You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I'm doing some major spring cleaning by re-designing/implementing all template handling in Hugo (with goals of making it more powerful but also much easier to understand). We have lots of integration tests, which in this case is both a curse and a blessing.
One of my now failing tests is basically this, in a site with 1 language (en):
By just looking at the above, it is obvious that these 3 are translations of the same document, but since we only have English defined, we get 3 pages (all with language set to en) with the following paths:
/doc1
/doc1.fr
/doc.sv
Which is obviously wrong. In my new setup, the last two will be thrown out as duplicates.
Note that there may be some examples in the wild somehow broken by this (e.g. /dot.net.md vs /dot.foo.md), but I'm not sure I'm prepared to go the extra mile to complicate this further. Those who think this is a considerable problem needs to raise an convincing argument in this issue..
The text was updated successfully, but these errors were encountered:
In the above example, I would have internally classified 234 and 5 as "unrecognized identifiers", so it would be possible to add a warnidf or something about it.
bep
added a commit
to bep/hugo
that referenced
this issue
Mar 24, 2025
Some notes:
* We now match using the full page Path and walk downwards from the root to find the most specific template.
* Outputformat definitions with different values of isPlainText is not compatible (a bug in the old setup)
Fixesgohugoio#13515
bep
added a commit
to bep/hugo
that referenced
this issue
Mar 25, 2025
Some notes:
* We now match using the full page Path and walk downwards from the root to find the most specific template.
* Outputformat definitions with different values of isPlainText is not compatible (a bug in the old setup)
Fixesgohugoio#13515
Some notes:
* We now match using the full page Path and walk downwards from the root to find the most specific template.
* Outputformat definitions with different values of isPlainText is not compatible (a bug in the old setup)
Fixesgohugoio#13515
I'm doing some major spring cleaning by re-designing/implementing all template handling in Hugo (with goals of making it more powerful but also much easier to understand). We have lots of integration tests, which in this case is both a curse and a blessing.
One of my now failing tests is basically this, in a site with 1 language (en):
By just looking at the above, it is obvious that these 3 are translations of the same document, but since we only have English defined, we get 3 pages (all with language set to
en
) with the following paths:Which is obviously wrong. In my new setup, the last two will be thrown out as duplicates.
Note that there may be some examples in the wild somehow broken by this (e.g.
/dot.net.md
vs/dot.foo.md
), but I'm not sure I'm prepared to go the extra mile to complicate this further. Those who think this is a considerable problem needs to raise an convincing argument in this issue..The text was updated successfully, but these errors were encountered: