Browse Source

Add HTML2text library. Fix extract formatting

pull/159/head
Richard Lewis 7 years ago
parent
commit
47060231d1
  1. 33
      src/github.com/matrix-org/go-neb/services/wikipedia/wikipedia.go
  2. 6
      vendor/manifest
  3. 22
      vendor/src/github.com/jaytaylor/html2text/LICENSE
  4. 116
      vendor/src/github.com/jaytaylor/html2text/README.md
  5. 304
      vendor/src/github.com/jaytaylor/html2text/html2text.go
  6. 630
      vendor/src/github.com/jaytaylor/html2text/html2text_test.go

33
src/github.com/matrix-org/go-neb/services/wikipedia/wikipedia.go

@ -7,15 +7,18 @@ import (
"io/ioutil"
"net/http"
"net/url"
"strconv"
"strings"
log "github.com/Sirupsen/logrus"
"github.com/jaytaylor/html2text"
"github.com/matrix-org/go-neb/types"
"github.com/matrix-org/gomatrix"
)
// ServiceType of the Wikipedia service
const ServiceType = "wikipedia"
const maxExtractLength = 1024 // Max length of extract string in bytes
var httpClient = &http.Client{}
@ -81,9 +84,25 @@ func (s *Service) cmdWikipediaSearch(client *gomatrix.Client, roomID, userID str
}, nil
}
extractText, err := html2text.FromString(searchResultPage.Extract)
if err != nil {
return gomatrix.TextMessage{
MsgType: "m.notice",
Body: "Failed to convert extract to plain text - " + err.Error(),
}, nil
}
// Truncate the extract text, if necessary
if len(extractText) > maxExtractLength {
extractText = extractText[:maxExtractLength] + "..."
}
// Add a link to the bottom of the extract
extractText += "\n" + "http://en.wikipedia.org/?curid=" + strconv.FormatInt(searchResultPage.PageID, 10)
return gomatrix.TextMessage{
MsgType: "m.notice",
Body: searchResultPage.Extract,
Body: extractText,
}, nil
}
@ -122,12 +141,18 @@ func (s *Service) text2Wikipedia(query string) (*wikipediaPage, error) {
// log.Info(response2String(res))
if err := json.NewDecoder(res.Body).Decode(&searchResults); err != nil {
return nil, fmt.Errorf("ERROR - %s", err.Error())
} else if len(searchResults.Pages) < 1 {
} else if len(searchResults.Query.Pages) < 1 {
return nil, fmt.Errorf("No articles found")
}
// Return only the first search result
return &searchResults.Pages[0], nil
// Return only the first search result with an extract
for _, page := range searchResults.Query.Pages {
if page.Extract != "" {
return &page, nil
}
}
return nil, fmt.Errorf("No articles with extracts found")
}
// response2String returns a string representation of an HTTP response body

6
vendor/manifest

@ -129,6 +129,12 @@
"revision": "d02018f006d98f58512bf3adfc171d88d17626df",
"branch": "master"
},
{
"importpath": "github.com/jaytaylor/html2text",
"repository": "https://github.com/jaytaylor/html2text",
"revision": "24f9b0f63599c6bbecc3b773636b54f8d302db67",
"branch": "master"
},
{
"importpath": "github.com/matrix-org/dugong",
"repository": "https://github.com/matrix-org/dugong",

22
vendor/src/github.com/jaytaylor/html2text/LICENSE

@ -0,0 +1,22 @@
The MIT License (MIT)
Copyright (c) 2015 Jay Taylor
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.

