New Go based XML reader

This includes a copy of Go's XML processing library until the build
system has Go version 1.8.

See golang/go#45628

doc/commands-xml/commands.xml

@@ -1924,6 +1924,14 @@
         </description>
       </choice>
     </attribute>
+    <attribute en="luaxmlreader" optional="yes" since="4.5.1" type="boolean">
+      <description xml:lang="en">
+        <para>Use old Lua based XML reader.</para>
+      </description>
+      <description xml:lang="de">
+        <para>Benutze alten Lua-basierten XML Parser.</para>
+      </description>
+    </attribute>
     <example xml:lang="en">
       <listing><![CDATA[<Compatibility
     movecursoronplaceobject="no"

schema/layoutschema-de.rng

@@ -973,6 +973,15 @@
                </choice>
             </attribute>
          </optional>
+         <optional>
+            <attribute name="luaxmlreader">
+               <a:documentation>Benutze alten Lua-basierten XML Parser.</a:documentation>
+               <choice>
+                  <value>yes</value>
+                  <value>no</value>
+               </choice>
+            </attribute>
+         </optional>
          <empty></empty>
       </element>
    </define>

schema/layoutschema-de.xsd

@@ -814,6 +814,17 @@
           </xs:restriction>
         </xs:simpleType>
       </xs:attribute>
+      <xs:attribute name="luaxmlreader">
+        <xs:annotation>
+          <xs:documentation>Benutze alten Lua-basierten XML Parser.</xs:documentation>
+        </xs:annotation>
+        <xs:simpleType>
+          <xs:restriction base="xs:token">
+            <xs:enumeration value="yes"/>
+            <xs:enumeration value="no"/>
+          </xs:restriction>
+        </xs:simpleType>
+      </xs:attribute>
     </xs:complexType>
   </xs:element>
   <xs:element name="Contents">

schema/layoutschema-en.rng

@@ -973,6 +973,15 @@
                </choice>
             </attribute>
          </optional>
+         <optional>
+            <attribute name="luaxmlreader">
+               <a:documentation>Use old Lua based XML reader.</a:documentation>
+               <choice>
+                  <value>yes</value>
+                  <value>no</value>
+               </choice>
+            </attribute>
+         </optional>
          <empty></empty>
       </element>
    </define>

schema/layoutschema-en.xsd

@@ -814,6 +814,17 @@
           </xs:restriction>
         </xs:simpleType>
       </xs:attribute>
+      <xs:attribute name="luaxmlreader">
+        <xs:annotation>
+          <xs:documentation>Use old Lua based XML reader.</xs:documentation>
+        </xs:annotation>
+        <xs:simpleType>
+          <xs:restriction base="xs:token">
+            <xs:enumeration value="yes"/>
+            <xs:enumeration value="no"/>
+          </xs:restriction>
+        </xs:simpleType>
+      </xs:attribute>
     </xs:complexType>
   </xs:element>
   <xs:element name="Contents">

src/go/splib/splib.go

@@ -240,4 +240,14 @@ func sdSegmentize(original string) *C.struct_splitvalues {
 
 	return returnStruct
 }
+
+//export sdReadXMLFile
+func sdReadXMLFile(filename string) *C.char {
+	str, err := splibaux.ReadXMLFile(filename)
+	if err != nil {
+		return s2c(errorpattern + err.Error())
+	}
+	return s2c(str)
+}
+
 func main() {}

src/go/splibaux/splibaux.go

@@ -18,6 +18,7 @@ var (
 	files      map[string]string
 	ignorefile string
 	verbosity  int
+	nr         *strings.Replacer
 )
 
 func init() {
@@ -27,6 +28,7 @@ func init() {
 	if v := os.Getenv("SP_VERBOSITY"); v != "" {
 		verbosity, _ = strconv.Atoi(v)
 	}
+	nr = strings.NewReplacer("\n", `\n`, `"`, `\"`, `\`, `\\`)
 }
 
 func downloadFile(resourceURL string, outfile io.Writer) error {
@@ -308,3 +310,22 @@ func ConvertSVGImage(filename string) (string, error) {
 	}
 	return pdffile, nil
 }
+
+func handleXInclude(href string, startindex, indent int) (string, error) {
+	fullpath := LookupFile(href)
+	f, err := os.Open(fullpath)
+	if err != nil {
+		return "", err
+	}
+	return readXMLFile(f, startindex, indent)
+}
+
+func ReadXMLFile(filename string) (string, error) {
+	fullpath := LookupFile(filename)
+	f, err := os.Open(fullpath)
+	if err != nil {
+		return "", err
+	}
+	str, err := readXMLFile(f, 1, 0)
+	return "tbl = {" + str + "}", err
+}

