From 3e8618a543cc942e5de4d4728b0f107b18b998ff Mon Sep 17 00:00:00 2001 From: Lauris BH Date: Fri, 19 Jun 2020 13:10:03 +0300 Subject: [PATCH] For language detection do not try to analyze big files by content (#11971) (#11975) --- modules/git/repo_language_stats.go | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/modules/git/repo_language_stats.go b/modules/git/repo_language_stats.go index 06d7d6aba..06ff70449 100644 --- a/modules/git/repo_language_stats.go +++ b/modules/git/repo_language_stats.go @@ -17,7 +17,8 @@ import ( "github.com/go-git/go-git/v5/plumbing/object" ) -const fileSizeLimit int64 = 16 * 1024 * 1024 +const fileSizeLimit int64 = 16 * 1024 // 16 KiB +const bigFileSize int64 = 1024 * 1024 // 1 MiB // specialLanguages defines list of languages that are excluded from the calculation // unless they are the only language present in repository. Only languages which under @@ -62,8 +63,11 @@ func (repo *Repository) GetLanguageStats(commitID string) (map[string]int64, err return nil } - // If content can not be read just do detection by filename - content, _ := readFile(f, fileSizeLimit) + // If content can not be read or file is too big just do detection by filename + var content []byte + if f.Size <= bigFileSize { + content, _ = readFile(f, fileSizeLimit) + } if enry.IsGenerated(f.Name, content) { return nil }