mail: new package.

Basic parsing, plus date parsing.

R=bradfitz, gary.burd, bsiegert, rsc
CC=golang-dev
https://golang.org/cl/4530079

src/pkg/Makefile

@@ -116,6 +116,7 @@ DIRS=\
 	io/ioutil\
 	json\
 	log\
+	mail\
 	math\
 	mime\
 	mime/multipart\

src/pkg/mail/Makefile

@@ -0,0 +1,11 @@
+# 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.
+
+include ../../Make.inc
+
+TARG=mail
+GOFILES=\
+	message.go\
+
+include ../../Make.pkg

src/pkg/mail/message.go

@@ -0,0 +1,95 @@
+// 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 mail implements parsing of mail messages according to RFC 5322.
+package mail
+
+import (
+	"bufio"
+	"io"
+	"net/textproto"
+	"os"
+	"time"
+)
+
+// A Message represents a parsed mail message.
+type Message struct {
+	Header Header
+	Body   io.Reader
+}
+
+// ReadMessage reads a message from r.
+// The headers are parsed, and the body of the message will be reading from r.
+func ReadMessage(r io.Reader) (msg *Message, err os.Error) {
+	tp := textproto.NewReader(bufio.NewReader(r))
+
+	hdr, err := tp.ReadMIMEHeader()
+	if err != nil {
+		return nil, err
+	}
+
+	return &Message{
+		Header: Header(hdr),
+		Body:   tp.R,
+	}, nil
+}
+
+// Layouts suitable for passing to time.Parse.
+// These are tried in order.
+var dateLayouts []string
+
+func init() {
+	// Generate layouts based on RFC 5322, section 3.3.
+
+	dows := [...]string{"", "Mon, "}     // day-of-week
+	days := [...]string{"2", "02"}       // day = 1*2DIGIT
+	years := [...]string{"2006", "06"}   // year = 4*DIGIT / 2*DIGIT
+	seconds := [...]string{":05", ""}    // second
+	zones := [...]string{"-0700", "MST"} // zone = (("+" / "-") 4DIGIT) / "GMT" / ...
+
+	for _, dow := range dows {
+		for _, day := range days {
+			for _, year := range years {
+				for _, second := range seconds {
+					for _, zone := range zones {
+						s := dow + day + " Jan " + year + " 15:04" + second + " " + zone
+						dateLayouts = append(dateLayouts, s)
+					}
+				}
+			}
+		}
+	}
+}
+
+func parseDate(date string) (*time.Time, os.Error) {
+	for _, layout := range dateLayouts {
+		t, err := time.Parse(layout, date)
+		if err == nil {
+			return t, nil
+		}
+	}
+	return nil, os.ErrorString("mail: header could not be parsed")
+}
+
+// TODO(dsymonds): Parsers for more specific headers such as To, From, etc.
+
+// A Header represents the key-value pairs in a mail message header.
+type Header map[string][]string
+
+// Get gets the first value associated with the given key.
+// If there are no values associated with the key, Get returns "".
+func (h Header) Get(key string) string {
+	return textproto.MIMEHeader(h).Get(key)
+}
+
+var ErrHeaderNotPresent = os.ErrorString("mail: header not in message")
+
+// Date parses the Date header field.
+func (h Header) Date() (*time.Time, os.Error) {
+	hdr := h.Get("Date")
+	if hdr == "" {
+		return nil, ErrHeaderNotPresent
+	}
+	return parseDate(hdr)
+}

src/pkg/mail/message_test.go

@@ -0,0 +1,129 @@
+// 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 mail
+
+import (
+	"bytes"
+	"io/ioutil"
+	"reflect"
+	"testing"
+	"time"
+)
+
+var parseTests = []struct {
+	in     string
+	header Header
+	body   string
+}{
+	{
+		// RFC 5322, Appendix A.1.1
+		in: `From: John Doe <jdoe@machine.example>
+To: Mary Smith <mary@example.net>
+Subject: Saying Hello
+Date: Fri, 21 Nov 1997 09:55:06 -0600
+Message-ID: <1234@local.machine.example>
+
+This is a message just to say hello.
+So, "Hello".
+`,
+		header: Header{
+			"From":       []string{"John Doe <jdoe@machine.example>"},
+			"To":         []string{"Mary Smith <mary@example.net>"},
+			"Subject":    []string{"Saying Hello"},
+			"Date":       []string{"Fri, 21 Nov 1997 09:55:06 -0600"},
+			"Message-Id": []string{"<1234@local.machine.example>"},
+		},
+		body: "This is a message just to say hello.\nSo, \"Hello\".\n",
+	},
+}
+
+func TestParsing(t *testing.T) {
+	for i, test := range parseTests {
+		msg, err := ReadMessage(bytes.NewBuffer([]byte(test.in)))
+		if err != nil {
+			t.Errorf("test #%d: Failed parsing message: %v", i, err)
+			continue
+		}
+		if !headerEq(msg.Header, test.header) {
+			t.Errorf("test #%d: Incorrectly parsed message header.\nGot:\n%+v\nWant:\n%+v",
+				i, msg.Header, test.header)
+		}
+		body, err := ioutil.ReadAll(msg.Body)
+		if err != nil {
+			t.Errorf("test #%d: Failed reading body: %v", i, err)
+			continue
+		}
+		bodyStr := string(body)
+		if bodyStr != test.body {
+			t.Errorf("test #%d: Incorrectly parsed message body.\nGot:\n%+v\nWant:\n%+v",
+				i, bodyStr, test.body)
+		}
+	}
+}
+
+func headerEq(a, b Header) bool {
+	if len(a) != len(b) {
+		return false
+	}
+	for k, as := range a {
+		bs, ok := b[k]
+		if !ok {
+			return false
+		}
+		if !reflect.DeepEqual(as, bs) {
+			return false
+		}
+	}
+	return true
+}
+
+func TestDateParsing(t *testing.T) {
+	tests := []struct {
+		dateStr string
+		exp     *time.Time
+	}{
+		// RFC 5322, Appendix A.1.1
+		{
+			"Fri, 21 Nov 1997 09:55:06 -0600",
+			&time.Time{
+				Year:       1997,
+				Month:      11,
+				Day:        21,
+				Hour:       9,
+				Minute:     55,
+				Second:     6,
+				Weekday:    5, // Fri
+				ZoneOffset: -6 * 60 * 60,
+			},
+		},
+		// RFC5322, Appendix A.6.2
+		// Obsolete date.
+		{
+			"21 Nov 97 09:55:06 GMT",
+			&time.Time{
+				Year:   1997,
+				Month:  11,
+				Day:    21,
+				Hour:   9,
+				Minute: 55,
+				Second: 6,
+				Zone:   "GMT",
+			},
+		},
+	}
+	for _, test := range tests {
+		hdr := Header{
+			"Date": []string{test.dateStr},
+		}
+		date, err := hdr.Date()
+		if err != nil {
+			t.Errorf("Failed parsing %q: %v", test.dateStr, err)
+			continue
+		}
+		if !reflect.DeepEqual(date, test.exp) {
+			t.Errorf("Parse of %q: got %+v, want %+v", test.dateStr, date, test.exp)
+		}
+	}
+}