From 90da7664bf1f3a0ca2e18144b5deacf532c6e3cf Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Bj=C3=B8rn=20Erik=20Pedersen?= Date: Sat, 11 Feb 2023 16:20:24 +0100 Subject: Add page fragments support to Related The main topic of this commit is that you can now index fragments (content heading identifiers) when calling `.Related`. You can do this by: * Configure one or more indices with type `fragments` * The name of those index configurations maps to an (optional) front matter slice with fragment references. This allows you to link page<->fragment and page<->page. * This also will index all the fragments (heading identifiers) of the pages. It's also possible to use type `fragments` indices in shortcode, e.g.: ``` {{ $related := site.RegularPages.Related .Page }} ``` But, and this is important, you need to include the shortcode using the `{{<` delimiter. Not doing so will create infinite loops and timeouts. This commit also: * Adds two new methods to Page: Fragments (can also be used to build ToC) and HeadingsFiltered (this is only used in Related Content with index type `fragments` and `enableFilter` set to true. * Consolidates all `.Related*` methods into one, which takes either a `Page` or an options map as its only argument. * Add `context.Context` to all of the content related Page API. Turns out it wasn't strictly needed for this particular feature, but it will soon become usefil, e.g. in #9339. Closes #10711 Updates #9339 Updates #10725 --- markup/asciidocext/convert.go | 26 +++--- markup/asciidocext/convert_test.go | 82 ++---------------- markup/converter/converter.go | 2 +- markup/goldmark/convert.go | 10 +-- markup/goldmark/toc.go | 10 +-- markup/tableofcontents/tableofcontents.go | 110 +++++++++++++++++++++---- markup/tableofcontents/tableofcontents_test.go | 85 ++++++++++++++++--- 7 files changed, 199 insertions(+), 126 deletions(-) (limited to 'markup') diff --git a/markup/asciidocext/convert.go b/markup/asciidocext/convert.go index 4c83e0e95..c9524778f 100644 --- a/markup/asciidocext/convert.go +++ b/markup/asciidocext/convert.go @@ -53,10 +53,10 @@ func (p provider) New(cfg converter.ProviderConfig) (converter.Provider, error) type asciidocResult struct { converter.Result - toc tableofcontents.Root + toc *tableofcontents.Fragments } -func (r asciidocResult) TableOfContents() tableofcontents.Root { +func (r asciidocResult) TableOfContents() *tableofcontents.Fragments { return r.toc } @@ -205,16 +205,16 @@ func hasAsciiDoc() bool { // extractTOC extracts the toc from the given src html. // It returns the html without the TOC, and the TOC data -func (a *asciidocConverter) extractTOC(src []byte) ([]byte, tableofcontents.Root, error) { +func (a *asciidocConverter) extractTOC(src []byte) ([]byte, *tableofcontents.Fragments, error) { var buf bytes.Buffer buf.Write(src) node, err := html.Parse(&buf) if err != nil { - return nil, tableofcontents.Root{}, err + return nil, nil, err } var ( f func(*html.Node) bool - toc tableofcontents.Root + toc *tableofcontents.Fragments toVisit []*html.Node ) f = func(n *html.Node) bool { @@ -242,12 +242,12 @@ func (a *asciidocConverter) extractTOC(src []byte) ([]byte, tableofcontents.Root } f(node) if err != nil { - return nil, tableofcontents.Root{}, err + return nil, nil, err } buf.Reset() err = html.Render(&buf, node) if err != nil { - return nil, tableofcontents.Root{}, err + return nil, nil, err } // ltrim and rtrim which are added by html.Render res := buf.Bytes()[25:] @@ -256,9 +256,9 @@ func (a *asciidocConverter) extractTOC(src []byte) ([]byte, tableofcontents.Root } // parseTOC returns a TOC root from the given toc Node -func parseTOC(doc *html.Node) tableofcontents.Root { +func parseTOC(doc *html.Node) *tableofcontents.Fragments { var ( - toc tableofcontents.Root + toc tableofcontents.Builder f func(*html.Node, int, int) ) f = func(n *html.Node, row, level int) { @@ -276,9 +276,9 @@ func parseTOC(doc *html.Node) tableofcontents.Root { continue } href := attr(c, "href")[1:] - toc.AddAt(tableofcontents.Heading{ - Text: nodeContent(c), - ID: href, + toc.AddAt(&tableofcontents.Heading{ + Title: nodeContent(c), + ID: href, }, row, level) } f(n.FirstChild, row, level) @@ -289,7 +289,7 @@ func parseTOC(doc *html.Node) tableofcontents.Root { } } f(doc.FirstChild, -1, 0) - return toc + return toc.Build() } func attr(node *html.Node, key string) string { diff --git a/markup/asciidocext/convert_test.go b/markup/asciidocext/convert_test.go index 3a350c5ce..47208c066 100644 --- a/markup/asciidocext/convert_test.go +++ b/markup/asciidocext/convert_test.go @@ -21,13 +21,13 @@ import ( "path/filepath" "testing" + "github.com/gohugoio/hugo/common/collections" "github.com/gohugoio/hugo/common/hexec" "github.com/gohugoio/hugo/common/loggers" "github.com/gohugoio/hugo/config" "github.com/gohugoio/hugo/config/security" "github.com/gohugoio/hugo/markup/converter" "github.com/gohugoio/hugo/markup/markup_config" - "github.com/gohugoio/hugo/markup/tableofcontents" qt "github.com/frankban/quicktest" ) @@ -343,49 +343,8 @@ testContent c.Assert(err, qt.IsNil) toc, ok := r.(converter.TableOfContentsProvider) c.Assert(ok, qt.Equals, true) - expected := tableofcontents.Root{ - Headings: tableofcontents.Headings{ - { - ID: "", - Text: "", - Headings: tableofcontents.Headings{ - { - ID: "_introduction", - Text: "Introduction", - Headings: nil, - }, - { - ID: "_section_1", - Text: "Section 1", - Headings: tableofcontents.Headings{ - { - ID: "_section_1_1", - Text: "Section 1.1", - Headings: tableofcontents.Headings{ - { - ID: "_section_1_1_1", - Text: "Section 1.1.1", - Headings: nil, - }, - }, - }, - { - ID: "_section_1_2", - Text: "Section 1.2", - Headings: nil, - }, - }, - }, - { - ID: "_section_2", - Text: "Section 2", - Headings: nil, - }, - }, - }, - }, - } - c.Assert(toc.TableOfContents(), qt.DeepEquals, expected) + + c.Assert(toc.TableOfContents().Identifiers, qt.DeepEquals, collections.SortedStringSlice{"_introduction", "_section_1", "_section_1_1", "_section_1_1_1", "_section_1_2", "_section_2"}) c.Assert(string(r.Bytes()), qt.Not(qt.Contains), "
") } @@ -404,22 +363,7 @@ func TestTableOfContentsWithCode(t *testing.T) { c.Assert(err, qt.IsNil) toc, ok := r.(converter.TableOfContentsProvider) c.Assert(ok, qt.Equals, true) - expected := tableofcontents.Root{ - Headings: tableofcontents.Headings{ - { - ID: "", - Text: "", - Headings: tableofcontents.Headings{ - { - ID: "_some_code_in_the_title", - Text: "Some code in the title", - Headings: nil, - }, - }, - }, - }, - } - c.Assert(toc.TableOfContents(), qt.DeepEquals, expected) + c.Assert(toc.TableOfContents().HeadingsMap["_some_code_in_the_title"].Title, qt.Equals, "Some code in the title") c.Assert(string(r.Bytes()), qt.Not(qt.Contains), "
") } @@ -443,21 +387,7 @@ func TestTableOfContentsPreserveTOC(t *testing.T) { c.Assert(err, qt.IsNil) toc, ok := r.(converter.TableOfContentsProvider) c.Assert(ok, qt.Equals, true) - expected := tableofcontents.Root{ - Headings: tableofcontents.Headings{ - { - ID: "", - Text: "", - Headings: tableofcontents.Headings{ - { - ID: "some-title", - Text: "Some title", - Headings: nil, - }, - }, - }, - }, - } - c.Assert(toc.TableOfContents(), qt.DeepEquals, expected) + + c.Assert(toc.TableOfContents().Identifiers, qt.DeepEquals, collections.SortedStringSlice{"some-title"}) c.Assert(string(r.Bytes()), qt.Contains, "
") } diff --git a/markup/converter/converter.go b/markup/converter/converter.go index c760381f4..7e5b56b07 100644 --- a/markup/converter/converter.go +++ b/markup/converter/converter.go @@ -101,7 +101,7 @@ type DocumentInfo interface { // TableOfContentsProvider provides the content as a ToC structure. type TableOfContentsProvider interface { - TableOfContents() tableofcontents.Root + TableOfContents() *tableofcontents.Fragments } // AnchorNameSanitizer tells how a converter sanitizes anchor names. diff --git a/markup/goldmark/convert.go b/markup/goldmark/convert.go index a179cd233..6c1c7ad0a 100644 --- a/markup/goldmark/convert.go +++ b/markup/goldmark/convert.go @@ -160,11 +160,11 @@ var _ identity.IdentitiesProvider = (*converterResult)(nil) type converterResult struct { converter.Result - toc tableofcontents.Root + toc *tableofcontents.Fragments ids identity.Identities } -func (c converterResult) TableOfContents() tableofcontents.Root { +func (c converterResult) TableOfContents() *tableofcontents.Fragments { return c.toc } @@ -228,9 +228,9 @@ type parserContext struct { parser.Context } -func (p *parserContext) TableOfContents() tableofcontents.Root { +func (p *parserContext) TableOfContents() *tableofcontents.Fragments { if v := p.Get(tocResultKey); v != nil { - return v.(tableofcontents.Root) + return v.(*tableofcontents.Fragments) } - return tableofcontents.Root{} + return nil } diff --git a/markup/goldmark/toc.go b/markup/goldmark/toc.go index 396c1d071..ac5040e85 100644 --- a/markup/goldmark/toc.go +++ b/markup/goldmark/toc.go @@ -41,8 +41,8 @@ func (t *tocTransformer) Transform(n *ast.Document, reader text.Reader, pc parse } var ( - toc tableofcontents.Root - tocHeading tableofcontents.Heading + toc tableofcontents.Builder + tocHeading = &tableofcontents.Heading{} level int row = -1 inHeading bool @@ -53,10 +53,10 @@ func (t *tocTransformer) Transform(n *ast.Document, reader text.Reader, pc parse s := ast.WalkStatus(ast.WalkContinue) if n.Kind() == ast.KindHeading { if inHeading && !entering { - tocHeading.Text = headingText.String() + tocHeading.Title = headingText.String() headingText.Reset() toc.AddAt(tocHeading, row, level-1) - tocHeading = tableofcontents.Heading{} + tocHeading = &tableofcontents.Heading{} inHeading = false return s, nil } @@ -106,7 +106,7 @@ func (t *tocTransformer) Transform(n *ast.Document, reader text.Reader, pc parse return s, nil }) - pc.Set(tocResultKey, toc) + pc.Set(tocResultKey, toc.Build()) } type tocExtension struct { diff --git a/markup/tableofcontents/tableofcontents.go b/markup/tableofcontents/tableofcontents.go index 2e7f47d20..bd0aaa801 100644 --- a/markup/tableofcontents/tableofcontents.go +++ b/markup/tableofcontents/tableofcontents.go @@ -14,35 +14,104 @@ package tableofcontents import ( + "sort" "strings" + + "github.com/gohugoio/hugo/common/collections" ) +// Empty is an empty ToC. +var Empty = &Fragments{ + Headings: Headings{}, + HeadingsMap: map[string]*Heading{}, +} + +// Builder is used to build the ToC data structure. +type Builder struct { + toc *Fragments +} + +// Add adds the heading to the ToC. +func (b *Builder) AddAt(h *Heading, row, level int) { + if b.toc == nil { + b.toc = &Fragments{} + } + b.toc.addAt(h, row, level) +} + +// Build returns the ToC. +func (b Builder) Build() *Fragments { + if b.toc == nil { + return Empty + } + b.toc.HeadingsMap = make(map[string]*Heading) + b.toc.walk(func(h *Heading) { + if h.ID != "" { + b.toc.HeadingsMap[h.ID] = h + b.toc.Identifiers = append(b.toc.Identifiers, h.ID) + } + }) + sort.Strings(b.toc.Identifiers) + return b.toc +} + // Headings holds the top level headings. -type Headings []Heading +type Headings []*Heading + +// FilterBy returns a new Headings slice with all headings that matches the given predicate. +// For internal use only. +func (h Headings) FilterBy(fn func(*Heading) bool) Headings { + var out Headings + + for _, h := range h { + h.walk(func(h *Heading) { + if fn(h) { + out = append(out, h) + } + }) + } + return out +} // Heading holds the data about a heading and its children. type Heading struct { - ID string - Text string + ID string + Title string Headings Headings } // IsZero is true when no ID or Text is set. func (h Heading) IsZero() bool { - return h.ID == "" && h.Text == "" + return h.ID == "" && h.Title == "" +} + +func (h *Heading) walk(fn func(*Heading)) { + fn(h) + for _, h := range h.Headings { + h.walk(fn) + } } -// Root implements AddAt, which can be used to build the -// data structure for the ToC. -type Root struct { +// Fragments holds the table of contents for a page. +type Fragments struct { + // Headings holds the top level headings. Headings Headings + + // Identifiers holds all the identifiers in the ToC as a sorted slice. + // Note that collections.SortedStringSlice has both a Contains and Count method + // that can be used to identify missing and duplicate IDs. + Identifiers collections.SortedStringSlice + + // HeadingsMap holds all the headings in the ToC as a map. + // Note that with duplicate IDs, the last one will win. + HeadingsMap map[string]*Heading } -// AddAt adds the heading into the given location. -func (toc *Root) AddAt(h Heading, row, level int) { +// addAt adds the heading into the given location. +func (toc *Fragments) addAt(h *Heading, row, level int) { for i := len(toc.Headings); i <= row; i++ { - toc.Headings = append(toc.Headings, Heading{}) + toc.Headings = append(toc.Headings, &Heading{}) } if level == 0 { @@ -50,19 +119,22 @@ func (toc *Root) AddAt(h Heading, row, level int) { return } - heading := &toc.Headings[row] + heading := toc.Headings[row] for i := 1; i < level; i++ { if len(heading.Headings) == 0 { - heading.Headings = append(heading.Headings, Heading{}) + heading.Headings = append(heading.Headings, &Heading{}) } - heading = &heading.Headings[len(heading.Headings)-1] + heading = heading.Headings[len(heading.Headings)-1] } heading.Headings = append(heading.Headings, h) } // ToHTML renders the ToC as HTML. -func (toc Root) ToHTML(startLevel, stopLevel int, ordered bool) string { +func (toc *Fragments) ToHTML(startLevel, stopLevel int, ordered bool) string { + if toc == nil { + return "" + } b := &tocBuilder{ s: strings.Builder{}, h: toc.Headings, @@ -74,6 +146,12 @@ func (toc Root) ToHTML(startLevel, stopLevel int, ordered bool) string { return b.s.String() } +func (toc Fragments) walk(fn func(*Heading)) { + for _, h := range toc.Headings { + h.walk(fn) + } +} + type tocBuilder struct { s strings.Builder h Headings @@ -133,11 +211,11 @@ func (b *tocBuilder) writeHeadings(level, indent int, h Headings) { } } -func (b *tocBuilder) writeHeading(level, indent int, h Heading) { +func (b *tocBuilder) writeHeading(level, indent int, h *Heading) { b.indent(indent) b.s.WriteString("
  • ") if !h.IsZero() { - b.s.WriteString("" + h.Text + "") + b.s.WriteString("" + h.Title + "") } b.writeHeadings(level, indent, h.Headings) b.s.WriteString("
  • \n") diff --git a/markup/tableofcontents/tableofcontents_test.go b/markup/tableofcontents/tableofcontents_test.go index daeb9f991..adbda4b00 100644 --- a/markup/tableofcontents/tableofcontents_test.go +++ b/markup/tableofcontents/tableofcontents_test.go @@ -17,18 +17,33 @@ import ( "testing" qt "github.com/frankban/quicktest" + "github.com/gohugoio/hugo/common/collections" ) +var newTestTocBuilder = func() Builder { + var b Builder + b.AddAt(&Heading{Title: "Heading 1", ID: "h1-1"}, 0, 0) + b.AddAt(&Heading{Title: "1-H2-1", ID: "1-h2-1"}, 0, 1) + b.AddAt(&Heading{Title: "1-H2-2", ID: "1-h2-2"}, 0, 1) + b.AddAt(&Heading{Title: "1-H3-1", ID: "1-h2-2"}, 0, 2) + b.AddAt(&Heading{Title: "Heading 2", ID: "h1-2"}, 1, 0) + return b +} + +var newTestToc = func() *Fragments { + return newTestTocBuilder().Build() +} + func TestToc(t *testing.T) { c := qt.New(t) - toc := &Root{} + toc := &Fragments{} - toc.AddAt(Heading{Text: "Heading 1", ID: "h1-1"}, 0, 0) - toc.AddAt(Heading{Text: "1-H2-1", ID: "1-h2-1"}, 0, 1) - toc.AddAt(Heading{Text: "1-H2-2", ID: "1-h2-2"}, 0, 1) - toc.AddAt(Heading{Text: "1-H3-1", ID: "1-h2-2"}, 0, 2) - toc.AddAt(Heading{Text: "Heading 2", ID: "h1-2"}, 1, 0) + toc.addAt(&Heading{Title: "Heading 1", ID: "h1-1"}, 0, 0) + toc.addAt(&Heading{Title: "1-H2-1", ID: "1-h2-1"}, 0, 1) + toc.addAt(&Heading{Title: "1-H2-2", ID: "1-h2-2"}, 0, 1) + toc.addAt(&Heading{Title: "1-H3-1", ID: "1-h2-2"}, 0, 2) + toc.addAt(&Heading{Title: "Heading 2", ID: "h1-2"}, 1, 0) got := toc.ToHTML(1, -1, false) c.Assert(got, qt.Equals, `