Revert to ReCAPTCHA 2 for Staticman

[pacollins]

Description

ReCAPTCHA reverted back to v2 Checkbox for Staticman

Motivation and Context

Closes #7

Screenshots (if appropriate):

Types of changes

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to change)

Checklist:

• I have read the Contributing Document.
• My code follows the code style of this project.
• My change requires a change to the documentation.
• I have updated the documentation, including theme.toml, accordingly.

[accesslint]

There are accessibility issues in these changes.

[VincentTam]

Simplify config.toml and avoid ...recaptcha.recaptcha since this doesn't look beautiful.

[VincentTam]

Staticman with reCAPTCHA worked on my demo site: https://framagit.org/vincenttam/fish-demo. @staticmanlab1 managed to push again the master branch of the corresponding repo.

[accesslint]

There are accessibility issues in these changes.

[pacollins]

Now that I have made the change and pushed it. I am torn.

...recaptcha.recaptcha served as a way to simply turn off the feature without deleting the keys. Theoretically, you could just comment out the keys now I suppose.

If we are doing this, should we change ...staticman.staticman to a similar system of {{ if and ...}}? The new method would be backwards compatible in theory (assuming people only fill in the information if they want to use the feature).

hugo-future-imperfect-slim/layouts/post/comments.html

Lines 5 to 9 in f0ec5ac

	{{ else if .Site.Params.staticman.staticman }}
	<article class="post">
	{{ .Render "staticman" }}
	</article>
	{{ end }}

[VincentTam]

Now that I have made the change and pushed it. I am torn.

Thanks for editing. It's Sunday, so you may take a rest and lemme do the testing and reviewing work for you.

...recaptcha.recaptcha served as a way to simply turn off the feature without deleting the keys. Theoretically, you could just comment out the keys now I suppose.

Even though this theme is called "Future Imperfect", when we look into the past, that's really a breaking change. But what's broken? Here's the only case.

    [params.staticman.recaptcha]
      recaptcha = false
      siteKey           = "<nonempty site key>" # Site Key
      encryptedKey      = "<nonempty encrypted key>"

Here're the repos using this theme with recaptcha = false: https://github.com/search?o=asc&q=theme+%3D+%22hugo-future-imperfect-slim%22+AND+%22recaptcha+%3D+false%22+filename%3Aconfig.toml+fork%3Atrue&s=indexed&type=Code. I haven't (and perhaps won't) check all of them, but from the samples that I've viewed, all of them has simply left this section untouched—perhaps due to error #7.

To sum up, we're breaking a broken stuff (so as to reassemble and fix it), so we don't need to worry about this.

If we are doing this, should we change ...staticman.staticman to a similar system of {{ if and ...}}? The new method would be backwards compatible in theory (assuming people only fill in the information if they want to use the feature).

hugo-future-imperfect-slim/layouts/post/comments.html

Lines 5 to 9 in f0ec5ac

	{{ else if .Site.Params.staticman.staticman }}
	<article class="post">
	{{ .Render "staticman" }}
	</article>
	{{ end }}

Lemme do that for you. From the last previous part, we can forget about the backward compatibility.

Upcoming: some Staticman changes out of the scope of this PR.

ℹ️ In the next time, when the local branch and the remote branch have diverged for just a few commits, please use git pull --rebase to avoid leaving a merge commit at the head of an unmerged PR. This linearise the commit graph of the branch.

[VincentTam]

🙏 Thanks for your great work this w.e. 🔬 I've just tested it again on my test site on Framagit 🌐 served by the Framagit instance @staticmanlab1. It worked !!!

📝 Test report [TL;DR]

1. Squash merged recent commits into my dev branch.

   Compare my dev branch with your Revert-staticman-recaptcha branch: https://github.com/pacollins/hugo-future-imperfect-slim/compare/Revert-staticman-recaptcha..VincentTam:dev?expand=1

   SHA-1 hash of my dev branch: 5a3e756

2. Updated the submodule of my test site.

   SHA-1 hash of the master branch: 18df4ac

3. Submitted a comment at a sample post.

4. Redirected to the post. @staticmanlab1 commited directly into the master branch.

   SHA-1 hash of the master branch: 19fb99d. Its parent is the commit mentioned in step 2. Note that the config.toml has not yet been changed. This shows the backward compatibility of this PR at ff52ad0.

5. Git pulled the Staticman comment in step 3 into my local repo.

6. Removed the lines recaptcha = true and staticman = true in config.toml.

   diff --git a/config.toml b/config.toml
   index 51fc23e..e3a379e 100644
   --- a/config.toml
   +++ b/config.toml
   @@ -92,7 +92,6 @@ disableLanguages        = [""]
        # Sets Statiman to be active
        # If using GitHub, go to https://github.com/apps/staticman-net
        # If using GitLab, just add the GitLab bot, NO need to hit `/connect/v3/...`
   -    staticman           = true
        # Sets the location for Staticman to connect to
        api                 = "staticman-frama.herokuapp.com"  # without trailing slash
        gitProvider         = "gitlab"  # either "github" or "gitlab"
   @@ -101,7 +100,6 @@ disableLanguages        = [""]
        branch              = "master"
    
        [params.staticman.recaptcha]
   -      recaptcha         = true
          siteKey           = "6Lcv8G8UAAAAAEqV1Y-XEPum00C_DxhD6O--qkFo" # Site Key
          encryptedKey      = "GLGBV5a+DHijkgBrPXRVvlK2CUrgyo0vOgpnVFTX76VUsTtrxqK8X5mJ9IjXiZbSmEwpflr6il1aLPueoV38nv9ZBIbPNUzj1xmt1CQ9HUOQFpB+n22HlwKJPqj29X8xkgCKnR190WK/KpPau6rF/PYzQn/eI+6hEEcCfOWUm9iQqUSk9vpCTtFD7LnFtxd+TjCxfY8R0uFpvs141aOAv/8wcR3ZzOjoWx3adAQkAgu6QLC1SVBQVnqDKiBrw4Tys/P8LhJE2Ul28tItI7XH5ZUpt8AV/38t5v+6cMEAk3sUr1VqnrPfjnQrLysXQlyeHMqqV48IZLLBWuMb0OqGog=="

   SHA-1 hash of this commit: 62c5562

7. Submitted a comment again at the post mentioned in step 3.

8. @staticmanlab1 received the comment, and regenerated the site at commit 3c1d2d9, whose parent is the commit mentioned in step 6.