-
Notifications
You must be signed in to change notification settings - Fork 214
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Completely drop stack.yaml and sha256map.nix #3328
Completely drop stack.yaml and sha256map.nix #3328
Conversation
@@ -32,7 +32,6 @@ steps: | |||
key: nix | |||
commands: | |||
- './nix/regenerate.sh' | |||
- 'nix build .#checkStackProject -o check-stack-project.sh && ./check-stack-project.sh' |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This line now exists inside ./nix/regenerate.sh
- stubs/plutus-ghc-stub | ||
- prettyprinter-configurable | ||
- word-array | ||
>>>>>>> adcfc0603 (Update node to 1.35.0-rc2) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Hm, had left this in a broken state anyway 😅
@@ -110,10 +110,6 @@ haskell-nix: haskell-nix.cabalProject' [ | |||
filter = haskell-nix.haskellSourceFilter; | |||
}; | |||
|
|||
# materialized = ./materialized/plan-nix; | |||
|
|||
sha256map = import ./sha256map.nix; |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Seems like the sha256map was unused anyway 🤷♂️ (unless there is some clever fallback)
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
No, it seems the use is somewhat "magical". On master
there is no more reference to sha256map
, yet if we remove this statement there we get:
direnv reload (12s 11ms)
direnv: loading ~/IOHK/cardano-wallet/.envrc
direnv: using flake
warning: Git tree '/Users/anviking/IOHK/cardano-wallet' is dirty
direnv: ([/usr/local/bin/direnv export fish]) is taking a while to execute. Use CTRL-C to give up.
error: Cannot find Git revision '7497a29cb998721a9068d5725d49461f2bba0e7a' in ref 'master' of repository 'https://github.com/input-output-hk/optparse-applicative'! Please make sure that the rev exists on the ref you've specified or add allRefs = true; to fetchGit.
(use '--show-trace' to show detailed location information)
warning: Git tree '/Users/anviking/IOHK/cardano-wallet' is dirty
direnv: renewed cache
direnv: export +XDG_DATA_DIRS
whereas it is redundant in this PR (at least with the --sha256
info in cabal.project)
@Anviking I am afraid I don't understand the purpose of the |
@Unisay As I understand it: Haskell.nix needs to know the content hash before fetching anything from GitHub or similar for security reasons. On In #3315 (or rather the included #3293) |
3c62f15
to
4f62922
Compare
2b34f06
to
603f7bb
Compare
FWIW here are some reasons not to do this:
|
But you should be able to generate without needing |
You mean like nix build .#checkCabalProject -o check-cabal-project.sh && ./check-cabal-project.sh which does update the sha256 lines, or is there a concrete example of what else you might want to do?
TBH it seems kind of neat to me to have the |
df70ec0
to
4cbf834
Compare
Tastes differ I guess :) |
9298e17
to
8eea1d3
Compare
I would normally be inclined just to do what Michael and the Plutus team are doing (safety in numbers), but I don't think it matters too much here. We can always switch to the sha256map if we need to, and this solution is less confusing for the rest of the team, as they've seen this syntax before, but have yet to be exposed to the Nice work Johannes, thank you! |
bors r+ |
3328: Completely drop stack.yaml and sha256map.nix r=sevanspowell a=Anviking - It seems `stack.yaml` and `sha256map.nix` can just be removed and replaced with [`checkCabalProject`](https://github.com/input-output-hk/iohk-nix/search?q=checkCabalProject) and `--sha256` lines in `cabal.project` ### Tested manually Sanity checking that it works similarly to with the stack and sha256map-based approach: #### Revision in cabal.project changed ``` $ ./nix/regenerate.sh Changes required to cabal.project --- cabal.project.old 2022-06-13 20:35:27.437617302 +0200 +++ cabal.project 2022-06-13 20:36:37.049474850 +0200 `@@` -90,7 +90,7 `@@` type: git location: https://github.com/input-output-hk/cardano-addresses tag: 4ff9c4022d4d257053c96509f20c9d931ab8f57b - --sha256: 0iqzppqqb1xqnp55yd0xil9qfy8fz8474w834nmfg2w692bynyb9 + --sha256: 0q93a2r4v8d3nfvix6yjmbh3v77hz2hpp6p11krklvnvw7h00lfk subdir: command-line core ``` (exit code 1) #### Invalid SHA256 hash ``` $ ./nix/regenerate.sh Changes required to cabal.project --- cabal.project.old 2022-06-13 20:42:25.934613264 +0200 +++ cabal.project 2022-06-13 20:43:38.737611148 +0200 `@@` -299,7 +299,7 `@@` type: git location: https://github.com/haskell-works/hw-aeson tag: d99d2f3e39a287607418ae605b132a3deb2b753f - --sha256: 0vxqcwjg9q37wbwi27y9ba5163lzfz51f1swbi0rp681yg63zvn4 + --sha256: 1vxqcwjg9q37wbwi27y9ba5163lzfz51f1swbi0rp681yg63zvn4 ``` (exit code 1) <!-- Additional comments, links, or screenshots to attach, if any. --> #### Invalid SHA256 hash in `cabal.project` + `direnv reload` The corresponding change on master would be having an incorrect hash in `sha256map.nix` and calling `direnv reload`. And if you on master changed a revision in `stack.yaml`, `direnv reload` would still work, only `./nix/regenerate.sh` wouldn't. Because `nix shell` now is derived directly from `cabal.project` without intermediate state like sha256map.nix, `direnv reload` now fails directly. ```bash direnv reload (4s 207ms) direnv: loading ~/IOHK/cardano-wallet/.envrc direnv: using flake warning: Git tree '/Users/anviking/IOHK/cardano-wallet' is dirty trace: To make project.plan-nix for cardano-wallet a fixed-output derivation but not materialized, set `plan-sha256` to the output of the 'calculateMaterializedSha' script in 'passthru'. trace: To materialize project.plan-nix for cardano-wallet entirely, pass a writable path as the `materialized` argument and run the 'updateMaterialized' script in 'passthru'. direnv: ([/usr/local/bin/direnv export fish]) is taking a while to execute. Use CTRL-C to give up. error: hash mismatch in fixed-output derivation '/nix/store/icv2hd3bghfg80fvppdnjp0b0h1wji7x-hw-aeson-d99d2f3.drv': specified: sha256-xO4/zPMBmZtBXFwHF8p3nw4TilrJHxH54mfg9CRnuG8= got: sha256-xO4/zPMBmZtBXFwHF8p3nw4TilrJHxH54mfg9CRnuO8= error: 1 dependencies of derivation '/nix/store/493wcz8r85zbq5yaimm74ldnwz62m6in-source-repository-package.drv' failed to build error: 1 dependencies of derivation '/nix/store/2ja2788s8ivgsv2mcrb2b9pdl7b894dx-cardano-wallet-plan-to-nix-pkgs.drv' failed to build (use '--show-trace' to show detailed location information) warning: Git tree '/Users/anviking/IOHK/cardano-wallet' is dirty direnv: renewed cache direnv: export +XDG_DATA_DIRS ``` ### Comments Targets #3280 ### Issue Number ADP-1744 <!-- Reference the Jira/GitHub issue that this PR relates to, and which requirements it tackles. Note: Jira issues of the form ADP- will be auto-linked. --> Co-authored-by: Johannes Lund <johannes.lund@iohk.io>
Build failed: |
3280: Update to cardano-node 1.35-rc2 r=Anviking a=Anviking ### Overview - Updates dependencies to target `cardano-node 1.35-rc2` - Adds support to make the wallet _mostly_ work in Babbage - Rework of nix setup: we now derive nix files from `cabal.project` rather than the weird `stack.yaml`-based approach we had before. ### Detailed todo-list - [x] Get nix happy with the new dependencies - [x] Make `cabal build all` work with modulo warnings - [x] Update to more recent revisions - [x] Fix compilation in tests again - [x] Make `cabal build all` work with _no_ warnings - [x] Fix puzzling persistent TH name shadowing warnings -> fixed by sam? - [x] Ensure existing tests pass - [x] unit - [x] Look closer at balanceTx golden unit failure after first seeing if balanceTx integration tests pass on both alonzo and Babbage - [x] Integration (on alonzo like on master) - [x] Was blocked by readTVar, fixed with input-output-hk/io-sim#6 (review) - [x] Fix failures relating to Babbage era -> fixed with a hack 9e2568e - [x] Fix