hugo/hugolib/page__per_output.go

723 lines
17 KiB
Go
Raw Normal View History

// Copyright 2019 The Hugo Authors. All rights reserved.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package hugolib
import (
"bytes"
"context"
"fmt"
"html/template"
"runtime/debug"
"strings"
"sync"
"unicode/utf8"
"github.com/gohugoio/hugo/common/text"
"github.com/gohugoio/hugo/common/types/hstring"
"github.com/gohugoio/hugo/identity"
"github.com/mitchellh/mapstructure"
"github.com/pkg/errors"
"github.com/spf13/cast"
"github.com/gohugoio/hugo/markup/converter/hooks"
"github.com/gohugoio/hugo/markup/converter"
"github.com/alecthomas/chroma/lexers"
"github.com/gohugoio/hugo/lazy"
bp "github.com/gohugoio/hugo/bufferpool"
"github.com/gohugoio/hugo/tpl"
"github.com/gohugoio/hugo/helpers"
"github.com/gohugoio/hugo/output"
"github.com/gohugoio/hugo/resources/page"
"github.com/gohugoio/hugo/resources/resource"
)
var (
nopTargetPath = targetPathsHolder{}
nopPagePerOutput = struct {
resource.ResourceLinksProvider
page.ContentProvider
page.PageRenderProvider
page.PaginatorProvider
page.TableOfContentsProvider
page.AlternativeOutputFormatsProvider
targetPather
}{
page.NopPage,
page.NopPage,
page.NopPage,
page.NopPage,
page.NopPage,
page.NopPage,
nopTargetPath,
}
)
var pageContentOutputDependenciesID = identity.KeyValueIdentity{Key: "pageOutput", Value: "dependencies"}
func newPageContentOutput(p *pageState, po *pageOutput) (*pageContentOutput, error) {
parent := p.init
var dependencyTracker identity.Manager
if p.s.running() {
dependencyTracker = identity.NewManager(pageContentOutputDependenciesID)
}
cp := &pageContentOutput{
dependencyTracker: dependencyTracker,
p: p,
f: po.f,
renderHooks: &renderHooks{},
}
initContent := func() (err error) {
p.s.h.IncrContentRender()
if p.cmap == nil {
// Nothing to do.
return nil
}
defer func() {
// See https://github.com/gohugoio/hugo/issues/6210
if r := recover(); r != nil {
err = fmt.Errorf("%s", r)
p.s.Log.Errorf("[BUG] Got panic:\n%s\n%s", r, string(debug.Stack()))
}
}()
if err := po.cp.initRenderHooks(); err != nil {
return err
}
var hasShortcodeVariants bool
f := po.f
cp.contentPlaceholders, hasShortcodeVariants, err = p.shortcodeState.renderShortcodesForPage(p, f)
if err != nil {
return err
}
if hasShortcodeVariants {
p.pageOutputTemplateVariationsState.Store(2)
}
cp.workContent = p.contentToRender(cp.contentPlaceholders)
isHTML := cp.p.m.markup == "html"
if !isHTML {
r, err := cp.renderContent(cp.workContent, true)
if err != nil {
return err
}
cp.workContent = r.Bytes()
if tocProvider, ok := r.(converter.TableOfContentsProvider); ok {
cfg := p.s.ContentSpec.Converters.GetMarkupConfig()
cp.tableOfContents = template.HTML(
tocProvider.TableOfContents().ToHTML(
cfg.TableOfContents.StartLevel,
cfg.TableOfContents.EndLevel,
cfg.TableOfContents.Ordered,
),
)
} else {
tmpContent, tmpTableOfContents := helpers.ExtractTOC(cp.workContent)
cp.tableOfContents = helpers.BytesToHTML(tmpTableOfContents)
cp.workContent = tmpContent
}
}
if cp.placeholdersEnabled {
// ToC was accessed via .Page.TableOfContents in the shortcode,
// at a time when the ToC wasn't ready.
