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
Laika supports setting a datePublished and dateModified attribute under the laika.metadata key in configuration headers in markup documents. Both of these values currently need to be set manually. There is an option to auto-populate the dateModified value from file system info, but this is at least beyond the scope of the 0.19 series, which is why it's currently in the backlog. Reasons for not doing it in 0.19 are:
The implementation is non-trivial as the information needs to be routed all the way from InputTreeBuilder to ParserRuntime and there is no obvious way to do that. Potential future implementations should try to avoid hard-coding a single, specific feature for this data transfer and think about a generic approach (e.g. passing a Config instance through these layers which could be used for other functionality in the future).
Secondly, the existing manual support is already a good option for most common cases. In my personal view it's actually a better option, because, while being mildly inconvenient, it gives the user more control: if you just added a comma, it is not a meaningful change that should be surfaced to the user and to search engines. That does not mean this is never going to happen, it just means that I see it as somewhat low-value, combined with high-effort described in the previous bullet, so it's something for 0.20 or 1.0, or whatever happens after 0.19.
The text was updated successfully, but these errors were encountered:
Laika supports setting a
datePublished
anddateModified
attribute under thelaika.metadata
key in configuration headers in markup documents. Both of these values currently need to be set manually. There is an option to auto-populate thedateModified
value from file system info, but this is at least beyond the scope of the 0.19 series, which is why it's currently in the backlog. Reasons for not doing it in 0.19 are:The implementation is non-trivial as the information needs to be routed all the way from
InputTreeBuilder
toParserRuntime
and there is no obvious way to do that. Potential future implementations should try to avoid hard-coding a single, specific feature for this data transfer and think about a generic approach (e.g. passing aConfig
instance through these layers which could be used for other functionality in the future).Secondly, the existing manual support is already a good option for most common cases. In my personal view it's actually a better option, because, while being mildly inconvenient, it gives the user more control: if you just added a comma, it is not a meaningful change that should be surfaced to the user and to search engines. That does not mean this is never going to happen, it just means that I see it as somewhat low-value, combined with high-effort described in the previous bullet, so it's something for 0.20 or 1.0, or whatever happens after 0.19.
The text was updated successfully, but these errors were encountered: