%s"`, k, escapeHTML(v)) }
|
||||
|
||||
var looseListItem = regexp.MustCompile(`<li>([^<]+)</li>`)
|
||||
|
||||
func loosifyLists(html string) string {
|
||||
return strings.ReplaceAll(
|
||||
looseListItem.ReplaceAllString(html, "<li>\n<p>$1</p>\n</li>"),
|
||||
"<li></li>", "<li>\n</li>")
|
||||
}
|
||||
|
||||
func dedent(text string) string {
|
||||
lines := strings.Split(strings.TrimPrefix(text, "\n"), "\n")
|
||||
line0 := lines[0]
|
||||
indent := line0[:len(line0)-len(strings.TrimLeft(lines[0], " \t"))]
|
||||
for i, line := range lines {
|
||||
if !strings.HasPrefix(line, indent) && line != "" {
|
||||
panic(fmt.Sprintf("line %d is not empty but doesn't start with %q", i, indent))
|
||||
}
|
||||
lines[i] = strings.TrimPrefix(line, indent)
|
||||
}
|
||||
return strings.Join(lines, "\n")
|
||||
}
|
||||
|
|
|
|||
263
pkg/md/testutils_test.go
Normal file
263
pkg/md/testutils_test.go
Normal file
|
|
@ -0,0 +1,263 @@
|
|||
package md_test
|
||||
|
||||
import (
|
||||
_ "embed"
|
||||
"encoding/json"
|
||||
"fmt"
|
||||
"regexp"
|
||||
"strings"
|
||||
"testing"
|
||||
|
||||
"src.elv.sh/pkg/md"
|
||||
"src.elv.sh/pkg/must"
|
||||
)
|
||||
|
||||
type testCase struct {
|
||||
Markdown string `json:"markdown"`
|
||||
HTML string `json:"html"`
|
||||
Example int `json:"example"`
|
||||
Section string `json:"section"`
|
||||
Name string // supplemental cases only
|
||||
}
|
||||
|
||||
func (tc *testCase) testName() string {
|
||||
if tc.Name != "" {
|
||||
return tc.Name
|
||||
}
|
||||
return fmt.Sprintf("%s/Example %d", tc.Section, tc.Example)
|
||||
}
|
||||
|
||||
func (tc *testCase) skipReason() string {
|
||||
switch tc.Section {
|
||||
case "Tabs",
|
||||
"Setext headings",
|
||||
"Link reference definitions":
|
||||
return "section not supported"
|
||||
}
|
||||
switch tc.Example {
|
||||
case 59, 115, 141, 300:
|
||||
return "setext heading not supported"
|
||||
case 23, 33, 317,
|
||||
// Link
|
||||
526, 527, 528, 529, 530, 531, 532, 533, 534, 535, 536, 537, 538, 539, 540, 541, 542, 543, 544, 548, 549, 552, 553, 554, 555, 556, 557, 558, 559, 560, 561, 562, 563, 564, 565, 566, 567, 568, 569, 570, 572, 575, 576,
|
||||
// Image
|
||||
581, 582, 583, 584, 585, 586, 587, 588, 590, 591, 592:
|
||||
return "link reference definitions not supported"
|
||||
case 294, 296, 307, 318, 319, 320, 321, 323:
|
||||
return "tight list not supported"
|
||||
}
|
||||
return ""
|
||||
}
|
||||
|
||||
func (tc *testCase) skipIfNotSupported(t *testing.T) {
|
||||
if reason := tc.skipReason(); reason != "" {
|
||||
t.Skip(reason)
|
||||
}
|
||||
}
|
||||
|
||||
//go:embed spec/spec.json
|
||||
var specJSON []byte
|
||||
|
||||
var testCases []testCase
|
||||
|
||||
// When adding supplemental test cases, check a reference implementation to
|
||||
// determine the expected output. https://spec.commonmark.org/dingus (which uses
|
||||
// https://github.com/commonmark/commonmark.js) is the most convenient.
|
||||
var supplementalCases = []testCase{
|
||||
{
|
||||
Name: "Fenced code blocks supplemental/Empty line in list item",
|
||||
Markdown: "- ```\n a\n\n ```\n",
|
||||
HTML: dedent(`
|
||||
<ul>
|
||||
<li>
|
||||
<pre><code>a
|
||||
|
||||
</code></pre>
|
||||
</li>
|
||||
</ul>
|
||||
`),
|
||||
},
|
||||
{
|
||||
Name: "HTML blocks supplemental/Closed by lack of blockquote marker",
|
||||
Markdown: dedent(`
|
||||
> <pre>
|
||||
|
||||
a
|
||||
`),
|
||||
HTML: dedent(`
|
||||
<blockquote>
|
||||
<pre>
|
||||
</blockquote>
|
||||
<p>a</p>
|
||||
`),
|
||||
},
|
||||
{
|
||||
Name: "HTML blocks supplemental/Closed by insufficient list item indentation",
|
||||
Markdown: dedent(`
|
||||
- <pre>
|
||||
a
|
||||
`),
|
||||
HTML: dedent(`
|
||||
<ul>
|
||||
<li>
|
||||
<pre>
|
||||
</li>
|
||||
</ul>
|
||||
<p>a</p>
|
||||
`),
|
||||
},
|
||||
{
|
||||
Name: "Blockquotes supplemental/Increasing level",
|
||||
Markdown: dedent(`
|
||||
> a
|
||||
>> b
|
||||
`),
|
||||
HTML: dedent(`
|
||||
<blockquote>
|
||||
<p>a</p>
|
||||
<blockquote>
|
||||
<p>b</p>
|
||||
</blockquote>
|
||||
</blockquote>
|
||||
`),
|
||||
},
|
||||
{
|
||||
