refactoring: remove un-used code, add go doc, fix ci (#199)

Refactoring, consisting of
 - remove unused method `isAuxiliaryLanguage` and `FileCountList`
   in order to reduce public API surfaces (go/java)
 - add GoDoc to public APIs
 - ci: java profile use latest go src
  It also now mimics https://docs.travis-ci.com/user/languages/go/#go-import-path
  for non-go build image, as code relies on internal imports.

TEST PLAN:
 - make test

.travis.yml

@@ -45,11 +45,19 @@ jobs:
       stage: test
       language: scala
       jdk: oraclejdk8
+      before_install:
+        # mimics exact behavior of 'go_import_path' for non-go build image
+        - export GOPATH=${TRAVIS_HOME}/gopath
+        - mkdir -p ${GOPATH}/src/gopkg.in/src-d/enry.v1
+        - tar -Pczf ${TRAVIS_TMPDIR}/src_archive.tar.gz -C ${TRAVIS_BUILD_DIR} . && tar -Pxzf ${TRAVIS_TMPDIR}/src_archive.tar.gz -C ${TRAVIS_HOME}/gopath/src/gopkg.in/src-d/enry.v1
+        - export TRAVIS_BUILD_DIR=${TRAVIS_HOME}/gopath/src/gopkg.in/src-d/enry.v1
+        - cd ${TRAVIS_HOME}/gopath/src/gopkg.in/src-d/enry.v1
       install:
         - gimme version
         - eval "$(curl -sL https://raw.githubusercontent.com/travis-ci/gimme/master/gimme | GIMME_GO_VERSION=$GO_VERSION bash)"
         - go version
-        - go get -v gopkg.in/src-d/enry.v1/...
+        - echo $PWD; echo $GOPATH
+        - go get -v ./...
       before_script:
         - cd java
         - make

cmd/enry/main.go

@@ -278,7 +278,7 @@ func printFileAnalysis(file string, limit int64, isJSON bool) error {
 	// functions below can work on a sample
 	fileType := getFileType(file, data)
 	language := enry.GetLanguage(file, data)
-	mimeType := enry.GetMimeType(file, language)
+	mimeType := enry.GetMIMEType(file, language)
 
 	if isJSON {
 		return json.NewEncoder(os.Stdout).Encode(map[string]interface{}{

common.go

@@ -26,6 +26,7 @@ var DefaultStrategies = []Strategy{
 	GetLanguagesByClassifier,
 }
 
+// DefaultClassifier is a naive Bayes classifier based on Linguist samples.
 var DefaultClassifier Classifier = &classifier{
 	languagesLogProbabilities: data.LanguagesLogProbabilities,
 	tokensLogProbabilities:    data.TokensLogProbabilities,

enry.go

@@ -1,3 +1,16 @@
+/*
+	Package enry implements multiple strategies for programming language identification.
+
+	Identification is made based on file name and file content using a seriece
+	of strategies to narrow down possible option.
+	Each strategy is available as a separate API call, as well as a main enty point
+
+		GetLanguage(filename string, content []byte) (language string)
+
+	It is a port of the https://github.com/github/linguist from Ruby.
+	Upstream Linguist YAML files are used to generate datastructures for data
+	package.
+*/
 package enry // import "gopkg.in/src-d/enry.v1"
 
 //go:generate make code-generate

java/Makefile

@@ -28,7 +28,7 @@ $(RESOURCES_DIR): os-shared-lib
 	cp -R $(RESOURCES_SRC) $(RESOURCES_DIR)
 
 $(JNAERATOR_JAR): $(RESOURCES_DIR)
-	mkdir $(JNAERATOR_DIR) && \
+	mkdir -p $(JNAERATOR_DIR) && \
 	wget $(JNAERATOR_JAR_URL) -O $(JNAERATOR_JAR)
 
 os-shared-lib:

java/src/main/java/tech/sourced/enry/Enry.java

@@ -9,16 +9,6 @@ public class Enry {
 
     private static final EnryLibrary nativeLib = EnryLibrary.INSTANCE;
 
-    /**
-     * Returns whether the given language is auxiliary or not.
-     *
-     * @param language name of the language, e.g. PHP, HTML, ...
-     * @return if it's an auxiliary language
-     */
-    public static synchronized boolean isAuxiliaryLanguage(String language) {
-        return toJavaBool(nativeLib.IsAuxiliaryLanguage(toGoString(language)));
-    }
-
     /**
      * Returns the language of the given file based on the filename and its
      * contents.

java/src/test/java/tech/sourced/enry/EnryTest.java

@@ -6,12 +6,6 @@ import static org.junit.Assert.*;
 
 public class EnryTest {
 
-    @Test
-    public void isAuxiliaryLanguage() {
-        assertTrue(Enry.isAuxiliaryLanguage("HTML"));
-        assertFalse(Enry.isAuxiliaryLanguage("Go"));
-    }
-
     @Test
     public void getLanguage() {
         String code = "<?php $foo = bar();";

shared/enry.go

@@ -93,12 +93,7 @@ func GetLanguagesByVimModeline(filename string, content []byte, candidates []str
 
 //export GetMimeType
 func GetMimeType(path string, language string) string {
-	return enry.GetMimeType(path, language)
+	return enry.GetMIMEType(path, language)
-}
-
-//export IsAuxiliaryLanguage
-func IsAuxiliaryLanguage(lang string) bool {
-	return enry.IsAuxiliaryLanguage(lang)
 }
 
 //export IsBinary

utils.go

@@ -8,53 +8,20 @@ import (
 	"gopkg.in/src-d/enry.v1/data"
 )
 
-var (
+const binSniffLen = 8000
-	auxiliaryLanguages = map[string]bool{
-		"Other": true, "XML": true, "YAML": true, "TOML": true, "INI": true,
-		"JSON": true, "TeX": true, "Public Key": true, "AsciiDoc": true,
-		"AGS Script": true, "VimL": true, "Diff": true, "CMake": true, "fish": true,
-		"Awk": true, "Graphviz (DOT)": true, "Markdown": true, "desktop": true,
-		"XSLT": true, "SQL": true, "RMarkdown": true, "IRC log": true,
-		"reStructuredText": true, "Twig": true, "CSS": true, "Batchfile": true,
-		"Text": true, "HTML+ERB": true, "HTML": true, "Gettext Catalog": true,
-		"Smarty": true, "Raw token data": true,
-	}
 
-	configurationLanguages = map[string]bool{
+var configurationLanguages = map[string]bool{
-		"XML": true, "JSON": true, "TOML": true, "YAML": true, "INI": true, "SQL": true,
+	"XML": true, "JSON": true, "TOML": true, "YAML": true, "INI": true, "SQL": true,
-	}
-)
-
-// IsAuxiliaryLanguage returns whether or not lang is an auxiliary language.
-func IsAuxiliaryLanguage(lang string) bool {
-	_, ok := auxiliaryLanguages[lang]
-	return ok
 }
 
-// IsConfiguration returns whether or not path is using a configuration language.
+// IsConfiguration tells if filename is in one of the configuration languages.
 func IsConfiguration(path string) bool {
 	language, _ := GetLanguageByExtension(path)
 	_, is := configurationLanguages[language]
 	return is
 }
 
-// IsDotFile returns whether or not path has dot as a prefix.
+// IsImage tells if a given file is an image (PNG, JPEG or GIF format).
-func IsDotFile(path string) bool {
-	path = filepath.Clean(path)
-	base := filepath.Base(path)
-	return strings.HasPrefix(base, ".") && base != "." && base != ".."
-}
-
-// IsVendor returns whether or not path is a vendor path.
-func IsVendor(path string) bool {
-	return data.VendorMatchers.Match(path)
-}
-
-// IsDocumentation returns whether or not path is a documentation path.
-func IsDocumentation(path string) bool {
-	return data.DocumentationMatchers.Match(path)
-}
-
 func IsImage(path string) bool {
 	extension := filepath.Ext(path)
 	if extension == ".png" || extension == ".jpg" || extension == ".jpeg" || extension == ".gif" {
@@ -64,7 +31,8 @@ func IsImage(path string) bool {
 	return false
 }
 
-func GetMimeType(path string, language string) string {
+// GetMIMEType returns a MIME type of a given file based on its languages.
+func GetMIMEType(path string, language string) string {
 	if mime, ok := data.LanguagesMime[language]; ok {
 		return mime
 	}
@@ -76,13 +44,27 @@ func GetMimeType(path string, language string) string {
 	return "text/plain"
 }
 
-const sniffLen = 8000
+// IsDocumentation returns whether or not path is a documentation path.
+func IsDocumentation(path string) bool {
+	return data.DocumentationMatchers.Match(path)
+}
+
+// IsDotFile returns whether or not path has dot as a prefix.
+func IsDotFile(path string) bool {
+	base := filepath.Base(filepath.Clean(path))
+	return strings.HasPrefix(base, ".") && base != "."
+}
+
+// IsVendor returns whether or not path is a vendor path.
+func IsVendor(path string) bool {
+	return data.VendorMatchers.Match(path)
+}
 