116
vendor/src/github.com/jaytaylor/html2text/README.md

@ -0,0 +1,116 @@
# html2text
[![Documentation](https://godoc.org/github.com/jaytaylor/html2text?status.svg)](https://godoc.org/github.com/jaytaylor/html2text)
[![Build Status](https://travis-ci.org/jaytaylor/html2text.svg?branch=master)](https://travis-ci.org/jaytaylor/html2text)
[![Report Card](https://goreportcard.com/badge/github.com/jaytaylor/html2text)](https://goreportcard.com/report/github.com/jaytaylor/html2text)
### Converts HTML into text
## Introduction
Ensure your emails are readable by all!
Turns HTML into raw text, useful for sending fancy HTML emails with a equivalently nicely formatted TXT document as a fallback (e.g. for people who don't allow HTML emails or have other display issues).
html2text is a simple golang package for rendering HTML into plaintext.
There are still lots of improvements to be had, but FWIW this has worked fine for my [basic] HTML-2-text needs.
It requires go 1.x or newer ;)
## Download the package
```bash
go get github.com/jaytaylor/html2text
```
## Example usage
```go
package main
import (
"fmt"
"github.com/jaytaylor/html2text"
)
func main() {
inputHtml := `
<html>
<head>
<title>My Mega Service</title>
<link rel=\"stylesheet\" href=\"main.css\">
<style type=\"text/css\">body { color: #fff; }</style>
</head>
<body>
<div class="logo">
<a href="http://mymegaservice.com/"><img src="/logo-image.jpg" alt="Mega Service"/></a>
</div>
<h1>Welcome to your new account on my service!</h1>
<p>
Here is some more information:
<ul>
<li>Link 1: <a href="https://example.com">Example.com</a></li>
<li>Link 2: <a href="https://example2.com">Example2.com</a></li>
<li>Something else</li>
</ul>
</p>
</body>
</html>
`
text, err := html2text.FromString(inputHtml)
if err != nil {
panic(err)
}
fmt.Println(text)
}
```
Output:
```
Mega Service ( http://mymegaservice.com/ )
******************************************
Welcome to your new account on my service!
******************************************
Here is some more information:
* Link 1: Example.com ( https://example.com )
* Link 2: Example2.com ( https://example2.com )
* Something else
```
## Unit-tests
Running the unit-tests is straightforward and standard:
```bash
go test
```
# License
Permissive MIT license.
## Contact
You are more than welcome to open issues and send pull requests if you find a bug or want a new feature.
If you appreciate this library please feel free to drop me a line and tell me! It's always nice to hear from people who have benefitted from my work.
Email: jay at (my github username).com
Twitter: [@jtaylor](https://twitter.com/jtaylor)

