Add GetArchiveReader() (#476)

This PR close #475

It complements the `Client.GetArchive` call, which returns a byte slice and hence is unsuitable for use with large repositories, with a `Client.GetArchiveReader` method that returns a `io.ReadCloser` that streams the retrieved archvie and, therefore, induces a much smaller memory footprint on the calling client.

Co-authored-by: Peter Gardfjäll <peter.gardfjall.work@gmail.com>
Reviewed-on: https://gitea.com/gitea/go-sdk/pulls/476
Reviewed-by: 6543 <6543@obermui.de>
Reviewed-by: Norwin <noerw@noreply.gitea.io>
Co-authored-by: petergardfjall <petergardfjall@noreply.gitea.io>
Co-committed-by: petergardfjall <petergardfjall@noreply.gitea.io>

gitea/client.go

@@ -198,6 +198,48 @@ func (c *Client) doRequest(method, path string, header http.Header, body io.Read
 	return &Response{resp}, nil
 }
 
+// Converts a response for a HTTP status code indicating an error condition
+// (non-2XX) to a well-known error value and response body. For non-problematic
+// (2XX) status codes nil will be returned. Note that on a non-2XX response, the
+// response body stream will have been read and, hence, is closed on return.
+func statusCodeToErr(resp *Response) (body []byte, err error) {
+	// no error
+	if resp.StatusCode/100 == 2 {
+		return nil, nil
+	}
+
+	//
+	// error: body will be read for details
+	//
+	defer resp.Body.Close()
+	data, err := ioutil.ReadAll(resp.Body)
+	if err != nil {
+		return nil, fmt.Errorf("body read on HTTP error %d: %v", resp.StatusCode, err)
+	}
+
+	switch resp.StatusCode {
+	case 403:
+		return data, errors.New("403 Forbidden")
+	case 404:
+		return data, errors.New("404 Not Found")
+	case 409:
+		return data, errors.New("409 Conflict")
+	case 422:
+		return data, fmt.Errorf("422 Unprocessable Entity: %s", string(data))
+	}
+
+	path := resp.Request.URL.Path
+	method := resp.Request.Method
+	header := resp.Request.Header
+	errMap := make(map[string]interface{})
+	if err = json.Unmarshal(data, &errMap); err != nil {
+		// when the JSON can't be parsed, data was probably empty or a
+		// plain string, so we try to return a helpful error anyway
+		return data, fmt.Errorf("Unknown API Error: %d\nRequest: '%s' with '%s' method '%s' header and '%s' body", resp.StatusCode, path, method, header, string(data))
+	}
+	return data, errors.New(errMap["message"].(string))
+}
+
 func (c *Client) getResponse(method, path string, header http.Header, body io.Reader) ([]byte, *Response, error) {
 	resp, err := c.doRequest(method, path, header, body)
 	if err != nil {
@@ -205,32 +247,18 @@ func (c *Client) getResponse(method, path string, header http.Header, body io.Re
 	}
 	defer resp.Body.Close()
 
-	data, err := ioutil.ReadAll(resp.Body)
+	// check for errors
+	data, err := statusCodeToErr(resp)
+	if err != nil {
+		return data, resp, err
+	}
+
+	// success (2XX), read body
+	data, err = ioutil.ReadAll(resp.Body)
 	if err != nil {
 		return nil, resp, err
 	}
 
-	switch resp.StatusCode {
-	case 403:
-		return data, resp, errors.New("403 Forbidden")
-	case 404:
-		return data, resp, errors.New("404 Not Found")
-	case 409:
-		return data, resp, errors.New("409 Conflict")
-	case 422:
-		return data, resp, fmt.Errorf("422 Unprocessable Entity: %s", string(data))
-	}
-
-	if resp.StatusCode/100 != 2 {
-		errMap := make(map[string]interface{})
-		if err = json.Unmarshal(data, &errMap); err != nil {
-			// when the JSON can't be parsed, data was probably empty or a plain string,
-			// so we try to return a helpful error anyway
-			return data, resp, fmt.Errorf("Unknown API Error: %d\nRequest: '%s' with '%s' method '%s' header and '%s' body", resp.StatusCode, path, method, header, string(data))
-		}
-		return data, resp, errors.New(errMap["message"].(string))
-	}
-
 	return data, resp, nil
 }
 

gitea/repo.go

@@ -9,6 +9,7 @@ import (
 	"bytes"
 	"encoding/json"
 	"fmt"
+	"io"
 	"net/url"
 	"strings"
 	"time"
@@ -420,3 +421,20 @@ const (
 func (c *Client) GetArchive(owner, repo, ref string, ext ArchiveType) ([]byte, *Response, error) {
 	return c.getResponse("GET", fmt.Sprintf("/repos/%s/%s/archive/%s%s", owner, repo, url.PathEscape(ref), ext), nil, nil)
 }
+
+// GetArchiveReader gets a `git archive` for a particular tree-ish git reference
+// such as a branch name (`master`), a commit hash (`70b7c74b33`), a tag
+// (`v1.2.1`). The archive is returned as a byte stream in a ReadCloser. It is
+// the responsibility of the client to close the reader.
+func (c *Client) GetArchiveReader(owner, repo, ref string, ext ArchiveType) (io.ReadCloser, *Response, error) {
+	resp, err := c.doRequest("GET", fmt.Sprintf("/repos/%s/%s/archive/%s%s", owner, repo, url.PathEscape(ref), ext), nil, nil)
+	if err != nil {
+		return nil, resp, err
+	}
+
+	if _, err := statusCodeToErr(resp); err != nil {
+		return nil, resp, err
+	}
+
+	return resp.Body, resp, nil
+}

gitea/repo_test.go

@@ -5,6 +5,8 @@
 package gitea
 
 import (
+	"bytes"
+	"io"
 	"log"
 	"testing"
 	"time"
@@ -139,6 +141,22 @@ func TestGetArchive(t *testing.T) {
 	assert.EqualValues(t, 1620, len(archive))
 }
 
+func TestGetArchiveReader(t *testing.T) {
+	log.Println("== TestGetArchiveReader ==")
+	c := newTestClient()
+	repo, _ := createTestRepo(t, "ToDownload", c)
+	time.Sleep(time.Second / 2)
+	r, _, err := c.GetArchiveReader(repo.Owner.UserName, repo.Name, "master", ZipArchive)
+	assert.NoError(t, err)
+	defer r.Close()
+
+	archive := bytes.NewBuffer(nil)
+	nBytes, err := io.Copy(archive, r)
+	assert.NoError(t, err)
+	assert.EqualValues(t, 1620, nBytes)
+	assert.EqualValues(t, 1620, len(archive.Bytes()))
+}
+
 // standard func to create a init repo for test routines
 func createTestRepo(t *testing.T, name string, c *Client) (*Repository, error) {
 	user, _, uErr := c.GetMyUserInfo()