Name: "Blockquotes supplemental/Reducing level",
|
||||
Markdown: dedent(`
|
||||
>> a
|
||||
>
|
||||
> b
|
||||
`),
|
||||
HTML: dedent(`
|
||||
<blockquote>
|
||||
<blockquote>
|
||||
<p>a</p>
|
||||
</blockquote>
|
||||
<p>b</p>
|
||||
</blockquote>
|
||||
`),
|
||||
},
|
||||
{
|
||||
Name: "List items supplemental/Two leading empty lines with spaces",
|
||||
Markdown: dedent(`
|
||||
-
|
||||
|
||||
a
|
||||
`),
|
||||
HTML: dedent(`
|
||||
<ul>
|
||||
<li></li>
|
||||
</ul>
|
||||
<p>a</p>
|
||||
`),
|
||||
},
|
||||
{
|
||||
Name: "Links supplemental/Backslash and entity in destination",
|
||||
Markdown: `[a](\>)`,
|
||||
HTML: `<p><a href="&gt;">a</a></p>` + "\n",
|
||||
},
|
||||
{
|
||||
Name: "Links supplemental/Backslash and entity in title",
|
||||
Markdown: `[a](b (\>))`,
|
||||
HTML: `<p><a href="b" title="&gt;">a</a></p>` + "\n",
|
||||
},
|
||||
{
|
||||
Name: "Links supplemental/Unmatched ( in destination, with title",
|
||||
Markdown: `[a](http://( "b")`,
|
||||
HTML: "<p>[a](http://( "b")</p>\n",
|
||||
},
|
||||
{
|
||||
Name: "Links supplemental/Unescaped ( in title started with (",
|
||||
Markdown: `[a](b (()))`,
|
||||
HTML: "<p>[a](b (()))</p>\n",
|
||||
},
|
||||
{
|
||||
Name: "Links supplemental/Literal & in destination",
|
||||
Markdown: `[a](http://b?c&d)`,
|
||||
HTML: `<p><a href="http://b?c&d">a</a></p>` + "\n",
|
||||
},
|
||||
{
|
||||
Name: "Image supplemental/Omit hard line break tag in alt",
|
||||
Markdown: dedent(`
|
||||
|
||||
`),
|
||||
HTML: dedent(`
|
||||
<p><img src="c.png" alt="a
|
||||
b" /></p>
|
||||
`),
|
||||
},
|
||||
// This behavior is intentionally under-specified in the spec. The reference
|
||||
// implementations puts the raw HTML in the alt attribute, so we match their
|
||||
// behavior.
|
||||
//
|
||||
// CommonMark.js is inconsistent here and does not escape the < and > in the
|
||||
// alt attribute: https://github.com/commonmark/commonmark.js/issues/264
|
||||
{
|
||||
Name: "Image supplemental/Keep raw HTML in alt",
|
||||
Markdown: "",
|
||||
HTML: `<p><img src="b.png" alt="a <a></a>" /></p>` + "\n",
|
||||
},
|
||||
// CommonMark.js has a bug and will not generate the expected output:
|
||||
// https://github.com/commonmark/commonmark.js/issues/263
|
||||
{
|
||||
Name: "Autolinks supplemental/Entity",
|
||||
Markdown: `<http://>>`,
|
||||
HTML: `<p><a href="http://%3E">http://></a></p>` + "\n",
|
||||
},
|
||||
{
|
||||
Name: "Raw HTML supplemental/unclosed <",
|
||||
Markdown: `a<`,
|
||||
HTML: "<p>a<</p>\n",
|
||||
},
|
||||
{
|
||||
Name: "Raw HTML supplemental/unclosed <!--",
|
||||
Markdown: `a<!--`,
|
||||
HTML: "<p>a<!--</p>\n",
|
||||
},
|
||||
{
|
||||
Name: "Soft line breaks supplemental/trailing spaces in last line",
|
||||
Markdown: "a \n",
|
||||
HTML: "<p>a</p>\n",
|
||||
},
|
||||
}
|
||||
|
||||
func init() {
|
||||
must.OK(json.Unmarshal(specJSON, &testCases))
|
||||
testCases = append(testCases, supplementalCases...)
|
||||
}
|
||||
|
||||
var (
|
||||
hr = strings.Repeat("═", 40)
|
||||
)
|
||||
|
||||
func dedent(text string) string {
|
||||
lines := strings.Split(strings.TrimPrefix(text, "\n"), "\n")
|
||||
line0 := lines[0]
|
||||
indent := line0[:len(line0)-len(strings.TrimLeft(lines[0], " \t"))]
|
||||
for i, line := range lines {
|
||||
if !strings.HasPrefix(line, indent) && line != "" {
|
||||
panic(fmt.Sprintf("line %d is not empty but doesn't start with %q", i, indent))
|
||||
}
|
||||
lines[i] = strings.TrimPrefix(line, indent)
|
||||
}
|
||||
return strings.Join(lines, "\n")
|
||||
}
|
||||
|
||||
var looseListItem = regexp.MustCompile(`<li>([^<]+)</li>`)
|
||||
|
||||
func loosifyLists(html string) string {
|
||||
return strings.ReplaceAll(
|
||||
looseListItem.ReplaceAllString(html, "<li>\n<p>$1</p>\n</li>"),
|
||||
"<li></li>", "<li>\n</li>")
|
||||
}
|
||||
|
||||
type codecStringer interface {
|
||||
md.Codec
|
||||
String() string
|
||||
}
|
||||
|
||||
func render(markdown string, codec codecStringer) string {
|
||||
md.Render(markdown, codec)
|
||||
return codec.String()
|
||||
}
|
||||
Loading…
Reference in New Issue
Block a user