aboutsummaryrefslogtreecommitdiffstats
path: root/rust
Commit message (Collapse)AuthorAgeFilesLines
* initial API schema rust implBryan Newbold2019-05-093-40/+70
|
* rust codegenBryan Newbold2019-05-099-10/+203
|
* extent release_date codegen hack to withdrawn_dateBryan Newbold2019-05-091-0/+1
|
* basic impl of new SQL schema in rust code (not wired to API)Bryan Newbold2019-05-083-19/+71
|
* codegen diesel rust schemaBryan Newbold2019-05-082-2/+14
| | | | | | | | The number of fields on release_rev went over 32, so we need the (slow) 64 column feature of diesel. Might be possible to get back under that if we don't use some of these columns after all.
* first draft of SQL schema changesBryan Newbold2019-05-082-0/+76
|
* more complete metadata dumpsBryan Newbold2019-04-301-1/+4
|
* rust: API lower-cases DOI lookupsBryan Newbold2019-04-182-2/+15
|
* rustfmtBryan Newbold2019-04-092-2/+7
|
* don't require auth for editgroup annotationsBryan Newbold2019-04-094-7/+1
| | | | | | | | | | Amazing that this bug found it's way through... because most testing is from Python, and was having serious auth config leakage with python_client. We're still in 0.x, and this is such a small/eggregious bug that i'm not going to tag as a backwards-incompatible schema update (but will note in CHANGELOG).
* tweak schema example wayback URLsBryan Newbold2019-04-091-2/+2
|
* rust: fix macaroon expiry checkBryan Newbold2019-04-092-10/+26
| | | | | | | | | | | | | There were two bugs with this code: the expiry timestamps were getting enclosed in double-quotes (which caused parse bugs), and the actual caveat check itself was backwards (expires < now instead of expires > now). An underlying issue was that these caveats weren't actually getting checked in the tests. Should fix a bug where users don't get auth'd correctly when logging in via mechanisms other than tokens.
* update dev setup instructionsBryan Newbold2019-03-153-15/+24
|
* example schema: more edge casesBryan Newbold2019-03-041-2/+16
|
* tweaks to schema examplesBryan Newbold2019-03-021-3/+3
|
* verify release_status fieldBryan Newbold2019-02-142-0/+35
|
* small number of additional release_typesBryan Newbold2019-02-141-0/+4
|
* target rust 1.32Bryan Newbold2019-02-071-2/+2
|
* more old rust tests removedBryan Newbold2019-02-052-302/+1
|
* tweak rust TODO (libs)Bryan Newbold2019-02-051-1/+2
|
* update rust/TODOBryan Newbold2019-02-051-13/+14
|
* remove test client helperBryan Newbold2019-02-051-36/+0
|
* update cargo depsBryan Newbold2019-02-052-516/+299
|
* rust: remove client tests (unused)Bryan Newbold2019-02-051-25/+0
| | | | Basically all of this has moved to python, which is fine.
* reformat and expand on example SQL entriesBryan Newbold2019-02-051-21/+65
|
* use pigz over gzip in more placesBryan Newbold2019-02-051-4/+4
|
* update dump and sort commandsBryan Newbold2019-02-011-5/+5
| | | | | Pipeline sorts are *so* starved and slow ; they only get a few MByte of RAM by default!
* be less arbitrary about octet_length checks: mostly just >= 1Bryan Newbold2019-01-291-2/+2
| | | | | We're just checking that strings aren't empty, not that they have any particular length.
* auth editor not-found is a 403 auth responseBryan Newbold2019-01-291-1/+13
|
* better database NotFound error propagationBryan Newbold2019-01-293-11/+51
|
* Fix Cargo version pinsBryan Newbold2019-01-282-4/+4
|
* update export READMEBryan Newbold2019-01-281-6/+7
|
* update rust Cargo lock (from version bump)Bryan Newbold2019-01-281-5/+5
|
* fix rust comment typosBryan Newbold2019-01-281-1/+1
|
* bump versions (0.1.0 to 0.2.0)Bryan Newbold2019-01-285-8/+8
|
* rustfmtBryan Newbold2019-01-281-3/+6
|
* get regular log lines flowing through slogBryan Newbold2019-01-283-1/+52
| | | | | The plan here is to replace slog with env_logger (much simpler!), but not there yet.
* wrote a whole thing to log Warning headers and not neededBryan Newbold2019-01-281-0/+3
|
* squelch some rust unused importsBryan Newbold2019-01-282-2/+0
|
* allow importing contrib/refs listsBryan Newbold2019-01-241-4/+25
| | | | | | The motivation here isn't really to support these gigantic lists on principle, but to be able to ingest large corpuses without having to decide whether to filter out or crop such lists.
* codegen schema tweaksBryan Newbold2019-01-244-20/+53
|
* ensure no zero-length strings in SQL schemaBryan Newbold2019-01-231-43/+43
|
* allow passing description+extra to batch endpointsBryan Newbold2019-01-2212-136/+519
| | | | | | Pretty messy, but I needed some way to do this. In particular, requires json.dumps() in python code, for now. Blech.
* add missing arxiv+jstor id indicesBryan Newbold2019-01-221-0/+2
|
* allow arxiv and jstor lookupsBryan Newbold2019-01-2110-12/+89
|
* rust impl of new fieldsBryan Newbold2019-01-213-14/+38
|
* codegenBryan Newbold2019-01-214-40/+101
|
* SQL schema bumpBryan Newbold2019-01-211-9/+12
|
* acutaly expand filesets/webcapturesBryan Newbold2019-01-181-1/+21
|
* include filesets and webcaptures in exportsBryan Newbold2019-01-181-1/+1
|
generated by cgit v1.2.3 (git 2.39.1) at 2024-11-27 19:27:25 +0000