304
vendor/src/github.com/jaytaylor/html2text/html2text.go

@ -0,0 +1,304 @@
package html2text
import (
"bytes"
"io"
"regexp"
"strings"
"unicode"
"golang.org/x/net/html"
"golang.org/x/net/html/atom"
)
var (
spacingRe = regexp.MustCompile(`[ \r\n\t]+`)
newlineRe = regexp.MustCompile(`\n\n+`)
)
type textifyTraverseCtx struct {
Buf bytes.Buffer
prefix string
blockquoteLevel int
lineLength int
endsWithSpace bool
endsWithNewline bool
justClosedDiv bool
}
func (ctx *textifyTraverseCtx) traverse(node *html.Node) error {
switch node.Type {
default:
return ctx.traverseChildren(node)
case html.TextNode:
data := strings.Trim(spacingRe.ReplaceAllString(node.Data, " "), " ")
return ctx.emit(data)
case html.ElementNode:
ctx.justClosedDiv = false
switch node.DataAtom {
case atom.Br:
return ctx.emit("\n")
case atom.H1, atom.H2, atom.H3:
subCtx := textifyTraverseCtx{}
if err := subCtx.traverseChildren(node); err != nil {
return err
}
str := subCtx.Buf.String()
dividerLen := 0
for _, line := range strings.Split(str, "\n") {
if lineLen := len([]rune(line)); lineLen-1 > dividerLen {
dividerLen = lineLen - 1
}
}
divider := ""
if node.DataAtom == atom.H1 {
divider = strings.Repeat("*", dividerLen)
} else {
divider = strings.Repeat("-", dividerLen)
}
if node.DataAtom == atom.H3 {
return ctx.emit("\n\n" + str + "\n" + divider + "\n\n")
}
return ctx.emit("\n\n" + divider + "\n" + str + "\n" + divider + "\n\n")
case atom.Blockquote:
ctx.blockquoteLevel++
ctx.prefix = strings.Repeat(">", ctx.blockquoteLevel) + " "
if err := ctx.emit("\n"); err != nil {
return err
}
if ctx.blockquoteLevel == 1 {
if err := ctx.emit("\n"); err != nil {
return err
}
}
if err := ctx.traverseChildren(node); err != nil {
return err
}
ctx.blockquoteLevel--
ctx.prefix = strings.Repeat(">", ctx.blockquoteLevel)
if ctx.blockquoteLevel > 0 {
ctx.prefix += " "
}
return ctx.emit("\n\n")
case atom.Div:
if ctx.lineLength > 0 {
if err := ctx.emit("\n"); err != nil {
return err
}
}
if err := ctx.traverseChildren(node); err != nil {
return err
}
var err error
if ctx.justClosedDiv == false {
err = ctx.emit("\n")
}
ctx.justClosedDiv = true
return err
case atom.Li:
if err := ctx.emit("* "); err != nil {
return err
}
if err := ctx.traverseChildren(node); err != nil {
return err
}
return ctx.emit("\n")
case atom.B, atom.Strong:
subCtx := textifyTraverseCtx{}
subCtx.endsWithSpace = true
if err := subCtx.traverseChildren(node); err != nil {
return err
}
str := subCtx.Buf.String()
return ctx.emit("*" + str + "*")
case atom.A:
// If image is the only child, take its alt text as the link text
if img := node.FirstChild; img != nil && node.LastChild == img && img.DataAtom == atom.Img {
if altText := getAttrVal(img, "alt"); altText != "" {
ctx.emit(altText)
}
} else if err := ctx.traverseChildren(node); err != nil {
return err
}
hrefLink := ""
if attrVal := getAttrVal(node, "href"); attrVal != "" {
attrVal = ctx.normalizeHrefLink(attrVal)
if attrVal != "" {
hrefLink = "( " + attrVal + " )"
}
}
return ctx.emit(hrefLink)
case atom.P, atom.Ul, atom.Table:
if err := ctx.emit("\n\n"); err != nil {
return err
}
if err := ctx.traverseChildren(node); err != nil {
return err
}
return ctx.emit("\n\n")
case atom.Tr:
if err := ctx.traverseChildren(node); err != nil {
return err
}
return ctx.emit("\n")
case atom.Style, atom.Script, atom.Head:
// Ignore the subtree
return nil
default:
return ctx.traverseChildren(node)
}
}
}
func (ctx *textifyTraverseCtx) traverseChildren(node *html.Node) error {
for c := node.FirstChild; c != nil; c = c.NextSibling {
if err := ctx.traverse(c); err != nil {
return err
}
}
return nil
}
func (ctx *textifyTraverseCtx) emit(data string) error {
if len(data) == 0 {
return nil
}
lines := ctx.breakLongLines(data)
var err error
for _, line := range lines {
runes := []rune(line)
startsWithSpace := unicode.IsSpace(runes[0])
if !startsWithSpace && !ctx.endsWithSpace {
ctx.Buf.WriteByte(' ')
ctx.lineLength++
}
ctx.endsWithSpace = unicode.IsSpace(runes[len(runes)-1])
for _, c := range line {
_, err = ctx.Buf.WriteString(string(c))
if err != nil {
return err
}
ctx.lineLength++
if c == '\n' {
ctx.lineLength = 0
if ctx.prefix != "" {
_, err = ctx.Buf.WriteString(ctx.prefix)
if err != nil {
return err
}
}
}
}
}
return nil
}
func (ctx *textifyTraverseCtx) breakLongLines(data string) []string {
// only break lines when we are in blockquotes
if ctx.blockquoteLevel == 0 {
return []string{data}
}
var ret []string
runes := []rune(data)
l := len(runes)
existing := ctx.lineLength
if existing >= 74 {
ret = append(ret, "\n")
existing = 0
}
for l+existing > 74 {
i := 74 - existing
for i >= 0 && !unicode.IsSpace(runes[i]) {
i--
}
if i == -1 {
// no spaces, so go the other way
i = 74 - existing
for i < l && !unicode.IsSpace(runes[i]) {
i++
}
}
ret = append(ret, string(runes[:i])+"\n")
for i < l && unicode.IsSpace(runes[i]) {
i++
}
runes = runes[i:]
l = len(runes)
existing = 0
}
if len(runes) > 0 {
ret = append(ret, string(runes))
}
return ret
}
func (ctx *textifyTraverseCtx) normalizeHrefLink(link string) string {
link = strings.TrimSpace(link)
link = strings.TrimPrefix(link, "mailto:")
return link
}
func getAttrVal(node *html.Node, attrName string) string {
for _, attr := range node.Attr {
if attr.Key == attrName {
return attr.Val
}
}
return ""
}
func FromHtmlNode(doc *html.Node) (string, error) {
ctx := textifyTraverseCtx{
Buf: bytes.Buffer{},
}
if err := ctx.traverse(doc); err != nil {
return "", err
}
text := strings.TrimSpace(newlineRe.ReplaceAllString(
strings.Replace(ctx.Buf.String(), "\n ", "\n", -1), "\n\n"))
return text, nil
}
func FromReader(reader io.Reader) (string, error) {
doc, err := html.Parse(reader)
if err != nil {
return "", err
}
return FromHtmlNode(doc)
}
func FromString(input string) (string, error) {
text, err := FromReader(strings.NewReader(input))
if err != nil {
return "", err
}
return text, nil
}

