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

Change behavior to output link after inner a tag text #1

Merged
merged 5 commits into from
Oct 27, 2022
Merged

Change behavior to output link after inner a tag text #1

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
efdf3ff
Change behavior to output link after inner a tag text
devstein Oct 27, 2022
fcb2e19
Better comment
devstein Oct 27, 2022
d3157e9
Add one more test to clarify nested behavior
devstein Oct 27, 2022
f0a0bb0
Update test again
devstein Oct 27, 2022
0e435a4
Add CI for tests
devstein Oct 27, 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
19 changes: 19 additions & 0 deletions .github/workflows/ci.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,19 @@
name: Run Tests

on:
- push

jobs:

build:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v3

- name: Set up Go
uses: actions/setup-go@v3
with:
go-version: 1.19

- name: Test
run: go test -v .
24 changes: 18 additions & 6 deletions html2text.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -13,7 +13,7 @@ const (
)

var lbr = WIN_LBR
var badTagnamesRE = regexp.MustCompile(`^(head|script|style|a)($|\s+)`)
var badTagnamesRE = regexp.MustCompile(`^(head|script|style)($|\s+)`)
var linkTagRE = regexp.MustCompile(`a.*href=('([^']*?)'|"([^"]*?)")`)
var badLinkHrefRE = regexp.MustCompile(`javascript:`)
var headersRE = regexp.MustCompile(`^(\/)?h[1-6]`)
Expand Down Expand Up @@ -118,6 +118,8 @@ func HTML2Text(html string) string {
inEnt := false
badTagStackDepth := 0 // if == 1 it means we are inside <head>...</head>
shouldOutput := true
// maintain a stack of <a> tag href links and output it after the tag's inner text
hrefs := []string{}
// new line cannot be printed at the beginning or
// for <p> after a new line created by previous <p></p>
canPrintNewline := false
Expand Down Expand Up @@ -199,11 +201,18 @@ func HTML2Text(html string) string {
outBuf.WriteString(lbr + lbr)
}
canPrintNewline = false
} else if badTagnamesRE.MatchString(tagNameLowercase) {
// unwanted block
badTagStackDepth++

} else if tagNameLowercase == "/a" {
// end of link
// links can be empty can happen if the link matches the badLinkHrefRE
if len(hrefs) > 0 {
outBuf.WriteString(" <")
outBuf.WriteString(HTMLEntitiesToText(hrefs[0]))
outBuf.WriteString(">")
hrefs = hrefs[1:]
}
} else if linkTagRE.MatchString(tagNameLowercase) {
// parse link href
// add special handling for a tags
m := linkTagRE.FindStringSubmatch(tag)
if len(m) == 4 {
link := m[2]
Expand All @@ -212,9 +221,12 @@ func HTML2Text(html string) string {
}

if !badLinkHrefRE.MatchString(link) {
outBuf.WriteString(HTMLEntitiesToText(link))
hrefs = append(hrefs, link)
}
}
} else if badTagnamesRE.MatchString(tagNameLowercase) {
// unwanted block
badTagStackDepth++
} else if len(tagNameLowercase) > 0 && tagNameLowercase[0] == '/' &&
badTagnamesRE.MatchString(tagNameLowercase[1:]) {
// end of unwanted block
Expand Down
15 changes: 8 additions & 7 deletions html2text_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,13 +12,14 @@ func TestHTML2Text(t *testing.T) {
Convey("Links", func() {
So(HTML2Text(`<div></div>`), ShouldEqual, "")
So(HTML2Text(`<div>simple text</div>`), ShouldEqual, "simple text")
So(HTML2Text(`click <a href="test">here</a>`), ShouldEqual, "click test")
So(HTML2Text(`click <a class="x" href="test">here</a>`), ShouldEqual, "click test")
So(HTML2Text(`click <a href="ents/&apos;x&apos;">here</a>`), ShouldEqual, "click ents/'x'")
So(HTML2Text(`click <a href="javascript:void(0)">here</a>`), ShouldEqual, "click ")
So(HTML2Text(`click <a href="test"><span>here</span> or here</a>`), ShouldEqual, "click test")
So(HTML2Text(`click <a href="http://bit.ly/2n4wXRs">news</a>`), ShouldEqual, "click http://bit.ly/2n4wXRs")
So(HTML2Text(`<a rel="mw:WikiLink" href="/wiki/yet#English" title="yet">yet</a>, <a rel="mw:WikiLink" href="/wiki/not_yet#English" title="not yet">not yet</a>`), ShouldEqual, "/wiki/yet#English, /wiki/not_yet#English")
So(HTML2Text(`click <a href="test">here</a>`), ShouldEqual, "click here <test>")
So(HTML2Text(`click <a class="x" href="test">here</a>`), ShouldEqual, "click here <test>")
So(HTML2Text(`click <a href="ents/&apos;x&apos;">here</a>`), ShouldEqual, "click here <ents/'x'>")
So(HTML2Text(`click <a href="javascript:void(0)">here</a>`), ShouldEqual, "click here")
So(HTML2Text(`click <a href="test"><span>here</span> or here</a>`), ShouldEqual, "click here or here <test>")
So(HTML2Text(`click <a href="http://bit.ly/2n4wXRs">news</a>`), ShouldEqual, "click news <http://bit.ly/2n4wXRs>")
So(HTML2Text(`<a rel="mw:WikiLink" href="/wiki/yet#English" title="yet">yet</a>, <a rel="mw:WikiLink" href="/wiki/not_yet#English" title="not yet">not yet</a>`), ShouldEqual, "yet </wiki/yet#English>, not yet </wiki/not_yet#English>")
So(HTML2Text(`click <a href="one">here<a href="two"> or</a><span> here</span></a>`), ShouldEqual, "click here or <one> here <two>")
})

Convey("Inlines", func() {
Expand Down