phoenix/hugo
1
0
Fork 0
mirror of https://github.com/gohugoio/hugo.git synced 2024-07-04 16:21:57 +00:00
Code Issues Packages Projects Releases Wiki Activity

parser/pageparser: Use []byte in page lexer

Browse source 
See #5324
This commit is contained in:
Bjørn Erik Pedersen 2018-10-18 09:04:48 +02:00
parent 2fdc4a24d5
commit 27f5a906a2
No known key found for this signature in database
GPG key ID: 330E6E2BD4859D8F
4 changed files with 115 additions and 103 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

4
parser/pageparser/item.go
View file

@ -18,7 +18,7 @@ import "fmt"
type Item struct { type Item struct {
typ itemType typ itemType
pos pos pos pos
Val string Val []byte
} }
func (i Item) IsText() bool { func (i Item) IsText() bool {
@ -70,7 +70,7 @@ func (i Item) String() string {
case i.typ == tEOF: case i.typ == tEOF:
return "EOF" return "EOF"
case i.typ == tError: case i.typ == tError:
return i.Val return string(i.Val)
case i.typ > tKeywordMarker: case i.typ > tKeywordMarker:
return fmt.Sprintf("<%s>", i.Val) return fmt.Sprintf("<%s>", i.Val)
case len(i.Val) > 50: case len(i.Val) > 50:

83
parser/pageparser/pagelexer.go
View file

@ -18,8 +18,8 @@
package pageparser package pageparser
import ( import (
"bytes"
"fmt" "fmt"
"strings"
"unicode" "unicode"
"unicode/utf8" "unicode/utf8"
) )
@ -44,7 +44,7 @@ type lexerShortcodeState struct {
} }
type pageLexer struct { type pageLexer struct {
input string input []byte
stateStart stateFunc stateStart stateFunc
state stateFunc state stateFunc
pos pos // input position pos pos // input position
@ -65,14 +65,16 @@ func Parse(s string) *Tokens {
} }
func ParseFrom(s string, from int) *Tokens { func ParseFrom(s string, from int) *Tokens {
lexer := newPageLexer(s, pos(from), lexMainSection) // TODO(bep) 2errors input := []byte(s)
lexer := newPageLexer(input, pos(from), lexMainSection) // TODO(bep) 2errors
lexer.run() lexer.run()
return &Tokens{lexer: lexer} return &Tokens{lexer: lexer}
} }
// note: the input position here is normally 0 (start), but // note: the input position here is normally 0 (start), but
// can be set if position of first shortcode is known // can be set if position of first shortcode is known
func newPageLexer(input string, inputPosition pos, stateStart stateFunc) *pageLexer { // TODO(bep) 2errors byte
func newPageLexer(input []byte, inputPosition pos, stateStart stateFunc) *pageLexer {
lexer := &pageLexer{ lexer := &pageLexer{
input: input, input: input,
pos: inputPosition, pos: inputPosition,
@ -97,19 +99,22 @@ func (l *pageLexer) run() *pageLexer {
} }
// Shortcode syntax // Shortcode syntax
const ( var (
leftDelimScNoMarkup = "{{<" leftDelimScNoMarkup = []byte("{{<")
rightDelimScNoMarkup = ">}}" rightDelimScNoMarkup = []byte(">}}")
leftDelimScWithMarkup = "{{%" leftDelimScWithMarkup = []byte("{{%")
rightDelimScWithMarkup = "%}}" rightDelimScWithMarkup = []byte("%}}")
leftComment = "/*" // comments in this context us used to to mark shortcodes as "not really a shortcode" leftComment = []byte("/*") // comments in this context us used to to mark shortcodes as "not really a shortcode"
rightComment = "*/" rightComment = []byte("*/")
) )
// Page syntax // Page syntax
const ( var (
summaryDivider = "<!--more-->" summaryDivider = []byte("<!--more-->")
summaryDividerOrg = "# more" summaryDividerOrg = []byte("# more")
delimTOML = []byte("+++")
delimYAML = []byte("---")
delimOrg = []byte("#+")
) )
func (l *pageLexer) next() rune { func (l *pageLexer) next() rune {
@ -118,9 +123,7 @@ func (l *pageLexer) next() rune {
return eof return eof
} }
// looks expensive, but should produce the same iteration sequence as the string range loop runeValue, runeWidth := utf8.DecodeRune(l.input[l.pos:])
// see: http://blog.golang.org/strings
runeValue, runeWidth := utf8.DecodeRuneInString(l.input[l.pos:])
l.width = pos(runeWidth) l.width = pos(runeWidth)