src/go/splibaux/xmlparser.go

@@ -0,0 +1,112 @@
+package splibaux
+
+import (
+	"fmt"
+	"io"
+	"strings"
+
+	"speedatapublisher/xml"
+)
+
+func luaescape(s string) string {
+	return nr.Replace(s)
+}
+
+func indent(i int) string {
+	return strings.Repeat("    ", i)
+}
+
+func readXMLFile(r io.Reader, startindex, extraindentlevel int) (string, error) {
+	var out strings.Builder
+
+	stackcounter := []int{startindex}
+
+	dec := xml.NewDecoder(r)
+	dec.Entity = xml.HTMLEntity
+	indentlevel := 0
+	for {
+		tok, err := dec.Token()
+		if err == io.EOF {
+			break
+		}
+		if err != nil {
+			return "", err
+		}
+		indentamount := indentlevel + extraindentlevel + 1
+
+		switch v := tok.(type) {
+		case xml.StartElement:
+			var href string
+			if v.Name.Space == "http://www.w3.org/2001/XInclude" && v.Name.Local == "include" {
+				for _, attr := range v.Attr {
+					if attr.Name.Local == "href" {
+						href = attr.Value
+					}
+				}
+				str, err := handleXInclude(href, indentlevel+1, indentamount-1)
+				if err != nil {
+					return "", err
+				}
+				fmt.Fprintf(&out, "%s", str)
+			} else {
+
+				fmt.Fprintf(&out, "%s[%d] = {", indent(indentamount-1), stackcounter[indentlevel])
+				fmt.Fprintln(&out)
+				fmt.Fprintf(&out, `%s[".__name"] = "%s",`, indent(indentamount), luaescape(v.Name.Local))
+				fmt.Fprintln(&out)
+				fmt.Fprintf(&out, `%s[".__type"] = "element",`, indent(indentamount))
+				fmt.Fprintln(&out)
+				fmt.Fprintf(&out, `%s[".__local_name"] = "%s",`, indent(indentamount), luaescape(v.Name.Local))
+				fmt.Fprintln(&out)
+				fmt.Fprintf(&out, `%s[".__namespace"] = "%s",`, indent(indentamount), luaescape(v.Name.Space))
+				fmt.Fprintln(&out)
+				line, col := dec.InputPos()
+				fmt.Fprintf(&out, `%s[".__line"] = %d,[".__col"] = %d,`, indent(indentamount), line, col)
+				fmt.Fprintln(&out)
+
+				attributes := make(map[string]string)
+
+				fmt.Fprintf(&out, `%s[".__ns"] = {`, indent(indentamount))
+				fmt.Fprintln(&out)
+				for _, attr := range v.Attr {
+					if attr.Name.Space == "xmlns" {
+						fmt.Fprintf(&out, `%s["%s"] = "%s",`, indent(indentamount+1), attr.Name.Local, attr.Value)
+						fmt.Fprintln(&out)
+					} else if attr.Name.Local == "xmlns" {
+						fmt.Fprintf(&out, `%s[""] = "%s",`, indent(indentamount+1), attr.Value)
+						fmt.Fprintln(&out)
+					} else {
+						attributes[attr.Name.Local] = attr.Value
+					}
+				}
+				fmt.Fprintf(&out, "%s},\n", indent(indentamount))
+				for k, val := range attributes {
+					fmt.Fprintf(&out, `%s["%s"] = "%s", `, indent(indentamount), luaescape(k), luaescape(val))
+					fmt.Fprintln(&out)
+				}
+			}
+			stackcounter[indentlevel]++
+			indentlevel++
+			stackcounter = append(stackcounter, 1)
+		case xml.CharData:
+			if indentlevel > 0 {
+
+				index := stackcounter[indentlevel]
+				stackcounter[indentlevel] = index + 1
+				fmt.Fprintf(&out, "%s[%d] = ", indent(indentlevel+extraindentlevel), index)
+				fmt.Fprintf(&out, `"%s",`, luaescape(string(v.Copy())))
+				fmt.Fprintln(&out)
+			}
+		case xml.EndElement:
+			if v.Name.Space == "http://www.w3.org/2001/XInclude" && v.Name.Local == "include" {
+				// ignore
+			} else {
+				fmt.Fprintf(&out, "%s},\n", indent(indentlevel+extraindentlevel))
+			}
+
+			stackcounter = stackcounter[:len(stackcounter)-1]
+			indentlevel--
+		}
+	}
+	return out.String(), nil
+}

