Generate pages with multi-level path

[migpimenta]

Let's say I want to generate pages on the following URLs:
/animal/mammal/1/whale
/animal/mammal/2/tiger
/animal/reptile/3/snake

The directory pattern would have to be /animal/{group}/{id}/{name} where input data is an array of animals. The directory does not support patterns.

I have tried to create an object property with the value /animal/mammal/1/whale. That doesn't work either, pages are placed in the folder animalmammal1whale, i.e., slashes are stripped.

[kateho]

+1 for this

[avillafiorita]

pull requests #63 seems to be doing what you are asking. Quite some time passed by, I hope the solution is still relevant.

[tjkohli]

pull requests #63 seems to be doing what you are asking. Quite some time passed by, I hope the solution is still relevant.

The way I understand it, @migpimenta is asking for a way to have /animal/mammal/whale/ all built from data, whereas the #63 merge pertains to creating a page called animal_mammal_whale.

How can we achieve the former result with proper nested directories?

[avillafiorita]

Thanks, now I get it, different label assigned (... easier than actually solving the issue ;-))

[fititnt]

@migpimenta @kateho @tjkohli

Its not fully tested ("it's working for me") but this diff may allow what is asking.

HXL-CPLP/Auxilium-Humanitarium-API@0891b72#diff-c1005e95bd3da19bb281fd99bbdd51ee66b5725a5e384d291bf30246310cbc6c

This link may show the difference https://www.diffchecker.com/ZsMpGYvp.

# _config.yml
# @see https://github.com/avillafiorita/jekyll-datapage_gen
page_gen-dirs: true
page_gen:
  - data: 'api'
    template: 'api'
    name_expr: "record['uid'].split('/').last"
    dir_expr: "record['uid'].chomp(record['uid'].split('/').last + '/')"

# _data/api.yml
### [linguam]/api/UN ___________________________________________________________

## -----------------------------------------------------------------------------
- x-default: /mul/api/UN/HDX/
  uid: /mul/api/UN/HDX/
  linguam: mul
  typum: api
  gid: UN
  # lid: HDX
  title: "Index: HDX"
  noindex: true

- x-default: /mul/api/UN/HDX/
  uid: /eng/api/UN/humanitarian-data-exchange/
  linguam: eng
  typum: api
  gid: UN
  # lid: HDX
  openapi_filum: /api/UN/HDX/eng/openapi.yaml
  slug: humanitarian-data-exchange
  title: HDX - The Humanitarian Data Exchange API
  noindex: true

## -----------------------------------------------------------------------------
- x-default: /mul/api/UN/reliefweb/
  uid: /mul/api/UN/reliefweb/
  linguam: mul
  typum: api
  gid: UN
  # lid: reliefweb
  title: "Index: ReliefWeb"
  noindex: true

- x-default: /mul/api/UN/reliefweb/
  uid: /eng/api/UN/reliefweb/
  linguam: eng
  typum: api
  gid: UN
  # lid: ReliefWeb
  openapi_filum: /api/UN/reliefweb/eng/openapi.yaml
  slug: reliefweb
  title: ReliefWeb - Informing humanitarians worldwide
  # title: ReliefWeb - Informing humanitarians worldwide
  noindex: true

(...)

Is generating a sitemap like this

  <!--
    typum: api
  -->
  
  <url>
    <loc>https://hapi.etica.ai/mul/api/UN/hdx/</loc>
    
    <lastmod>2021-05-17T22:44:58-03:00</lastmod>
    
    <changefreq>weekly</changefreq>
    
    <priority>0.3</priority>
    
    
    <xhtml:link
      rel="alternate"
      hreflang="x-default"
      href="https://hapi.etica.ai/mul/api/UN/HDX/" />
  </url>
    
  <url>
    <loc>https://hapi.etica.ai/eng/api/UN/humanitarian-data-exchange/</loc>
    
    <lastmod>2021-05-17T22:44:58-03:00</lastmod>
    
    <changefreq>weekly</changefreq>
    
    <priority>0.3</priority>
    
    
    <xhtml:link
      rel="alternate"
      hreflang="x-default"
      href="https://hapi.etica.ai/mul/api/UN/HDX/" />
  </url>
    
  <url>
    <loc>https://hapi.etica.ai/mul/api/UN/reliefweb/</loc>
    
    <lastmod>2021-05-17T22:44:58-03:00</lastmod>
    
    <changefreq>weekly</changefreq>
    
    <priority>0.3</priority>
    
    
    <xhtml:link
      rel="alternate"
      hreflang="x-default"
      href="https://hapi.etica.ai/mul/api/UN/reliefweb/" />
  </url>
    
  <url>
    <loc>https://hapi.etica.ai/eng/api/UN/reliefweb/</loc>
    
    <lastmod>2021-05-17T22:44:58-03:00</lastmod>
    
    <changefreq>weekly</changefreq>
    
    <priority>0.3</priority>
    
    
    <xhtml:link
      rel="alternate"
      hreflang="x-default"
      href="https://hapi.etica.ai/mul/api/UN/reliefweb/" />
  </url>

Comments

I may need to make a few other changes (so lastest file on the the repository may have some other changes), but if someone is in a hurry,this approach could work.

Note that the

    name_expr: "record['uid'].split('/').last"
    dir_expr: "record['uid'].chomp(record['uid'].split('/').last + '/')"

is a bit hacky way to use a single variable to split between the folder and the name (so most people may not need to need it. Also, this likely to break with the datapage_url (at least I did not tested)

Also, great project @avillafiorita !

[fititnt]

avillafiorita on 16 Oct 2020
Thanks, now I get it, different label assigned (... easier than actually solving the issue ;-))

Yes! Actually, looking at the code, it was the same logic made by @DanielJDufour, so mostly is his credits.

[DanielJDufour]

@fititnt , thank you. I appreciate the kind words :-)

[ryanwoldatwork]

+1

[probablykory]

This was very useful, thank you. I was able to output multiple related renderings of records, grouped by directory, which really simplifies the output layout.