l.pos += l.width l.pos += l.width
return runeValue return runeValue
@ -146,7 +149,7 @@ func (l *pageLexer) emit(t itemType) {
// special case, do not send '\\' back to client // special case, do not send '\\' back to client
func (l *pageLexer) ignoreEscapesAndEmit(t itemType) { func (l *pageLexer) ignoreEscapesAndEmit(t itemType) {
val := strings.Map(func(r rune) rune { val := bytes.Map(func(r rune) rune {
if r == '\\' { if r == '\\' {
return -1 return -1
} }
@ -157,7 +160,7 @@ func (l *pageLexer) ignoreEscapesAndEmit(t itemType) {
} }
// gets the current value (for debugging and error handling) // gets the current value (for debugging and error handling)
func (l *pageLexer) current() string { func (l *pageLexer) current() []byte {
return l.input[l.start:l.pos] return l.input[l.start:l.pos]
} }
@ -166,14 +169,16 @@ func (l *pageLexer) ignore() {
l.start = l.pos l.start = l.pos
} }
var lf = []byte("\n")
// nice to have in error logs // nice to have in error logs
func (l *pageLexer) lineNum() int { func (l *pageLexer) lineNum() int {
return strings.Count(l.input[:l.lastPos], "\n") + 1 return bytes.Count(l.input[:l.lastPos], lf) + 1
} }
// nil terminates the parser // nil terminates the parser
func (l *pageLexer) errorf(format string, args ...interface{}) stateFunc { func (l *pageLexer) errorf(format string, args ...interface{}) stateFunc {
l.items = append(l.items, Item{tError, l.start, fmt.Sprintf(format, args...)}) l.items = append(l.items, Item{tError, l.start, []byte(fmt.Sprintf(format, args...))})
return nil return nil
} }
@ -203,7 +208,7 @@ func lexMainSection(l *pageLexer) stateFunc {
if l.pos > l.start { if l.pos > l.start {
l.emit(tText) l.emit(tText)
} }
if strings.HasPrefix(l.input[l.pos:], leftDelimScWithMarkup) { if bytes.HasPrefix(l.input[l.pos:], leftDelimScWithMarkup) {
l.currLeftDelimItem = tLeftDelimScWithMarkup l.currLeftDelimItem = tLeftDelimScWithMarkup
l.currRightDelimItem = tRightDelimScWithMarkup l.currRightDelimItem = tRightDelimScWithMarkup
} else { } else {
@ -214,14 +219,14 @@ func lexMainSection(l *pageLexer) stateFunc {
} }
if l.contentSections <= 1 { if l.contentSections <= 1 {
if strings.HasPrefix(l.input[l.pos:], summaryDivider) { if bytes.HasPrefix(l.input[l.pos:], summaryDivider) {
if l.pos > l.start { if l.pos > l.start {
l.emit(tText) l.emit(tText)
} }
l.contentSections++ l.contentSections++
l.pos += pos(len(summaryDivider)) l.pos += pos(len(summaryDivider))
l.emit(tSummaryDivider) l.emit(tSummaryDivider)
} else if strings.HasPrefix(l.input[l.pos:], summaryDividerOrg) { } else if bytes.HasPrefix(l.input[l.pos:], summaryDividerOrg) {
if l.pos > l.start { if l.pos > l.start {
l.emit(tText) l.emit(tText)
} }
@ -243,7 +248,7 @@ func lexMainSection(l *pageLexer) stateFunc {
} }
func (l *pageLexer) isShortCodeStart() bool { func (l *pageLexer) isShortCodeStart() bool {
return strings.HasPrefix(l.input[l.pos:], leftDelimScWithMarkup) || strings.HasPrefix(l.input[l.pos:], leftDelimScNoMarkup) return bytes.HasPrefix(l.input[l.pos:], leftDelimScWithMarkup) || bytes.HasPrefix(l.input[l.pos:], leftDelimScNoMarkup)
} }
func lexIntroSection(l *pageLexer) stateFunc { func lexIntroSection(l *pageLexer) stateFunc {
@ -256,9 +261,9 @@ LOOP:
switch { switch {
case r == '+': case r == '+':
return l.lexFrontMatterSection(tFrontMatterTOML, r, "TOML", "+++") return l.lexFrontMatterSection(tFrontMatterTOML, r, "TOML", delimTOML)
case r == '-': case r == '-':
return l.lexFrontMatterSection(tFrontMatterYAML, r, "YAML", "---") return l.lexFrontMatterSection(tFrontMatterYAML, r, "YAML", delimYAML)
case r == '{': case r == '{':
return lexFrontMatterJSON return lexFrontMatterJSON
case r == '#': case r == '#':
@ -342,11 +347,9 @@ func lexFrontMatterOrgMode(l *pageLexer) stateFunc {
#+DESCRIPTION: Just another golang parser for org content! #+DESCRIPTION: Just another golang parser for org content!
*/ */
const prefix = "#+"
l.backup() l.backup()
if !strings.HasPrefix(l.input[l.pos:], prefix) { if !bytes.HasPrefix(l.input[l.pos:], delimOrg) {
// TODO(bep) consider error // TODO(bep) consider error
return lexMainSection return lexMainSection
} }
@ -359,7 +362,7 @@ LOOP:
switch { switch {
case r == '\n': case r == '\n':
if !strings.HasPrefix(l.input[l.pos:], prefix) { if !bytes.HasPrefix(l.input[l.pos:], delimOrg) {
break LOOP break LOOP
} }
case r == eof: case r == eof:
@ -375,7 +378,7 @@ LOOP:
} }
// Handle YAML or TOML front matter. // Handle YAML or TOML front matter.
func (l *pageLexer) lexFrontMatterSection(tp itemType, delimr rune, name, delim string) stateFunc { func (l *pageLexer) lexFrontMatterSection(tp itemType, delimr rune, name string, delim []byte) stateFunc {
for i := 0; i < 2; i++ { for i := 0; i < 2; i++ {
if r := l.next(); r != delimr { if r := l.next(); r != delimr {
return l.errorf("invalid %s delimiter", name) return l.errorf("invalid %s delimiter", name)
@ -395,7 +398,7 @@ func (l *pageLexer) lexFrontMatterSection(tp itemType, delimr rune, name, delim
return l.errorf("EOF looking for end %s front matter delimiter", name) return l.errorf("EOF looking for end %s front matter delimiter", name)
} }
if isEndOfLine(r) { if isEndOfLine(r) {
if strings.HasPrefix(l.input[l.pos:], delim) { if bytes.HasPrefix(l.input[l.pos:], delim) {
l.emit(tp) l.emit(tp)
l.pos += 3 l.pos += 3
l.consumeCRLF() l.consumeCRLF()
@ -410,7 +413,7 @@ func (l *pageLexer) lexFrontMatterSection(tp itemType, delimr rune, name, delim
func lexShortcodeLeftDelim(l *pageLexer) stateFunc { func lexShortcodeLeftDelim(l *pageLexer) stateFunc {
l.pos += pos(len(l.currentLeftShortcodeDelim())) l.pos += pos(len(l.currentLeftShortcodeDelim()))
if strings.HasPrefix(l.input[l.pos:], leftComment) { if bytes.HasPrefix(l.input[l.pos:], leftComment) {
return lexShortcodeComment return lexShortcodeComment
} }
l.emit(l.currentLeftShortcodeDelimItem()) l.emit(l.currentLeftShortcodeDelimItem())
@ -420,7 +423,7 @@ func lexShortcodeLeftDelim(l *pageLexer) stateFunc {
} }
func lexShortcodeComment(l *pageLexer) stateFunc { func lexShortcodeComment(l *pageLexer) stateFunc {
posRightComment := strings.Index(l.input[l.pos:], rightComment+l.currentRightShortcodeDelim()) posRightComment := bytes.Index(l.input[l.pos:], append(rightComment, l.currentRightShortcodeDelim()...))
if posRightComment <= 1 { if posRightComment <= 1 {
return l.errorf("comment must be closed") return l.errorf("comment must be closed")
} }
@ -576,7 +579,7 @@ Loop:
case r == '/': case r == '/':
default: default:
l.backup() l.backup()
word := l.input[l.start:l.pos] word := string(l.input[l.start:l.pos])
if l.closingState > 0 && !l.openShortcodes[word] { if l.closingState > 0 && !l.openShortcodes[word] {
return l.errorf("closing tag for shortcode '%s' does not match start tag", word) return l.errorf("closing tag for shortcode '%s' does not match start tag", word)
} else if l.closingState > 0 { } else if l.closingState > 0 {
@ -600,7 +603,7 @@ Loop:
} }
func lexEndOfShortcode(l *pageLexer) stateFunc { func lexEndOfShortcode(l *pageLexer) stateFunc {
if strings.HasPrefix(l.input[l.pos:], l.currentRightShortcodeDelim()) { if bytes.HasPrefix(l.input[l.pos:], l.currentRightShortcodeDelim()) {
return lexShortcodeRightDelim return lexShortcodeRightDelim
} }
switch r := l.next(); { switch r := l.next(); {
@ -614,7 +617,7 @@ func lexEndOfShortcode(l *pageLexer) stateFunc {
// scans the elements inside shortcode tags // scans the elements inside shortcode tags
func lexInsideShortcode(l *pageLexer) stateFunc { func lexInsideShortcode(l *pageLexer) stateFunc {