cardano-wallet-core Aeson failures - [x] More closely review code-changes - [x] Clean up git commit history - [x] Flatten merge commits - [x] Squash back-and-forth nix and revision changes - [x] Rebase off master - [x] Ensure revisions in stack.yaml and cabal.project are updated consistently - Alternatively remove `stack.yaml` and ensure ./nix/regenerate.sh still works and that we know how to update sha256map.nix — #3328 but without the sha256map.nix - [x] Ensure CI is completely green: [Hydra jobset](https://hydra.iohk.io/jobset/Cardano/cardano-wallet-pr-3280), ADP-1888 - There may be CI specific problems with nix, windows and more (there probably are) - [x] fix hydra wine TH error e.g. `$(TH.lift =<< TH.runIO (readFile "app/foo.txt"))` - seems relevant input-output-hk/haskell.nix#1508 - seems gone now - [x] `Ambiguous module name` https://hydra.iohk.io/build/15469467 - https://input-output-rnd.slack.com/archives/C48HA39RS/p1654388106509579 - Bumping Haskell.nix again should help? - Didn't seem like it worked... - https://input-output-rnd.slack.com/archives/CAP8NM7N0/p1654828014483969 - [x] Remove-iserv problem https://hydra.iohk.io/build/15604435 - Update from Hamish https://input-output-rnd.slack.com/archives/CAP8NM7N0/p1655213059547499?thread_ts=1654828014.483969&cid=CAP8NM7N0 - [x] Docs shell - we don't need explicit emanote input in nix config - [x] Profiled shell: `recursionzmschemeszm5zi2zi2zi2zm6NC6xZZx2tI3826DrVhqI9v_DataziFunctorziFoldableziTH_zdfMakeBaseFunctorNamezuzdcmakeBaseFunctor_closure` https://hydra.iohk.io/build/15694577/nixlog/2/tail - [x] musl cli goldens https://hydra.iohk.io/build/15694618/nixlog/82 - [x] `hie-bios: hie-bios: this operation requires -fexternal-interpreter - [x] `Ambiguous module name 'Options.Applicative':` in HLS - now we at least have an ugly `-XPackageImport` workaround. - [x] Update node revision in readme.md ### Things we need eventually, but not strictly for this PR > - Update to final 1.35 tag once it is released - can be done in separate PR > - Run existing integration tests on both Alonzo and Babbage eras in CI > - Depend on cardano-addresses with aeson2 support merged, tagged and released > - Depend on a cardano-sl-x509 where aeson 2 support is merged to master > - Consider appropriateness Tx deserialisation hack 9e2568e Moved to jira tasks in https://input-output.atlassian.net/browse/ADP-1538 ### Comments Includes #3258, #3293 <!-- Additional comments, links, or screenshots to attach, if any. --> ### Issue Number ADP-1744 <!-- Reference the Jira/GitHub issue that this PR relates to, and which requirements it tackles. Note: Jira issues of the form ADP- will be auto-linked. --> Co-authored-by: Johannes Lund <johannes.lund@iohk.io> Co-authored-by: Pawel Jakubas <jakubas.pawel@gmail.com> Co-authored-by: Samuel Evans-Powell <mail@sevanspowell.net> Co-authored-by: Jonathan Knowles <jonathan.knowles@iohk.io>
3280: Update to cardano-node 1.35-rc2 r=Anviking a=Anviking ### Overview - Updates dependencies to target `cardano-node 1.35-rc2` - Adds support to make the wallet _mostly_ work in Babbage - Rework of nix setup: we now derive nix files from `cabal.project` rather than the weird `stack.yaml`-based approach we had before. ### Detailed todo-list - [x] Get nix happy with the new dependencies - [x] Make `cabal build all` work with modulo warnings - [x] Update to more recent revisions - [x] Fix compilation in tests again - [x] Make `cabal build all` work with _no_ warnings - [x] Fix puzzling persistent TH name shadowing warnings -> fixed by sam? - [x] Ensure existing tests pass - [x] unit - [x] Look closer at balanceTx golden unit failure after first seeing if balanceTx integration tests pass on both alonzo and Babbage - [x] Integration (on alonzo like on master) - [x] Was blocked by readTVar, fixed with input-output-hk/io-sim#6 (review) - [x] Fix failures