cp.contentPlaceholders[tocShortcodePlaceholder] = string(cp.tableOfContents)
}
if p.cmap.hasNonMarkdownShortcode || cp.placeholdersEnabled {
// There are one or more replacement tokens to be replaced.
cp.workContent, err = replaceShortcodeTokens(cp.workContent, cp.contentPlaceholders)
if err != nil {
return err
}
}
if cp.p.source.hasSummaryDivider {
if isHTML {
src := p.source.parsed.Input()
// Use the summary sections as they are provided by the user.
if p.source.posSummaryEnd != -1 {
cp.summary = helpers.BytesToHTML(src[p.source.posMainContent:p.source.posSummaryEnd])
}
if cp.p.source.posBodyStart != -1 {
cp.workContent = src[cp.p.source.posBodyStart:]
}
} else {
summary, content, err := splitUserDefinedSummaryAndContent(cp.p.m.markup, cp.workContent)
if err != nil {
cp.p.s.Log.Errorf("Failed to set user defined summary for page %q: %s", cp.p.pathOrTitle(), err)
} else {
cp.workContent = content
cp.summary = helpers.BytesToHTML(summary)
}
}
} else if cp.p.m.summary != "" {
b, err := cp.renderContent([]byte(cp.p.m.summary), false)
if err != nil {
return err
}
html := cp.p.s.ContentSpec.TrimShortHTML(b.Bytes())
cp.summary = helpers.BytesToHTML(html)
tpl/tplimpl: Rework template management to get rid of concurrency issues This more or less completes the simplification of the template handling code in Hugo started in v0.62. The main motivation was to fix a long lasting issue about a crash in HTML content files without front matter. But this commit also comes with a big functional improvement. As we now have moved the base template evaluation to the build stage we now use the same lookup rules for `baseof` as for `list` etc. type of templates. This means that in this simple example you can have a `baseof` template for the `blog` section without having to duplicate the others: ``` layouts ├── _default │   ├── baseof.html │   ├── list.html │   └── single.html └── blog └── baseof.html ``` Also, when simplifying code, you often get rid of some double work, as shown in the "site building" benchmarks below. These benchmarks looks suspiciously good, but I have repeated the below with ca. the same result. Compared to master: ``` name old time/op new time/op delta SiteNew/Bundle_with_image-16 13.1ms ± 1% 10.5ms ± 1% -19.34% (p=0.029 n=4+4) SiteNew/Bundle_with_JSON_file-16 13.0ms ± 0% 10.7ms ± 1% -18.05% (p=0.029 n=4+4) SiteNew/Tags_and_categories-16 46.4ms ± 2% 43.1ms ± 1% -7.15% (p=0.029 n=4+4) SiteNew/Canonify_URLs-16 52.2ms ± 2% 47.8ms ± 1% -8.30% (p=0.029 n=4+4) SiteNew/Deep_content_tree-16 77.9ms ± 1% 70.9ms ± 1% -9.01% (p=0.029 n=4+4) SiteNew/Many_HTML_templates-16 43.0ms ± 0% 37.2ms ± 1% -13.54% (p=0.029 n=4+4) SiteNew/Page_collections-16 58.2ms ± 1% 52.4ms ± 1% -9.95% (p=0.029 n=4+4) name old alloc/op new alloc/op delta SiteNew/Bundle_with_image-16 3.81MB ± 0% 2.22MB ± 0% -41.70% (p=0.029 n=4+4) SiteNew/Bundle_with_JSON_file-16 3.60MB ± 0% 2.01MB ± 0% -44.20% (p=0.029 n=4+4) SiteNew/Tags_and_categories-16 19.3MB ± 1% 14.1MB ± 0% -26.91% (p=0.029 n=4+4) SiteNew/Canonify_URLs-16 70.7MB ± 0% 69.0MB ± 0% -2.40% (p=0.029 n=4+4) SiteNew/Deep_content_tree-16 37.1MB ± 0% 31.2MB ± 0% -15.94% (p=0.029 n=4+4) SiteNew/Many_HTML_templates-16 17.6MB ± 0% 10.6MB ± 0% -39.92% (p=0.029 