[feature] support processing of (many) more media types (#3090)

* initial work replacing our media decoding / encoding pipeline with ffprobe + ffmpeg

* specify the video codec to use when generating static image from emoji

* update go-storage library (fixes incompatibility after updating go-iotools)

* maintain image aspect ratio when generating a thumbnail for it

* update readme to show go-ffmpreg

* fix a bunch of media tests, move filesize checking to callers of media manager for more flexibility

* remove extra debug from error message

* fix up incorrect function signatures

* update PutFile to just use regular file copy, as changes are file is on separate partition

* fix remaining tests, remove some unneeded tests now we're working with ffmpeg/ffprobe

* update more tests, add more code comments

* add utilities to generate processed emoji / media outputs

* fix remaining tests

* add test for opus media file, add license header to utility cmds

* limit the number of concurrently available ffmpeg / ffprobe instances

* reduce number of instances

* further reduce number of instances

* fix envparsing test with configuration variables

* update docs and configuration with new media-{local,remote}-max-size variables
This commit is contained in:
kim 2024-07-12 09:39:47 +00:00 committed by GitHub
commit cde2fb6244
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
376 changed files with 8026 additions and 54091 deletions

View file

@ -31,7 +31,6 @@ import (
"strings"
"time"
"codeberg.org/gruf/go-bytesize"
"codeberg.org/gruf/go-cache/v3"
errorsv2 "codeberg.org/gruf/go-errors/v2"
"codeberg.org/gruf/go-iotools"
@ -89,9 +88,6 @@ type Config struct {
// WriteBufferSize: see http.Transport{}.WriteBufferSize.
WriteBufferSize int
// MaxBodySize determines the maximum fetchable body size.
MaxBodySize int64
// Timeout: see http.Client{}.Timeout.
Timeout time.Duration
@ -111,7 +107,6 @@ type Config struct {
type Client struct {
client http.Client
badHosts cache.TTLCache[string, struct{}]
bodyMax int64
retries uint
}
@ -137,11 +132,6 @@ func New(cfg Config) *Client {
cfg.MaxIdleConns = cfg.MaxOpenConnsPerHost * 10
}
if cfg.MaxBodySize <= 0 {
// By default set this to a reasonable 40MB.
cfg.MaxBodySize = int64(40 * bytesize.MiB)
}
// Protect the dialer
// with IP range sanitizer.
d.Control = (&Sanitizer{
@ -151,7 +141,6 @@ func New(cfg Config) *Client {
// Prepare client fields.
c.client.Timeout = cfg.Timeout
c.bodyMax = cfg.MaxBodySize
// Prepare transport TLS config.
tlsClientConfig := &tls.Config{
@ -377,31 +366,15 @@ func (c *Client) do(r *Request) (rsp *http.Response, retry bool, err error) {
rbody := (io.Reader)(rsp.Body)
cbody := (io.Closer)(rsp.Body)
var limit int64
if limit = rsp.ContentLength; limit < 0 {
// If unknown, use max as reader limit.
limit = c.bodyMax
}
// Don't trust them, limit body reads.
rbody = io.LimitReader(rbody, limit)
// Wrap closer to ensure entire body drained BEFORE close.
// Wrap closer to ensure body drained BEFORE close.
cbody = iotools.CloserAfterCallback(cbody, func() {
_, _ = discard.ReadFrom(rbody)
})
// Wrap body with limit.
rsp.Body = &struct {
io.Reader
io.Closer
}{rbody, cbody}
// Check response body not too large.
if rsp.ContentLength > c.bodyMax {
_ = rsp.Body.Close()
return nil, false, ErrBodyTooLarge
// Set the wrapped response body.
rsp.Body = &iotools.ReadCloserType{
Reader: rbody,
Closer: cbody,
}
return rsp, true, nil

View file

@ -48,44 +48,19 @@ var bodies = []string{
"body with\r\nnewlines",
}
func TestHTTPClientSmallBody(t *testing.T) {
func TestHTTPClientBody(t *testing.T) {
for _, body := range bodies {
_TestHTTPClientWithBody(t, []byte(body), int(^uint16(0)))
testHTTPClientWithBody(t, []byte(body))
}
}
func TestHTTPClientExactBody(t *testing.T) {
for _, body := range bodies {
_TestHTTPClientWithBody(t, []byte(body), len(body))
}
}
func TestHTTPClientLargeBody(t *testing.T) {
for _, body := range bodies {
_TestHTTPClientWithBody(t, []byte(body), len(body)-1)
}
}
func _TestHTTPClientWithBody(t *testing.T, body []byte, max int) {
func testHTTPClientWithBody(t *testing.T, body []byte) {
var (
handler http.HandlerFunc
expect []byte
expectErr error
)
// If this is a larger body, reslice and
// set error so we know what to expect
expect = body
if max < len(body) {
expect = expect[:max]
expectErr = httpclient.ErrBodyTooLarge
}
// Create new HTTP client with maximum body size
client := httpclient.New(httpclient.Config{
MaxBodySize: int64(max),
DisableCompression: true,
AllowRanges: []netip.Prefix{
// Loopback (used by server)
@ -110,10 +85,8 @@ func _TestHTTPClientWithBody(t *testing.T, body []byte, max int) {
// Perform the test request
rsp, err := client.Do(req)
if !errors.Is(err, expectErr) {
if err != nil {
t.Fatalf("error performing client request: %v", err)
} else if err != nil {
return // expected error
}
defer rsp.Body.Close()
@ -124,8 +97,8 @@ func _TestHTTPClientWithBody(t *testing.T, body []byte, max int) {
}
// Check actual response body matches expected
if !bytes.Equal(expect, check) {
t.Errorf("response body did not match expected: expect=%q actual=%q", string(expect), string(check))
if !bytes.Equal(body, check) {
t.Errorf("response body did not match expected: expect=%q actual=%q", string(body), string(check))
}
}