feat(docs): put-txt preserves line breaks in DOCX
TxtToMarkdown uses markdown hard breaks so pandoc keeps each source line on its own row. Adds oo docs put-txt and projects files alias. Co-authored-by: Cursor <cursoragent@cursor.com>
This commit is contained in:
@@ -76,6 +76,48 @@ func (t Tools) ConvertFile(inPath, outPath string) error {
|
||||
return nil
|
||||
}
|
||||
|
||||
// TXTToDOCX converts plain text to DOCX preserving line breaks (via markdown hard breaks).
|
||||
func (t Tools) TXTToDOCX(txtPath, docxPath string) error {
|
||||
if Ext(txtPath) != ".txt" {
|
||||
return fmt.Errorf("expected .txt input, got %q", txtPath)
|
||||
}
|
||||
if docxPath == "" {
|
||||
docxPath = strings.TrimSuffix(txtPath, Ext(txtPath)) + ".docx"
|
||||
}
|
||||
b, err := os.ReadFile(txtPath)
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
dir := filepath.Dir(docxPath)
|
||||
if dir == "" || dir == "." {
|
||||
dir = os.TempDir()
|
||||
}
|
||||
tmpMD := filepath.Join(dir, trimExt(filepath.Base(txtPath))+".txt2docx.md")
|
||||
md := TxtToMarkdown(string(b))
|
||||
if err := os.WriteFile(tmpMD, []byte(md), 0o644); err != nil {
|
||||
return err
|
||||
}
|
||||
defer os.Remove(tmpMD)
|
||||
return t.MDToDOCX(tmpMD, docxPath)
|
||||
}
|
||||
|
||||
// TxtToMarkdown converts plain text to Markdown that preserves line breaks in DOCX output.
|
||||
// Single newlines become hard breaks; blank lines stay paragraph separators.
|
||||
func TxtToMarkdown(content string) string {
|
||||
content = strings.ReplaceAll(content, "\r\n", "\n")
|
||||
content = strings.ReplaceAll(content, "\r", "\n")
|
||||
var b strings.Builder
|
||||
for _, line := range strings.Split(content, "\n") {
|
||||
if strings.TrimSpace(line) == "" {
|
||||
b.WriteByte('\n')
|
||||
continue
|
||||
}
|
||||
b.WriteString(line)
|
||||
b.WriteString(" \n")
|
||||
}
|
||||
return b.String()
|
||||
}
|
||||
|
||||
// MDToDOCX writes a DOCX next to or at outPath from a Markdown file.
|
||||
func (t Tools) MDToDOCX(mdPath, docxPath string) error {
|
||||
if Ext(mdPath) != ".md" && Ext(mdPath) != ".markdown" {
|
||||
|
||||
@@ -34,6 +34,51 @@ func TestNeedsOCR_Image(t *testing.T) {
|
||||
}
|
||||
}
|
||||
|
||||
func TestTxtToMarkdownPreservesLines(t *testing.T) {
|
||||
in := "line1\nline2\n\nline4"
|
||||
md := TxtToMarkdown(in)
|
||||
if !strings.Contains(md, "line1 \n") || !strings.Contains(md, "line2 \n") {
|
||||
t.Fatalf("hard breaks missing: %q", md)
|
||||
}
|
||||
if !strings.Contains(md, "line4 \n") {
|
||||
t.Fatalf("last line: %q", md)
|
||||
}
|
||||
}
|
||||
|
||||
func TestTXTToDOCXPreservesLines(t *testing.T) {
|
||||
tools := LookPath()
|
||||
if tools.Pandoc == "" {
|
||||
t.Skip("pandoc not installed")
|
||||
}
|
||||
dir := t.TempDir()
|
||||
txt := filepath.Join(dir, "sample.txt")
|
||||
docx := filepath.Join(dir, "sample.docx")
|
||||
body := "MyBox Auto — resumen\nNº contrato: 123\n\nEstado: Vigente\n"
|
||||
if err := os.WriteFile(txt, []byte(body), 0o644); err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
if err := tools.TXTToDOCX(txt, docx); err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
mdOut := filepath.Join(dir, "out.md")
|
||||
if err := tools.DOCXToMD(docx, mdOut); err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
got, err := os.ReadFile(mdOut)
|
||||
if err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
s := string(got)
|
||||
for _, want := range []string{"MyBox Auto", "Nº contrato", "Estado: Vigente"} {
|
||||
if !strings.Contains(s, want) {
|
||||
t.Fatalf("missing %q in %q", want, s)
|
||||
}
|
||||
}
|
||||
if strings.Contains(s, "MyBox Auto — resumen Nº") {
|
||||
t.Fatalf("lines collapsed: %q", s)
|
||||
}
|
||||
}
|
||||
|
||||
func TestMDDocxRoundTrip(t *testing.T) {
|
||||
tools := LookPath()
|
||||
if tools.Pandoc == "" {
|
||||
|
||||
Reference in New Issue
Block a user