1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
|
# CHANGELOG
Notable technical changes to the fatcat schemas, Rust API server, Python web
interface, and Python client library will be recorded here.
Intent is to follow semantic versioning, with all components kept in lock-step,
but note that everything is pre-1.0, so there aren't any actual backwards
compatibility guarantees. The API prefix (eg, `/v0/`) will probably roughly
track major version number, but TBD.
See also:
- [Keep a Changelog](https://keepachangelog.com/en/1.0.0/)
- [Semantic Versioning](https://semver.org/spec/v2.0.0.html)
## [Unreleased]
## Changed
- Switch from swagger-codegen to openapi-generator for python client generation
## [0.3.1] - 2019-09-18
Many small web interface improvements, bug fixes, and importer additions are
not explicitly noted below.
### Changed
- renamed python client library `fatcat-openapi-client` and rust api-spec
library `fatcat-openapi`
- updated rust (cargo) dependencies; depend on rust 1.34+ (Debian buster)
- entity view page design and backend code structure completely refactored
### Fixed
- compilation of `fatcat-openapi` rust crate (including examples) now actually
works using 2018 edition. Many lint warnings were patched, making rust
compilation cleaner with recent compiler versions.
### Added
- container "coverage" page, showing elasticsearch-generated chart of IA
release archival coverage
- `create_auth_token` API endpoint, for generation of API tokens via web
interface
- API token generation in web interface
- quick web interface search (front page and top bar) now detects additional
identifiers, such as ISBN-13, SHA-1, and arxiv
## [0.3.0] - 2019-05-23
This version includes both a SQL migration and backwards-incompatible API
changes.
### Fixed
- `edit_extra` on individual entity create was not being written to database
### Changed
- `release_status` field (on releases) renamed to `release_stage`
- moved external identifier fields on release entities into a new sub-namespace
called `ext_ids`. Eg, instead of `release.doi`, it's now
`release.ext_ids.doi`. This field is always required for release entities, so
fields can be accessed directly without a null/option check on the `ext_ids`
field itself. This impacts `doi`, `pmid`, `pmcid`, etc.
- `pmcid` field now accepts versioned identifiers, with a trailing dot and
version number, like `PMC1234.2`.
- many more restrictions on external identifiers at creation time.
- API endpoints that mutate entities are now prefixed with
`/endpoint/{endpoint_id}/...`, which changes the `endpoint_id` parameter from
optional to required (in the path). In client libraries, this also changed
the order of parameters (`endpoint_id` now comes first). Batch edits now only
work in "auto batch" mode, and an editgroup object must be included in the
body (not just `extra` and `description` fields as query parameters).
- several additional editgroup/changelog/history endpoints now expand `editor`
in editgroup objects by default.
- `created` timestamp is included in editgroup objects (when GET) by default.
- in client libraries, `FileEntityUrls` renamed `FileUrl`, and several similar
singlular/plural and `Entity` removed renamings.
- elasticsearch release schema updated to match API schema changes
### Removed
- Non-auto batch mode no longer implemented.
### Added
- release contribs may include `given_name` and `surname` fields, like
`creator` objects, in addition to existing `raw_name`.
- add `withdrawn_status`, `withdrawn_year`, and `withdrawn_date` fields to
releases
- added `retraction` as an allowable `release_type`, for a publication. When
used, the `release_stage` should also be `retraction`.
- `subtitle` added as a top-level release field, along with `version` and
`number`
- `ark_id` (for ARK identifiers) and `mag_id` (for Microsoft Academic Graph
identifiers) added to releases (under `ext_id`)
- allow expanding `releases` for file, fileset, and webcapture entities.
Expanded release entities have their abstracts and refs hidden by default
(for performance)
- `creator_ids` in release elasticsearch schema, for lookups
- new importers: arxiv, pubmed, JALC, and JSTOR. Most still need refactoring
and polish
## [0.2.2] - 2019-05-08
### Fixed
- fixed major authn/authz bug with fatcat python client which leaked API tokens
between API client handles. Almost all tests/webfact/etc were potentially
running with the privileged (superuser) webface-bot privileges. Yikes!
- API `get_editgroup_annotations` endpoint was requiring auth; this was a typo.
Going to call this a very minor/backwards-compatible API change and not do a
minor version bump for it.
- DOI lookups (for release entities) were not being lower-cased as intended
(DOIs are stored lower-case internally)
### Added
- ORCID OAuth login provider
### Changed
- small tweaks to URL domain/rel mapping list
## [0.2.1] - 2019-04-09
No API or SQL schema changes in this release. Macaroon generation and
verification was broken; all non-CLI-generated tokens will need to be
regenerated (eg, log-out, log-in).
### Fixed
- fix macaroon 'expires' caveat check (in rust), allowing actual login via OAuth/IA
### Added
- basic release, container, file webface editing
- basic editgroup annotation and control (submit/accept) in webface
- CSL/BibTeX citation endpoints
- "review bot" framework (for CI-style edit checks)
- wikipedia OAuth login provider
### Changed
- expanded example entities in SQL schema
- updated rust (cargo) dependencies; depend on rust 1.32+
- many tweaks to webface and guide
## [0.2.0] - 2019-02-05
First tagged release, and start of notable change tracking.
|
