tokenizers 0.6.2 → 0.6.3
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- checksums.yaml +4 -4
- data/CHANGELOG.md +6 -0
- data/Cargo.lock +37 -31
- data/ext/tokenizers/Cargo.toml +4 -3
- data/ext/tokenizers/src/lib.rs +1 -1
- data/lib/tokenizers/from_pretrained.rb +23 -10
- data/lib/tokenizers/version.rb +1 -1
- metadata +2 -2
checksums.yaml
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
---
|
|
2
2
|
SHA256:
|
|
3
|
-
metadata.gz:
|
|
4
|
-
data.tar.gz:
|
|
3
|
+
metadata.gz: b7e76174884c06417a6829e89ef1f5785957d92f10d8a9acd6586e19ec84737c
|
|
4
|
+
data.tar.gz: 04b8e76f25f59e404978f2af68829142890c7bdd4ffbe7cd86a5af4f72de2d5e
|
|
5
5
|
SHA512:
|
|
6
|
-
metadata.gz:
|
|
7
|
-
data.tar.gz:
|
|
6
|
+
metadata.gz: b9b2583c6c2aac22d835c045f6674a95a1f1a9dcdddd7d2406b34c7f64cb04bd8900de05feaa85a8f9d5601392636b41ed300326f362e9e5de29098506801cc6
|
|
7
|
+
data.tar.gz: '08f6e8e3c4187a5f3bd57e75141add2973857a1fe4211b86acd8ab43a3ec5fc5550309910c606d09267dfe3242c7088b7ffa2d5ccb0f94c5ee922deb0fd0c943'
|
data/CHANGELOG.md
CHANGED
data/Cargo.lock
CHANGED
|
@@ -124,9 +124,9 @@ dependencies = [
|
|
|
124
124
|
|
|
125
125
|
[[package]]
|
|
126
126
|
name = "console"
|
|
127
|
-
version = "0.
|
|
127
|
+
version = "0.16.1"
|
|
128
128
|
source = "registry+https://github.com/rust-lang/crates.io-index"
|
|
129
|
-
checksum = "
|
|
129
|
+
checksum = "b430743a6eb14e9764d4260d4c0d8123087d504eeb9c48f2b2a5e810dd369df4"
|
|
130
130
|
dependencies = [
|
|
131
131
|
"encode_unicode",
|
|
132
132
|
"libc",
|
|
@@ -288,14 +288,14 @@ checksum = "b9e0384b61958566e926dc50660321d12159025e767c18e043daf26b70104c39"
|
|
|
288
288
|
|
|
289
289
|
[[package]]
|
|
290
290
|
name = "indicatif"
|
|
291
|
-
version = "0.
|
|
291
|
+
version = "0.18.3"
|
|
292
292
|
source = "registry+https://github.com/rust-lang/crates.io-index"
|
|
293
|
-
checksum = "
|
|
293
|
+
checksum = "9375e112e4b463ec1b1c6c011953545c65a30164fbab5b581df32b3abf0dcb88"
|
|
294
294
|
dependencies = [
|
|
295
295
|
"console",
|
|
296
|
-
"number_prefix",
|
|
297
296
|
"portable-atomic",
|
|
298
297
|
"unicode-width",
|
|
298
|
+
"unit-prefix",
|
|
299
299
|
"web-time",
|
|
300
300
|
]
|
|
301
301
|
|
|
@@ -449,12 +449,6 @@ dependencies = [
|
|
|
449
449
|
"minimal-lexical",
|
|
450
450
|
]
|
|
451
451
|
|
|
452
|
-
[[package]]
|
|
453
|
-
name = "number_prefix"
|
|
454
|
-
version = "0.4.0"
|
|
455
|
-
source = "registry+https://github.com/rust-lang/crates.io-index"
|
|
456
|
-
checksum = "830b246a0e5f20af87141b25c173cd1b609bd7779a4617d6ec582abaf90870f3"
|
|
457
|
-
|
|
458
452
|
[[package]]
|
|
459
453
|
name = "once_cell"
|
|
460
454
|
version = "1.21.3"
|
|
@@ -596,18 +590,18 @@ dependencies = [
|
|
|
596
590
|
|
|
597
591
|
[[package]]
|
|
598
592
|
name = "rb-sys"
|
|
599
|
-
version = "0.9.
