internal/fetch/internal/doc: remove unused code

Change-Id: I50918f4d9984c6d01649cc5674d9e35af360d0d9
Reviewed-on: https://go-review.googlesource.com/c/pkgsite/+/242857
Reviewed-by: Julie Qiu <julie@golang.org>
This commit is contained in:
Jonathan Amsterdam 2020-07-15 17:16:00 -04:00
Родитель 57c5aa6ebf
Коммит 71ca2575a1
4 изменённых файлов: 8 добавлений и 760 удалений

Просмотреть файл

@ -1,513 +0,0 @@
// Copyright 2009 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// Godoc comment extraction and comment -> HTML formatting.
package doc
import (
"bytes"
"io"
"regexp"
"strings"
"text/template" // for HTMLEscape
"unicode"
"unicode/utf8"
)
const (
ldquo = "&ldquo;"
rdquo = "&rdquo;"
ulquo = "“"
urquo = "”"
)
var (
htmlQuoteReplacer = strings.NewReplacer(ulquo, ldquo, urquo, rdquo)
unicodeQuoteReplacer = strings.NewReplacer("``", ulquo, "''", urquo)
)
// Escape comment text for HTML. If nice is set,
// also turn `` into &ldquo; and '' into &rdquo;.
func commentEscape(w io.Writer, text string, nice bool) {
if nice {
// In the first pass, we convert `` and '' into their unicode equivalents.
// This prevents them from being escaped in HTMLEscape.
text = convertQuotes(text)
var buf bytes.Buffer
template.HTMLEscape(&buf, []byte(text))
// Now we convert the unicode quotes to their HTML escaped entities to maintain old behavior.
// We need to use a temp buffer to read the string back and do the conversion,
// otherwise HTMLEscape will escape & to &amp;
htmlQuoteReplacer.WriteString(w, buf.String())
return
}
template.HTMLEscape(w, []byte(text))
}
func convertQuotes(text string) string {
return unicodeQuoteReplacer.Replace(text)
}
const (
// Regexp for Go identifiers
identRx = `[\pL_][\pL_0-9]*`
// Regexp for URLs
// Match parens, and check later for balance - see #5043, #22285
// Match .,:;?! within path, but not at end - see #18139, #16565
// This excludes some rare yet valid urls ending in common punctuation
// in order to allow sentences ending in URLs.
// protocol (required) e.g. http
protoPart = `(https?|ftp|file|gopher|mailto|nntp)`
// host (required) e.g. www.example.com or [::1]:8080
hostPart = `([a-zA-Z0-9_@\-.\[\]:]+)`
// path+query+fragment (optional) e.g. /path/index.html?q=foo#bar
pathPart = `([.,:;?!]*[a-zA-Z0-9$'()*+&#=@~_/\-\[\]%])*`
urlRx = protoPart + `://` + hostPart + pathPart
)
var matchRx = regexp.MustCompile(`(` + urlRx + `)|(` + identRx + `)`)
var (
html_a = []byte(`<a href="`)
html_aq = []byte(`">`)
html_enda = []byte("</a>")
html_i = []byte("<i>")
html_endi = []byte("</i>")
html_p = []byte("<p>\n")
html_endp = []byte("</p>\n")
html_pre = []byte("<pre>")
html_endpre = []byte("</pre>\n")
html_h = []byte(`<h3 id="`)
html_hq = []byte(`">`)
html_endh = []byte("</h3>\n")
)
// Emphasize and escape a line of text for HTML. URLs are converted into links;
// if the URL also appears in the words map, the link is taken from the map (if
// the corresponding map value is the empty string, the URL is not converted
// into a link). Go identifiers that appear in the words map are italicized; if
// the corresponding map value is not the empty string, it is considered a URL
// and the word is converted into a link. If nice is set, the remaining text's
// appearance is improved where it makes sense (e.g., `` is turned into &ldquo;
// and '' into &rdquo;).
