[v6] Make hash usage extensible to enable concurrent support of multiple hash types #1362
Conversation
| const ( | ||
| // CryptoType defines what hash algorithm is being used. | ||
| CryptoType = crypto.SHA1 | ||
| // Size defines the amount of bytes the hash yields. | ||
| Size = 20 | ||
| Size = SHA1Size | ||
| // HexSize defines the strings size of the hash when represented in hexadecimal. | ||
| HexSize = 40 | ||
| HexSize = SHA1HexSize | ||
| ) |
There was a problem hiding this comment.
This is still being used, instead of replacing all the refs to SHA1Size, to then have to change them again later, I think we are better off leaving as is and replace as the specific packages get multi-hash support.
plumbing/hash/hash_zero.go
Outdated
| func Zero() SHA1Hash { | ||
| return SHA1Hash{} | ||
| } | ||
|
|
||
| func ZeroSHA256() SHA256Hash { | ||
| return SHA256Hash{} | ||
| } |
There was a problem hiding this comment.
Can we replace these with var zero hash.SHA1Hash and var zero256 hash.SHA256Hash?
There was a problem hiding this comment.
The main reason for a func was to avoid the awkard position were API users are able to change the meaning of zero, using a single var will revert to that. However, I agree with the change for the time being, but it is likely this code will change again before it being ready for merging back to v6-exp.
plumbing/hash/hash_sha256.go
Outdated
|
|
||
| func (ih SHA256Hash) IsZero() bool { | ||
| var empty SHA256Hash | ||
| return ih == empty |
There was a problem hiding this comment.
| func (ih SHA256Hash) IsZero() bool { | |
| var empty SHA256Hash | |
| return ih == empty | |
| var empty256 SHA256Hash | |
| func (ih SHA256Hash) IsZero() bool { | |
| return ih == empty256 |
plumbing/hash/hash_sha1.go
Outdated
|
|
||
| func (ih SHA1Hash) IsZero() bool { | ||
| var empty SHA1Hash | ||
| return ih == empty | ||
| } |
There was a problem hiding this comment.
| func (ih SHA1Hash) IsZero() bool { | |
| var empty SHA1Hash | |
| return ih == empty | |
| } | |
| var empty SHA1Hash | |
| func (ih SHA1Hash) IsZero() bool { | |
| return ih == empty | |
| } |
plumbing/hash/new.go
Outdated
| // ZeroFromHash returns a zeroed hash based on the given hash.Hash. | ||
| // | ||
| // Defaults to SHA1-sized hash if the provided hash is not supported. | ||
| func ZeroFromHash(h hash.Hash) ObjectID { | ||
| switch h.Size() { | ||
| case SHA256Size: | ||
| return SHA256Hash{} | ||
| default: | ||
| return SHA1Hash{} | ||
| } | ||
| } | ||
|
|
||
| // ZeroFromHash returns a zeroed hash based on the given ObjectFormat. | ||
| // | ||
| // Defaults to SHA1-sized hash if the provided format is not supported. | ||
| func ZeroFromObjectFormat(f format.ObjectFormat) ObjectID { | ||
| switch f { | ||
| case format.SHA256: | ||
| return SHA256Hash{} | ||
| default: | ||
| return SHA1Hash{} | ||
| } | ||
| } |
There was a problem hiding this comment.
We don't use these funcs now. We can add them later when the need arises
| import "io" | ||
|
|
||
| // ObjectID represents a calculated hash, which is immutable by nature. | ||
| type ObjectID interface { |
There was a problem hiding this comment.
I'd rename this to Hash to keep naming consistent
There was a problem hiding this comment.
I am holding back on making this change while plumbing.Hash exists, to avoid additional confusion on having multiple "Hash" types. At present plumbing.Hash has more than one use which is leading me to think that potentially it could be split into two different types. Either way, let's make that decision together with the final design for the API (as per discord thread) - which will be in a follow-up PR.
|
|
||
| // LazyObjectID represents an object hash which may not be known at the time | ||
| // the object is created. Or an objectID which changes during its lifetime. | ||
| type LazyObjectID interface { |
There was a problem hiding this comment.