|
|
593
|
+
version = "0.9.124"
|
|
600
594
|
source = "registry+https://github.com/rust-lang/crates.io-index"
|
|
601
|
-
checksum = "
|
|
595
|
+
checksum = "c85c4188462601e2aa1469def389c17228566f82ea72f137ed096f21591bc489"
|
|
602
596
|
dependencies = [
|
|
603
597
|
"rb-sys-build",
|
|
604
598
|
]
|
|
605
599
|
|
|
606
600
|
[[package]]
|
|
607
601
|
name = "rb-sys-build"
|
|
608
|
-
version = "0.9.
|
|
602
|
+
version = "0.9.124"
|
|
609
603
|
source = "registry+https://github.com/rust-lang/crates.io-index"
|
|
610
|
-
checksum = "
|
|
604
|
+
checksum = "568068db4102230882e6d4ae8de6632e224ca75fe5970f6e026a04e91ed635d3"
|
|
611
605
|
dependencies = [
|
|
612
606
|
"bindgen",
|
|
613
607
|
"lazy_static",
|
|
@@ -784,20 +778,9 @@ dependencies = [
|
|
|
784
778
|
|
|
785
779
|
[[package]]
|
|
786
780
|
name = "tokenizers"
|
|
787
|
-
version = "0.
|
|
788
|
-
dependencies = [
|
|
789
|
-
"ahash",
|
|
790
|
-
"magnus",
|
|
791
|
-
"onig",
|
|
792
|
-
"serde",
|
|
793
|
-
"tokenizers 0.22.1",
|
|
794
|
-
]
|
|
795
|
-
|
|
796
|
-
[[package]]
|
|
797
|
-
name = "tokenizers"
|
|
798
|
-
version = "0.22.1"
|
|
781
|
+
version = "0.22.2"
|
|
799
782
|
source = "registry+https://github.com/rust-lang/crates.io-index"
|
|
800
|
-
checksum = "
|
|
783
|
+
checksum = "b238e22d44a15349529690fb07bd645cf58149a1b1e44d6cb5bd1641ff1a6223"
|
|
801
784
|
dependencies = [
|
|
802
785
|
"ahash",
|
|
803
786
|
"aho-corasick",
|
|
@@ -827,6 +810,17 @@ dependencies = [
|
|
|
827
810
|
"unicode_categories",
|
|
828
811
|
]
|
|
829
812
|
|
|
813
|
+
[[package]]
|
|
814
|
+
name = "tokenizers-ruby"
|
|
815
|
+
version = "0.6.3"
|
|
816
|
+
dependencies = [
|
|
817
|
+
"ahash",
|
|
818
|
+
"magnus",
|
|
819
|
+
"onig",
|
|
820
|
+
"serde",
|
|
821
|
+
"tokenizers",
|
|
822
|
+
]
|
|
823
|
+
|
|
830
824
|
[[package]]
|
|
831
825
|
name = "unicode-ident"
|
|
832
826
|
version = "1.0.18"
|
|
@@ -860,6 +854,12 @@ version = "0.1.1"
|
|
|
860
854
|
source = "registry+https://github.com/rust-lang/crates.io-index"
|
|
861
855
|
checksum = "39ec24b3121d976906ece63c9daad25b85969647682eee313cb5779fdd69e14e"
|
|
862
856
|
|
|
857
|
+
[[package]]
|
|
858
|
+
name = "unit-prefix"
|
|
859
|
+
version = "0.5.2"
|
|
860
|
+
source = "registry+https://github.com/rust-lang/crates.io-index"
|
|
861
|
+
checksum = "81e544489bf3d8ef66c953931f56617f423cd4b5494be343d9b9d3dda037b9a3"
|
|
862
|
+
|
|
863
863
|
[[package]]
|
|
864
864
|
name = "version_check"
|
|
865
865
|
version = "0.9.5"
|
|
@@ -942,13 +942,19 @@ dependencies = [
|
|
|
942
942
|
"wasm-bindgen",
|
|
943
943
|
]
|
|
944
944
|
|
|
945
|
+
[[package]]
|
|
946
|
+
name = "windows-link"
|
|
947
|
+
version = "0.2.1"
|
|
948
|
+
source = "registry+https://github.com/rust-lang/crates.io-index"
|
|
949
|
+
checksum = "f0805222e57f7521d6a62e36fa9163bc891acd422f971defe97d64e70d0a4fe5"
|
|
950
|
+
|
|
945
951
|
[[package]]
|
|
946
952
|
name = "windows-sys"
|
|
947
|
-
version = "0.
