Exclude generated files from language statistics (go-gitea#11653)

* Update go-enry to v2.5.2

go.mod

@@ -37,7 +37,7 @@ require (
  github.com/facebookgo/subset v0.0.0-20150612182917-8dac2c3c4870 // indirect
  github.com/gliderlabs/ssh v0.2.2
  github.com/glycerine/go-unsnap-stream v0.0.0-20190901134440-81cf024a9e0a // indirect
- github.com/go-enry/go-enry/v2 v2.3.0
+ github.com/go-enry/go-enry/v2 v2.5.2
  github.com/go-git/go-billy/v5 v5.0.0
  github.com/go-git/go-git/v5 v5.0.0
  github.com/go-openapi/jsonreference v0.19.3 // indirect

go.sum

@@ -193,10 +193,10 @@ github.com/glycerine/go-unsnap-stream v0.0.0-20190901134440-81cf024a9e0a h1:FQqo
 github.com/glycerine/go-unsnap-stream v0.0.0-20190901134440-81cf024a9e0a/go.mod h1:/20jfyN9Y5QPEAprSgKAUr+glWDY39ZiUEAYOEv5dsE=
 github.com/glycerine/goconvey v0.0.0-20190410193231-58a59202ab31 h1:gclg6gY70GLy3PbkQ1AERPfmLMMagS60DKF78eWwLn8=
 github.com/glycerine/goconvey v0.0.0-20190410193231-58a59202ab31/go.mod h1:Ogl1Tioa0aV7gstGFO7KhffUsb9M4ydbEbbxpcEDc24=
-github.com/go-enry/go-enry/v2 v2.3.0 h1:o8KwgY6uSplysrIpj+Y42J/xGPp90ogVpxE2Z3s8Unk=
-github.com/go-enry/go-enry/v2 v2.3.0/go.mod h1:+xFJwbqWi15bvqFHb2ELUWVRKFQtwB61+sDrkvvxxGI=
-github.com/go-enry/go-oniguruma v1.2.0 h1:oBO9XC1IDT9+AoWW5oFsa/7gFeOPacEqDbyXZKWXuDs=
-github.com/go-enry/go-oniguruma v1.2.0/go.mod h1:bWDhYP+S6xZQgiRL7wlTScFYBe023B6ilRZbCAD5Hf4=
+github.com/go-enry/go-enry/v2 v2.5.2 h1:3f3PFAO6JitWkPi1GQ5/m6Xu4gNL1U5soJ8QaYqJ0YQ=
+github.com/go-enry/go-enry/v2 v2.5.2/go.mod h1:GVzIiAytiS5uT/QiuakK7TF1u4xDab87Y8V5EJRpsIQ=
+github.com/go-enry/go-oniguruma v1.2.1 h1:k8aAMuJfMrqm/56SG2lV9Cfti6tC4x8673aHCcBk+eo=
+github.com/go-enry/go-oniguruma v1.2.1/go.mod h1:bWDhYP+S6xZQgiRL7wlTScFYBe023B6ilRZbCAD5Hf4=
 github.com/go-git/gcfg v1.5.0 h1:Q5ViNfGF8zFgyJWPqYwA7qGFoMTEiBmdlkcfRmpIMa4=
 github.com/go-git/gcfg v1.5.0/go.mod h1:5m20vg6GwYabIxaOonVkTdrILxQMpEShl1xiMF4ua+E=
 github.com/go-git/go-billy/v5 v5.0.0 h1:7NQHvd9FVid8VL4qVUMm8XifBK+2xCoZ2lSk0agRrHM=
@@ -616,8 +616,6 @@ github.com/tinylib/msgp v1.1.0/go.mod h1:+d+yLhGm8mzTaHzB+wgMYrodPfmZrzkirds8fDW
 github.com/tinylib/msgp v1.1.2 h1:gWmO7n0Ys2RBEb7GPYB9Ujq8Mk5p2U08lRnmMcGy6BQ=
 github.com/tinylib/msgp v1.1.2/go.mod h1:+d+yLhGm8mzTaHzB+wgMYrodPfmZrzkirds8fDWklFE=
 github.com/tmc/grpc-websocket-proxy v0.0.0-20190109142713-0ad062ec5ee5/go.mod h1:ncp9v5uamzpCO7NfCPTXjqaC+bZgJeR0sMTm6dMHP7U=
-github.com/toqueteos/trie v1.0.0 h1:8i6pXxNUXNRAqP246iibb7w/pSFquNTQ+uNfriG7vlk=
-github.com/toqueteos/trie v1.0.0/go.mod h1:Ywk48QhEqhU1+DwhMkJ2x7eeGxDHiGkAdc9+0DYcbsM=
 github.com/toqueteos/webbrowser v1.2.0 h1:tVP/gpK69Fx+qMJKsLE7TD8LuGWPnEV71wBN9rrstGQ=
 github.com/toqueteos/webbrowser v1.2.0/go.mod h1:XWoZq4cyp9WeUeak7w7LXRUQf1F1ATJMir8RTqb4ayM=
 github.com/tstranex/u2f v1.0.0 h1:HhJkSzDDlVSVIVt7pDJwCHQj67k7A5EeBgPmeD+pVsQ=
@@ -876,8 +874,6 @@ gopkg.in/testfixtures.v2 v2.5.0 h1:N08B7l2GzFQenyYbzqthDnKAA+cmb17iAZhhFxr7JHw=
 gopkg.in/testfixtures.v2 v2.5.0/go.mod h1:vyAq+MYCgNpR29qitQdLZhdbLFf4mR/2MFJRFoQZZ2M=
 gopkg.in/tomb.v1 v1.0.0-20141024135613-dd632973f1e7 h1:uRGJdciOHaEIrze2W8Q3AKkepLTh2hOroT7a+7czfdQ=
 gopkg.in/tomb.v1 v1.0.0-20141024135613-dd632973f1e7/go.mod h1:dt/ZhP58zS4L8KSrWDmTeBkI65Dw0HsyUHuEVlX15mw=
-gopkg.in/toqueteos/substring.v1 v1.0.2 h1:urLqCeMm6x/eTuQa1oZerNw8N1KNOIp5hD5kGL7lFsE=
-gopkg.in/toqueteos/substring.v1 v1.0.2/go.mod h1:Eb2Z1UYehlVK8LYW2WBVR2rwbujsz3aX8XDrM1vbNew=
 gopkg.in/warnings.v0 v0.1.2 h1:wFXVbFY8DY5/xOe1ECiWdKCzZlxgshcYVNkBHstARME=
 gopkg.in/warnings.v0 v0.1.2/go.mod h1:jksf8JmL6Qr/oQM2OXTHunEvvTAsrWBLb6OOjuVWRNI=
 gopkg.in/yaml.v2 v2.0.0-20170812160011-eb3733d160e7/go.mod h1:JAlM8MvJe8wmxCU4Bli9HhUf9+ttbYbLASfIpnQbh74=