if strings.HasPrefix(l.input[l.pos:], l.currentRightShortcodeDelim()) { if bytes.HasPrefix(l.input[l.pos:], l.currentRightShortcodeDelim()) {
return lexShortcodeRightDelim return lexShortcodeRightDelim
} }
switch r := l.next(); { switch r := l.next(); {
@ -659,7 +662,7 @@ func (l *pageLexer) currentRightShortcodeDelimItem() itemType {
return l.currRightDelimItem return l.currRightDelimItem
} }
func (l *pageLexer) currentLeftShortcodeDelim() string { func (l *pageLexer) currentLeftShortcodeDelim() []byte {
if l.currLeftDelimItem == tLeftDelimScWithMarkup { if l.currLeftDelimItem == tLeftDelimScWithMarkup {
return leftDelimScWithMarkup return leftDelimScWithMarkup
} }
@ -667,7 +670,7 @@ func (l *pageLexer) currentLeftShortcodeDelim() string {
} }
func (l *pageLexer) currentRightShortcodeDelim() string { func (l *pageLexer) currentRightShortcodeDelim() []byte {
if l.currRightDelimItem == tRightDelimScWithMarkup { if l.currRightDelimItem == tRightDelimScWithMarkup {
return rightDelimScWithMarkup return rightDelimScWithMarkup
} }

31
parser/pageparser/pageparser_intro_test.go
View file

@ -15,6 +15,7 @@ package pageparser
import ( import (
"fmt" "fmt"
"reflect"
"strings" "strings"
"testing" "testing"
) )
@ -25,23 +26,27 @@ type lexerTest struct {
items []Item items []Item
} }
func nti(tp itemType, val string) Item {
return Item{tp, 0, []byte(val)}
}
var ( var (
tstJSON = `{ "a": { "b": "\"Hugo\"}" } }` tstJSON = `{ "a": { "b": "\"Hugo\"}" } }`
tstHTMLLead = Item{tHTMLLead, 0, " <"} tstHTMLLead = nti(tHTMLLead, " <")
tstFrontMatterTOML = Item{tFrontMatterTOML, 0, "foo = \"bar\"\n"} tstFrontMatterTOML = nti(tFrontMatterTOML, "foo = \"bar\"\n")
tstFrontMatterYAML = Item{tFrontMatterYAML, 0, "foo: \"bar\"\n"} tstFrontMatterYAML = nti(tFrontMatterYAML, "foo: \"bar\"\n")
tstFrontMatterYAMLCRLF = Item{tFrontMatterYAML, 0, "foo: \"bar\"\r\n"} tstFrontMatterYAMLCRLF = nti(tFrontMatterYAML, "foo: \"bar\"\r\n")
tstFrontMatterJSON = Item{tFrontMatterJSON, 0, tstJSON + "\r\n"} tstFrontMatterJSON = nti(tFrontMatterJSON, tstJSON+"\r\n")
tstSomeText = Item{tText, 0, "\nSome text.\n"} tstSomeText = nti(tText, "\nSome text.\n")
tstSummaryDivider = Item{tSummaryDivider, 0, "<!--more-->"} tstSummaryDivider = nti(tSummaryDivider, "<!--more-->")
tstSummaryDividerOrg = Item{tSummaryDividerOrg, 0, "# more"} tstSummaryDividerOrg = nti(tSummaryDividerOrg, "# more")
tstORG = ` tstORG = `
#+TITLE: T1 #+TITLE: T1
#+AUTHOR: A1 #+AUTHOR: A1
#+DESCRIPTION: D1 #+DESCRIPTION: D1
` `
tstFrontMatterORG = Item{tFrontMatterORG, 0, tstORG} tstFrontMatterORG = nti(tFrontMatterORG, tstORG)
) )
var crLfReplacer = strings.NewReplacer("\r", "#", "\n", "$") var crLfReplacer = strings.NewReplacer("\r", "#", "\n", "$")
@ -49,7 +54,7 @@ var crLfReplacer = strings.NewReplacer("\r", "#", "\n", "$")
// TODO(bep) a way to toggle ORG mode vs the rest. // TODO(bep) a way to toggle ORG mode vs the rest.
var frontMatterTests = []lexerTest{ var frontMatterTests = []lexerTest{
{"empty", "", []Item{tstEOF}}, {"empty", "", []Item{tstEOF}},
{"HTML Document", ` <html> `, []Item{tstHTMLLead, Item{tText, 0, "html> "}, tstEOF}}, {"HTML Document", ` <html> `, []Item{tstHTMLLead, nti(tText, "html> "), tstEOF}},
{"YAML front matter", "---\nfoo: \"bar\"\n---\n\nSome text.\n", []Item{tstFrontMatterYAML, tstSomeText, tstEOF}}, {"YAML front matter", "---\nfoo: \"bar\"\n---\n\nSome text.\n", []Item{tstFrontMatterYAML, tstSomeText, tstEOF}},
// Note that we keep all bytes as they are, but we need to handle CRLF // Note that we keep all bytes as they are, but we need to handle CRLF
{"YAML front matter CRLF", "---\r\nfoo: \"bar\"\r\n---\n\nSome text.\n", []Item{tstFrontMatterYAMLCRLF, tstSomeText, tstEOF}}, {"YAML front matter CRLF", "---\r\nfoo: \"bar\"\r\n---\n\nSome text.\n", []Item{tstFrontMatterYAMLCRLF, tstSomeText, tstEOF}},
