Add Tabular Diff for CSV files (#14661)

Implements request #14320 The rendering of CSV files does match the diff style.

* Moved CSV logic into base package.

* Added method to create a tabular diff.

* Added CSV compare context.

* Added CSV diff template.

* Use new table style in CSV markup.

* Added file size limit for CSV rendering.

* Display CSV parser errors in diff.

* Lazy read single file.

* Lazy read rows for full diff.

* Added unit tests for various CSV changes.

routers/repo/commit.go

@@ -336,9 +336,8 @@ func Diff(ctx *context.Context) {
 			return
 		}
 	}
-	setImageCompareContext(ctx, parentCommit, commit)
 	headTarget := path.Join(userName, repoName)
-	setPathsCompareContext(ctx, parentCommit, commit, headTarget)
+	setCompareContext(ctx, parentCommit, commit, headTarget)
 	ctx.Data["Title"] = commit.Summary() + " · " + base.ShortSha(commitID)
 	ctx.Data["Commit"] = commit
 	verification := models.ParseCommitWithSignature(commit)

routers/repo/compare.go

@@ -6,14 +6,20 @@ package repo
 
 import (
 	"bufio"
+	"encoding/csv"
+	"errors"
 	"fmt"
 	"html"
+	"io/ioutil"
 	"path"
+	"path/filepath"
 	"strings"
 
 	"code.gitea.io/gitea/models"
 	"code.gitea.io/gitea/modules/base"
+	"code.gitea.io/gitea/modules/charset"
 	"code.gitea.io/gitea/modules/context"
+	csv_module "code.gitea.io/gitea/modules/csv"
 	"code.gitea.io/gitea/modules/git"
 	"code.gitea.io/gitea/modules/log"
 	"code.gitea.io/gitea/modules/setting"
@@ -26,6 +32,16 @@ const (
 	tplBlobExcerpt base.TplName = "repo/diff/blob_excerpt"
 )
 
+// setCompareContext sets context data.
+func setCompareContext(ctx *context.Context, base *git.Commit, head *git.Commit, headTarget string) {
+	ctx.Data["BaseCommit"] = base
+	ctx.Data["HeadCommit"] = head
+
+	setPathsCompareContext(ctx, base, head, headTarget)
+	setImageCompareContext(ctx, base, head)
+	setCsvCompareContext(ctx)
+}
+
 // setPathsCompareContext sets context data for source and raw paths
 func setPathsCompareContext(ctx *context.Context, base *git.Commit, head *git.Commit, headTarget string) {
 	sourcePath := setting.AppSubURL + "/%s/src/commit/%s"
@@ -65,6 +81,73 @@ func setImageCompareContext(ctx *context.Context, base *git.Commit, head *git.Co
 	}
 }
 
+// setCsvCompareContext sets context data that is required by the CSV compare template
+func setCsvCompareContext(ctx *context.Context) {
+	ctx.Data["IsCsvFile"] = func(diffFile *gitdiff.DiffFile) bool {
+		extension := strings.ToLower(filepath.Ext(diffFile.Name))
+		return extension == ".csv" || extension == ".tsv"
+	}
+
+	type CsvDiffResult struct {
+		Sections []*gitdiff.TableDiffSection
+		Error    string
+	}
+
+	ctx.Data["CreateCsvDiff"] = func(diffFile *gitdiff.DiffFile, baseCommit *git.Commit, headCommit *git.Commit) CsvDiffResult {
+		if diffFile == nil || baseCommit == nil || headCommit == nil {
+			return CsvDiffResult{nil, ""}
+		}
+
+		errTooLarge := errors.New(ctx.Locale.Tr("repo.error.csv.too_large"))
+
+		csvReaderFromCommit := func(c *git.Commit) (*csv.Reader, error) {
+			blob, err := c.GetBlobByPath(diffFile.Name)
+			if err != nil {
+				return nil, err
+			}
+
+			if setting.UI.CSV.MaxFileSize != 0 && setting.UI.CSV.MaxFileSize < blob.Size() {
+				return nil, errTooLarge
+			}
+
+			reader, err := blob.DataAsync()
+			if err != nil {
+				return nil, err
+			}
+			defer reader.Close()
+
+			b, err := ioutil.ReadAll(reader)
+			if err != nil {
+				return nil, err
+			}
+
+			b = charset.ToUTF8WithFallback(b)
+
+			return csv_module.CreateReaderAndGuessDelimiter(b), nil
+		}
+
+		baseReader, err := csvReaderFromCommit(baseCommit)
+		if err == errTooLarge {
+			return CsvDiffResult{nil, err.Error()}
+		}
+		headReader, err := csvReaderFromCommit(headCommit)
+		if err == errTooLarge {
+			return CsvDiffResult{nil, err.Error()}
+		}
+
+		sections, err := gitdiff.CreateCsvDiff(diffFile, baseReader, headReader)
+		if err != nil {
+			errMessage, err := csv_module.FormatError(err, ctx.Locale)
+			if err != nil {
+				log.Error("RenderCsvDiff failed: %v", err)
+				return CsvDiffResult{nil, ""}
+			}
+			return CsvDiffResult{nil, errMessage}
+		}
+		return CsvDiffResult{sections, ""}
+	}
+}
+
 // ParseCompareInfo parse compare info between two commit for preparing comparing references
 func ParseCompareInfo(ctx *context.Context) (*models.User, *models.Repository, *git.Repository, *git.CompareInfo, string, string) {
 	baseRepo := ctx.Repo.Repository
@@ -490,9 +573,8 @@ func PrepareCompareDiff(
 	ctx.Data["Username"] = headUser.Name
 	ctx.Data["Reponame"] = headRepo.Name
 
-	setImageCompareContext(ctx, baseCommit, headCommit)
 	headTarget := path.Join(headUser.Name, repo.Name)
-	setPathsCompareContext(ctx, baseCommit, headCommit, headTarget)
+	setCompareContext(ctx, baseCommit, headCommit, headTarget)
 
 	return false
 }

routers/repo/pull.go

@@ -591,7 +591,6 @@ func ViewPullFiles(ctx *context.Context) {
 		gitRepo       *git.Repository
 	)
 
-	var headTarget string
 	var prInfo *git.CompareInfo
 	if pull.HasMerged {
 		prInfo = PrepareMergedViewPullInfo(ctx, issue)
@@ -618,7 +617,6 @@ func ViewPullFiles(ctx *context.Context) {
 	startCommitID = prInfo.MergeBase
 	endCommitID = headCommitID
 
-	headTarget = path.Join(ctx.Repo.Owner.Name, ctx.Repo.Repository.Name)
 	ctx.Data["Username"] = ctx.Repo.Owner.Name
 	ctx.Data["Reponame"] = ctx.Repo.Repository.Name
 	ctx.Data["AfterCommitID"] = endCommitID
@@ -672,8 +670,8 @@ func ViewPullFiles(ctx *context.Context) {
 		}
 	}
 
-	setImageCompareContext(ctx, baseCommit, commit)
-	setPathsCompareContext(ctx, baseCommit, commit, headTarget)
+	headTarget := path.Join(ctx.Repo.Owner.Name, ctx.Repo.Repository.Name)
+	setCompareContext(ctx, baseCommit, commit, headTarget)
 
 	ctx.Data["RequireHighlightJS"] = true
 	ctx.Data["RequireSimpleMDE"] = true