|
|
953
|
+
version = "0.61.2"
|
|
948
954
|
source = "registry+https://github.com/rust-lang/crates.io-index"
|
|
949
|
-
checksum = "
|
|
955
|
+
checksum = "ae137229bcbd6cdf0f7b80a31df61766145077ddf49416a728b02cb3921ff3fc"
|
|
950
956
|
dependencies = [
|
|
951
|
-
"windows-
|
|
957
|
+
"windows-link",
|
|
952
958
|
]
|
|
953
959
|
|
|
954
960
|
[[package]]
|
data/ext/tokenizers/Cargo.toml
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
[package]
|
|
2
|
-
name = "tokenizers"
|
|
3
|
-
version = "0.6.
|
|
2
|
+
name = "tokenizers-ruby"
|
|
3
|
+
version = "0.6.3"
|
|
4
4
|
license = "Apache-2.0"
|
|
5
5
|
authors = ["Andrew Kane <andrew@ankane.org>"]
|
|
6
6
|
edition = "2021"
|
|
@@ -8,6 +8,7 @@ rust-version = "1.63.0"
|
|
|
8
8
|
publish = false
|
|
9
9
|
|
|
10
10
|
[lib]
|
|
11
|
+
name = "tokenizers"
|
|
11
12
|
crate-type = ["cdylib"]
|
|
12
13
|
|
|
13
14
|
[dependencies]
|
|
@@ -17,6 +18,6 @@ onig = { version = "6", default-features = false }
|
|
|
17
18
|
serde = { version = "1", features = ["rc", "derive"] }
|
|
18
19
|
|
|
19
20
|
[dependencies.tokenizers]
|
|
20
|
-
version = "=0.22.
|
|
21
|
+
version = "=0.22.2" # also update in from_pretrained.rb
|
|
21
22
|
default-features = false
|
|
22
23
|
features = ["progressbar", "onig", "esaxx_fast"]
|
data/ext/tokenizers/src/lib.rs
CHANGED
|
@@ -49,7 +49,7 @@ static PROCESSORS: Lazy<RModule> =
|
|
|
49
49
|
static TRAINERS: Lazy<RModule> =
|
|
50
50
|
Lazy::new(|ruby| ruby.get_inner(&TOKENIZERS).const_get("Trainers").unwrap());
|
|
51
51
|
|
|
52
|
-
#[magnus::init]
|
|
52
|
+
#[magnus::init(name = "tokenizers")]
|
|
53
53
|
fn init(ruby: &Ruby) -> RbResult<()> {
|
|
54
54
|
let module = ruby.define_module("Tokenizers")?;
|
|
55
55
|
|
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
module Tokenizers
|
|
2
2
|
module FromPretrained
|
|
3
3
|
# for user agent
|
|
4
|
-
TOKENIZERS_VERSION = "0.22.
