Merge branch 'master' into parse

Conflicts:
	db/load.go

.travis.yml

@@ -17,5 +17,5 @@ install:
   - go get github.com/syndtr/goleveldb/leveldb/iterator
   - go get github.com/syndtr/goleveldb/leveldb/opt
   - go get github.com/syndtr/goleveldb/leveldb/util
-  - go get labix.org/v2/mgo
-  - go get labix.org/v2/mgo/bson
+  - go get gopkg.in/mgo.v2
+  - go get gopkg.in/mgo.v2/bson

README.md

@@ -72,14 +72,13 @@ cayley> graph.Vertex("dani").Out("follows").All()
 For somewhat more interesting data, a sample of 30k movies from Freebase comes in the checkout.
 
 ```
-gzip -cd 30kmoviedata.nt.gz > 30kmovies.nt
-./cayley repl --dbpath=30kmovies.nt
+./cayley repl --dbpath=30kmoviedata.nt.gz
 ```
 
 To run the web frontend, replace the "repl" command with "http"
 
 ```
-./cayley http --dbpath=30kmovies.nt
+./cayley http --dbpath=30kmoviedata.nt.gz
 ```
 
 And visit port 64210 on your machine, commonly [http://localhost:64210](http://localhost:64210)

db/load.go

@@ -15,10 +15,14 @@
 package db
 
 import (
+	"bytes"
+	"compress/bzip2"
+	"compress/gzip"
 	"fmt"
 	"io"
 	"os"
 
+	"github.com/barakmich/glog"
 	"github.com/google/cayley/config"
 	"github.com/google/cayley/graph"
 	"github.com/google/cayley/nquads"
@@ -31,7 +35,12 @@ func Load(ts graph.TripleStore, cfg *config.Config, path string) error {
 	}
 	defer f.Close()
 
-	dec := nquads.NewDecoder(f)
+	r, err := decompressor(f)
+	if err != nil {
+		glog.Fatalln(err)
+	}
+
+	dec := nquads.NewDecoder(r)
 
 	bulker, canBulk := ts.(graph.BulkLoader)
 	if canBulk {
@@ -66,3 +75,29 @@ func Load(ts graph.TripleStore, cfg *config.Config, path string) error {
 
 	return nil
 }
+
+const (
+	gzipMagic  = "\x1f\x8b"
+	b2zipMagic = "BZh"
+)
+
+type readAtReader interface {
+	io.Reader
+	io.ReaderAt
+}
+
+func decompressor(r readAtReader) (io.Reader, error) {
+	var buf [3]byte
+	_, err := r.ReadAt(buf[:], 0)
+	if err != nil {
+		return nil, err
+	}
+	switch {
+	case bytes.Compare(buf[:2], []byte(gzipMagic)) == 0:
+		return gzip.NewReader(r)
+	case bytes.Compare(buf[:3], []byte(b2zipMagic)) == 0:
+		return bzip2.NewReader(r), nil
+	default:
+		return r, nil
+	}
+}