func emphasize(w io.Writer, line string, words map[string]string, nice bool) {
for {
m := matchRx.FindStringSubmatchIndex(line)
if m == nil {
break
}
// m >= 6 (two parenthesized sub-regexps in matchRx, 1st one is urlRx)
// write text before match
commentEscape(w, line[0:m[0]], nice)
// adjust match for URLs
match := line[m[0]:m[1]]
if strings.Contains(match, "://") {
m0, m1 := m[0], m[1]
for _, s := range []string{"()", "{}", "[]"} {
open, close := s[:1], s[1:] // E.g., "(" and ")"
// require opening parentheses before closing parentheses (#22285)
if i := strings.Index(match, close); i >= 0 && i < strings.Index(match, open) {
m1 = m0 + i
match = line[m0:m1]
}
// require balanced pairs of parentheses (#5043)
for i := 0; strings.Count(match, open) != strings.Count(match, close) && i < 10; i++ {
m1 = strings.LastIndexAny(line[:m1], s)
match = line[m0:m1]
}
}
if m1 != m[1] {
// redo matching with shortened line for correct indices
m = matchRx.FindStringSubmatchIndex(line[:m[0]+len(match)])
}
}
// analyze match
url := ""
italics := false
if words != nil {
url, italics = words[match]
}
if m[2] >= 0 {
// match against first parenthesized sub-regexp; must be match against urlRx
if !italics {
// no alternative URL in words list, use match instead
url = match
}
italics = false // don't italicize URLs
}
// write match
if len(url) > 0 {
w.Write(html_a)
template.HTMLEscape(w, []byte(url))
w.Write(html_aq)
}
if italics {
w.Write(html_i)
}
commentEscape(w, match, nice)
if italics {
w.Write(html_endi)
}
if len(url) > 0 {
w.Write(html_enda)
}
// advance
line = line[m[1]:]
}
commentEscape(w, line, nice)
}
func indentLen(s string) int {
i := 0
for i < len(s) && (s[i] == ' ' || s[i] == '\t') {
i++
}
return i
}
func isBlank(s string) bool {
return len(s) == 0 || (len(s) == 1 && s[0] == '\n')
}
func commonPrefix(a, b string) string {
i := 0
for i < len(a) && i < len(b) && a[i] == b[i] {
i++
}
return a[0:i]
}
func unindent(block []string) {
if len(block) == 0 {
return
}
// compute maximum common white prefix
prefix := block[0][0:indentLen(block[0])]
for _, line := range block {
if !isBlank(line) {
prefix = commonPrefix(prefix, line[0:indentLen(line)])
}
}
n := len(prefix)
// remove
for i, line := range block {
if !isBlank(line) {
block[i] = line[n:]
}
}
}
// heading returns the trimmed line if it passes as a section heading;
// otherwise it returns the empty string.
func heading(line string) string {
line = strings.TrimSpace(line)
if len(line) == 0 {
return ""
}
// a heading must start with an uppercase letter
r, _ := utf8.DecodeRuneInString(line)
if !unicode.IsLetter(r) || !unicode.IsUpper(r) {
return ""
}
// it must end in a letter or digit:
r, _ = utf8.DecodeLastRuneInString(line)
if !unicode.IsLetter(r) && !unicode.IsDigit(r) {
return ""
}
// exclude lines with illegal characters. we allow "(),"
if strings.ContainsAny(line, ";:!?+*/=[]{}_^°&§~%#@<\">\\") {
return ""
}
// allow "'" for possessive "'s" only
for b := line; ; {
i := strings.IndexRune(b, '\'')
if i < 0 {
break
}
if i+1 >= len(b) || b[i+1] != 's' || (i+2 < len(b) && b[i+2] != ' ') {
return "" // not followed by "s "
}
b = b[i+2:]
}
// allow "." when followed by non-space
for b := line; ; {
i := strings.IndexRune(b, '.')
if i < 0 {
break
}
if i+1 >= len(b) || b[i+1] == ' ' {
return "" // not followed by non-space
}
b = b[i+1:]
}
return line
}
type op int
const (
opPara op = iota
opHead
opPre
)
type block struct {
op op
lines []string
}
var nonAlphaNumRx = regexp.MustCompile(`[^a-zA-Z0-9]`)
func anchorID(line string) string {
// Add a "hdr-" prefix to avoid conflicting with IDs used for package symbols.
return "hdr-" + nonAlphaNumRx.ReplaceAllString(line, "_")
}
// ToHTML converts comment text to formatted HTML.