@ -63,7 +68,7 @@ var frontMatterTests = []lexerTest{
func TestFrontMatter(t *testing.T) { func TestFrontMatter(t *testing.T) {
t.Parallel() t.Parallel()
for i, test := range frontMatterTests { for i, test := range frontMatterTests {
items := collect(test.name, test.input, false, lexIntroSection) items := collect([]byte(test.input), false, lexIntroSection)
if !equal(items, test.items) { if !equal(items, test.items) {
got := crLfReplacer.Replace(fmt.Sprint(items)) got := crLfReplacer.Replace(fmt.Sprint(items))
expected := crLfReplacer.Replace(fmt.Sprint(test.items)) expected := crLfReplacer.Replace(fmt.Sprint(test.items))
@ -72,7 +77,7 @@ func TestFrontMatter(t *testing.T) {
} }
} }
func collect(name, input string, skipFrontMatter bool, stateStart stateFunc) (items []Item) { func collect(input []byte, skipFrontMatter bool, stateStart stateFunc) (items []Item) {
l := newPageLexer(input, 0, stateStart) l := newPageLexer(input, 0, stateStart)
l.run() l.run()
@ -95,7 +100,7 @@ func equal(i1, i2 []Item) bool {
if i1[k].typ != i2[k].typ { if i1[k].typ != i2[k].typ {
return false return false
} }
if i1[k].Val != i2[k].Val { if !reflect.DeepEqual(i1[k].Val, i2[k].Val) {
return false return false
} }
} }

100
parser/pageparser/pageparser_shortcode_test.go
View file

@ -16,25 +16,25 @@ package pageparser
import "testing" import "testing"
var ( var (
tstEOF = Item{tEOF, 0, ""} tstEOF = nti(tEOF, "")
tstLeftNoMD = Item{tLeftDelimScNoMarkup, 0, "{{<"} tstLeftNoMD = nti(tLeftDelimScNoMarkup, "{{<")
tstRightNoMD = Item{tRightDelimScNoMarkup, 0, ">}}"} tstRightNoMD = nti(tRightDelimScNoMarkup, ">}}")
tstLeftMD = Item{tLeftDelimScWithMarkup, 0, "{{%"} tstLeftMD = nti(tLeftDelimScWithMarkup, "{{%")
tstRightMD = Item{tRightDelimScWithMarkup, 0, "%}}"} tstRightMD = nti(tRightDelimScWithMarkup, "%}}")
tstSCClose = Item{tScClose, 0, "/"} tstSCClose = nti(tScClose, "/")
tstSC1 = Item{tScName, 0, "sc1"} tstSC1 = nti(tScName, "sc1")
tstSC2 = Item{tScName, 0, "sc2"} tstSC2 = nti(tScName, "sc2")
tstSC3 = Item{tScName, 0, "sc3"} tstSC3 = nti(tScName, "sc3")
tstSCSlash = Item{tScName, 0, "sc/sub"} tstSCSlash = nti(tScName, "sc/sub")
tstParam1 = Item{tScParam, 0, "param1"} tstParam1 = nti(tScParam, "param1")
tstParam2 = Item{tScParam, 0, "param2"} tstParam2 = nti(tScParam, "param2")
tstVal = Item{tScParamVal, 0, "Hello World"} tstVal = nti(tScParamVal, "Hello World")
) )
var shortCodeLexerTests = []lexerTest{ var shortCodeLexerTests = []lexerTest{
{"empty", "", []Item{tstEOF}}, {"empty", "", []Item{tstEOF}},
{"spaces", " \t\n", []Item{{tText, 0, " \t\n"}, tstEOF}}, {"spaces", " \t\n", []Item{nti(tText, " \t\n"), tstEOF}},
{"text", `to be or not`, []Item{{tText, 0, "to be or not"}, tstEOF}}, {"text", `to be or not`, []Item{nti(tText, "to be or not"), tstEOF}},
{"no markup", `{{< sc1 >}}`, []Item{tstLeftNoMD, tstSC1, tstRightNoMD, tstEOF}}, {"no markup", `{{< sc1 >}}`, []Item{tstLeftNoMD, tstSC1, tstRightNoMD, tstEOF}},
{"with EOL", "{{< sc1 \n >}}", []Item{tstLeftNoMD, tstSC1, tstRightNoMD, tstEOF}}, {"with EOL", "{{< sc1 \n >}}", []Item{tstLeftNoMD, tstSC1, tstRightNoMD, tstEOF}},
@ -43,12 +43,12 @@ var shortCodeLexerTests = []lexerTest{
{"simple with markup", `{{% sc1 %}}`, []Item{tstLeftMD, tstSC1, tstRightMD, tstEOF}}, {"simple with markup", `{{% sc1 %}}`, []Item{tstLeftMD, tstSC1, tstRightMD, tstEOF}},
{"with spaces", `{{< sc1 >}}`, []Item{tstLeftNoMD, tstSC1, tstRightNoMD, tstEOF}}, {"with spaces", `{{< sc1 >}}`, []Item{tstLeftNoMD, tstSC1, tstRightNoMD, tstEOF}},
{"mismatched rightDelim", `{{< sc1 %}}`, []Item{tstLeftNoMD, tstSC1, {"mismatched rightDelim", `{{< sc1 %}}`, []Item{tstLeftNoMD, tstSC1,