modules/analyze/code_langauge.go

@@ -10,8 +10,8 @@ import (
  "github.com/go-enry/go-enry/v2"
 )
 
-// GetCodeLanguageWithCallback detects code language based on file name and content using callback
-func GetCodeLanguageWithCallback(filename string, contentFunc func() ([]byte, error)) string {
+// GetCodeLanguage detects code language based on file name and content
+func GetCodeLanguage(filename string, content []byte) string {
  if language, ok := enry.GetLanguageByExtension(filename); ok {
  return language
  }
@@ -20,17 +20,9 @@ func GetCodeLanguageWithCallback(filename string, contentFunc func() ([]byte, er
  return language
  }
 
- content, err := contentFunc()
- if err != nil {
+ if len(content) == 0 {
  return enry.OtherLanguage
  }
 
  return enry.GetLanguage(filepath.Base(filename), content)
 }
-
-// GetCodeLanguage detects code language based on file name and content
-func GetCodeLanguage(filename string, content []byte) string {
- return GetCodeLanguageWithCallback(filename, func() ([]byte, error) {
- return content, nil
- })
-}

modules/git/repo_language_stats.go

@@ -50,11 +50,15 @@ func (repo *Repository) GetLanguageStats(commitID string) (map[string]float32, e
  return nil
  }
 
+ // If content can not be read just do detection by filename
+ content, _ := readFile(f, fileSizeLimit)
+ if enry.IsGenerated(f.Name, content) {
+ return nil
+ }
+
  // TODO: Use .gitattributes file for linguist overrides
 
- language := analyze.GetCodeLanguageWithCallback(f.Name, func() ([]byte, error) {
- return readFile(f, fileSizeLimit)
- })
+ language := analyze.GetCodeLanguage(f.Name, content)
  if language == enry.OtherLanguage || language == "" {
  return nil
  }