Ehundu 0.11.0

Images with several widths come out with their srcset. Ehundu still does not process images, but now it reads the variants that another tool (an editor when you upload the photo, a separate program) leaves in publico/. With the imagenes section of sitio.yml (anchos: [480, 960, 1600], formatos: [avif, webp] and a sizes if needed), the variantes('img/foto') function gives src, ancho, alto, srcset and fuentes to write the <img> or the <picture>, and the [imagen …] shortcode and the figures in Markdown do it by themselves. The srcset carries the real width of each file, because whoever generates the variants does not enlarge a narrow photo; an image without variants comes out as before, without a warning; every file that is looked for, whether it exists or not, is a dependency of the page in the incremental build, and the feed turns the addresses in each srcset into complete ones (docs/formato.md (opens in a new tab), §2, §7.3 and §8.2, decision 120).

And a long list is split into pages with paginar. The page itself is the first one, at its address, and the engine generates the rest: paginar: {coleccion: blog, porPagina: 10, orden: fecha desc}. Each one sees the entries of its chunk in pagina.paginas and where it is in pagina.paginacion (actual, total, anterior, siguiente, enlaces…). By default they come out at /blog/pagina/2/, and paginar.url with {{ numero }} changes the form for whoever migrates and does not want to lose addresses. A new entry that overflows the last page opens the next one by itself. The pages it generates do not enter collections or the sitemap. Splitting each agrupar group is reserved (§6.4, decision 121).

0.10 sites build the same: nothing changes that does not use imagenes or paginar. Built from scratch with 0.10.0 and with this one, the twenty-three test sites (PDFs included) give the same files, one by one.

This version on GitHub (opens in a new tab), with the ehundu.phar.