feat(tidy-up): remove double spaces and save Markdown file

.golangci.yaml

@@ -10,8 +10,10 @@ linters:
     - gci
     - gochecknoinits
     - godox
+    - gofmt
+    - gofumpt
+    - goimports
     - intrange
     - ireturn
     - paralleltest
-    - testpackage
     # keep-sorted end

Makefile

@@ -16,6 +16,13 @@ build-go: # Build the Golang executable
 	$(MAKE) list-go
 .PHONY: build-go
 
+refresh-go: # Refresh the Go dependencies from the last commit
+	LAST_COMMIT=$$(cd ../logseq-go; git log -1 --format=%h); \
+		echo "LAST_COMMIT: $$LAST_COMMIT"; \
+		go get -u github.com/andreoliwa/logseq-go@$$LAST_COMMIT
+	go mod tidy
+.PHONY: refresh-go
+
 clean: # Clean the build artifacts
 	cargo clean
 	-rm `go env GOPATH`/bin/logseq-doctor

cmd/testdata/graph/pages/spaces.md

@@ -5,3 +5,5 @@
   - child 2
   - child 3
     - Here we have #[[some  tag with   spaces]]
+    - top level 2
+- [[some  page   title  with  spaces]]

cmd/testdata/graph/pages/spaces.md.golden

@@ -0,0 +1,9 @@
+- Regular   text with  spaces
+  - child
+    - nested [Link With Spaces ](https://example.com)
+- top level
+  - child 2
+  - child 3
+    - Here we have #[[some tag with spaces]]
+    - top level 2
+- [[some page title with spaces]]

cmd/tidy_up.go

@@ -5,6 +5,7 @@ import (
 	"fmt"
 	"log"
 	"os"
+	"regexp"
 	"sort"
 	"strings"
 
@@ -35,36 +36,50 @@ var tidyUpCmd = &cobra.Command{ //nolint:exhaustruct,gochecknoglobals
 			log.Fatalln("error opening graph: %w", err)
 		}
 
+		transaction := graph.NewTransaction()
+		shouldSave := false
+
 		exitCode := 0
 		for _, path := range args {
 			if !isValidMarkdownFile(path) {
 				fmt.Printf("%s: skipping, not a Markdown file\n", path)
 			} else {
-				page, err := graph.OpenViaPath(path)
+				page, err := transaction.OpenViaPath(path)
 				if err != nil {
 					log.Fatalf("%s: error opening file via path: %s\n", path, err)
 				}
 				if page == nil {
 					log.Fatalf("%s: error opening file via path: page is nil\n", path)
 				}
 
-				changes := make([]string, 0)
+				messages := make([]string, 0)
 
-				functions := []func(logseq.Page) string{checkForbiddenReferences, checkRunningTasks, checkDoubleSpaces}
+				functions := []func(logseq.Page) tidyInfo{
+					checkForbiddenReferences, checkRunningTasks, removeDoubleSpaces}
 				for _, f := range functions {
-					if msg := f(page); msg != "" {
-						changes = append(changes, msg)
+					info := f(page)
+					if info.msg != "" {
+						messages = append(messages, info.msg)
+					}
+					if info.changed {
+						shouldSave = true
 					}
 				}
 
-				if len(changes) > 0 {
+				if len(messages) > 0 {
 					exitCode = 1
-					for _, change := range changes {
-						fmt.Printf("%s: %s\n", path, change)
+					for _, msg := range messages {
+						fmt.Printf("%s: %s\n", path, msg)
 					}
 				}
 			}
 		}
+		if shouldSave {
+			err = transaction.Save()
+			if err != nil {
+				log.Fatalf("error saving transaction: %s\n", err)
+			}
+		}
 		os.Exit(exitCode)
 	},
 }
@@ -91,12 +106,17 @@ func isValidMarkdownFile(filePath string) bool {
 	return !info.IsDir()
 }
 
+type tidyInfo struct {
+	msg     string
+	changed bool
+}
+
 // checkForbiddenReferences checks if a page has forbidden references to other pages or tags.
-func checkForbiddenReferences(page logseq.Page) string {
+func checkForbiddenReferences(page logseq.Page) tidyInfo {
 	all := make([]string, 0)
 
 	for _, block := range page.Blocks() {
-		block.Children().FilterDeep(func(n content.Node) bool {
+		block.Children().FindDeep(func(n content.Node) bool {
 			var reference string
 			if pageLink, ok := n.(*content.PageLink); ok {
 				reference = pageLink.To
@@ -125,10 +145,11 @@ func checkForbiddenReferences(page logseq.Page) string {
 	if count := len(all); count > 0 {
 		unique := sortAndRemoveDuplicates(all)
 
-		return fmt.Sprintf("remove %d forbidden references to pages/tags: %s", count, strings.Join(unique, ", "))
+		return tidyInfo{fmt.Sprintf("remove %d forbidden references to pages/tags: %s",
+			count, strings.Join(unique, ", ")), false}
 	}
 
-	return ""
+	return tidyInfo{"", false}
 }
 
 func sortAndRemoveDuplicates(elements []string) []string {
@@ -149,11 +170,11 @@ func sortAndRemoveDuplicates(elements []string) []string {
 }
 
 // checkRunningTasks checks if a page has running tasks (DOING, etc.).
-func checkRunningTasks(page logseq.Page) string {
+func checkRunningTasks(page logseq.Page) tidyInfo {
 	all := make([]string, 0)
 
 	for _, block := range page.Blocks() {
-		block.Children().FilterDeep(func(n content.Node) bool {
+		block.Children().FindDeep(func(n content.Node) bool {
 			if task, ok := n.(*content.TaskMarker); ok {
 				status := task.Status
 				// TODO: convert to strings "DOING"/"IN-PROGRESS" in logseq-go
@@ -173,29 +194,41 @@ func checkRunningTasks(page logseq.Page) string {
 	if count := len(all); count > 0 {
 		unique := sortAndRemoveDuplicates(all)
 
-		return fmt.Sprintf("stop %d running task(s): %s", count, strings.Join(unique, ", "))
+		return tidyInfo{fmt.Sprintf("stop %d running task(s): %s", count, strings.Join(unique, ", ")), false}
 	}
 
-	return ""
+	return tidyInfo{"", false}
 }
 
-func checkDoubleSpaces(page logseq.Page) string {
+func removeDoubleSpaces(page logseq.Page) tidyInfo {
 	all := make([]string, 0)
+	doubleSpaceRegex := regexp.MustCompile(`\s{2,}`)
+	fixed := false
 
 	for _, block := range page.Blocks() {
-		block.Children().FilterDeep(func(node content.Node) bool {
-			var value string
+		block.Children().FindDeep(func(node content.Node) bool {
+			var oldValue string
 
 			if text, ok := node.(*content.Text); ok {
-				value = text.Value
+				oldValue = text.Value
 			} else if pageLink, ok := node.(*content.PageLink); ok {
-				value = pageLink.To
+				oldValue = pageLink.To
 			} else if tag, ok := node.(*content.Hashtag); ok {
-				value = tag.To
+				oldValue = tag.To
 			}
 
-			if strings.Contains(value, "  ") {
-				all = append(all, fmt.Sprintf("'%s'", value))
+			if strings.Contains(oldValue, "  ") {
+				all = append(all, fmt.Sprintf("'%s'", oldValue))
+				newValue := doubleSpaceRegex.ReplaceAllString(oldValue, " ")
+				fixed = true
+
+				if text, ok := node.(*content.Text); ok {
+					text.Value = newValue
+				} else if pageLink, ok := node.(*content.PageLink); ok {
+					pageLink.To = newValue
+				} else if tag, ok := node.(*content.Hashtag); ok {
+					tag.To = newValue
+				}
 			}
 
 			return false
@@ -205,8 +238,8 @@ func checkDoubleSpaces(page logseq.Page) string {
 	if count := len(all); count > 0 {
 		unique := sortAndRemoveDuplicates(all)
 
-		return fmt.Sprintf("%d double spaces: %s", count, strings.Join(unique, ", "))
+		return tidyInfo{fmt.Sprintf("fixed %d double spaces: %s", count, strings.Join(unique, ", ")), fixed}
 	}
 
-	return ""
+	return tidyInfo{"", fixed}
 }

cmd/tidy_up_test.go

@@ -1,4 +1,4 @@
-package cmd
+package cmd //nolint:testpackage
 
 import (
 	"context"
@@ -93,27 +93,34 @@ func setupPage(t *testing.T, name string) logseq.Page {
 
 func TestCheckForbiddenReferences(t *testing.T) {
 	invalid := setupPage(t, "forbidden")
-	assert.Equal(t, "remove 4 forbidden references to pages/tags: Inbox, quick capture",
+	assert.Equal(t, tidyInfo{"remove 4 forbidden references to pages/tags: Inbox, quick capture", false},
 		checkForbiddenReferences(invalid))
 
 	valid := setupPage(t, "valid")
-	assert.Equal(t, "", checkForbiddenReferences(valid))
+	assert.Equal(t, tidyInfo{"", false}, checkForbiddenReferences(valid))
 }
 
 func TestCheckRunningTasks(t *testing.T) {
 	invalid := setupPage(t, "running")
-	assert.Equal(t, "stop 2 running task(s): DOING, IN-PROGRESS", checkRunningTasks(invalid))
+	assert.Equal(t, tidyInfo{"stop 2 running task(s): DOING, IN-PROGRESS", false}, checkRunningTasks(invalid))
 
 	valid := setupPage(t, "valid")
-	assert.Equal(t, "", checkRunningTasks(valid))
+	assert.Equal(t, tidyInfo{"", false}, checkRunningTasks(valid))
 }
 
