refactor(parser): default subs on paragraphs and delimited blocks (by…

…tesparadise#688) Splitting the parsing in 3 stages: - the first parsing phase returns a `RawDocument` (new type) which contains `RawLine` elements as the content of delimited blocks, standalone paragraphs and paragraphs in continued list items (but NOT in list items) - the second phase applies all delimited block, paragraph and attribute substitutions and produces a `DraftDocument` (which now holds some attributes at its top-level) - the final phase takes care about sections, lists, preamble and table of contents (as before) Also: - using the new "MatchInlineElements" to check inline elements - paragraphs lines are now `[]interface{}` instead of `[][]interface{}`, which allows for containing a singular `types.RawLine` after the first parsing phase. Fixes bytesparadise#633 Signed-off-by: Xavier Coulon <[email protected]>
gdamore · Jul 5, 2020 · fe6293f · fe6293f
1 parent e6437ac
commit fe6293f
Show file tree

Hide file tree

Showing 82 changed files with 15,794 additions and 17,303 deletions.
diff --git a/go.mod b/go.mod
@@ -7,21 +7,23 @@ require (
 	github.com/davecgh/go-spew v1.1.1
 	github.com/golang/protobuf v1.3.2 // indirect
 	github.com/kr/text v0.2.0 // indirect
-	github.com/mattn/go-isatty v0.0.8 // indirect
+	github.com/mattn/go-isatty v0.0.12 // indirect
 	github.com/mna/pigeon v1.0.1-0.20200224192238-18953b277063
 	github.com/modocache/gover v0.0.0-20171022184752-b58185e213c5 // indirect
 	github.com/niemeyer/pretty v0.0.0-20200227124842-a10e7caefd8e // indirect
 	github.com/onsi/ginkgo v1.12.0
 	github.com/onsi/gomega v1.9.0
-	github.com/pkg/errors v0.8.1
+	github.com/pkg/errors v0.9.1
 	github.com/sergi/go-diff v1.0.0
-	github.com/sirupsen/logrus v1.4.2
+	github.com/sirupsen/logrus v1.6.0
 	github.com/sozorogami/gover v0.0.0-20171022184752-b58185e213c5
-	github.com/spf13/cobra v0.0.5
+	github.com/spf13/cobra v1.0.0
 	github.com/spf13/pflag v1.0.5 // indirect
-	github.com/stretchr/testify v1.5.1
+	github.com/stretchr/testify v1.6.1
+	golang.org/x/mod v0.3.0 // indirect
+	golang.org/x/net v0.0.0-20200324143707-d3edc9973b7e // indirect
 	golang.org/x/text v0.3.2 // indirect
-	golang.org/x/tools v0.0.0-20200502202811-ed308ab3e770 // indirect
+	golang.org/x/tools v0.0.0-20200702044944-0cc1aa72b347 // indirect
 	gopkg.in/check.v1 v1.0.0-20200227125254-8fa46927fb4f // indirect
-	gopkg.in/yaml.v2 v2.2.8
+	gopkg.in/yaml.v2 v2.3.0
 )
diff --git a/go.sum b/go.sum
diff --git a/libasciidoc.go b/libasciidoc.go
@@ -67,7 +67,7 @@ func Convert(r io.Reader, output io.Writer, config configuration.Configuration)
 		log.Debugf("rendered the output in %v", duration)
 	}()
 	log.Debugf("parsing the asciidoc source...")
-	doc, err := parser.ParseDocument(r, config) //, parser.Debug(true))
+	doc, err := parser.ParseDocument(r, config)
 	if err != nil {
 		return types.Metadata{}, err
 	}

diff --git a/make/go.mk b/make/go.mk
@@ -42,7 +42,7 @@ generate: install-pigeon
 generate-optimized: install-pigeon
 	@echo "generating the parser (optimized)..."
 	@pigeon -optimize-parser \
-		-alternate-entrypoints AsciidocDocument,VerbatimDocument,TextDocument,DocumentBlock,FileLocation,IncludedFileLine,InlineLinks,LabeledListItemTerm,NormalBlockContent,VerseBlockContent,MarkdownQuoteBlockAttribution \
+		-alternate-entrypoints AsciidocRawDocument,RawFile,TextDocument,DocumentRawBlock,FileLocation,IncludedFileLine,InlineLinks,LabeledListItemTerm,NormalBlockContent,NormalParagraphContent,VerseBlockContent,MarkdownQuoteBlockAttribution,InlineElements \
 		-o ./pkg/parser/parser.go ./pkg/parser/parser.peg
 
 .PHONY: build

diff --git a/pkg/parser/attributes_test.go b/pkg/parser/attributes_test.go
@@ -18,6 +18,9 @@ var _ = Describe("attributes", func() {
 			expected := types.DraftDocument{
 				Blocks: []interface{}{
 					types.ImageBlock{
+						Attributes: types.Attributes{
+							types.AttrImageAlt: "foo",
+						},
 						Location: types.Location{
 							Path: []interface{}{
 								types.StringElement{Content: "foo.png"},
@@ -33,6 +36,9 @@ var _ = Describe("attributes", func() {
 			expected := types.DraftDocument{
 				Blocks: []interface{}{
 					types.ImageBlock{
+						Attributes: types.Attributes{
+							types.AttrImageAlt: "foo",
+						},
 						Location: types.Location{
 							Path: []interface{}{
 								types.StringElement{Content: "foo.png"},
@@ -49,6 +55,9 @@ var _ = Describe("attributes", func() {
 			expected := types.DraftDocument{
 				Blocks: []interface{}{
 					types.ImageBlock{
+						Attributes: types.Attributes{
+							types.AttrImageAlt: "foo",
+						},
 						Location: types.Location{
 							Path: []interface{}{
 								types.StringElement{Content: "foo.png"},
@@ -66,7 +75,8 @@ var _ = Describe("attributes", func() {
 				Blocks: []interface{}{
 					types.ImageBlock{
 						Attributes: types.Attributes{
-							types.AttrWidth: "200",
+							types.AttrImageAlt: "foo",
+							types.AttrWidth:    "200",
 						},
 						Location: types.Location{
 							Path: []interface{}{

diff --git a/pkg/parser/blank_line_test.go b/pkg/parser/blank_line_test.go
@@ -8,61 +8,127 @@ import (
 	. "github.com/onsi/gomega" //nolint golint
 )
 
-var _ = Describe("blank lines - draft", func() {
-	It("blank line between 2 paragraphs", func() {
-		source := `first paragraph
+var _ = Describe("blank lines", func() {
+
+	Context("draft document", func() {
+
+		It("blank line between 2 paragraphs", func() {
+			source := `first paragraph
  
 second paragraph`
-		expected := types.DraftDocument{
-			Blocks: []interface{}{
-				types.Paragraph{
-					Lines: [][]interface{}{
-						{
-							types.StringElement{Content: "first paragraph"},
+			expected := types.DraftDocument{
+				Blocks: []interface{}{
+					types.Paragraph{
+						Lines: []interface{}{
+							[]interface{}{
+								types.StringElement{
+									Content: "first paragraph",
+								},
+							},
+						},
+					},
+					types.BlankLine{},
+					types.Paragraph{
+						Lines: []interface{}{
+							[]interface{}{
+								types.StringElement{
+									Content: "second paragraph",
+								},
+							},
 						},
 					},
 				},
-				types.BlankLine{},
-				types.Paragraph{
-					Lines: [][]interface{}{
-						{
-							types.StringElement{Content: "second paragraph"},
+			}
+			Expect(ParseDraftDocument(source)).To(MatchDraftDocument(expected))
+		})
+		It("blank line with spaces and tabs between 2 paragraphs and after second paragraph", func() {
+			source := `first paragraph
+		 
+
+		
+second paragraph
+`
+			expected := types.DraftDocument{
+				Blocks: []interface{}{
+					types.Paragraph{
+						Lines: []interface{}{
+							[]interface{}{
+								types.StringElement{
+									Content: "first paragraph",
+								},
+							},
+						},
+					},
+					types.BlankLine{},
+					types.BlankLine{},
+					types.BlankLine{},
+					types.Paragraph{
+						Lines: []interface{}{
+							[]interface{}{
+								types.StringElement{
+									Content: "second paragraph",
+								},
+							},
 						},
 					},
 				},
-			},
-		}
-		Expect(ParseDraftDocument(source)).To(MatchDraftDocument(expected))
+			}
+			Expect(ParseDraftDocument(source)).To(MatchDraftDocument(expected))
+		})
 	})
-	It("blank line with spaces and tabs between 2 paragraphs and after second paragraph", func() {
-		source := `first paragraph
+
+	Context("final document", func() {
+
+		It("blank line between 2 paragraphs", func() {
+			source := `first paragraph
+ 
+second paragraph`
+			expected := types.Document{
+				Elements: []interface{}{
+					types.Paragraph{
+						Lines: []interface{}{
+							[]interface{}{
+								types.StringElement{Content: "first paragraph"},
+							},
+						},
+					},
+					types.Paragraph{
+						Lines: []interface{}{
+							[]interface{}{
+								types.StringElement{Content: "second paragraph"},
+							},
+						},
+					},
+				},
+			}
+			Expect(ParseDocument(source)).To(MatchDocument(expected))
+		})
+		It("blank line with spaces and tabs between 2 paragraphs and after second paragraph", func() {
+			source := `first paragraph
 		 
 
 		
 second paragraph
 `
-		expected := types.DraftDocument{
-			Blocks: []interface{}{
-				types.Paragraph{
-					Lines: [][]interface{}{
-						{
-							types.StringElement{Content: "first paragraph"},
+			expected := types.Document{
+				Elements: []interface{}{
+					types.Paragraph{
+						Lines: []interface{}{
+							[]interface{}{
+								types.StringElement{Content: "first paragraph"},
+							},
 						},
 					},
-				},
-				types.BlankLine{},
-				types.BlankLine{},
-				types.BlankLine{},
-				types.Paragraph{
-					Lines: [][]interface{}{
-						{
-							types.StringElement{Content: "second paragraph"},
+					types.Paragraph{
+						Lines: []interface{}{
+							[]interface{}{
+								types.StringElement{Content: "second paragraph"},
+							},
 						},
 					},
 				},
-			},
-		}
-		Expect(ParseDraftDocument(source)).To(MatchDraftDocument(expected))
+			}
+			Expect(ParseDocument(source)).To(MatchDocument(expected))
+		})
 	})
-
 })