webui-client merge requestshttps://gitlab.fit.cvut.cz/algorithms-library-toolkit/webui-client/-/merge_requests2022-12-04T12:36:45+01:00https://gitlab.fit.cvut.cz/algorithms-library-toolkit/webui-client/-/merge_requests/44webui: add input form for grammar2022-12-04T12:36:45+01:00Hana Litavskáwebui: add input form for grammarBPJan TrávníčekJan Trávníčekhttps://gitlab.fit.cvut.cz/algorithms-library-toolkit/webui-client/-/merge_requests/66worker: bump core version2022-10-27T10:47:52+02:00Tomáš Peckaworker: bump core versionWe are now building against core commit
bd3eddd7cf1645d2d54113c5ffa2ecedffd4f0fe[1].
This brings several new overloads for unranked trees and Determinize NFA
now does not create total DFA.
[1] https://gitlab.fit.cvut.cz/algorithms-libr...We are now building against core commit
bd3eddd7cf1645d2d54113c5ffa2ecedffd4f0fe[1].
This brings several new overloads for unranked trees and Determinize NFA
now does not create total DFA.
[1] https://gitlab.fit.cvut.cz/algorithms-library-toolkit/automata-library/-/commit/bd3eddd7cf1645d2d54113c5ffa2ecedffd4f0fehttps://gitlab.fit.cvut.cz/algorithms-library-toolkit/webui-client/-/merge_requests/65worker: Hopefully fix move semantics for abstraction Values2022-10-19T21:56:58+02:00Tomáš Peckaworker: Hopefully fix move semantics for abstraction Values* https://gitlab.fit.cvut.cz/algorithms-library-toolkit/webui-client/-/issues/91
* https://gitlab.fit.cvut.cz/algorithms-library-toolkit/automata-library/-/issues/231* https://gitlab.fit.cvut.cz/algorithms-library-toolkit/webui-client/-/issues/91
* https://gitlab.fit.cvut.cz/algorithms-library-toolkit/automata-library/-/issues/231Tomáš PeckaTomáš Peckahttps://gitlab.fit.cvut.cz/algorithms-library-toolkit/webui-client/-/merge_requests/64Dev tp2022-05-26T08:32:33+02:00Tomáš PeckaDev tpTomáš PeckaTomáš Peckahttps://gitlab.fit.cvut.cz/algorithms-library-toolkit/webui-client/-/merge_requests/63Dev tp2022-05-25T13:55:32+02:00Tomáš PeckaDev tphttps://gitlab.fit.cvut.cz/algorithms-library-toolkit/webui-client/-/merge_requests/62Dev tp2022-05-15T23:04:47+02:00Tomáš PeckaDev tpOutputDialog enhancements for 0.4.1OutputDialog enhancements for 0.4.1Tomáš PeckaTomáš Peckahttps://gitlab.fit.cvut.cz/algorithms-library-toolkit/webui-client/-/merge_requests/61Dev tp2022-05-15T15:28:12+02:00Tomáš PeckaDev tpTomáš PeckaTomáš Peckahttps://gitlab.fit.cvut.cz/algorithms-library-toolkit/webui-client/-/merge_requests/60release 0.4.02022-05-14T19:30:14+02:00Tomáš Peckarelease 0.4.0Tomáš PeckaTomáš Peckahttps://gitlab.fit.cvut.cz/algorithms-library-toolkit/webui-client/-/merge_requests/53webui: use one box for all overloads2022-05-14T19:07:44+02:00Hana Litavskáwebui: use one box for all overloadsOverloads with the same number of parameters are represented by one box, all overloads with docs can be listed with info popover. The box representing only one overload uses its own parameters and docs. Squashed boxes use common docs. Pa...Overloads with the same number of parameters are represented by one box, all overloads with docs can be listed with info popover. The box representing only one overload uses its own parameters and docs. Squashed boxes use common docs. Parameters that are common for all overloads are displayed on the box. When the types are different, at least namespace is used as name for the parameter - it was extremely confusing with only generic names and types like "parameter1, parameter2" during testing, so I thought we should give a little hint.BPJan TrávníčekJan Trávníčekhttps://gitlab.fit.cvut.cz/algorithms-library-toolkit/webui-client/-/merge_requests/59Dev output compose2022-05-14T18:56:26+02:00Tomáš PeckaDev output composeAfter recent big changes to the output handling in webui we were hitting
an issue how to decide whether resulting datatype has a string::Compose
overload. This could be handled only in webui but I *think* it is
unnecessarily hard and ...After recent big changes to the output handling in webui we were hitting
an issue how to decide whether resulting datatype has a string::Compose
overload. This could be handled only in webui but I *think* it is
unnecessarily hard and unclean in current code. Also, the dance around
serializing the graph when we were adding "invisible" string::Compose
nodes was far from something optimal.
I was thinking and it is easily possible to run string::Compose *always*
on the worker side and return the result in the webui (see previous
commit). We can easily handle when the Compose overload exists and when
it does not. We always return both the outputs (plain and pretty) and
let the webui decide how to handle that.
Could we do the same for DotConverter?
cc @litavhan - what do you think about this solution?Jan TrávníčekJan Trávníčekhttps://gitlab.fit.cvut.cz/algorithms-library-toolkit/webui-client/-/merge_requests/57webui: custom functions2022-05-13T19:24:00+02:00Hana Litavskáwebui: custom functionsBPJan TrávníčekJan Trávníčekhttps://gitlab.fit.cvut.cz/algorithms-library-toolkit/webui-client/-/merge_requests/55Worker cleanup2022-05-13T12:52:21+02:00Tomáš PeckaWorker cleanupFirst phase of worker cleanup.
