Document extractor service (#15665)

* Document extractor service

* Fixing vendor modules

* Addressing PR Review comments

* Some small simplifications

* Fixing a linter complain

* simplifying a bit the code using package variables

Co-authored-by: Mattermod <mattermod@users.noreply.github.com>

vendor/code.sajari.com/docconv/pdf.go

@@ -0,0 +1,30 @@
+// +build !ocr
+
+package docconv
+
+import (
+	"fmt"
+	"io"
+)
+
+func ConvertPDF(r io.Reader) (string, map[string]string, error) {
+
+	f, err := NewLocalFile(r)
+	if err != nil {
+		return "", nil, fmt.Errorf("error creating local file: %v", err)
+	}
+	defer f.Done()
+
+	bodyResult, metaResult, convertErr := ConvertPDFText(f.Name())
+	if convertErr != nil {
+		return "", nil, convertErr
+	}
+	if bodyResult.err != nil {
+		return "", nil, bodyResult.err
+	}
+	if metaResult.err != nil {
+		return "", nil, metaResult.err
+	}
+	return bodyResult.body, metaResult.meta, nil
+
+}