// The comment was prepared by DocReader,
// so it is known not to have leading, trailing blank lines
// nor to have trailing spaces at the end of lines.
// The comment markers have already been removed.
//
// Each span of unindented non-blank lines is converted into
// a single paragraph. There is one exception to the rule: a span that
// consists of a single line, is followed by another paragraph span,
// begins with a capital letter, and contains no punctuation
// other than parentheses and commas is formatted as a heading.
//
// A span of indented lines is converted into a <pre> block,
// with the common indent prefix removed.
//
// URLs in the comment text are converted into links; if the URL also appears
// in the words map, the link is taken from the map (if the corresponding map
// value is the empty string, the URL is not converted into a link).
//
// A pair of (consecutive) backticks (`) is converted to a unicode left quote (“), and a pair of (consecutive)
// single quotes (') is converted to a unicode right quote (”).
//
// Go identifiers that appear in the words map are italicized; if the corresponding
// map value is not the empty string, it is considered a URL and the word is converted
// into a link.
func ToHTML(w io.Writer, text string, words map[string]string) {
for _, b := range blocks(text) {
switch b.op {
case opPara:
w.Write(html_p)
for _, line := range b.lines {
emphasize(w, line, words, true)
}
w.Write(html_endp)
case opHead:
w.Write(html_h)
id := ""
for _, line := range b.lines {
if id == "" {
id = anchorID(line)
w.Write([]byte(id))
w.Write(html_hq)
}
commentEscape(w, line, true)
}
if id == "" {
w.Write(html_hq)
}
w.Write(html_endh)
case opPre:
w.Write(html_pre)
for _, line := range b.lines {
emphasize(w, line, nil, false)
}
w.Write(html_endpre)
}
}
}
func blocks(text string) []block {
var (
out []block
para []string
lastWasBlank = false
lastWasHeading = false
)
close := func() {
if para != nil {
out = append(out, block{opPara, para})
para = nil
}
}
lines := strings.SplitAfter(text, "\n")
unindent(lines)
for i := 0; i < len(lines); {
line := lines[i]
if isBlank(line) {
// close paragraph
close()
i++
lastWasBlank = true
continue
}
if indentLen(line) > 0 {
// close paragraph
close()
// count indented or blank lines
j := i + 1
for j < len(lines) && (isBlank(lines[j]) || indentLen(lines[j]) > 0) {
j++
}
// but not trailing blank lines
for j > i && isBlank(lines[j-1]) {
j--
}
pre := lines[i:j]
i = j
unindent(pre)
// put those lines in a pre block
out = append(out, block{opPre, pre})
lastWasHeading = false
continue
}
if lastWasBlank && !lastWasHeading && i+2 < len(lines) &&
isBlank(lines[i+1]) && !isBlank(lines[i+2]) && indentLen(lines[i+2]) == 0 {
// current line is non-blank, surrounded by blank lines
// and the next non-blank line is not indented: this
// might be a heading.
if head := heading(line); head != "" {
close()
out = append(out, block{opHead, []string{head}})
i += 2
lastWasHeading = true
continue
}
}
// open paragraph
lastWasBlank = false
lastWasHeading = false
para = append(para, lines[i])
i++
}
close()
return out
}
// ToText prepares comment text for presentation in textual output.
// It wraps paragraphs of text to width or fewer Unicode code points
// and then prefixes each line with the indent. In preformatted sections
// (such as program text), it prefixes each non-blank line with preIndent.