No major changes on the outside. Internally, I have done some restructualization and replaced evaluation timeout handler with `boost::{asio,process}`. Also I had to bump ALT worker to alpine3.15 because 3....First phase of worker cleanup.
No major changes on the outside. Internally, I have done some restructualization and replaced evaluation timeout handler with `boost::{asio,process}`. Also I had to bump ALT worker to alpine3.15 because 3.13 had old boost. This is another MR in core.Jan TrávníčekJan Trávníčekhttps://gitlab.fit.cvut.cz/algorithms-library-toolkit/webui-client/-/merge_requests/54Introduce parser generators for ALT formats2022-05-13T12:04:50+02:00Tomáš PeckaIntroduce parser generators for ALT formatsCloses #82
I have absolutely no idea what did I break with this :-)
Please just check the grammars.Closes #82
I have absolutely no idea what did I break with this :-)
Please just check the grammars.Jan TrávníčekJan Trávníčekhttps://gitlab.fit.cvut.cz/algorithms-library-toolkit/webui-client/-/merge_requests/56worker: add filter2022-05-10T19:53:38+02:00Hana Litavskáworker: add filterIgnore algorithms matched with regex in introspect.
Closes: https://gitlab.fit.cvut.cz/algorithms-library-toolkit/webui-client/issues/50Ignore algorithms matched with regex in introspect.
Closes: https://gitlab.fit.cvut.cz/algorithms-library-toolkit/webui-client/issues/50Jan TrávníčekJan Trávníčekhttps://gitlab.fit.cvut.cz/algorithms-library-toolkit/webui-client/-/merge_requests/58webui: fix epsilon symbol input in automaton table2022-05-10T19:28:36+02:00Hana Litavskáwebui: fix epsilon symbol input in automaton tableHash symbol was prohibited in symbols too. The check for epsilon symbol is added and now allowed.Hash symbol was prohibited in symbols too. The check for epsilon symbol is added and now allowed.BPJan TrávníčekJan Trávníčekhttps://gitlab.fit.cvut.cz/algorithms-library-toolkit/webui-client/-/merge_requests/47webui: output and parse2022-05-09T21:14:59+02:00Hana Litavskáwebui: output and parseBPJan TrávníčekJan Trávníčekhttps://gitlab.fit.cvut.cz/algorithms-library-toolkit/webui-client/-/merge_requests/52webui: automaton table input2022-05-02T23:07:38+02:00Hana Litavskáwebui: automaton table input* table input was added
* input dialog size and resizing was changed
* a few console errors related to input and output dialog were fixed (the rest will be fixed with #81)* table input was added
* input dialog size and resizing was changed
* a few console errors related to input and output dialog were fixed (the rest will be fixed with #81)BPJan TrávníčekJan Trávníčekhttps://gitlab.fit.cvut.cz/algorithms-library-toolkit/webui-client/-/merge_requests/51Resolve "Let output nodes decide their features from the actual visualized type"2022-04-25T17:15:15+02:00Tomáš PeckaResolve "Let output nodes decide their features from the actual visualized type"Closes #80
perhaps let's wait for @litavhan's opinion?Closes #80
perhaps let's wait for @litavhan's opinion?Jan TrávníčekJan Trávníčekhttps://gitlab.fit.cvut.cz/algorithms-library-toolkit/webui-client/-/merge_requests/50ci: remove staging branch2022-04-19T16:03:21+02:00Tomáš Peckaci: remove staging branchSince 2fe187214 we are no longer auto-deploying current master tree. We
create docker 'latest' images only from tags.
There is probably no need to keep the staging branch alive. We can send
MRs directly to master and we can autodeploy ma...Since 2fe187214 we are no longer auto-deploying current master tree. We
create docker 'latest' images only from tags.
There is probably no need to keep the staging branch alive. We can send
MRs directly to master and we can autodeploy master branch to the
staging environment.Jan TrávníčekJan Trávníčekhttps://gitlab.fit.cvut.cz/algorithms-library-toolkit/webui-client/-/merge_requests/48ci: speed up, possibly remove docker TOOMANYREQUESTS errors2022-04-17T22:05:20+02:00Tomáš Peckaci: speed up, possibly remove docker TOOMANYREQUESTS errorsThis series of commits both simplifies and complicates the CI.
* The test jobs for worker and webui are squashed into build jobs. I think there is no benefit in having them separate. if separated, they have to be in separate stages an...This series of commits both simplifies and complicates the CI.
* The test jobs for worker and webui are squashed into build jobs. I think there is no benefit in having them separate. if separated, they have to be in separate stages and therefore the CI must run longer (pulling images, waiting for prev stage to end, ...). So let's add the tests into the build process.
* All the remaining test jobs (clang-format only for now) is run in build phase as well
* The runner runs in gitlab CI on docker runners. We use *a lot of* docker pulls and the underlying FIT infrastructure does not login to docker hub. This results in a lot of unauthenticated pulls (from all the projects on this gitlab instance) and we fastly reach pull limit. Let's bypass that by creating local copies of the most used images in https://gitlab.fit.cvut.cz/algorithms-library-toolkit/infrastructure/ci-docker-images repository and use those.
cc @litavhanJan TrávníčekJan Trávníček