630
vendor/src/github.com/jaytaylor/html2text/html2text_test.go

@ -0,0 +1,630 @@
package html2text
import (
"fmt"
"regexp"
"testing"
)
func TestStrippingWhitespace(t *testing.T) {
testCases := []struct {
input string
output string
}{
{
"test text",
"test text",
},
{
" \ttext\ntext\n",
"text text",
},
{
" \na \n\t \n \n a \t",
"a a",
},
{
"test text",
"test text",
},
{
"test&nbsp;&nbsp;&nbsp; text&nbsp;",
"test    text",
},
}
for _, testCase := range testCases {
assertString(t, testCase.input, testCase.output)
}
}
func TestParagraphsAndBreaks(t *testing.T) {
testCases := []struct {
input string
output string
}{
{
"Test text",
"Test text",
},
{
"Test text<br>",
"Test text",
},
{
"Test text<br>Test",
"Test text\nTest",
},
{
"<p>Test text</p>",
"Test text",
},
{
"<p>Test text</p><p>Test text</p>",
"Test text\n\nTest text",
},
{
"\n<p>Test text</p>\n\n\n\t<p>Test text</p>\n",
"Test text\n\nTest text",
},
{
"\n<p>Test text<br/>Test text</p>\n",
"Test text\nTest text",
},
{
"\n<p>Test text<br> \tTest text<br></p>\n",
"Test text\nTest text",
},
{
"Test text<br><BR />Test text",
"Test text\n\nTest text",
},
}
for _, testCase := range testCases {
assertString(t, testCase.input, testCase.output)
}
}
func TestTables(t *testing.T) {
testCases := []struct {
input string
output string
}{
{
"<table><tr><td></td><td></td></tr></table>",
"",
},
{
"<table><tr><td>cell1</td><td>cell2</td></tr></table>",
"cell1 cell2",
},
{
"<table><tr><td>row1</td></tr><tr><td>row2</td></tr></table>",
"row1\nrow2",
},
{
`<table>
<tr><td>cell1-1</td><td>cell1-2</td></tr>
<tr><td>cell2-1</td><td>cell2-2</td></tr>
</table>`,
"cell1-1 cell1-2\ncell2-1 cell2-2",
},
{
"_<table><tr><td>cell</td></tr></table>_",
"_\n\ncell\n\n_",
},
}
for _, testCase := range testCases {
assertString(t, testCase.input, testCase.output)
}
}
func TestStrippingLists(t *testing.T) {
testCases := []struct {
input string
output string
}{
{
"<ul></ul>",
"",
},
{
"<ul><li>item</li></ul>_",
"* item\n\n_",
},
{
"<li class='123'>item 1</li> <li>item 2</li>\n_",
"* item 1\n* item 2\n_",
},
{
"<li>item 1</li> \t\n <li>item 2</li> <li> item 3</li>\n_",
"* item 1\n* item 2\n* item 3\n_",
},
}
for _, testCase := range testCases {
assertString(t, testCase.input, testCase.output)
}
}
func TestLinks(t *testing.T) {
testCases := []struct {
input string
output string
}{
{
`<a></a>`,
``,
},
{
`<a href=""></a>`,
``,
},
{
`<a href="http://example.com/"></a>`,
`( http://example.com/ )`,
},
{
`<a href="">Link</a>`,
`Link`,
},
{
`<a href="http://example.com/">Link</a>`,
`Link ( http://example.com/ )`,
},
{
`<a href="http://example.com/"><span class="a">Link</span></a>`,
`Link ( http://example.com/ )`,
},
{
"<a href='http://example.com/'>\n\t<span class='a'>Link</span>\n\t</a>",
`Link ( http://example.com/ )`,
},
{
"<a href='mailto:contact@example.org'>Contact Us</a>",
`Contact Us ( contact@example.org )`,
},
{
"<a href=\"http://example.com:80/~user?aaa=bb&amp;c=d,e,f#foo\">Link</a>",
`Link ( http://example.com:80/~user?aaa=bb&c=d,e,f#foo )`,
},
{
"<a title='title' href=\"http://example.com/\">Link</a>",
`Link ( http://example.com/ )`,
},
{
"<a href=\" http://example.com/ \"> Link </a>",
`Link ( http://example.com/ )`,
},
{
"<a href=\"http://example.com/a/\">Link A</a> <a href=\"http://example.com/b/\">Link B</a>",