relating to Babbage era -> fixed with a hack 9e2568e - [x] Fix cardano-wallet-core Aeson failures - [x] More closely review code-changes - [x] Clean up git commit history - [x] Flatten merge commits - [x] Squash back-and-forth nix and revision changes - [x] Rebase off master - [x] Ensure revisions in stack.yaml and cabal.project are updated consistently - Alternatively remove `stack.yaml` and ensure ./nix/regenerate.sh still works and that we know how to update sha256map.nix — #3328 but without the sha256map.nix - [x] Ensure CI is completely green: [Hydra jobset](https://hydra.iohk.io/jobset/Cardano/cardano-wallet-pr-3280), ADP-1888 - There may be CI specific problems with nix, windows and more (there probably are) - [x] fix hydra wine TH error e.g. `$(TH.lift =<< TH.runIO (readFile "app/foo.txt"))` - seems relevant input-output-hk/haskell.nix#1508 - seems gone now - [x] `Ambiguous module name` https://hydra.iohk.io/build/15469467 - https://input-output-rnd.slack.com/archives/C48HA39RS/p1654388106509579 - Bumping Haskell.nix again should help? - Didn't seem like it worked... - https://input-output-rnd.slack.com/archives/CAP8NM7N0/p1654828014483969 - [x] Remove-iserv problem https://hydra.iohk.io/build/15604435 - Update from Hamish https://input-output-rnd.slack.com/archives/CAP8NM7N0/p1655213059547499?thread_ts=1654828014.483969&cid=CAP8NM7N0 - [x] Docs shell - we don't need explicit emanote input in nix config - [x] Profiled shell: `recursionzmschemeszm5zi2zi2zi2zm6NC6xZZx2tI3826DrVhqI9v_DataziFunctorziFoldableziTH_zdfMakeBaseFunctorNamezuzdcmakeBaseFunctor_closure` https://hydra.iohk.io/build/15694577/nixlog/2/tail - [x] musl cli goldens https://hydra.iohk.io/build/15694618/nixlog/82 - [x] `hie-bios: hie-bios: this operation requires -fexternal-interpreter - [x] `Ambiguous module name 'Options.Applicative':` in HLS - now we at least have an ugly `-XPackageImport` workaround. - [x] Update node revision in readme.md ### Things we need eventually, but not strictly for this PR > - Update to final 1.35 tag once it is released - can be done in separate PR > - Run existing integration tests on both Alonzo and Babbage eras in CI > - Depend on cardano-addresses with aeson2 support merged, tagged and released > - Depend on a cardano-sl-x509 where aeson 2 support is merged to master > - Consider appropriateness Tx deserialisation hack 9e2568e Moved to jira tasks in https://input-output.atlassian.net/browse/ADP-1538 ### Comments Includes #3258, #3293 <!-- Additional comments, links, or screenshots to attach, if any. --> ### Issue Number ADP-1744 <!-- Reference the Jira/GitHub issue that this PR relates to, and which requirements it tackles. Note: Jira issues of the form ADP- will be auto-linked. --> Co-authored-by: Johannes Lund <johannes.lund@iohk.io> Co-authored-by: Pawel Jakubas <jakubas.pawel@gmail.com> Co-authored-by: Samuel Evans-Powell <mail@sevanspowell.net> Co-authored-by: Jonathan Knowles <jonathan.knowles@iohk.io>
3280: Update to cardano-node 1.35-rc2 r=Anviking a=Anviking ### Overview - Updates dependencies to target `cardano-node 1.35-rc2` - Adds support to make the wallet _mostly_ work in Babbage - Rework of nix setup: we now derive nix files from `cabal.project` rather than the weird `stack.yaml`-based approach we had before. ### Detailed todo-list - [x] Get nix happy with the new dependencies - [x] Make `cabal build all` work with modulo warnings - [x] Update to more recent revisions - [x] Fix compilation in tests again - [x] Make `cabal build all` work with _no_ warnings - [x] Fix puzzling persistent TH name shadowing warnings -> fixed by sam? - [x] Ensure existing tests pass - [x] unit - [x] Look closer at balanceTx golden unit failure after first seeing if balanceTx integration tests pass on both alonzo and Babbage - [x] Integration (on alonzo like on master) - [x] Was blocked by readTVar, fixed