{tError, 0, "unrecognized character in shortcode action: U+0025 '%'. Note: Parameters with non-alphanumeric args must be quoted"}}}, nti(tError, "unrecognized character in shortcode action: U+0025 '%'. Note: Parameters with non-alphanumeric args must be quoted")}},
{"inner, markup", `{{% sc1 %}} inner {{% /sc1 %}}`, []Item{ {"inner, markup", `{{% sc1 %}} inner {{% /sc1 %}}`, []Item{
tstLeftMD, tstLeftMD,
tstSC1, tstSC1,
tstRightMD, tstRightMD,
{tText, 0, " inner "}, nti(tText, " inner "),
tstLeftMD, tstLeftMD,
tstSCClose, tstSCClose,
tstSC1, tstSC1,
@ -56,20 +56,20 @@ var shortCodeLexerTests = []lexerTest{
tstEOF, tstEOF,
}}, }},
{"close, but no open", `{{< /sc1 >}}`, []Item{ {"close, but no open", `{{< /sc1 >}}`, []Item{
tstLeftNoMD, {tError, 0, "got closing shortcode, but none is open"}}}, tstLeftNoMD, nti(tError, "got closing shortcode, but none is open")}},
{"close wrong", `{{< sc1 >}}{{< /another >}}`, []Item{ {"close wrong", `{{< sc1 >}}{{< /another >}}`, []Item{
tstLeftNoMD, tstSC1, tstRightNoMD, tstLeftNoMD, tstSCClose, tstLeftNoMD, tstSC1, tstRightNoMD, tstLeftNoMD, tstSCClose,
{tError, 0, "closing tag for shortcode 'another' does not match start tag"}}}, nti(tError, "closing tag for shortcode 'another' does not match start tag")}},
{"close, but no open, more", `{{< sc1 >}}{{< /sc1 >}}{{< /another >}}`, []Item{ {"close, but no open, more", `{{< sc1 >}}{{< /sc1 >}}{{< /another >}}`, []Item{
tstLeftNoMD, tstSC1, tstRightNoMD, tstLeftNoMD, tstSCClose, tstSC1, tstRightNoMD, tstLeftNoMD, tstSCClose, tstLeftNoMD, tstSC1, tstRightNoMD, tstLeftNoMD, tstSCClose, tstSC1, tstRightNoMD, tstLeftNoMD, tstSCClose,
{tError, 0, "closing tag for shortcode 'another' does not match start tag"}}}, nti(tError, "closing tag for shortcode 'another' does not match start tag")}},
{"close with extra keyword", `{{< sc1 >}}{{< /sc1 keyword>}}`, []Item{ {"close with extra keyword", `{{< sc1 >}}{{< /sc1 keyword>}}`, []Item{
tstLeftNoMD, tstSC1, tstRightNoMD, tstLeftNoMD, tstSCClose, tstSC1, tstLeftNoMD, tstSC1, tstRightNoMD, tstLeftNoMD, tstSCClose, tstSC1,
{tError, 0, "unclosed shortcode"}}}, nti(tError, "unclosed shortcode")}},
{"Youtube id", `{{< sc1 -ziL-Q_456igdO-4 >}}`, []Item{ {"Youtube id", `{{< sc1 -ziL-Q_456igdO-4 >}}`, []Item{
tstLeftNoMD, tstSC1, {tScParam, 0, "-ziL-Q_456igdO-4"}, tstRightNoMD, tstEOF}}, tstLeftNoMD, tstSC1, nti(tScParam, "-ziL-Q_456igdO-4"), tstRightNoMD, tstEOF}},
{"non-alphanumerics param quoted", `{{< sc1 "-ziL-.%QigdO-4" >}}`, []Item{ {"non-alphanumerics param quoted", `{{< sc1 "-ziL-.%QigdO-4" >}}`, []Item{
tstLeftNoMD, tstSC1, {tScParam, 0, "-ziL-.%QigdO-4"}, tstRightNoMD, tstEOF}}, tstLeftNoMD, tstSC1, nti(tScParam, "-ziL-.%QigdO-4"), tstRightNoMD, tstEOF}},
{"two params", `{{< sc1 param1 param2 >}}`, []Item{ {"two params", `{{< sc1 param1 param2 >}}`, []Item{
tstLeftNoMD, tstSC1, tstParam1, tstParam2, tstRightNoMD, tstEOF}}, tstLeftNoMD, tstSC1, tstParam1, tstParam2, tstRightNoMD, tstEOF}},
@ -94,64 +94,64 @@ var shortCodeLexerTests = []lexerTest{
tstLeftNoMD, tstSCClose, tstSC1, tstRightNoMD, tstEOF}}, tstLeftNoMD, tstSCClose, tstSC1, tstRightNoMD, tstEOF}},
{"nested complex", `{{< sc1 >}}ab{{% sc2 param1 %}}cd{{< sc3 >}}ef{{< /sc3 >}}gh{{% /sc2 %}}ij{{< /sc1 >}}kl`, []Item{ {"nested complex", `{{< sc1 >}}ab{{% sc2 param1 %}}cd{{< sc3 >}}ef{{< /sc3 >}}gh{{% /sc2 %}}ij{{< /sc1 >}}kl`, []Item{
tstLeftNoMD, tstSC1, tstRightNoMD, tstLeftNoMD, tstSC1, tstRightNoMD,
{tText, 0, "ab"}, nti(tText, "ab"),
tstLeftMD, tstSC2, tstParam1, tstRightMD, tstLeftMD, tstSC2, tstParam1, tstRightMD,
{tText, 0, "cd"}, nti(tText, "cd"),
