Skip to content

Fix CSV render error #17406

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 6 commits into from
Oct 24, 2021
Merged
Show file tree
Hide file tree
Changes from 1 commit
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
24 changes: 8 additions & 16 deletions modules/csv/csv.go
Original file line number Diff line number Diff line change
Expand Up @@ -27,35 +27,27 @@ func CreateReader(input io.Reader, delimiter rune) *stdcsv.Reader {
}

// CreateReaderAndGuessDelimiter tries to guess the field delimiter from the content and creates a csv.Reader.
// Reads at most 10k bytes.
func CreateReaderAndGuessDelimiter(rd io.Reader) (*stdcsv.Reader, error) {
var data = make([]byte, 1e4)
size, err := rd.Read(data)
var buf = make([]byte, 1e4)
n, err := rd.Read(buf)
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Are you trying to fill that buffer if you at all can? There's no guarantee even if the size is 16Mb that the buffer will contain more than one byte. You need to use ReadAtLeast or ReadFull.

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Will add this in a follow up PR because that needs to be fixed in more places like LFS pointer detection.

if err != nil {
if err == io.EOF {
return CreateReader(bytes.NewReader([]byte{}), rune(',')), nil
}
return nil, err
}

delimiter := guessDelimiter(data[:size])

var newInput io.Reader
if size < 1e4 {
newInput = bytes.NewReader(data[:size])
} else {
newInput = io.MultiReader(bytes.NewReader(data), rd)
}

return CreateReader(newInput, delimiter), nil
return CreateReader(
io.MultiReader(bytes.NewReader(buf[:n]), rd),
guessDelimiter(buf[:n]),
), nil
}

// guessDelimiter scores the input CSV data against delimiters, and returns the best match.
// Reads at most 10k bytes & 10 lines.
func guessDelimiter(data []byte) rune {
maxLines := 10
maxBytes := util.Min(len(data), 1e4)
text := string(data[:maxBytes])
text = quoteRegexp.ReplaceAllLiteralString(text, "")
text := quoteRegexp.ReplaceAllLiteralString(string(data), "")
lines := strings.SplitN(text, "\n", maxLines+1)
lines = lines[:util.Min(maxLines, len(lines))]

Expand Down
23 changes: 15 additions & 8 deletions routers/web/repo/compare.go
Original file line number Diff line number Diff line change
Expand Up @@ -10,6 +10,7 @@ import (
"errors"
"fmt"
"html"
"io"
"net/http"
"path"
"path/filepath"
Expand Down Expand Up @@ -105,30 +106,36 @@ func setCsvCompareContext(ctx *context.Context) {

errTooLarge := errors.New(ctx.Locale.Tr("repo.error.csv.too_large"))

csvReaderFromCommit := func(c *git.Commit) (*csv.Reader, error) {
csvReaderFromCommit := func(c *git.Commit) (*csv.Reader, io.Closer, error) {
blob, err := c.GetBlobByPath(diffFile.Name)
if err != nil {
return nil, err
return nil, nil, err
}

if setting.UI.CSV.MaxFileSize != 0 && setting.UI.CSV.MaxFileSize < blob.Size() {
return nil, errTooLarge
return nil, nil, errTooLarge
}

reader, err := blob.DataAsync()
if err != nil {
return nil, err
return nil, nil, err
}
defer reader.Close()

return csv_module.CreateReaderAndGuessDelimiter(charset.ToUTF8WithFallbackReader(reader))
csvReader, err := csv_module.CreateReaderAndGuessDelimiter(charset.ToUTF8WithFallbackReader(reader))
return csvReader, reader, err
}

baseReader, err := csvReaderFromCommit(baseCommit)
baseReader, baseBlobCloser, err := csvReaderFromCommit(baseCommit)
if baseBlobCloser != nil {
defer baseBlobCloser.Close()
}
if err == errTooLarge {
return CsvDiffResult{nil, err.Error()}
}
headReader, err := csvReaderFromCommit(headCommit)
headReader, headBlobCloser, err := csvReaderFromCommit(headCommit)
if headBlobCloser != nil {
defer headBlobCloser.Close()
}
if err == errTooLarge {
return CsvDiffResult{nil, err.Error()}
}
Expand Down