REGRESSION(r265601): invalid blame URL copied when `include_revision` is `false`

[Devin Rousso]

# STEPS TO REPRODUCE:
1. copy `CopyWebKitPermalink.sublime-settings` into `Packages/User`
2. set `"include_revision"` to `false`
3. attempt to "Copy WebKit Permalink to Blame" in any file in the tree
 => URL is missing "?"

[Devin Rousso]

Created attachment 407172 [details]
Patch

[Daniel Bates]

Comment on attachment 407172 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=407172&action=review

> Tools/CopyPermalink/Sublime Text/CopyWebKitPermalink/CopyWebKitPermalink.py:194
> +        return 'https://trac.webkit.org/browser/{}/{}?{}{}{}'.format(branch, path, revision, annotate_blame, line_number)

This doesn't look correct. Looks like ?&rev can be formed.

[Devin Rousso]

Comment on attachment 407172 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=407172&action=review

>> Tools/CopyPermalink/Sublime Text/CopyWebKitPermalink/CopyWebKitPermalink.py:194
>> +        return 'https://trac.webkit.org/browser/{}/{}?{}{}{}'.format(branch, path, revision, annotate_blame, line_number)
> 
> This doesn't look correct. Looks like ?&rev can be formed.

Yes, that was somewhat intentional.  I figured this would be easier and more straightforward than adding a bunch of `if`s to determine whether to put the `?` on the `revision` or on the `annotate_blame`.  AFAIK, having `?&` in a URL is valid, and it worked just fine with <https://trac.webkit.org/>.

[EWS]

Committed r266135: <https://trac.webkit.org/changeset/266135>

All reviewed patches have been landed. Closing bug and clearing flags on attachment 407172 [details].

[Radar WebKit Bug Importer]

<rdar://problem/67749448>