//
// A pair of (consecutive) backticks (`) is converted to a unicode left quote (“), and a pair of (consecutive)
// single quotes (') is converted to a unicode right quote (”).
func ToText(w io.Writer, text string, indent, preIndent string, width int) {
l := lineWrapper{
out: w,
width: width,
indent: indent,
}
for _, b := range blocks(text) {
switch b.op {
case opPara:
// l.write will add leading newline if required
for _, line := range b.lines {
line = convertQuotes(line)
l.write(line)
}
l.flush()
case opHead:
w.Write(nl)
for _, line := range b.lines {
line = convertQuotes(line)
l.write(line + "\n")
}
l.flush()
case opPre:
w.Write(nl)
for _, line := range b.lines {
if isBlank(line) {
w.Write([]byte("\n"))
} else {
w.Write([]byte(preIndent))
w.Write([]byte(line))
}
}
}
}
}
type lineWrapper struct {
out io.Writer
printed bool
width int
indent string
n int
pendSpace int
}
var nl = []byte("\n")
var space = []byte(" ")
var prefix = []byte("// ")
func (l *lineWrapper) write(text string) {
if l.n == 0 && l.printed {
l.out.Write(nl) // blank line before new paragraph
}
l.printed = true
needsPrefix := false
isComment := strings.HasPrefix(text, "//")
for _, f := range strings.Fields(text) {
w := utf8.RuneCountInString(f)
// wrap if line is too long
if l.n > 0 && l.n+l.pendSpace+w > l.width {
l.out.Write(nl)
l.n = 0
l.pendSpace = 0
needsPrefix = isComment
}
if l.n == 0 {
l.out.Write([]byte(l.indent))
}
if needsPrefix {
l.out.Write(prefix)
needsPrefix = false
}
l.out.Write(space[:l.pendSpace])
l.out.Write([]byte(f))
l.n += l.pendSpace + w
l.pendSpace = 1
}
}
func (l *lineWrapper) flush() {
if l.n == 0 {
return
}
l.out.Write(nl)
l.pendSpace = 0
l.n = 0
}

Просмотреть файл

@ -1,236 +0,0 @@
// Copyright 2011 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package doc
import (
"bytes"
"reflect"
"strings"
"testing"
)
var headingTests = []struct {
line string
ok bool
}{
{"Section", true},
{"A typical usage", true},
{"ΔΛΞ is Greek", true},
{"Foo 42", true},
{"", false},
{"section", false},
{"A typical usage:", false},
{"This code:", false},
{"δ is Greek", false},
{"Foo §", false},
{"Fermat's Last Sentence", true},
{"Fermat's", true},
{"'sX", false},
{"Ted 'Too' Bar", false},
{"Use n+m", false},
{"Scanning:", false},
{"N:M", false},
}
func TestIsHeading(t *testing.T) {
for _, tt := range headingTests {
if h := heading(tt.line); (len(h) > 0) != tt.ok {
t.Errorf("isHeading(%q) = %v, want %v", tt.line, h, tt.ok)
}
}
}
var blocksTests = []struct {
in string
out []block
text string
}{
{
in: `Para 1.
Para 1 line 2.
Para 2.
Section
Para 3.
pre
pre1
Para 4.
pre
pre1
pre2
Para 5.
pre
pre1
pre2
Para 6.
pre
pre2
`,
out: []block{
{opPara, []string{"Para 1.\n", "Para 1 line 2.\n"}},
{opPara, []string{"Para 2.\n"}},
{opHead, []string{"Section"}},
{opPara, []string{"Para 3.\n"}},
{opPre, []string{"pre\n", "pre1\n"}},
{opPara, []string{"Para 4.\n"}},
{opPre, []string{"pre\n", "pre1\n", "\n", "pre2\n"}},
{opPara, []string{"Para 5.\n"}},
{opPre, []string{"pre\n", "\n", "\n", "pre1\n", "pre2\n"}},
{opPara, []string{"Para 6.\n"}},
{opPre, []string{"pre\n", "pre2\n"}},
},
text: `. Para 1. Para 1 line 2.
. Para 2.
. Section
. Para 3.
$ pre
$ pre1
. Para 4.
$ pre
$ pre1
$ pre2
. Para 5.
$ pre
$ pre1
$ pre2
. Para 6.
$ pre
$ pre2
`,
},
{
in: "Para.\n\tshould not be ``escaped''",
out: []block{
{opPara, []string{"Para.\n"}},
{opPre, []string{"should not be ``escaped''"}},
},
text: ". Para.\n\n$ should not be ``escaped''",
},
{
in: "// A very long line of 46 char for line wrapping.",
out: []block{
{opPara, []string{"// A very long line of 46 char for line wrapping."}},
},
text: `. // A very long line of 46 char for line
. // wrapping.
`,
},
{
in: `/* A very long line of 46 char for line wrapping.
A very long line of 46 char for line wrapping. */`,
out: []block{
{opPara, []string{"/* A very long line of 46 char for line wrapping.\n", "A very long line of 46 char for line wrapping. */"}},
},
text: `. /* A very long line of 46 char for line
. wrapping. A very long line of 46 char
. for line wrapping. */
`,
},
}
func TestBlocks(t *testing.T) {
for i, tt := range blocksTests {
b := blocks(tt.in)
if !reflect.DeepEqual(b, tt.out) {
t.Errorf("#%d: mismatch\nhave: %v\nwant: %v", i, b, tt.out)
}
}
}
func TestToText(t *testing.T) {
var buf bytes.Buffer
for i, tt := range blocksTests {
ToText(&buf, tt.in, ". ", "$\t", 40)
if have := buf.String(); have != tt.text {
t.Errorf("#%d: mismatch\nhave: %s\nwant: %s\nhave vs want:\n%q\n%q", i, have, tt.text, have, tt.text)
}
buf.Reset()
}
}
var emphasizeTests = []struct {
in, out string
}{
{"", ""},
{"http://[::1]:8080/foo.txt", `<a href="http://[::1]:8080/foo.txt">http://[::1]:8080/foo.txt</a>`},
{"before (https://www.google.com) after", `before (<a href="https://www.google.com">https://www.google.com</a>) after`},
{"before https://www.google.com:30/x/y/z:b::c. After", `before <a href="https://www.google.com:30/x/y/z:b::c">https://www.google.com:30/x/y/z:b::c</a>. After`},
{"http://www.google.com/path/:;!-/?query=%34b#093124", `<a href="http://www.google.com/path/:;!-/?query=%34b#093124">http://www.google.com/path/:;!-/?query=%34b#093124</a>`},
{"http://www.google.com/path/:;!-/?query=%34bar#093124", `<a href="http://www.google.com/path/:;!-/?query=%34bar#093124">http://www.google.com/path/:;!-/?query=%34bar#093124</a>`},
{"http://www.google.com/index.html! After", `<a href="http://www.google.com/index.html">http://www.google.com/index.html</a>! After`},
{"http://www.google.com/", `<a href="http://www.google.com/">http://www.google.com/</a>`},
{"https://www.google.com/", `<a href="https://www.google.com/">https://www.google.com/</a>`},
{"http://www.google.com/path.", `<a href="http://www.google.com/path">http://www.google.com/path</a>.`},
{"http://en.wikipedia.org/wiki/Camellia_(cipher)", `<a href="http://en.wikipedia.org/wiki/Camellia_(cipher)">http://en.wikipedia.org/wiki/Camellia_(cipher)</a>`},
{"(http://www.google.com/)", `(<a href="http://www.google.com/">http://www.google.com/</a>)`},
{"http://gmail.com)", `<a href="http://gmail.com">http://gmail.com</a>)`},
{"((http://gmail.com))", `((<a href="http://gmail.com">http://gmail.com</a>))`},