`Link A ( http://example.com/a/ ) Link B ( http://example.com/b/ )`,
},
{
"<a href=\"%%LINK%%\">Link</a>",
`Link ( %%LINK%% )`,
},
{
"<a href=\"[LINK]\">Link</a>",
`Link ( [LINK] )`,
},
{
"<a href=\"{LINK}\">Link</a>",
`Link ( {LINK} )`,
},
{
"<a href=\"[[!unsubscribe]]\">Link</a>",
`Link ( [[!unsubscribe]] )`,
},
{
"<p>This is <a href=\"http://www.google.com\" >link1</a> and <a href=\"http://www.google.com\" >link2 </a> is next.</p>",
`This is link1 ( http://www.google.com ) and link2 ( http://www.google.com ) is next.`,
},
}
for _, testCase := range testCases {
assertString(t, testCase.input, testCase.output)
}
}
func TestImageAltTags(t *testing.T) {
testCases := []struct {
input string
output string
}{
{
`<img />`,
``,
},
{
`<img src="http://example.ru/hello.jpg" />`,
``,
},
{
`<img alt="Example"/>`,
``,
},
{
`<img src="http://example.ru/hello.jpg" alt="Example"/>`,
``,
},
// Images do matter if they are in a link
{
`<a href="http://example.com/"><img src="http://example.ru/hello.jpg" alt="Example"/></a>`,
`Example ( http://example.com/ )`,
},
{
`<a href="http://example.com/"><img src="http://example.ru/hello.jpg" alt="Example"></a>`,
`Example ( http://example.com/ )`,
},
{
`<a href='http://example.com/'><img src='http://example.ru/hello.jpg' alt='Example'/></a>`,
`Example ( http://example.com/ )`,
},
{
`<a href='http://example.com/'><img src='http://example.ru/hello.jpg' alt='Example'></a>`,
`Example ( http://example.com/ )`,
},
}
for _, testCase := range testCases {
assertString(t, testCase.input, testCase.output)
}
}
func TestHeadings(t *testing.T) {
testCases := []struct {
input string
output string
}{
{
"<h1>Test</h1>",
"****\nTest\n****",
},
{
"\t<h1>\nTest</h1> ",
"****\nTest\n****",
},
{
"\t<h1>\nTest line 1<br>Test 2</h1> ",
"***********\nTest line 1\nTest 2\n***********",
},
{
"<h1>Test</h1> <h1>Test</h1>",
"****\nTest\n****\n\n****\nTest\n****",
},
{
"<h2>Test</h2>",
"----\nTest\n----",
},
{
"<h1><a href='http://example.com/'>Test</a></h1>",
"****************************\nTest ( http://example.com/ )\n****************************",
},
{
"<h3> <span class='a'>Test </span></h3>",
"Test\n----",
},
}
for _, testCase := range testCases {
assertString(t, testCase.input, testCase.output)
}
}
func TestBold(t *testing.T) {
testCases := []struct {
input string
output string
}{
{
"<b>Test</b>",
"*Test*",
},
{
"\t<b>Test</b> ",
"*Test*",
},
{
"\t<b>Test line 1<br>Test 2</b> ",
"*Test line 1\nTest 2*",
},
{
"<b>Test</b> <b>Test</b>",
"*Test* *Test*",
},
}
for _, testCase := range testCases {
assertString(t, testCase.input, testCase.output)
}
}
func TestDiv(t *testing.T) {
testCases := []struct {
input string
output string
}{
{
"<div>Test</div>",
"Test",
},
{
"\t<div>Test</div> ",
"Test",
},
{
"<div>Test line 1<div>Test 2</div></div>",
"Test line 1\nTest 2",
},
{
"Test 1<div>Test 2</div> <div>Test 3</div>Test 4",
"Test 1\nTest 2\nTest 3\nTest 4",
},
}
for _, testCase := range testCases {
assertString(t, testCase.input, testCase.output)
}
}
func TestBlockquotes(t *testing.T) {
testCases := []struct {
input string
output string
}{
{
"<div>level 0<blockquote>level 1<br><blockquote>level 2</blockquote>level 1</blockquote><div>level 0</div></div>",