db/load_test.go

@@ -0,0 +1,81 @@
+package db
+
+import (
+	"bytes"
+	"compress/bzip2"
+	"compress/gzip"
+	"testing"
+)
+
+var testDecompressor = []struct {
+	message string
+	input   []byte
+	expect  []byte
+	err     error
+	readErr error
+}{
+	{
+		message: "text input",
+		input:   []byte("cayley data\n"),
+		err:     nil,
+		expect:  []byte("cayley data\n"),
+		readErr: nil,
+	},
+	{
+		message: "gzip input",
+		input: []byte{
+			0x1f, 0x8b, 0x08, 0x00, 0x5c, 0xbc, 0xcd, 0x53, 0x00, 0x03, 0x4b, 0x4e, 0xac, 0xcc, 0x49, 0xad,
+			0x54, 0x48, 0x49, 0x2c, 0x49, 0xe4, 0x02, 0x00, 0x03, 0xe1, 0xfc, 0xc3, 0x0c, 0x00, 0x00, 0x00,
+		},
+		err:     nil,
+		expect:  []byte("cayley data\n"),
+		readErr: nil,
+	},
+	{
+		message: "bzip2 input",
+		input: []byte{
+			0x42, 0x5a, 0x68, 0x39, 0x31, 0x41, 0x59, 0x26, 0x53, 0x59, 0xb5, 0x4b, 0xe3, 0xc4, 0x00, 0x00,
+			0x02, 0xd1, 0x80, 0x00, 0x10, 0x40, 0x00, 0x2e, 0x04, 0x04, 0x20, 0x20, 0x00, 0x31, 0x06, 0x4c,
+			0x41, 0x4c, 0x1e, 0xa7, 0xa9, 0x2a, 0x18, 0x26, 0xb1, 0xc2, 0xee, 0x48, 0xa7, 0x0a, 0x12, 0x16,
+			0xa9, 0x7c, 0x78, 0x80,
+		},
+		err:     nil,
+		expect:  []byte("cayley data\n"),
+		readErr: nil,
+	},
+	{
+		message: "bad gzip input",
+		input:   []byte{0x1f, 0x8b, 'c', 'a', 'y', 'l', 'e', 'y', ' ', 'd', 'a', 't', 'a', '\n'},
+		err:     gzip.ErrHeader,
+		expect:  nil,
+		readErr: nil,
+	},
+	{
+		message: "bad bzip2 input",
+		input:   []byte{0x42, 0x5a, 0x68, 'c', 'a', 'y', 'l', 'e', 'y', ' ', 'd', 'a', 't', 'a', '\n'},
+		err:     nil,
+		expect:  nil,
+		readErr: bzip2.StructuralError("invalid compression level"),
+	},
+}
+
+func TestDecompressor(t *testing.T) {
+	for _, test := range testDecompressor {
+		buf := bytes.NewReader(test.input)
+		r, err := decompressor(buf)
+		if err != test.err {
+			t.Fatalf("Unexpected error for %s, got:%v expect:%v", test.message, err, test.err)
+		}
+		if err != nil {
+			continue
+		}
+		p := make([]byte, len(test.expect)*2)
+		n, err := r.Read(p)
+		if err != test.readErr {
+			t.Fatalf("Unexpected error for reading %s, got:%v expect:%v", test.message, err, test.err)
+		}
+		if bytes.Compare(p[:n], test.expect) != 0 {
+			t.Errorf("Unexpected read result for %s, got:%q expect:%q", test.message, p[:n], test.expect)
+		}
+	}
+}

docs/Overview.md

@@ -25,22 +25,16 @@ You can repeat the `--db` and `--dbpath` flags from here forward instead of the
 
 ### Load Data Into A Graph
 
-Let's extract the sample data, a couple hundred thousand movie triples, that comes in the checkout:
+First we load the data.
 
 ```bash
-zcat 30kmoviedatauniq.n3.gz > 30k.n3
-```
-
-Then, we can load the data.
-
-```bash
-./cayley load --config=cayley.cfg.overview --triples=30k.n3
+./cayley load --config=cayley.cfg.overview --triples=30kmoviedata.nt.gz
 ```
 
 And wait. It will load. If you'd like to watch it load, you can run
 
 ```bash
-./cayley load --config=cayley.cfg.overview --triples=30k.n3 --alsologtostderr
+./cayley load --config=cayley.cfg.overview --triples=30kmoviedata.nt.gz --alsologtostderr
 ```
 
 And watch the log output go by.

graph/mongo/iterator.go

@@ -19,8 +19,8 @@ import (
 	"strings"
 
 	"github.com/barakmich/glog"
-	"labix.org/v2/mgo"
-	"labix.org/v2/mgo/bson"
+	"gopkg.in/mgo.v2"
+	"gopkg.in/mgo.v2/bson"
 
 	"github.com/google/cayley/graph"
 	"github.com/google/cayley/graph/iterator"

graph/mongo/triplestore.go

@@ -20,8 +20,8 @@ import (
 	"hash"
 	"log"
 
-	"labix.org/v2/mgo"
-	"labix.org/v2/mgo/bson"
+	"gopkg.in/mgo.v2"
+	"gopkg.in/mgo.v2/bson"
 
 	"github.com/barakmich/glog"
 	"github.com/google/cayley/graph"