1
0
Fork 1
mirror of https://github.com/vbatts/tar-split.git synced 2025-03-20 23:34:08 +00:00
tar-split/tar/asm/iterate.go
Vincent Batts c8b16f6803
*.go: move the carrierd archive/tar to internal/
The feature of golang having source in a project path with "internal"
makes the code only able to be imported by the project itself.
https://docs.google.com/document/d/1e8kOo3r51b2BWtTs_1uADIA5djfXhPT36s6eHVRIvaU/edit?tab=t.0

Since we have this carried version of `archive/tar` with our byte
accounting patches, it is also means the version of archive/tar has not
kept up with upstream.

our tar-split libraries do not even utilize all the function calls of
our carried archive/tar, so might as well limit anyone else in the world
from calling them as well.

Signed-off-by: Vincent Batts <vbatts@hashbangbash.com>
2025-01-28 11:04:47 -05:00

57 lines
1.7 KiB
Go
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

package asm
import (
"bytes"
"fmt"
"io"
"github.com/vbatts/tar-split/internal/archive/tar"
"github.com/vbatts/tar-split/tar/storage"
)
// IterateHeaders calls handler for each tar header provided by Unpacker
func IterateHeaders(unpacker storage.Unpacker, handler func(hdr *tar.Header) error) error {
// We assume about NewInputTarStream:
// - There is a separate SegmentType entry for every tar header, but only one SegmentType entry for the full header incl. any extensions
// - (There is a FileType entry for every tar header, we ignore it)
// - Trailing padding of a file, if any, is included in the next SegmentType entry
// - At the end, there may be SegmentType entries just for the terminating zero blocks.
var pendingPadding int64 = 0
for {
tsEntry, err := unpacker.Next()
if err != nil {
if err == io.EOF {
return nil
}
return fmt.Errorf("reading tar-split entries: %w", err)
}
switch tsEntry.Type {
case storage.SegmentType:
payload := tsEntry.Payload
if int64(len(payload)) < pendingPadding {
return fmt.Errorf("expected %d bytes of padding after previous file, but next SegmentType only has %d bytes", pendingPadding, len(payload))
}
payload = payload[pendingPadding:]
pendingPadding = 0
tr := tar.NewReader(bytes.NewReader(payload))
hdr, err := tr.Next()
if err != nil {
if err == io.EOF { // Probably the last entry, but lets let the unpacker drive that.
break
}
return fmt.Errorf("decoding a tar header from a tar-split entry: %w", err)
}
if err := handler(hdr); err != nil {
return err
}
pendingPadding = tr.ExpectedPadding()
case storage.FileType:
// Nothing
default:
return fmt.Errorf("unexpected tar-split entry type %q", tsEntry.Type)
}
}
}