 // IsBinary detects if data is a binary value based on:
 // http://git.kernel.org/cgit/git/git.git/tree/xdiff-interface.c?id=HEAD#n198
 func IsBinary(data []byte) bool {
-	if len(data) > sniffLen {
+	if len(data) > binSniffLen {
-		data = data[:sniffLen]
+		data = data[:binSniffLen]
 	}
 
 	if bytes.IndexByte(data, byte(0)) == -1 {
@@ -91,17 +73,3 @@ func IsBinary(data []byte) bool {
 
 	return true
 }
-
-// FileCount type stores language name and count of files belonging to the
-// language.
-type FileCount struct {
-	Name  string
-	Count int
-}
-
-// FileCountList type is a list of FileCounts.
-type FileCountList []FileCount
-
-func (fcl FileCountList) Len() int           { return len(fcl) }
-func (fcl FileCountList) Less(i, j int) bool { return fcl[i].Count < fcl[j].Count }
-func (fcl FileCountList) Swap(i, j int)      { fcl[i], fcl[j] = fcl[j], fcl[i] }

utils_test.go

@@ -3,38 +3,11 @@ package enry
 import (
 	"bytes"
 	"fmt"
-	"sort"
 	"testing"
 
 	"github.com/stretchr/testify/assert"
 )
 
-func TestIsAuxiliaryLanguage(t *testing.T) {
-	type testType struct {
-		name     string
-		lang     string
-		expected bool
-	}
-
-	tests := []testType{
-		{name: "TestIsAuxiliaryLanguage_Invalid", lang: "invalid", expected: false},
-	}
-	for k := range auxiliaryLanguages {
-		t := testType{
-			name:     fmt.Sprintf("TestIsAuxiliaryLanguage_%s", k),
-			lang:     k,
-			expected: true,
-		}
-		tests = append(tests, t)
-	}
-
-	for _, test := range tests {
-		is := IsAuxiliaryLanguage(test.lang)
-		assert.Equal(t, is, test.expected,
-			fmt.Sprintf("%v: is = %v, expected: %v", test.name, is, test.expected))
-	}
-}
-
 func TestIsVendor(t *testing.T) {
 	tests := []struct {
 		name     string
@@ -106,7 +79,7 @@ func TestGetMimeType(t *testing.T) {
 	}
 
 	for _, test := range tests {
-		is := GetMimeType(test.path, test.lang)
+		is := GetMIMEType(test.path, test.lang)
 		assert.Equal(t, is, test.expected, fmt.Sprintf("%v: is = %v, expected: %v", test.name, is, test.expected))
 	}
 }
@@ -160,43 +133,3 @@ func TestIsDotFile(t *testing.T) {
 		assert.Equal(t, test.expected, is, fmt.Sprintf("%v: is = %v, expected: %v", test.name, is, test.expected))
 	}
 }
-
-func TestFileCountListSort(t *testing.T) {
-	sampleData := FileCountList{{"a", 8}, {"b", 65}, {"c", 20}, {"d", 90}}
-	const ascending = "ASC"
-	const descending = "DESC"
-
-	tests := []struct {
-		name         string
-		data         FileCountList
-		order        string
-		expectedData FileCountList
-	}{
-		{
-			name:         "ascending order",
-			data:         sampleData,
-			order:        ascending,
-			expectedData: FileCountList{{"a", 8}, {"c", 20}, {"b", 65}, {"d", 90}},
-		},
-		{
-			name:         "descending order",
-			data:         sampleData,
-			order:        descending,
-			expectedData: FileCountList{{"d", 90}, {"b", 65}, {"c", 20}, {"a", 8}},
-		},
-	}
-
-	for _, test := range tests {
-		t.Run(test.name, func(t *testing.T) {
-			if test.order == descending {
-				sort.Sort(sort.Reverse(test.data))
-			} else {
-				sort.Sort(test.data)
-			}
-
-			for i := 0; i < len(test.data); i++ {
-				assert.Equal(t, test.data[i], test.expectedData[i], fmt.Sprintf("%v: FileCount at position %d = %v, expected: %v", test.name, i, test.data[i], test.expectedData[i]))
-			}
-		})
-	}
-}