-func TestCheckDoubleSpaces(t *testing.T) {
+func TestRemoveDoubleSpaces(t *testing.T) {
 	invalid := setupPage(t, "spaces")
 	assert.Equal(t,
-		"3 double spaces: 'Link   With  Spaces  ', 'Regular   text with  spaces', 'some  tag with   spaces'",
-		checkDoubleSpaces(invalid))
+		tidyInfo{"fixed 4 double spaces: 'Link   With  Spaces  ', 'Regular   text with  spaces'," +
+			" 'some  page   title  with  spaces', 'some  tag with   spaces'", true},
+		removeDoubleSpaces(invalid))
+
+	// TODO: compare the saved Markdown file with spaces.md.golden.
+	//  I tested manually and it works but I need to do something like:
+	// actual := os.ReadFile(invalid.Path())
+	// expected := os.ReadFile(invalid.Path() + ".golden")
+	// assert.Equal(t, expected, actual)
 
 	valid := setupPage(t, "valid")
-	assert.Equal(t, "", checkDoubleSpaces(valid))
+	assert.Equal(t, tidyInfo{"", false}, removeDoubleSpaces(valid))
 }

go.mod

@@ -1,40 +1,44 @@
 module github.com/andreoliwa/logseq-doctor
 
-go 1.21
+go 1.23
+
+toolchain go1.23.4
 
 require (
-	github.com/andreoliwa/logseq-go v0.0.0-20241226195503-ec30920a9c1c
+	github.com/andreoliwa/logseq-go v0.0.0-20241228161444-32088a5a9491
 	github.com/spf13/cobra v1.8.1
 	github.com/stretchr/testify v1.10.0
 )
 
 require (
-	github.com/RoaringBitmap/roaring v0.9.4 // indirect
-	github.com/axiomhq/hyperloglog v0.0.0-20191112132149-a4c4c47bc57f // indirect
-	github.com/bits-and-blooms/bitset v1.2.0 // indirect
+	github.com/RoaringBitmap/roaring v1.9.4 // indirect
+	github.com/axiomhq/hyperloglog v0.2.2 // indirect
+	github.com/bits-and-blooms/bitset v1.20.0 // indirect
 	github.com/blevesearch/go-porterstemmer v1.0.3 // indirect
 	github.com/blevesearch/mmap-go v1.0.4 // indirect
-	github.com/blevesearch/segment v0.9.0 // indirect
+	github.com/blevesearch/segment v0.9.1 // indirect
 	github.com/blevesearch/snowballstem v0.9.0 // indirect
-	github.com/blevesearch/vellum v1.0.7 // indirect
+	github.com/blevesearch/vellum v1.1.0 // indirect
 	github.com/blugelabs/bluge v0.2.2 // indirect
 	github.com/blugelabs/bluge_segment_api v0.2.0 // indirect
 	github.com/blugelabs/ice v1.0.0 // indirect
 	github.com/blugelabs/ice/v2 v2.0.1 // indirect
 	github.com/caio/go-tdigest v3.1.0+incompatible // indirect
 	github.com/davecgh/go-spew v1.1.1 // indirect
-	github.com/dgryski/go-metro v0.0.0-20180109044635-280f6062b5bc // indirect
-	github.com/fsnotify/fsnotify v1.7.0 // indirect
-	github.com/golang/snappy v0.0.1 // indirect
-	github.com/google/uuid v1.5.0 // indirect
+	github.com/dgryski/go-metro v0.0.0-20211217172704-adc40b04c140 // indirect
+	github.com/fsnotify/fsnotify v1.8.0 // indirect
+	github.com/golang/snappy v0.0.4 // indirect
+	github.com/google/uuid v1.6.0 // indirect
 	github.com/inconshreveable/mousetrap v1.1.0 // indirect
-	github.com/klauspost/compress v1.15.2 // indirect
+	github.com/kamstrup/intmap v0.5.1 // indirect
+	github.com/klauspost/compress v1.17.11 // indirect
+	github.com/kr/text v0.2.0 // indirect
 	github.com/mschoch/smat v0.2.0 // indirect
 	github.com/pmezard/go-difflib v1.0.0 // indirect
 	github.com/spf13/pflag v1.0.5 // indirect
-	github.com/yuin/goldmark v1.6.0 // indirect
-	golang.org/x/sys v0.8.0 // indirect
-	golang.org/x/text v0.9.0 // indirect
+	github.com/yuin/goldmark v1.7.8 // indirect
+	golang.org/x/sys v0.28.0 // indirect
+	golang.org/x/text v0.21.0 // indirect
 	gopkg.in/yaml.v3 v3.0.1 // indirect
 	olympos.io/encoding/edn v0.0.0-20201019073823-d3554ca0b0a3 // indirect
 )