n=4+4) SiteNew/Page_collections-16 25.9MB ± 0% 21.2MB ± 0% -17.99% (p=0.029 n=4+4) name old allocs/op new allocs/op delta SiteNew/Bundle_with_image-16 52.3k ± 0% 26.1k ± 0% -50.18% (p=0.029 n=4+4) SiteNew/Bundle_with_JSON_file-16 52.3k ± 0% 26.1k ± 0% -50.16% (p=0.029 n=4+4) SiteNew/Tags_and_categories-16 336k ± 1% 269k ± 0% -19.90% (p=0.029 n=4+4) SiteNew/Canonify_URLs-16 422k ± 0% 395k ± 0% -6.43% (p=0.029 n=4+4) SiteNew/Deep_content_tree-16 401k ± 0% 313k ± 0% -21.79% (p=0.029 n=4+4) SiteNew/Many_HTML_templates-16 247k ± 0% 143k ± 0% -42.17% (p=0.029 n=4+4) SiteNew/Page_collections-16 282k ± 0% 207k ± 0% -26.55% (p=0.029 n=4+4) ``` Fixes #6716 Fixes #6760 Fixes #6768 Fixes #6778
2020-01-15 09:59:56 -05:00
}
cp.content = helpers.BytesToHTML(cp.workContent)
return nil
}
// There may be recursive loops in shortcodes and render hooks.
cp.initMain = parent.BranchWithTimeout(p.s.siteCfg.timeout, func(ctx context.Context) (any, error) {
return nil, initContent()
})
cp.initPlain = cp.initMain.Branch(func() (any, error) {
cp.plain = helpers.StripHTML(string(cp.content))
cp.plainWords = strings.Fields(cp.plain)
cp.setWordCounts(p.m.isCJKLanguage)
if err := cp.setAutoSummary(); err != nil {
return err, nil
}
return nil, nil
})
return cp, nil
}
type renderHooks struct {
getRenderer hooks.GetRendererFunc
init sync.Once
}
// pageContentOutput represents the Page content for a given output format.
type pageContentOutput struct {
f output.Format
p *pageState
// Lazy load dependencies
initMain *lazy.Init
initPlain *lazy.Init
placeholdersEnabled bool
placeholdersEnabledInit sync.Once
// Renders Markdown hooks.
renderHooks *renderHooks
workContent []byte
dependencyTracker identity.Manager // Set in server mode.
// Temporary storage of placeholders mapped to their content.
// These are shortcodes etc. Some of these will need to be replaced
// after any markup is rendered, so they share a common prefix.
contentPlaceholders map[string]string
// Content sections
content template.HTML
summary template.HTML
tableOfContents template.HTML
truncated bool
plainWords []string
plain string
fuzzyWordCount int
wordCount int
readingTime int
}
func (p *pageContentOutput) trackDependency(id identity.Provider) {
if p.dependencyTracker != nil {
p.dependencyTracker.Add(id)
}
}
func (p *pageContentOutput) Reset() {
if p.dependencyTracker != nil {
p.dependencyTracker.Reset()
}
p.initMain.Reset()
p.initPlain.Reset()
p.renderHooks = &renderHooks{}
}
func (p *pageContentOutput) Content() (any, error) {
if p.p.s.initInit(p.initMain, p.p) {
return p.content, nil
}
return nil, nil
}
func (p *pageContentOutput) FuzzyWordCount() int {
p.p.s.initInit(p.initPlain, p.p)
return p.fuzzyWordCount
}
func (p *pageContentOutput) Len() int {
p.p.s.initInit(p.initMain, p.p)
return len(p.content)
}
func (p *pageContentOutput) Plain() string {
p.p.s.initInit(p.initPlain, p.p)
return p.plain
}
func (p *pageContentOutput) PlainWords() []string {
p.p.s.initInit(p.initPlain, p.p)
return p.plainWords
}
func (p *pageContentOutput) ReadingTime() int {
p.p.s.initInit(p.initPlain, p.p)
return p.readingTime
}
func (p *pageContentOutput) Summary() template.HTML {
p.p.s.initInit(p.initMain, p.p)
if !p.p.source.hasSummaryDivider {