tstLeftNoMD, tstSC3, tstRightNoMD, tstLeftNoMD, tstSC3, tstRightNoMD,
{tText, 0, "ef"}, nti(tText, "ef"),
tstLeftNoMD, tstSCClose, tstSC3, tstRightNoMD, tstLeftNoMD, tstSCClose, tstSC3, tstRightNoMD,
{tText, 0, "gh"}, nti(tText, "gh"),
tstLeftMD, tstSCClose, tstSC2, tstRightMD, tstLeftMD, tstSCClose, tstSC2, tstRightMD,
{tText, 0, "ij"}, nti(tText, "ij"),
tstLeftNoMD, tstSCClose, tstSC1, tstRightNoMD, tstLeftNoMD, tstSCClose, tstSC1, tstRightNoMD,
{tText, 0, "kl"}, tstEOF, nti(tText, "kl"), tstEOF,
}}, }},
{"two quoted params", `{{< sc1 "param nr. 1" "param nr. 2" >}}`, []Item{ {"two quoted params", `{{< sc1 "param nr. 1" "param nr. 2" >}}`, []Item{
tstLeftNoMD, tstSC1, {tScParam, 0, "param nr. 1"}, {tScParam, 0, "param nr. 2"}, tstRightNoMD, tstEOF}}, tstLeftNoMD, tstSC1, nti(tScParam, "param nr. 1"), nti(tScParam, "param nr. 2"), tstRightNoMD, tstEOF}},
{"two named params", `{{< sc1 param1="Hello World" param2="p2Val">}}`, []Item{ {"two named params", `{{< sc1 param1="Hello World" param2="p2Val">}}`, []Item{
tstLeftNoMD, tstSC1, tstParam1, tstVal, tstParam2, {tScParamVal, 0, "p2Val"}, tstRightNoMD, tstEOF}}, tstLeftNoMD, tstSC1, tstParam1, tstVal, tstParam2, nti(tScParamVal, "p2Val"), tstRightNoMD, tstEOF}},
{"escaped quotes", `{{< sc1 param1=\"Hello World\" >}}`, []Item{ {"escaped quotes", `{{< sc1 param1=\"Hello World\" >}}`, []Item{
tstLeftNoMD, tstSC1, tstParam1, tstVal, tstRightNoMD, tstEOF}}, tstLeftNoMD, tstSC1, tstParam1, tstVal, tstRightNoMD, tstEOF}},
{"escaped quotes, positional param", `{{< sc1 \"param1\" >}}`, []Item{ {"escaped quotes, positional param", `{{< sc1 \"param1\" >}}`, []Item{
tstLeftNoMD, tstSC1, tstParam1, tstRightNoMD, tstEOF}}, tstLeftNoMD, tstSC1, tstParam1, tstRightNoMD, tstEOF}},
{"escaped quotes inside escaped quotes", `{{< sc1 param1=\"Hello \"escaped\" World\" >}}`, []Item{ {"escaped quotes inside escaped quotes", `{{< sc1 param1=\"Hello \"escaped\" World\" >}}`, []Item{
tstLeftNoMD, tstSC1, tstParam1, tstLeftNoMD, tstSC1, tstParam1,
{tScParamVal, 0, `Hello `}, {tError, 0, `got positional parameter 'escaped'. Cannot mix named and positional parameters`}}}, nti(tScParamVal, `Hello `), nti(tError, `got positional parameter 'escaped'. Cannot mix named and positional parameters`)}},
{"escaped quotes inside nonescaped quotes", {"escaped quotes inside nonescaped quotes",
`{{< sc1 param1="Hello \"escaped\" World" >}}`, []Item{ `{{< sc1 param1="Hello \"escaped\" World" >}}`, []Item{
tstLeftNoMD, tstSC1, tstParam1, {tScParamVal, 0, `Hello "escaped" World`}, tstRightNoMD, tstEOF}}, tstLeftNoMD, tstSC1, tstParam1, nti(tScParamVal, `Hello "escaped" World`), tstRightNoMD, tstEOF}},
{"escaped quotes inside nonescaped quotes in positional param", {"escaped quotes inside nonescaped quotes in positional param",
`{{< sc1 "Hello \"escaped\" World" >}}`, []Item{ `{{< sc1 "Hello \"escaped\" World" >}}`, []Item{
tstLeftNoMD, tstSC1, {tScParam, 0, `Hello "escaped" World`}, tstRightNoMD, tstEOF}}, tstLeftNoMD, tstSC1, nti(tScParam, `Hello "escaped" World`), tstRightNoMD, tstEOF}},
{"unterminated quote", `{{< sc1 param2="Hello World>}}`, []Item{ {"unterminated quote", `{{< sc1 param2="Hello World>}}`, []Item{
tstLeftNoMD, tstSC1, tstParam2, {tError, 0, "unterminated quoted string in shortcode parameter-argument: 'Hello World>}}'"}}}, tstLeftNoMD, tstSC1, tstParam2, nti(tError, "unterminated quoted string in shortcode parameter-argument: 'Hello World>}}'")}},
{"one named param, one not", `{{< sc1 param1="Hello World" p2 >}}`, []Item{ {"one named param, one not", `{{< sc1 param1="Hello World" p2 >}}`, []Item{
tstLeftNoMD, tstSC1, tstParam1, tstVal, tstLeftNoMD, tstSC1, tstParam1, tstVal,
