Restrict repository indexing by glob match (#7767)

* Restrict repository indexing by file extension

* Use REPO_EXTENSIONS_LIST_INCLUDE instead of REPO_EXTENSIONS_LIST_EXCLUDE and have a more flexible extension pattern

* Corrected to pass lint gosimple

* Add wildcard support to REPO_INDEXER_EXTENSIONS

* This reverts commit 72a650c8e42f4abf59d5df7cd5dc27b451494cc6.

* Add wildcard support to REPO_INDEXER_EXTENSIONS (no make vendor)

* Simplify isIndexable() for better clarity

* Add gobwas/glob to vendors

* manually set appengine new release

* Implement better REPO_INDEXER_INCLUDE and REPO_INDEXER_EXCLUDE

* Add unit and integration tests

* Update app.ini.sample and reword config-cheat-sheet

* Add doc page and correct app.ini.sample

* Some polish on the doc

* Simplify code as suggested by @lafriks

2 files changed, 99 insertions, 0 deletions

diff --git a/modules/setting/indexer.go b/modules/setting/indexer.go
index 36fd4a020b..30c670d407 100644
--- a/modules/setting/indexer.go
+++ b/modules/setting/indexer.go
@@ -7,6 +7,11 @@ package setting
 import (
 	"path"
 	"path/filepath"
+	"strings"
+
+	"code.gitea.io/gitea/modules/log"
+
+	"github.com/gobwas/glob"
 )
 
 // enumerates all the indexer queue types
@@ -29,6 +34,8 @@ var (
 		IssueQueueDir         string
 		IssueQueueConnStr     string
 		IssueQueueBatchNumber int
+		IncludePatterns       []glob.Glob
+		ExcludePatterns       []glob.Glob
 	}{
 		IssueType:             "bleve",
 		IssuePath:             "indexers/issues.bleve",
@@ -51,6 +58,9 @@ func newIndexerService() {
 	if !filepath.IsAbs(Indexer.RepoPath) {
 		Indexer.RepoPath = path.Join(AppWorkPath, Indexer.RepoPath)
 	}
+	Indexer.IncludePatterns = IndexerGlobFromString(sec.Key("REPO_INDEXER_INCLUDE").MustString(""))
+	Indexer.ExcludePatterns = IndexerGlobFromString(sec.Key("REPO_INDEXER_EXCLUDE").MustString(""))
+
 	Indexer.UpdateQueueLength = sec.Key("UPDATE_BUFFER_LEN").MustInt(20)
 	Indexer.MaxIndexerFileSize = sec.Key("MAX_FILE_SIZE").MustInt64(1024 * 1024)
 	Indexer.IssueQueueType = sec.Key("ISSUE_INDEXER_QUEUE_TYPE").MustString(LevelQueueType)
@@ -58,3 +68,19 @@ func newIndexerService() {
 	Indexer.IssueQueueConnStr = sec.Key("ISSUE_INDEXER_QUEUE_CONN_STR").MustString(path.Join(AppDataPath, ""))
 	Indexer.IssueQueueBatchNumber = sec.Key("ISSUE_INDEXER_QUEUE_BATCH_NUMBER").MustInt(20)
 }
+
+// IndexerGlobFromString parses a comma separated list of patterns and returns a glob.Glob slice suited for repo indexing
+func IndexerGlobFromString(globstr string) []glob.Glob {
+	extarr := make([]glob.Glob, 0, 10)
+	for _, expr := range strings.Split(strings.ToLower(globstr), ",") {
+		expr = strings.TrimSpace(expr)
+		if expr != "" {
+			if g, err := glob.Compile(expr, '.', '/'); err != nil {
+				log.Info("Invalid glob expresion '%s' (skipped): %v", expr, err)
+			} else {
+				extarr = append(extarr, g)
+			}
+		}
+	}
+	return extarr
+}
diff --git a/modules/setting/indexer_test.go b/modules/setting/indexer_test.go
new file mode 100644
index 0000000000..ed631747dc
--- /dev/null
+++ b/modules/setting/indexer_test.go
@@ -0,0 +1,73 @@
+// Copyright 2019 The Gitea Authors. All rights reserved.
+// Use of this source code is governed by a MIT-style
+// license that can be found in the LICENSE file.
+
+package setting
+
+import (
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+)
+
+type indexerMatchList struct {
+	value    string
+	position int
+}
+
+func Test_newIndexerGlobSettings(t *testing.T) {
+
+	checkGlobMatch(t, "", []indexerMatchList{})
+	checkGlobMatch(t, "     ", []indexerMatchList{})
+	checkGlobMatch(t, "data, */data, */data/*, **/data/*, **/data/**", []indexerMatchList{
+		{"", -1},
+		{"don't", -1},
+		{"data", 0},
+		{"/data", 1},
+		{"x/data", 1},
+		{"x/data/y", 2},
+		{"a/b/c/data/z", 3},
+		{"a/b/c/data/x/y/z", 4},
+	})
+	checkGlobMatch(t, "*.txt, txt, **.txt, **txt, **txt*", []indexerMatchList{
+		{"my.txt", 0},
+		{"don't", -1},
+		{"mytxt", 3},
+		{"/data/my.txt", 2},
+		{"data/my.txt", 2},
+		{"data/txt", 3},
+		{"data/thistxtfile", 4},
+		{"/data/thistxtfile", 4},
+	})
+	checkGlobMatch(t, "data/**/*.txt, data/**.txt", []indexerMatchList{
+		{"data/a/b/c/d.txt", 0},
+		{"data/a.txt", 1},
+	})
+	checkGlobMatch(t, "**/*.txt, data/**.txt", []indexerMatchList{
+		{"data/a/b/c/d.txt", 0},
+		{"data/a.txt", 0},
+		{"a.txt", -1},
+	})
+}
+
+func checkGlobMatch(t *testing.T, globstr string, list []indexerMatchList) {
+	glist := IndexerGlobFromString(globstr)
+	if len(list) == 0 {
+		assert.Empty(t, glist)
+		return
+	}
+	assert.NotEmpty(t, glist)
+	for _, m := range list {
+		found := false
+		for pos, g := range glist {
+			if g.Match(m.value) {
+				assert.Equal(t, m.position, pos, "Test string `%s` doesn't match `%s`@%d, but matches @%d", m.value, globstr, m.position, pos)
+				found = true
+				break
+			}
+		}
+		if !found {
+			assert.Equal(t, m.position, -1, "Test string `%s` doesn't match `%s` anywhere; expected @%d", m.value, globstr, m.position)
+		}
+	}
+}