p.p.s.initInit(p.initPlain, p.p)
}
return p.summary
}
func (p *pageContentOutput) TableOfContents() template.HTML {
p.p.s.initInit(p.initMain, p.p)
return p.tableOfContents
}
func (p *pageContentOutput) Truncated() bool {
if p.p.truncated {
return true
}
p.p.s.initInit(p.initPlain, p.p)
return p.truncated
}
func (p *pageContentOutput) WordCount() int {
p.p.s.initInit(p.initPlain, p.p)
return p.wordCount
}
func (p *pageContentOutput) RenderString(args ...any) (template.HTML, error) {
if len(args) < 1 || len(args) > 2 {
return "", errors.New("want 1 or 2 arguments")
}
var s string
opts := defaultRenderStringOpts
sidx := 1
if len(args) == 1 {
sidx = 0
} else {
m, ok := args[0].(map[string]any)
if !ok {
return "", errors.New("first argument must be a map")
}
if err := mapstructure.WeakDecode(m, &opts); err != nil {
return "", errors.WithMessage(err, "failed to decode options")
}
}
contentToRender := args[sidx]
if _, ok := contentToRender.(hstring.RenderedString); ok {
// This content is already rendered, this is potentially
// a infinite recursion.
return "", errors.New("text is already rendered, repeating it may cause infinite recursion")
}
var err error
s, err = cast.ToStringE(contentToRender)
if err != nil {
return "", err
}
if err = p.initRenderHooks(); err != nil {
return "", err
}
conv := p.p.getContentConverter()
if opts.Markup != "" && opts.Markup != p.p.m.markup {
var err error
// TODO(bep) consider cache
conv, err = p.p.m.newContentConverter(p.p, opts.Markup, nil)
if err != nil {
return "", p.p.wrapError(err)
}
}
c, err := p.renderContentWithConverter(conv, []byte(s), false)
if err != nil {
return "", p.p.wrapError(err)
}
b := c.Bytes()
if opts.Display == "inline" {
// We may have to rethink this in the future when we get other
// renderers.
b = p.p.s.ContentSpec.TrimShortHTML(b)
}
return template.HTML(string(b)), nil
}
func (p *pageContentOutput) RenderWithTemplateInfo(info tpl.Info, layout ...string) (template.HTML, error) {
p.p.addDependency(info)
return p.Render(layout...)
}
func (p *pageContentOutput) Render(layout ...string) (template.HTML, error) {
templ, found, err := p.p.resolveTemplate(layout...)
if err != nil {
return "", p.p.wrapError(err)
}
if !found {
return "", nil
}
p.p.addDependency(templ.(tpl.Info))
// Make sure to send the *pageState and not the *pageContentOutput to the template.
res, err := executeToString(p.p.s.Tmpl(), templ, p.p)
if err != nil {
return "", p.p.wrapError(errors.Wrapf(err, "failed to execute template %q v", layout))
}
return template.HTML(res), nil
}
func (p *pageContentOutput) initRenderHooks() error {
if p == nil {
return nil
}
p.renderHooks.init.Do(func() {
if p.p.pageOutputTemplateVariationsState.Load() == 0 {
p.p.pageOutputTemplateVariationsState.Store(1)
}
type cacheKey struct {
tp hooks.RendererType
id any
f output.Format
}
renderCache := make(map[cacheKey]any)
var renderCacheMu sync.Mutex
resolvePosition := func(ctx any) text.Position {
var offset int
switch v := ctx.(type) {
case hooks.CodeblockContext:
offset = bytes.Index(p.p.source.parsed.Input(), []byte(v.Inner()))
}
pos := p.p.posFromInput(p.p.source.parsed.Input(), offset)
if pos.LineNumber > 0 {
// Move up to the code fence delimiter.
// This is in line with how we report on shortcodes.
pos.LineNumber = pos.LineNumber - 1
}
return pos
}
p.renderHooks.getRenderer = func(tp hooks.RendererType, id any) any {
renderCacheMu.Lock()
defer renderCacheMu.Unlock()
key := cacheKey{tp: tp, id: id, f: p.f}
if r, ok := renderCache[key]; ok {
return r
}
layoutDescriptor := p.p.getLayoutDescriptor()
layoutDescriptor.RenderingHook = true
layoutDescriptor.LayoutOverride = false
layoutDescriptor.Layout = ""
switch tp {
case hooks.LinkRendererType:
layoutDescriptor.Kind = "render-link"
case hooks.ImageRendererType:
layoutDescriptor.Kind = "render-image"
case hooks.HeadingRendererType:
layoutDescriptor.Kind = "render-heading"
case hooks.CodeBlockRendererType:
layoutDescriptor.Kind = "render-codeblock"
if id != nil {
lang := id.(string)
lexer := lexers.Get(lang)
if lexer != nil {
layoutDescriptor.KindVariants = strings.Join(lexer.Config().Aliases, ",")
} else {
layoutDescriptor.KindVariants = lang
}
}
}
getHookTemplate := func(f output.Format) (tpl.Template, bool) {
templ, found, err := p.p.s.Tmpl().LookupLayout(layoutDescriptor, f)
if err != nil {
panic(err)
}
return templ, found
}
templ, found1 := getHookTemplate(p.f)
if p.p.reusePageOutputContent() {
// Check if some of the other output formats would give a different template.
for _, f := range p.p.s.renderFormats {
if f.Name == p.f.Name {
continue
}
templ2, found2 := getHookTemplate(f)
if found2 {
if !found1 {
templ = templ2
found1 = true
break
}
if templ != templ2 {
p.p.pageOutputTemplateVariationsState.Store(2)
break
}
}
}
}
if !found1 {
if tp == hooks.CodeBlockRendererType {
// No user provided tempplate for code blocks, so we use the native Go code version -- which is also faster.
r := p.p.s.ContentSpec.Converters.GetHighlighter()
renderCache[key] = r
return r
}
return nil
}
r := hookRendererTemplate{
templateHandler: p.p.s.Tmpl(),
SearchProvider: templ.(identity.SearchProvider),
templ: templ,
resolvePosition: resolvePosition,
}
renderCache[key] = r
return r
}
})
return nil
}
func (p *pageContentOutput) setAutoSummary() error {
if p.p.source.hasSummaryDivider || p.p.m.summary != "" {
return nil
}
var summary string
var truncated bool
if p.p.m.isCJKLanguage {
summary, truncated = p.p.s.ContentSpec.TruncateWordsByRune(p.plainWords)
} else {
summary, truncated = p.p.s.ContentSpec.TruncateWordsToWholeSentence(p.plain)
}
p.summary = template.HTML(summary)
p.truncated = truncated
return nil
}
func (cp *pageContentOutput) renderContent(content []byte, renderTOC bool) (converter.Result, error) {
if err := cp.initRenderHooks(); err != nil {
return nil, err
}
c := cp.p.getContentConverter()
return cp.renderContentWithConverter(c, content, renderTOC)
}
func (cp *pageContentOutput) renderContentWithConverter(c converter.Converter, content []byte, renderTOC bool) (converter.Result, error) {
r, err := c.Convert(
converter.RenderContext{
Src: content,
RenderTOC: renderTOC,
GetRenderer: cp.renderHooks.getRenderer,
})
if err == nil {
if ids, ok := r.(identity.IdentitiesProvider); ok {
for _, v := range ids.GetIdentities() {
cp.trackDependency(v)
}
}
}
return r, err
}
func (p *pageContentOutput) setWordCounts(isCJKLanguage bool) {
if isCJKLanguage {
p.wordCount = 0
for _, word := range p.plainWords {
runeCount := utf8.RuneCountInString(word)
if len(word) == runeCount {
p.wordCount++
} else {
p.wordCount += runeCount
}
}
} else {
p.wordCount = helpers.TotalWords(p.plain)
}
// TODO(bep) is set in a test. Fix that.
