GODRIVER-3587 Use raw bytes in valueReader #2120
There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,128 @@ | ||
| // Copyright (C) MongoDB, Inc. 2025-present. | ||
| // | ||
| // Licensed under the Apache License, Version 2.0 (the "License"); you may | ||
| // not use this file except in compliance with the License. You may obtain | ||
| // a copy of the License at http://www.apache.org/licenses/LICENSE-2.0 | ||
|
|
||
| package bson | ||
|
|
||
| import ( | ||
| "bytes" | ||
| "io" | ||
| ) | ||
|
|
||
| // bufferedValueReader implements the low-level byteSrc interface by reading | ||
| // directly from an in-memory byte slice. It provides efficient, zero-copy | ||
| // access for parsing BSON when the entire document is buffered in memory. | ||
| type bufferedValueReader struct { | ||
| buf []byte // entire BSON document | ||
| offset int64 // Current read index into buf | ||
| } | ||
|
|
||
| var _ valueReaderByteSrc = (*bufferedValueReader)(nil) | ||
|
|
||
| // Read reads up to len(p) bytes from the in-memory buffer, advancing the offset | ||
| // by the number of bytes read. | ||
| func (b *bufferedValueReader) readExact(p []byte) (int, error) { | ||
| if b.offset >= int64(len(b.buf)) { | ||
| return 0, io.EOF | ||
| } | ||
| n := copy(p, b.buf[b.offset:]) | ||
| b.offset += int64(n) | ||
| return n, nil | ||
| } | ||
|
|
||
| // ReadByte returns the single byte at buf[offset] and advances offset by 1. | ||
| func (b *bufferedValueReader) ReadByte() (byte, error) { | ||
| if b.offset >= int64(len(b.buf)) { | ||
| return 0, io.EOF | ||
| } | ||
|
Comment on lines
+37
to
+39
There was a problem hiding this comment. What are your thoughts about returning the remaining buf as well as the EOF for |
||
| b.offset++ | ||
| return b.buf[b.offset-1], nil | ||
| } | ||
|
|
||
| // peek returns buf[offset:offset+n] without advancing offset. | ||
| func (b *bufferedValueReader) peek(n int) ([]byte, error) { | ||
| // Ensure we don't read past the end of the buffer. | ||
| if int64(n)+b.offset > int64(len(b.buf)) { | ||
| return b.buf[b.offset:], io.EOF | ||
| } | ||
|
|
||
| // Return the next n bytes without advancing the offset | ||
| return b.buf[b.offset : b.offset+int64(n)], nil | ||
| } | ||
|
|
||
| // discard advances offset by n bytes, returning the number of bytes discarded. | ||
| func (b *bufferedValueReader) discard(n int) (int, error) { | ||
| // Ensure we don't read past the end of the buffer. | ||
| if int64(n)+b.offset > int64(len(b.buf)) { | ||
| // If we have exceeded the buffer length, discard only up to the end. | ||
| left := len(b.buf) - int(b.offset) | ||
| b.offset = int64(len(b.buf)) | ||
|
|
||
| return left, io.EOF | ||
| } | ||
|
|
||
| // Advance the read position | ||
| b.offset += int64(n) | ||
| return n, nil | ||
| } | ||
|
|
||
| // readSlice scans buf[offset:] for the first occurrence of delim, returns | ||
| // buf[offset:idx+1], and advances offset past it; errors if delim not found. | ||
| func (b *bufferedValueReader) readSlice(delim byte) ([]byte, error) { | ||
| // Ensure we don't read past the end of the buffer. | ||
| if b.offset >= int64(len(b.buf)) { | ||
| return nil, io.EOF | ||
| } | ||
|
|
||
| // Look for the delimiter in the remaining bytes | ||
| rem := b.buf[b.offset:] | ||
| idx := bytes.IndexByte(rem, delim) | ||
| if idx < 0 { | ||
| return nil, io.EOF | ||
| } | ||
|
|
||
| // Build the result slice up through the delimiter. | ||
| result := rem[:idx+1] | ||
|
|
||
| // Advance the offset past the delimiter. | ||
| b.offset += int64(idx + 1) | ||
|
|
||
| return result, nil | ||
| } | ||
|
|
||
| // pos returns the current read position in the buffer. | ||
| func (b *bufferedValueReader) pos() int64 { | ||
| return b.offset | ||
| } | ||
|
|
||
| // regexLength will return the total byte length of a BSON regex value. | ||
| func (b *bufferedValueReader) regexLength() (int32, error) { | ||
| rem := b.buf[b.offset:] | ||
|
|
||
| // Find end of the first C-string (pattern). | ||
| i := bytes.IndexByte(rem, 0x00) | ||
| if i < 0 { | ||
| return 0, io.EOF | ||
| } | ||
|
|
||
| // Find end of second C-string (options). | ||
| j := bytes.IndexByte(rem[i+1:], 0x00) | ||
| if j < 0 { | ||
| return 0, io.EOF | ||
| } | ||
|
|
||
| // Total length = first C-string length (pattern) + second C-string length | ||
| // (options) + 2 null terminators | ||
| return int32(i + j + 2), nil | ||
| } | ||
|
|
||
| func (*bufferedValueReader) streamable() bool { | ||
| return false | ||
| } | ||
|
|
||
| func (b *bufferedValueReader) reset() { | ||
