The problem was that it would download the CRLs again and again for every signing commit, even if it's the same URL as the commits before.
This PR changes the behaviour, so that already downloaded CRLs are cached in a file called crl_tmp.<hash of URL>.clr. So if we come across the same URL again, it will not be downloaded again.
Further optimizations could be made (in separate PRs)
check if we really have to go up and check the CRLs for EVERY commit, or would it be enough to check only the latest
do not do regenerate $OUTPUT_FILE if it already exists for the same data
Whenever the last commit in a branch is not a signing commit, the script walks up the chain to the top and downloads revocation lists (CRLs) from the signing authorities, to store them in the new commit (see https://git.jankaritech.eu/JankariTechUG/GitTrustedTimestamps#ltv---long-term-validation).
The problem was that it would download the CRLs again and again for every signing commit, even if it's the same URL as the commits before.
This PR changes the behaviour, so that already downloaded CRLs are cached in a file called `crl_tmp.<hash of URL>.clr`. So if we come across the same URL again, it will not be downloaded again.
Further optimizations could be made (in separate PRs)
1. check if we really have to go up and check the CRLs for EVERY commit, or would it be enough to check only the latest
2. do not do regenerate $OUTPUT_FILE if it already exists for the same data
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Whenever the last commit in a branch is not a signing commit, the script walks up the chain to the top and downloads revocation lists (CRLs) from the signing authorities, to store them in the new commit (see https://git.jankaritech.eu/JankariTechUG/GitTrustedTimestamps#ltv---long-term-validation).
The problem was that it would download the CRLs again and again for every signing commit, even if it's the same URL as the commits before.
This PR changes the behaviour, so that already downloaded CRLs are cached in a file called
crl_tmp.<hash of URL>.clr. So if we come across the same URL again, it will not be downloaded again.Further optimizations could be made (in separate PRs)
Works for me - JankariTechUG/accounting#160
Validate Trusted Timestamps Actions - IMO this repo has had various commits that do not have timestamps.