{tError, 0, "got positional parameter 'p2'. Cannot mix named and positional parameters"}}}, nti(tError, "got positional parameter 'p2'. Cannot mix named and positional parameters")}},
{"one named param, one quoted positional param", `{{< sc1 param1="Hello World" "And Universe" >}}`, []Item{ {"one named param, one quoted positional param", `{{< sc1 param1="Hello World" "And Universe" >}}`, []Item{
tstLeftNoMD, tstSC1, tstParam1, tstVal, tstLeftNoMD, tstSC1, tstParam1, tstVal,
{tError, 0, "got quoted positional parameter. Cannot mix named and positional parameters"}}}, nti(tError, "got quoted positional parameter. Cannot mix named and positional parameters")}},
{"one quoted positional param, one named param", `{{< sc1 "param1" param2="And Universe" >}}`, []Item{ {"one quoted positional param, one named param", `{{< sc1 "param1" param2="And Universe" >}}`, []Item{
tstLeftNoMD, tstSC1, tstParam1, tstLeftNoMD, tstSC1, tstParam1,
{tError, 0, "got named parameter 'param2'. Cannot mix named and positional parameters"}}}, nti(tError, "got named parameter 'param2'. Cannot mix named and positional parameters")}},
{"ono positional param, one not", `{{< sc1 param1 param2="Hello World">}}`, []Item{ {"ono positional param, one not", `{{< sc1 param1 param2="Hello World">}}`, []Item{
tstLeftNoMD, tstSC1, tstParam1, tstLeftNoMD, tstSC1, tstParam1,
{tError, 0, "got named parameter 'param2'. Cannot mix named and positional parameters"}}}, nti(tError, "got named parameter 'param2'. Cannot mix named and positional parameters")}},
{"commented out", `{{</* sc1 */>}}`, []Item{ {"commented out", `{{</* sc1 */>}}`, []Item{
{tText, 0, "{{<"}, {tText, 0, " sc1 "}, {tText, 0, ">}}"}, tstEOF}}, nti(tText, "{{<"), nti(tText, " sc1 "), nti(tText, ">}}"), tstEOF}},
{"commented out, with asterisk inside", `{{</* sc1 "**/*.pdf" */>}}`, []Item{ {"commented out, with asterisk inside", `{{</* sc1 "**/*.pdf" */>}}`, []Item{
{tText, 0, "{{<"}, {tText, 0, " sc1 \"**/*.pdf\" "}, {tText, 0, ">}}"}, tstEOF}}, nti(tText, "{{<"), nti(tText, " sc1 \"**/*.pdf\" "), nti(tText, ">}}"), tstEOF}},
{"commented out, missing close", `{{</* sc1 >}}`, []Item{ {"commented out, missing close", `{{</* sc1 >}}`, []Item{
{tError, 0, "comment must be closed"}}}, nti(tError, "comment must be closed")}},
{"commented out, misplaced close", `{{</* sc1 >}}*/`, []Item{ {"commented out, misplaced close", `{{</* sc1 >}}*/`, []Item{
{tError, 0, "comment must be closed"}}}, nti(tError, "comment must be closed")}},
} }
func TestShortcodeLexer(t *testing.T) { func TestShortcodeLexer(t *testing.T) {
t.Parallel() t.Parallel()
for i, test := range shortCodeLexerTests { for i, test := range shortCodeLexerTests {
items := collect(test.name, test.input, true, lexMainSection) items := collect([]byte(test.input), true, lexMainSection)
if !equal(items, test.items) { if !equal(items, test.items) {
t.Errorf("[%d] %s: got\n\t%v\nexpected\n\t%v", i, test.name, items, test.items) t.Errorf("[%d] %s: got\n\t%v\nexpected\n\t%v", i, test.name, items, test.items)
} }
@ -159,13 +159,17 @@ func TestShortcodeLexer(t *testing.T) {
} }
func BenchmarkShortcodeLexer(b *testing.B) { func BenchmarkShortcodeLexer(b *testing.B) {
testInputs := make([][]byte, len(shortCodeLexerTests))
for i, input := range shortCodeLexerTests {
testInputs[i] = []byte(input.input)
}
b.ResetTimer() b.ResetTimer()
for i := 0; i < b.N; i++ { for i := 0; i < b.N; i++ {
for _, test := range shortCodeLexerTests { for _, input := range testInputs {
items := collect(test.name, test.input, true, lexMainSection) items := collect(input, true, lexMainSection)
if !equal(items, test.items) { if len(items) == 0 {
b.Errorf("%s: got\n\t%v\nexpected\n\t%v", test.name, items, test.items)
} }
} }
} }
} }