{"http://gmail.com ((http://gmail.com)) ()", `<a href="http://gmail.com">http://gmail.com</a> ((<a href="http://gmail.com">http://gmail.com</a>)) ()`},
{"Foo bar http://example.com/ quux!", `Foo bar <a href="http://example.com/">http://example.com/</a> quux!`},
{"Hello http://example.com/%2f/ /world.", `Hello <a href="http://example.com/%2f/">http://example.com/%2f/</a> /world.`},
{"Lorem http: ipsum //host/path", "Lorem http: ipsum //host/path"},
{"javascript://is/not/linked", "javascript://is/not/linked"},
{"http://foo", `<a href="http://foo">http://foo</a>`},
{"art by [[https://www.example.com/person/][Person Name]]", `art by [[<a href="https://www.example.com/person/">https://www.example.com/person/</a>][Person Name]]`},
{"please visit (http://golang.org/)", `please visit (<a href="http://golang.org/">http://golang.org/</a>)`},
{"please visit http://golang.org/hello())", `please visit <a href="http://golang.org/hello()">http://golang.org/hello()</a>)`},
{"http://git.qemu.org/?p=qemu.git;a=blob;f=qapi-schema.json;hb=HEAD", `<a href="http://git.qemu.org/?p=qemu.git;a=blob;f=qapi-schema.json;hb=HEAD">http://git.qemu.org/?p=qemu.git;a=blob;f=qapi-schema.json;hb=HEAD</a>`},
{"https://foo.bar/bal/x(])", `<a href="https://foo.bar/bal/x(">https://foo.bar/bal/x(</a>])`}, // inner ] causes (]) to be cut off from URL
{"foo [ http://bar(])", `foo [ <a href="http://bar(">http://bar(</a>])`}, // outer [ causes ]) to be cut off from URL
}
func TestEmphasize(t *testing.T) {
for i, tt := range emphasizeTests {
var buf bytes.Buffer
emphasize(&buf, tt.in, nil, true)
out := buf.String()
if out != tt.out {
t.Errorf("#%d: mismatch\nhave: %v\nwant: %v", i, out, tt.out)
}
}
}
func TestCommentEscape(t *testing.T) {
commentTests := []struct {
in, out string
}{
{"typically invoked as ``go tool asm'',", "typically invoked as " + ldquo + "go tool asm" + rdquo + ","},
{"For more detail, run ``go help test'' and ``go help testflag''", "For more detail, run " + ldquo + "go help test" + rdquo + " and " + ldquo + "go help testflag" + rdquo},
}
for i, tt := range commentTests {
var buf strings.Builder
commentEscape(&buf, tt.in, true)
out := buf.String()
if out != tt.out {
t.Errorf("#%d: mismatch\nhave: %q\nwant: %q", i, out, tt.out)
}
}
}

Просмотреть файл

@ -154,15 +154,6 @@ func Test(t *testing.T) {
test(t, AllMethods) test(t, AllMethods)
} }
func TestAnchorID(t *testing.T) {
const in = "Important Things 2 Know & Stuff"
const want = "hdr-Important_Things_2_Know___Stuff"
got := anchorID(in)
if got != want {
t.Errorf("anchorID(%q) = %q; want %q", in, got, want)
}
}
func TestPackageName(t *testing.T) { func TestPackageName(t *testing.T) {
for _, test := range []struct { for _, test := range []struct {
path string path string

Просмотреть файл

@ -72,10 +72,16 @@ func Synopsis(s string) string {
return "" return ""
} }
} }
s = convertQuotes(s) return unicodeQuoteReplacer.Replace(s)
return s
} }
const (
ulquo = "“"
urquo = "”"
)
var unicodeQuoteReplacer = strings.NewReplacer("``", ulquo, "''", urquo)
var IllegalPrefixes = []string{ var IllegalPrefixes = []string{
"copyright", "copyright",
"all rights", "all rights",