| b.buf = nil | ||
| b.offset = 0 | ||
| } | ||
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,153 @@ | ||
| // Copyright (C) MongoDB, Inc. 2025-present. | ||
| // | ||
| // Licensed under the Apache License, Version 2.0 (the "License"); you may | ||
| // not use this file except in compliance with the License. You may obtain | ||
| // a copy of the License at http://www.apache.org/licenses/LICENSE-2.0 | ||
|
|
||
| package bson | ||
|
|
||
| import ( | ||
| "bytes" | ||
| "io" | ||
| "testing" | ||
|
|
||
| "go.mongodb.org/mongo-driver/v2/internal/assert" | ||
| "go.mongodb.org/mongo-driver/v2/internal/require" | ||
| ) | ||
|
|
||
| func TestBufferedvalueReader_discard(t *testing.T) { | ||
| tests := []struct { | ||
| name string | ||
| buf []byte | ||
| n int | ||
| want int | ||
| wantOffset int64 | ||
| wantErr error | ||
| }{ | ||
| { | ||
| name: "nothing", | ||
| buf: bytes.Repeat([]byte("a"), 1024), | ||
| n: 0, | ||
| want: 0, | ||
| wantOffset: 0, | ||
| wantErr: nil, | ||
| }, | ||
| { | ||
| name: "amount less than buffer size", | ||
| buf: bytes.Repeat([]byte("a"), 1024), | ||
| n: 100, | ||
| want: 100, | ||
| wantOffset: 100, | ||
| wantErr: nil, | ||
| }, | ||
| { | ||
| name: "amount greater than buffer size", | ||
| buf: bytes.Repeat([]byte("a"), 1024), | ||
| n: 10000, | ||
| want: 1024, | ||
| wantOffset: 1024, | ||
| wantErr: io.EOF, | ||
| }, | ||
| { | ||
| name: "exact buffer size", | ||
| buf: bytes.Repeat([]byte("a"), 1024), | ||
| n: 1024, | ||
| want: 1024, | ||
| wantOffset: 1024, | ||
| wantErr: nil, | ||
| }, | ||
| { | ||
| name: "from empty buffer", | ||
| buf: []byte{}, | ||
| n: 10, | ||
| want: 0, | ||
| wantOffset: 0, | ||
| wantErr: io.EOF, | ||
| }, | ||
| } | ||
|
|
||
| for _, tt := range tests { | ||
| t.Run(tt.name, func(t *testing.T) { | ||
| reader := &bufferedValueReader{buf: tt.buf, offset: 0} | ||
| n, err := reader.discard(tt.n) | ||
| if tt.wantErr != nil { | ||
| assert.ErrorIs(t, err, tt.wantErr, "Expected error %v, got %v", tt.wantErr, err) | ||
| } else { | ||
| require.NoError(t, err, "Expected no error when discarding %d bytes", tt.n) | ||
| } | ||
|
|
||
| assert.Equal(t, tt.want, n, "Expected to discard %d bytes, got %d", tt.want, n) | ||
| assert.Equal(t, tt.wantOffset, reader.offset, "Expected offset to be %d, got %d", tt.wantOffset, reader.offset) | ||
| }) | ||
| } | ||
| } | ||
|
|
||
| func TestBufferedvalueReader_peek(t *testing.T) { | ||
| tests := []struct { | ||
| name string | ||
| buf []byte | ||
| n int | ||
| offset int64 | ||
| want []byte | ||
| wantErr error | ||
| }{ | ||
| { | ||
| name: "nothing", | ||
| buf: bytes.Repeat([]byte("a"), 1024), | ||
| n: 0, | ||
| want: []byte{}, | ||
| wantErr: nil, | ||
| }, | ||
| { | ||
| name: "amount less than buffer size", | ||
| buf: bytes.Repeat([]byte("a"), 1024), | ||
| n: 100, | ||
| want: bytes.Repeat([]byte("a"), 100), | ||
| wantErr: nil, | ||
| }, | ||
| { | ||
| name: "amount greater than buffer size", | ||
| buf: bytes.Repeat([]byte("a"), 1024), | ||
| n: 10000, | ||
| want: bytes.Repeat([]byte("a"), 1024), | ||
| wantErr: io.EOF, | ||
| }, | ||
| { | ||
| name: "exact buffer size", | ||
| buf: bytes.Repeat([]byte("a"), 1024), | ||
| n: 1024, | ||
| want: bytes.Repeat([]byte("a"), 1024), | ||
| wantErr: nil, | ||
| }, | ||
| { | ||
| name: "from empty buffer", | ||
| buf: []byte{}, | ||
| n: 10, | ||
| want: []byte{}, | ||
| wantErr: io.EOF, | ||
| }, | ||
| { | ||
| name: "peek with offset", | ||
| buf: append(bytes.Repeat([]byte("a"), 100), bytes.Repeat([]byte("b"), 100)...), | ||
| offset: 100, | ||
| n: 100, | ||
| want: bytes.Repeat([]byte("b"), 100), | ||
| wantErr: nil, | ||
| }, | ||
| } | ||
|
|
||
| for _, tt := range tests { | ||
| t.Run(tt.name, func(t *testing.T) { | ||
| reader := &bufferedValueReader{buf: tt.buf, offset: tt.offset} | ||
| n, err := reader.peek(tt.n) | ||
| if tt.wantErr != nil { | ||
| assert.ErrorIs(t, err, tt.wantErr, "Expected error %v, got %v", tt.wantErr, err) | ||
| } else { | ||
| require.NoError(t, err, "Expected no error when peeking %d bytes", tt.n) | ||
| } | ||
|
|
||
| assert.Equal(t, tt.want, n, "Expected to peek %d bytes, got %d", len(tt.want), len(n)) | ||
| assert.Equal(t, tt.offset, reader.offset, "Expected offset to be %d, got %d", tt.offset, reader.offset) | ||
| }) | ||
| } | ||
| } |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Oops, something went wrong.
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Optional: This isn't a
ValueReader, so the name is a bit confusing. Consider a name likebufferedByteSrc.