aboutsummaryrefslogtreecommitdiffstats
path: root/rust
Commit message (Collapse)AuthorAgeFilesLines
* gitignore: don't skip run bin filesBryan Newbold2022-08-121-1/+0
|
* update rust README dependency section (ubuntu focal)Bryan Newbold2022-01-261-4/+6
|
* upgrade gitlab CI, docs, and various scripts from postgresql 11 to 13Bryan Newbold2022-01-263-2/+3
| | | | | | | | Expecting few if any compatibility issues with this transition. Actually upgrading the production databases will be some work, but will hopefully yield disk savings from changes in the B-Tree index on-disk format.
* rust: handle database read-only mode (as 5xx errors, not 4xx)Bryan Newbold2022-01-212-8/+16
|
* rust: swith postgresql database to 'fatcat_dev' (distinct from 'fatcat_test')Bryan Newbold2022-01-213-3/+3
|
* codespell fixes to various other docsBryan Newbold2021-11-243-3/+3
|
* bump rust code to 0.5.0Bryan Newbold2021-11-175-6/+7
|
* rust: bump crate version and lockfileBryan Newbold2021-11-172-3/+3
|
* rust: implement content_scopeBryan Newbold2021-11-175-0/+22
|
* SQL implementation of content_scopeBryan Newbold2021-11-172-0/+36
|
* codegen rust code for content_scopeBryan Newbold2021-11-173-4/+19
|
* api: add SQL 'ORDER BY' to many reads to stabilize API array orderingBryan Newbold2021-11-051-3/+14
| | | | | | | | | | | | | | The hope is to make things like file entity URLs, fileset manifests, and other arrays in the JSON API "stable", meaning that if you create an entity with a list of a given order, a read back (in any environment, including prod/QA, bulk dumps, etc) will return the array with the same sort order. This was informally happening most of the time, but occasionally not (!) Assumption is that these sorts will have little or no performance impact, as the common case is less than a dozen elements, and the hard cases are a few thousand at most, and there is already a sorted index.
* rust: expand valid hdl identifiersBryan Newbold2021-10-131-3/+7
| | | | Based on real examples found on dblp.org
* rust: partial test coverage of schema changesBryan Newbold2021-10-131-2/+32
|
* rust: prep for possible DOI lowercase enforcementBryan Newbold2021-10-131-1/+5
| | | | | | See also: https://github.com/internetarchive/fatcat/issues/83 This commit is no behavior change, just leaving a note to self.
* rust: implement scheman and API changesBryan Newbold2021-10-135-38/+353
|
* rust: handle new migrations in test helperBryan Newbold2021-10-131-1/+1
|
* rust: implement recent SQL changesBryan Newbold2021-10-132-0/+12
|
* fatcatd: display version correctly, and at startupBryan Newbold2021-10-131-2/+8
|
* fatcat-api: enforce more release ext_id checks at create/updateBryan Newbold2021-10-131-2/+15
| | | | Not enforcing these was a serious bug!
* sql: v0.4 schema implementation (as diesel migration)Bryan Newbold2021-10-132-0/+58
|
* bump rust code version to v0.4.0Bryan Newbold2021-10-133-5/+7
|
* rust codegen for v0.4Bryan Newbold2021-10-138-25/+307
|
* guide and openapi schema: fix QA URLs, and disclaim QA instanceBryan Newbold2021-04-121-2/+2
|
* rust: update Cargo.lock (openapi dep was updated)Bryan Newbold2021-01-051-1/+1
|
* rust makefile: rename 'dev' to 'serve'Bryan Newbold2021-01-051-2/+2
|
* rust openapi lib: bump version to v0.3.3Bryan Newbold2020-12-241-1/+1
|
* rust: update lazy_static dependencyBryan Newbold2020-12-243-35/+26
| | | | | The motivation for this is to quiet very verbose warnings about some deprecated use of std::sync. Expect no actual runtime/behavior change.
* rust: fix malformed ext id error typeBryan Newbold2020-12-171-2/+2
| | | | This bug was due to copy/paste of SHA-1 check
* rust: rename and improve dblp key (id) syntax checkBryan Newbold2020-12-172-9/+17
|
* cargo: update sentry to fix memory initialization issueBryan Newbold2020-11-202-274/+332
| | | | | | | | | | | | | | | | | | Older sentry had an unsafe memory initialization error, which wasn't caught by older compilers. Rust 1.48 catches the problem at runtime and raises a panic. This meant that new builds (eg, on QA machine after update) were panic-ing. Newest versions of sentry have modern dependencies, which breaks our crufty old 'iron' dependency tree. Work-around is to only partially update (v0.12 to v0.15). This is a fairly frustrating situation. I'm hopeful that when we update to a different web framework and openapi generator 5.0 (not yet released), many of these dependency issues will be resolved, but i'm not certain. I did notice that if we entirely remove Sentry, which has not really been used much (only a small handful of issues reported over several years), we might be able to resolve openssl dependency issues.
* update fatcatd rust code for 'oai' external identifierBryan Newbold2020-11-194-11/+189
|
* codegen rust schema crateBryan Newbold2020-11-196-3/+20
|
* rustfmtBryan Newbold2020-11-195-87/+138
|
* rust: fatcatd changes for DOAJ+dblp identifiersBryan Newbold2020-11-196-949/+1062
|
* codegen rust crate for v0.3.3Bryan Newbold2020-11-198-227/+244
|
* fatcat export: flush after batch, not per-lineBryan Newbold2020-08-051-1/+1
| | | | Good catch, thanks Martin
* group-by-work mode for fatcat-exportBryan Newbold2020-08-041-15/+157
|
* rust Makefile: fix test commandBryan Newbold2020-08-041-2/+1
|
* updates to MakefileBryan Newbold2020-07-011-1/+1
|
* remove accidentally-commited lines from rust MakefileBryan Newbold2020-06-261-3/+0
|
* add work-in-progress Rust makefileBryan Newbold2020-05-262-2/+29
|
* more TODO/proposal cleanupBryan Newbold2020-01-221-0/+4
|
* fatcatd: fix corner-case in abstracts insertionBryan Newbold2020-01-151-1/+4
| | | | | | | | | | | | | Both the release_abstract relation table and the abstracts table inserts were being gated by a check on new abstracts table rows. I *think* the chance of this having caused problems is low. The most likely would have been updates to exiting entities that somehow removed the abstracts content, but not the sha1 keys. This is not the default behavior of the API: either the entire abstract (content and hash) is returned, or the abstracts are hidden entirely (via `hide` flag). Still, best to be careful!
* bump required rust to 1.36Bryan Newbold2019-12-031-1/+1
| | | | | | | | | | | | This isn't a fatcat rust requirement, but instead a diesel requirement, via rust-smallvec, which in v1.0 uses the alloc crate: https://github.com/servo/rust-smallvec/issues/73 I think the reason this came up now is that diesel-cli is an application and doesn't have a Cargo.lock file, and the build was updated. Using some binary mechanism to install these dependencies would be more robust, but feels like a yak shave right now.
* add 'component' as a release_typeBryan Newbold2019-11-041-0/+1
|
* update rust README with fatcat_test db creation noteBryan Newbold2019-10-031-1/+4
|
* update rust README re: opensslBryan Newbold2019-10-011-17/+1
| | | | | | | | | I believe an openssl library is still required locally, but with the SSL/TLS removal it now doesn't matter if it is OpenSSL 1.0 or 1.1. This is only a temporary work-around. When we update rust code generation, we will need to revisit these changes. The current version of swagger-rs still depends on HTTPS and OpenSSL 1.0 (via dependencies).
* entirely remove unused https flag to fatcatdBryan Newbold2019-09-291-15/+6
|
* cargo update fatcat rust after openssl removalBryan Newbold2019-09-291-76/+32
|