with input-output-hk/io-sim#6 (review) - [x] Fix failures relating to Babbage era -> fixed with a hack 9e2568e - [x] Fix cardano-wallet-core Aeson failures - [x] More closely review code-changes - [x] Clean up git commit history - [x] Flatten merge commits - [x] Squash back-and-forth nix and revision changes - [x] Rebase off master - [x] Ensure revisions in stack.yaml and cabal.project are updated consistently - Alternatively remove `stack.yaml` and ensure ./nix/regenerate.sh still works and that we know how to update sha256map.nix — #3328 but without the sha256map.nix - [x] Ensure CI is completely green: [Hydra jobset](https://hydra.iohk.io/jobset/Cardano/cardano-wallet-pr-3280), ADP-1888 - There may be CI specific problems with nix, windows and more (there probably are) - [x] fix hydra wine TH error e.g. `$(TH.lift =<< TH.runIO (readFile "app/foo.txt"))` - seems relevant input-output-hk/haskell.nix#1508 - seems gone now - [x] `Ambiguous module name` https://hydra.iohk.io/build/15469467 - https://input-output-rnd.slack.com/archives/C48HA39RS/p1654388106509579 - Bumping Haskell.nix again should help? - Didn't seem like it worked... - https://input-output-rnd.slack.com/archives/CAP8NM7N0/p1654828014483969 - [x] Remove-iserv problem https://hydra.iohk.io/build/15604435 - Update from Hamish https://input-output-rnd.slack.com/archives/CAP8NM7N0/p1655213059547499?thread_ts=1654828014.483969&cid=CAP8NM7N0 - [x] Docs shell - we don't need explicit emanote input in nix config - [x] Profiled shell: `recursionzmschemeszm5zi2zi2zi2zm6NC6xZZx2tI3826DrVhqI9v_DataziFunctorziFoldableziTH_zdfMakeBaseFunctorNamezuzdcmakeBaseFunctor_closure` https://hydra.iohk.io/build/15694577/nixlog/2/tail - [x] musl cli goldens https://hydra.iohk.io/build/15694618/nixlog/82 - [x] `hie-bios: hie-bios: this operation requires -fexternal-interpreter - [x] `Ambiguous module name 'Options.Applicative':` in HLS - now we at least have an ugly `-XPackageImport` workaround. - [x] Update node revision in readme.md ### Things we need eventually, but not strictly for this PR > - Update to final 1.35 tag once it is released - can be done in separate PR > - Run existing integration tests on both Alonzo and Babbage eras in CI > - Depend on cardano-addresses with aeson2 support merged, tagged and released > - Depend on a cardano-sl-x509 where aeson 2 support is merged to master > - Consider appropriateness Tx deserialisation hack 9e2568e Moved to jira tasks in https://input-output.atlassian.net/browse/ADP-1538 ### Comments Includes #3258, #3293 <!-- Additional comments, links, or screenshots to attach, if any. --> ### Issue Number ADP-1744 <!-- Reference the Jira/GitHub issue that this PR relates to, and which requirements it tackles. Note: Jira issues of the form ADP- will be auto-linked. --> Co-authored-by: Johannes Lund <johannes.lund@iohk.io> Co-authored-by: Pawel Jakubas <jakubas.pawel@gmail.com> Co-authored-by: Samuel Evans-Powell <mail@sevanspowell.net> Co-authored-by: Jonathan Knowles <jonathan.knowles@iohk.io>
3280: Update to cardano-node 1.35-rc2 r=Anviking a=Anviking ### Overview - Updates dependencies to target `cardano-node 1.35-rc2` - Adds support to make the wallet _mostly_ work in Babbage - Rework of nix setup: we now derive nix files from `cabal.project` rather than the weird `stack.yaml`-based approach we had before. ### Detailed todo-list - [x] Get nix happy with the new dependencies - [x] Make `cabal build all` work with modulo warnings - [x] Update to more recent revisions - [x] Fix compilation in tests again - [x] Make `cabal build all` work with _no_ warnings - [x] Fix puzzling persistent TH name shadowing warnings -> fixed by sam? - [x] Ensure existing tests pass - [x] unit - [x] Look closer at balanceTx golden unit failure after first seeing if balanceTx integration tests pass on both alonzo and Babbage - [x] Integration (on alonzo like on master) - [x] Was blocked by readTVar, fixed with input-output-hk/io-sim#6 (review) - [x] Fix failures relating to Babbage era -> fixed with a hack 9e2568e - [x] Fix cardano-wallet-core Aeson failures - [x] More closely review code-changes - [x] Clean up git commit history - [x] Flatten merge commits - [x] Squash back-and-forth nix and revision changes - [x] Rebase off master - [x] Ensure revisions in stack.yaml and cabal.project are updated consistently - Alternatively remove `stack.yaml` and ensure ./nix/regenerate.sh still works and that we know how to update sha256map.nix — #3328 but without the sha256map.nix - [x] Ensure CI is completely green: [Hydra jobset](https://hydra.iohk.io/jobset/Cardano/cardano-wallet-pr-3280), ADP-1888 - There may be CI specific problems with nix, windows and more (there probably are) - [x] fix hydra wine TH error e.g. `$(TH.lift =<< TH.runIO (readFile "app/foo.txt"))` - seems relevant input-output-hk/haskell.nix#1508 - seems gone now - [x] `Ambiguous module name` https://hydra.iohk.io/build/15469467 - https://input-output-rnd.slack.com/archives/C48HA39RS/p1654388106509579 - Bumping Haskell.nix again should help? - Didn't seem like it worked... - https://input-output-rnd.slack.com/archives/CAP8NM7N0/p1654828014483969 - [x] Remove-iserv problem https://hydra.iohk.io/build/15604435 - Update from Hamish https://input-output-rnd.slack.com/archives/CAP8NM7N0/p1655213059547499?thread_ts=1654828014.483969&cid=CAP8NM7N0 - [x] Docs shell - we don't need explicit emanote input in nix config - [x] Profiled shell: `recursionzmschemeszm5zi2zi2zi2zm6NC6xZZx2tI3826DrVhqI9v_DataziFunctorziFoldableziTH_zdfMakeBaseFunctorNamezuzdcmakeBaseFunctor_closure` https://hydra.iohk.io/build/15694577/nixlog/2/tail - [x] musl cli goldens https://hydra.iohk.io/build/15694618/nixlog/82 - [x] `hie-bios: hie-bios: this operation requires -fexternal-interpreter - [x] `Ambiguous module name 'Options.Applicative':` in HLS - now we at least have an ugly `-XPackageImport` workaround. - [x] Update node revision in readme.md ### Things we need eventually, but not strictly for this PR > - Update to final 1.35 tag once it is released - can be done in separate PR > - Run existing integration tests on both Alonzo and Babbage eras in CI > - Depend on cardano-addresses with aeson2 support merged, tagged and released > - Depend on a cardano-sl-x509 where aeson 2 support is merged to master > - Consider appropriateness Tx deserialisation hack 9e2568e Moved to jira tasks in https://input-output.atlassian.net/browse/ADP-1538 ### Comments Includes #3258, #3293 <!-- Additional comments, links, or screenshots to attach, if any. --> ### Issue Number ADP-1744 <!-- Reference the Jira/GitHub issue that this PR relates to, and which requirements it tackles. Note: Jira issues of the form ADP- will be auto-linked. --> Co-authored-by: Johannes Lund <johannes.lund@iohk.io> Co-authored-by: Pawel Jakubas <jakubas.pawel@gmail.com> Co-authored-by: Samuel Evans-Powell <mail@sevanspowell.net> Co-authored-by: Jonathan Knowles <jonathan.knowles@iohk.io>
stack.yaml
andsha256map.nix
can just be removed and replaced withcheckCabalProject
and--sha256
lines incabal.project
Tested manually
Sanity checking that it works similarly to with the stack and sha256map-based approach:
Revision in cabal.project changed
(exit code 1)
Invalid SHA256 hash
(exit code 1)
Invalid SHA256 hash in
cabal.project
+direnv reload
The corresponding change on master would be having an incorrect hash in
sha256map.nix
and callingdirenv reload
. And if you on master changed a revision instack.yaml
,direnv reload
would still work, only./nix/regenerate.sh
wouldn't. Becausenix shell
now is derived directly fromcabal.project
without intermediate state like sha256map.nix,direnv reload
now fails directly.Comments
Targets #3280
Issue Number
ADP-1744