if p.fuzzyWordCount == 0 {
p.fuzzyWordCount = (p.wordCount + 100) / 100 * 100
}
if isCJKLanguage {
p.readingTime = (p.wordCount + 500) / 501
} else {
p.readingTime = (p.wordCount + 212) / 213
}
}
// A callback to signal that we have inserted a placeholder into the rendered
// content. This avoids doing extra replacement work.
func (p *pageContentOutput) enablePlaceholders() {
p.placeholdersEnabledInit.Do(func() {
p.placeholdersEnabled = true
})
}
// these will be shifted out when rendering a given output format.
type pagePerOutputProviders interface {
targetPather
page.PaginatorProvider
resource.ResourceLinksProvider
}
type targetPather interface {
targetPaths() page.TargetPaths
}
type targetPathsHolder struct {
paths page.TargetPaths
page.OutputFormat
}
func (t targetPathsHolder) targetPaths() page.TargetPaths {
return t.paths
}
func executeToString(h tpl.TemplateHandler, templ tpl.Template, data any) (string, error) {
b := bp.GetBuffer()
defer bp.PutBuffer(b)
if err := h.Execute(templ, b, data); err != nil {
return "", err
}
return b.String(), nil
}
func splitUserDefinedSummaryAndContent(markup string, c []byte) (summary []byte, content []byte, err error) {
defer func() {
if r := recover(); r != nil {
err = fmt.Errorf("summary split failed: %s", r)
}
}()
startDivider := bytes.Index(c, internalSummaryDividerBaseBytes)
if startDivider == -1 {
return
}
startTag := "p"
switch markup {
Rework external asciidoctor integration This commit solves the relative path problem with asciidoctor tooling. An include will resolve relatively, so you can refer easily to files in the same folder. Also `asciidoctor-diagram` and PlantUML rendering works now, because the created temporary files will be placed in the correct folder. This patch covers just the Ruby version of asciidoctor. The old AsciiDoc CLI EOLs in Jan 2020, so this variant is removed from code. The configuration is completely rewritten and now available in `config.toml` under the key `[markup.asciidocext]`: ```toml [markup.asciidocext] extensions = ["asciidoctor-html5s", "asciidoctor-diagram"] workingFolderCurrent = true trace = true [markup.asciidocext.attributes] my-base-url = "https://example.com/" my-attribute-name = "my value" ``` - backends, safe-modes, and extensions are now whitelisted to the popular (ruby) extensions and valid values. - the default for extensions is to not enable any, because they're all external dependencies so the build would break if the user didn't install them beforehand. - the default backend is html5 because html5s is an external gem dependency. - the default safe-mode is safe, explanations of the modes: https://asciidoctor.org/man/asciidoctor/ - the config is namespaced under asciidocext_config and the parser looks at asciidocext to allow a future native Go asciidoc. - `uglyUrls=true` option and `--source` flag are supported - `--destination` flag is required Follow the updated documentation under `docs/content/en/content-management/formats.md`. This patch would be a breaking change, because you need to correct all your absolute include pathes to relative paths, so using relative paths must be configured explicitly by setting `workingFolderCurrent = true`.
2020-06-25 03:51:33 -04:00
case "asciidocext":
startTag = "div"
}
// Walk back and forward to the surrounding tags.
start := bytes.LastIndex(c[:startDivider], []byte("<"+startTag))
end := bytes.Index(c[startDivider:], []byte("</"+startTag))
if start == -1 {
start = startDivider
} else {
start = startDivider - (startDivider - start)
}
if end == -1 {
end = startDivider + len(internalSummaryDividerBase)
} else {
end = startDivider + end + len(startTag) + 3
}
var addDiv bool
switch markup {
case "rst":
addDiv = true
}
withoutDivider := append(c[:start], bytes.Trim(c[end:], "\n")...)
if len(withoutDivider) > 0 {
summary = bytes.TrimSpace(withoutDivider[:start])
}
if addDiv {
// For the rst
summary = append(append([]byte(nil), summary...), []byte("</div>")...)
}
if err != nil {
return
}
content = bytes.TrimSpace(withoutDivider)
return
}