Same, I'd rename this to LazyHash
Signed-off-by: Paulo Gomes <paulo.gomes@suse.com>
Signed-off-by: Paulo Gomes <paulo.gomes@suse.com>
Signed-off-by: Paulo Gomes <paulo.gomes@suse.com>
The introduction of both new types are a stepping stone to enable SHA256 support concurrently - without the need for a build tag. ObjectID provides a way to represent varied sized hashes (for SHA1 or SHA256) with a single type, while keeping similar performance of the existing plumbing.Hash. The names were picked in order to provide a clearer distinction on what they do. ObjectID is the result of a hash operation and can't be changed once calculated. ObjectHasher, adds the Git object header before a normal hash operation that hash.Hash or plumbing/hash.Hash do. The performance results shows that SHA1 operations could be slower, however for SHA256 it can be over 3 times faster: ~Hasher/hasher-sha1-100B-16 2202226 574.7 ns/op 174.00 MB/s 272 B/op 7 allocs/op ~Hasher/objecthash-sha1-100B-16 1511851 772.6 ns/op 129.43 MB/s 272 B/op 9 allocs/op ~Hasher/objecthash-sha256-100B-16 5057584 247.4 ns/op 404.21 MB/s 96 B/op 7 allocs/op ~Hasher/hasher-sha1-5000B-16 96476 12523 ns/op 399.25 MB/s 5536 B/op 7 allocs/op ~Hasher/objecthash-sha1-5000B-16 105376 10983 ns/op 455.27 MB/s 272 B/op 9 allocs/op ~Hasher/objecthash-sha256-5000B-16 420741 2828 ns/op 1767.77 MB/s 96 B/op 7 allocs/op ~HashFromHex/hasher-parse-sha1-valid-16 23243548 64.65 ns/op 618.69 MB/s 48 B/op 1 allocs/op ~HashFromHex/objecthash-fromhex-sha1-valid-16 18120699 79.62 ns/op 502.36 MB/s 72 B/op 2 allocs/op ~HashFromHex/objecthash-fromhex-sha256-valid-16 8969871 124.2 ns/op 515.22 MB/s 96 B/op 2 allocs/op Signed-off-by: Paulo Gomes <paulo.gomes@suse.com>
Signed-off-by: Paulo Gomes <pjbgf@linux.com>
Signed-off-by: Paulo Gomes <pjbgf@linux.com>
Signed-off-by: Paulo Gomes <pjbgf@linux.com>
Signed-off-by: Paulo Gomes <pjbgf@linux.com>
Signed-off-by: Paulo Gomes <pjbgf@linux.com>
Signed-off-by: Paulo Gomes <pjbgf@linux.com>
Signed-off-by: Paulo Gomes <pjbgf@linux.com>
Signed-off-by: Paulo Gomes <pjbgf@linux.com>
Signed-off-by: Paulo Gomes <pjbgf@linux.com>
Improves plumbing.Hash extensibility, which is part of the sha256 support. Signed-off-by: Paulo Gomes <pjbgf@linux.com>
Signed-off-by: Paulo Gomes <pjbgf@linux.com>
Removes redundant logic, and keep all hash specific code inside its own package. The refactoring of Hasher will be done separately. Signed-off-by: Paulo Gomes <pjbgf@linux.com>
The initial implementation required build tags in order to support SHA256. This will no longer be the case, and therefore this code is redundant. Signed-off-by: Paulo Gomes <pjbgf@linux.com>
Signed-off-by: Paulo Gomes <pjbgf@linux.com>
Signed-off-by: Paulo Gomes <pjbgf@linux.com>
|
@aymanbagabas changes made and code rebased. PTAL |
Refactor the codebase so that it stops relying on
plumbing.Hash's implementation details, and instead rely on exported methods.This is a stepping stone to enable support of multiple hash types, which will be dealt with as a follow-up PR.
The
hashpackage now contains all code regarding hash, with exception toHasherrelated logic, which is spread betweenplumbingandhasher.plumbing.Hashnow redirects tohash.SHA1Hash, which may (or may not) be dropped at a later stage.The previous sha256 specific tests that relied on build tags were removed as they are now redundant.
Relates to #706 #899.
Supersedes to #920.