gen: same alias/language name lookup as in linguis

Includes generated code, to keep commits atomic.

Consits of:
 - code generator for alias produces new API
 - retrofiting all clients to a new API
 - generated code data/aliases.go

Signed-off-by: Alexander Bezzubov <bzz@apache.org>

common.go

@@ -455,9 +455,8 @@ func GetLanguageType(language string) (langType Type) {
 // GetLanguageByAlias returns either the language related to the given alias and ok set to true
 // or Otherlanguage and ok set to false if the alias is not recognized.
 func GetLanguageByAlias(alias string) (lang string, ok bool) {
-	a := strings.Split(alias, `,`)[0]
-	a = strings.ToLower(a)
-	lang, ok = data.LanguagesByAlias[a]
+	a := strings.SplitN(alias, `,`, 2)[0]
+	lang, ok = data.LanguagesByAlias(a)
 	if !ok {
 		lang = OtherLanguage
 	}

data/heuristics.go

@@ -22,7 +22,13 @@ func (h *Heuristics) Match(data []byte) []string {
 	var matchedLangs []string
 	for _, matcher := range *h {
 		if matcher.Match(data) {
-			for _, lang := range matcher.(Rule).GetLanguages() {
+			for _, langOrAlias := range matcher.(Rule).GetLanguages() {
+				lang, ok := LanguagesByAlias(langOrAlias)
+				if !ok { // should never happen
+					// language name/alias in heuristics.yml is not consistent with languages.yml
+					// but we do not surface any error on the API
+					continue
+				}
 				matchedLangs = append(matchedLangs, lang)
 			}
 			break

internal/code-generator/assets/alias.go.tmpl

@@ -1,9 +1,34 @@
 package data
 
-// LanguagesByAlias keeps alias for different languages and use the name of the languages as an alias too.
+import "strings"
+
+// LanguagesByAliasMap keeps alias for different languages and use the name of the languages as an alias too.
 // All the keys (alias or not) are written in lower case and the whitespaces has been replaced by underscores.
-var LanguagesByAlias = map[string]string{
+var LanguagesByAliasMap = map[string]string{
 	{{range $alias, $language := . -}}
-	"{{ $alias }}":	{{ printf "%q" $language -}},
+		"{{ $alias }}":	{{ printf "%q" $language -}},
 	{{end -}}
 }
+
+// LanguagesByAlias looks up the language name by it's alias or name.
+// It mirrors the logic of github linguist and is needed e.g for heuristcs.yml
+// that mixes names and aliases in a language field (see XPM example).
+func LanguagesByAlias(langOrAlias string) (lang string, ok bool) {
+	k := ConvertToAliasKey(langOrAlias)
+	lang, ok = LanguagesByAliasMap[k]
+	return
+}
+
+
+// ConvertToAliasKey converts language name to a key in LanguagesByAliasMap.
+// Following
+//  - internal.code-generator.generator.convertToAliasKey()
+//  - GetLanguageByAlias()
+// conventions.
+// It is here to avoid dependency on "generate" and "enry" packages.
+func ConvertToAliasKey(langName string) string {
+	ak := strings.SplitN(langName, `,`, 2)[0]
+	ak = strings.Replace(ak, ` `, `_`, -1)
+	ak = strings.ToLower(ak)
+	return ak
+}