|
|
4
|
+
TOKENIZERS_VERSION = "0.22.2"
|
|
5
5
|
|
|
6
6
|
# use Ruby for downloads
|
|
7
7
|
# this avoids the need to vendor OpenSSL on Linux
|
|
@@ -27,7 +27,8 @@ module Tokenizers
|
|
|
27
27
|
headers["Authorization"] = "Bearer #{auth_token}"
|
|
28
28
|
end
|
|
29
29
|
|
|
30
|
-
|
|
30
|
+
escaped_identifier = identifier.split("/", 2).map { |v| CGI.escape(v) }.join("/")
|
|
31
|
+
url = "https://huggingface.co/%s/resolve/%s/tokenizer.json" % [escaped_identifier, CGI.escape(revision)]
|
|
31
32
|
|
|
32
33
|
path =
|
|
33
34
|
begin
|
|
@@ -53,14 +54,7 @@ module Tokenizers
|
|
|
53
54
|
esum = Digest::SHA256.hexdigest(etag)
|
|
54
55
|
resource_path = File.join(cache_dir, "#{fsum}.#{esum}")
|
|
55
56
|
if File.exist?(resource_path)
|
|
56
|
-
|
|
57
|
-
req = Net::HTTP::Head.new(uri)
|
|
58
|
-
headers.each do |k, v|
|
|
59
|
-
req[k] = v
|
|
60
|
-
end
|
|
61
|
-
res = Net::HTTP.start(uri.hostname, uri.port, options.merge(use_ssl: true)) do |http|
|
|
62
|
-
http.request(req)
|
|
63
|
-
end
|
|
57
|
+
res = head_request(url, headers, options)
|
|
64
58
|
if res["etag"] == etag
|
|
65
59
|
return resource_path
|
|
66
60
|
end
|
|
@@ -93,6 +87,25 @@ module Tokenizers
|
|
|
93
87
|
resource_path
|
|
94
88
|
end
|
|
95
89
|
|
|
90
|
+
def head_request(url, headers, options, redirects = 0)
|
|
91
|
+
uri = URI(url)
|
|
92
|
+
req = Net::HTTP::Head.new(uri)
|
|
93
|
+
headers.each do |k, v|
|
|
94
|
+
req[k] = v
|
|
95
|
+
end
|
|
96
|
+
res = Net::HTTP.start(uri.hostname, uri.port, options.merge(use_ssl: true)) do |http|
|
|
97
|
+
http.request(req)
|
|
98
|
+
end
|
|
99
|
+
if res.is_a?(Net::HTTPRedirection) && redirects < 3
|
|
100
|
+
location = URI.parse(res["location"])
|
|
101
|
+
# follow relative redirects only
|
|
102
|
+
if location.relative?
|
|
103
|
+
return head_request(uri.merge(location), headers, options, redirects + 1)
|
|
104
|
+
end
|
|
105
|
+
end
|
|
106
|
+
res
|
|
107
|
+
end
|
|
108
|
+
|
|
96
109
|
def cache_dir
|
|
97
110
|
cache_dir =
|
|
98
111
|
if ENV["TOKENIZERS_CACHE"]
|
data/lib/tokenizers/version.rb
CHANGED
metadata
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
|
2
2
|
name: tokenizers
|
|
3
3
|
version: !ruby/object:Gem::Version
|
|
4
|
-
version: 0.6.
|
|
4
|
+
version: 0.6.3
|
|
5
5
|
platform: ruby
|
|
6
6
|
authors:
|
|
7
7
|
- Andrew Kane
|
|
@@ -98,7 +98,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
|
|
|
98
98
|
- !ruby/object:Gem::Version
|
|
99
99
|
version: '0'
|
|
100
100
|
requirements: []
|
|
101
|
-
rubygems_version:
|
|
101
|
+
rubygems_version: 4.0.3
|
|
102
102
|
specification_version: 4
|
|
103
103
|
summary: Fast state-of-the-art tokenizers for Ruby
|
|
104
104
|
test_files: []
|