"level 0\n> \n> level 1\n> \n>> level 2\n> \n> level 1\n\nlevel 0",
},
{
"<blockquote>Test</blockquote>Test",
"> \n> Test\n\nTest",
},
{
"\t<blockquote> \nTest<br></blockquote> ",
"> \n> Test\n>",
},
{
"\t<blockquote> \nTest line 1<br>Test 2</blockquote> ",
"> \n> Test line 1\n> Test 2",
},
{
"<blockquote>Test</blockquote> <blockquote>Test</blockquote> Other Test",
"> \n> Test\n\n> \n> Test\n\nOther Test",
},
{
"<blockquote>Lorem ipsum Commodo id consectetur pariatur ea occaecat minim aliqua ad sit consequat quis ex commodo Duis incididunt eu mollit consectetur fugiat voluptate dolore in pariatur in commodo occaecat Ut occaecat velit esse labore aute quis commodo non sit dolore officia Excepteur cillum amet cupidatat culpa velit labore ullamco dolore mollit elit in aliqua dolor irure do</blockquote>",
"> \n> Lorem ipsum Commodo id consectetur pariatur ea occaecat minim aliqua ad\n> sit consequat quis ex commodo Duis incididunt eu mollit consectetur fugiat\n> voluptate dolore in pariatur in commodo occaecat Ut occaecat velit esse\n> labore aute quis commodo non sit dolore officia Excepteur cillum amet\n> cupidatat culpa velit labore ullamco dolore mollit elit in aliqua dolor\n> irure do",
},
{
"<blockquote>Lorem<b>ipsum</b><b>Commodo</b><b>id</b><b>consectetur</b><b>pariatur</b><b>ea</b><b>occaecat</b><b>minim</b><b>aliqua</b><b>ad</b><b>sit</b><b>consequat</b><b>quis</b><b>ex</b><b>commodo</b><b>Duis</b><b>incididunt</b><b>eu</b><b>mollit</b><b>consectetur</b><b>fugiat</b><b>voluptate</b><b>dolore</b><b>in</b><b>pariatur</b><b>in</b><b>commodo</b><b>occaecat</b><b>Ut</b><b>occaecat</b><b>velit</b><b>esse</b><b>labore</b><b>aute</b><b>quis</b><b>commodo</b><b>non</b><b>sit</b><b>dolore</b><b>officia</b><b>Excepteur</b><b>cillum</b><b>amet</b><b>cupidatat</b><b>culpa</b><b>velit</b><b>labore</b><b>ullamco</b><b>dolore</b><b>mollit</b><b>elit</b><b>in</b><b>aliqua</b><b>dolor</b><b>irure</b><b>do</b></blockquote>",
"> \n> Lorem *ipsum* *Commodo* *id* *consectetur* *pariatur* *ea* *occaecat* *minim*\n> *aliqua* *ad* *sit* *consequat* *quis* *ex* *commodo* *Duis* *incididunt* *eu*\n> *mollit* *consectetur* *fugiat* *voluptate* *dolore* *in* *pariatur* *in* *commodo*\n> *occaecat* *Ut* *occaecat* *velit* *esse* *labore* *aute* *quis* *commodo*\n> *non* *sit* *dolore* *officia* *Excepteur* *cillum* *amet* *cupidatat* *culpa*\n> *velit* *labore* *ullamco* *dolore* *mollit* *elit* *in* *aliqua* *dolor* *irure*\n> *do*",
},
}
for _, testCase := range testCases {
assertString(t, testCase.input, testCase.output)
}
}
func TestIgnoreStylesScriptsHead(t *testing.T) {
testCases := []struct {
input string
output string
}{
{
"<style>Test</style>",
"",
},
{
"<style type=\"text/css\">body { color: #fff; }</style>",
"",
},
{
"<link rel=\"stylesheet\" href=\"main.css\">",
"",
},
{
"<script>Test</script>",
"",
},
{
"<script src=\"main.js\"></script>",
"",
},
{
"<script type=\"text/javascript\" src=\"main.js\"></script>",
"",
},
{
"<script type=\"text/javascript\">Test</script>",
"",
},
{
"<script type=\"text/ng-template\" id=\"template.html\"><a href=\"http://google.com\">Google</a></script>",
"",
},
{
"<script type=\"bla-bla-bla\" id=\"template.html\">Test</script>",
"",
},
{
`<html><head><title>Title</title></head><body></body></html>`,
"",
},
}
for _, testCase := range testCases {
assertString(t, testCase.input, testCase.output)
}
}
func TestText(t *testing.T) {
testCases := []struct {
input string
expr string
}{
{
`<li>
<a href="/new" data-ga-click="Header, create new repository, icon:repo"><span class="octicon octicon-repo"></span> New repository</a>
</li>`,
`\* New repository \( /new \)`,
},
{
`hi
<br>
hello <a href="https://google.com">google</a>
<br><br>
test<p>List:</p>
<ul>
<li><a href="foo">Foo</a></li>
<li><a href="http://www.microshwhat.com/bar/soapy">Barsoap</a></li>
<li>Baz</li>
</ul>
`,
`hi
hello google \( https://google.com \)
test
List:
\* Foo \( foo \)
\* Barsoap \( http://www.microshwhat.com/bar/soapy \)
\* Baz`,
},
// Malformed input html.
{
`hi
hello <a href="https://google.com">google</a>
test<p>List:</p>
<ul>
<li><a href="foo">Foo</a>
<li><a href="/
bar/baz">Bar</a>
<li>Baz</li>
</ul>
`,
`hi hello google \( https://google.com \) test
List:
\* Foo \( foo \)
\* Bar \( /\n[ \t]+bar/baz \)
\* Baz`,
},
}
for _, testCase := range testCases {
assertRegexp(t, testCase.input, testCase.expr)
}
}
type StringMatcher interface {
MatchString(string) bool
String() string
}
type RegexpStringMatcher string
func (m RegexpStringMatcher) MatchString(str string) bool {
return regexp.MustCompile(string(m)).MatchString(str)
}
func (m RegexpStringMatcher) String() string {
return string(m)
}
type ExactStringMatcher string
func (m ExactStringMatcher) MatchString(str string) bool {
return string(m) == str
}
func (m ExactStringMatcher) String() string {
return string(m)
}
func assertRegexp(t *testing.T, input string, outputRE string) {
assertPlaintext(t, input, RegexpStringMatcher(outputRE))
}
func assertString(t *testing.T, input string, output string) {
assertPlaintext(t, input, ExactStringMatcher(output))
}
func assertPlaintext(t *testing.T, input string, matcher StringMatcher) {
text, err := FromString(input)
if err != nil {
t.Error(err)
}
if !matcher.MatchString(text) {
t.Errorf("Input did not match expression\n"+
"Input:\n>>>>\n%s\n<<<<\n\n"+
"Output:\n>>>>\n%s\n<<<<\n\n"+
"Expected output:\n>>>>\n%s\n<<<<\n\n",
input, text, matcher.String())
} else {
t.Logf("input:\n\n%s\n\n\n\noutput:\n\n%s\n", input, text)
}
}
func Example() {
inputHtml := `
<html>
<head>
<title>My Mega Service</title>
<link rel=\"stylesheet\" href=\"main.css\">
<style type=\"text/css\">body { color: #fff; }</style>
</head>
<body>
<div class="logo">
<a href="http://mymegaservice.com/"><img src="/logo-image.jpg" alt="Mega Service"/></a>
</div>
<h1>Welcome to your new account on my service!</h1>
<p>
Here is some more information:
<ul>
<li>Link 1: <a href="https://example.com">Example.com</a></li>
<li>Link 2: <a href="https://example2.com">Example2.com</a></li>
<li>Something else</li>
</ul>
</p>
</body>
</html>
`
text, err := FromString(inputHtml)
if err != nil {
panic(err)
}
fmt.Println(text)
// Output:
// Mega Service ( http://mymegaservice.com/ )
//
// ******************************************
// Welcome to your new account on my service!
// ******************************************
//
// Here is some more information:
//
// * Link 1: Example.com ( https://example.com )
// * Link 2: Example2.com ( https://example2.com )
// * Something else
}
Loading…
Cancel
Save