src/go/xml/atom_test.go

@@ -0,0 +1,56 @@
+// Copyright 2011 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package xml
+
+import "time"
+
+var atomValue = &Feed{
+	XMLName: Name{"http://www.w3.org/2005/Atom", "feed"},
+	Title:   "Example Feed",
+	Link:    []Link{{Href: "http://example.org/"}},
+	Updated: ParseTime("2003-12-13T18:30:02Z"),
+	Author:  Person{Name: "John Doe"},
+	ID:      "urn:uuid:60a76c80-d399-11d9-b93C-0003939e0af6",
+
+	Entry: []Entry{
+		{
+			Title:   "Atom-Powered Robots Run Amok",
+			Link:    []Link{{Href: "http://example.org/2003/12/13/atom03"}},
+			ID:      "urn:uuid:1225c695-cfb8-4ebb-aaaa-80da344efa6a",
+			Updated: ParseTime("2003-12-13T18:30:02Z"),
+			Summary: NewText("Some text."),
+		},
+	},
+}
+
+var atomXML = `` +
+	`<feed xmlns="http://www.w3.org/2005/Atom" updated="2003-12-13T18:30:02Z">` +
+	`<title>Example Feed</title>` +
+	`<id>urn:uuid:60a76c80-d399-11d9-b93C-0003939e0af6</id>` +
+	`<link href="http://example.org/"></link>` +
+	`<author><name>John Doe</name><uri></uri><email></email></author>` +
+	`<entry>` +
+	`<title>Atom-Powered Robots Run Amok</title>` +
+	`<id>urn:uuid:1225c695-cfb8-4ebb-aaaa-80da344efa6a</id>` +
+	`<link href="http://example.org/2003/12/13/atom03"></link>` +
+	`<updated>2003-12-13T18:30:02Z</updated>` +
+	`<author><name></name><uri></uri><email></email></author>` +
+	`<summary>Some text.</summary>` +
+	`</entry>` +
+	`</feed>`
+
+func ParseTime(str string) time.Time {
+	t, err := time.Parse(time.RFC3339, str)
+	if err != nil {
+		panic(err)
+	}
+	return t
+}
+
+func NewText(text string) Text {
+	return Text{
+		Body: text,
+	}
+}

src/go/xml/example_marshaling_test.go

@@ -0,0 +1,84 @@
+// Copyright 2018 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package xml_test
+
+import (
+	"encoding/xml"
+	"fmt"
+	"log"
+	"strings"
+)
+
+type Animal int
+
+const (
+	Unknown Animal = iota
+	Gopher
+	Zebra
+)
+
+func (a *Animal) UnmarshalXML(d *xml.Decoder, start xml.StartElement) error {
+	var s string
+	if err := d.DecodeElement(&s, &start); err != nil {
+		return err
+	}
+	switch strings.ToLower(s) {
+	default:
+		*a = Unknown
+	case "gopher":
+		*a = Gopher
+	case "zebra":
+		*a = Zebra
+	}
+
+	return nil
+}
+
+func (a Animal) MarshalXML(e *xml.Encoder, start xml.StartElement) error {
+	var s string
+	switch a {
+	default:
+		s = "unknown"
+	case Gopher:
+		s = "gopher"
+	case Zebra:
+		s = "zebra"
+	}
+	return e.EncodeElement(s, start)
+}
+
+func Example_customMarshalXML() {
+	blob := `
+	<animals>
+		<animal>gopher</animal>
+		<animal>armadillo</animal>
+		<animal>zebra</animal>
+		<animal>unknown</animal>
+		<animal>gopher</animal>
+		<animal>bee</animal>
+		<animal>gopher</animal>
+		<animal>zebra</animal>
+	</animals>`
+	var zoo struct {
+		Animals []Animal `xml:"animal"`
+	}
+	if err := xml.Unmarshal([]byte(blob), &zoo); err != nil {
+		log.Fatal(err)
+	}
+
+	census := make(map[Animal]int)
+	for _, animal := range zoo.Animals {
+		census[animal] += 1
+	}
+
+	fmt.Printf("Zoo Census:\n* Gophers: %d\n* Zebras:  %d\n* Unknown: %d\n",
+		census[Gopher], census[Zebra], census[Unknown])
+
+	// Output:
+	// Zoo Census:
+	// * Gophers: 3
+	// * Zebras:  2
+	// * Unknown: 3
+}