powered by
cDoc
FTD
FPM
dF
Home
Author Manual
Builder Manual
Development
About Author Manual
Sitemap

Sitemap
“FPM Sitemap” is the structure we recommend for sites. This recommendation is optional, you can create any sitemap you want for your site.

The Structure
A site should have “sections” on top level. Each section should ideally be listed in the header of the site.

Each section may have one or more sub-sections. Sub-section should be listed as second level navigation in the header of the site. Each sub-section has one or more documents organised in a “table of content”, and TOC should be shown on the left.

How To Configure Sitemap For Your Site
FPM.ftd contains the sitemap of your site.

Sitemap Example (in FPM.ftd)
-- fpm.sitemap:

# Section: /section/url/
nav-title: Optional Longer Section

# If Section: Has A Colon In The Name
url: sectionURL

## Sub Section: /sub/url/
nav-title: Longer Sub Section

## If Sub Section: Has A Colon In The Name
url: whatever

- ToC Item: toc/
  nav-title: Shorter title
Note: The URLs in sitemap start with slash, but we remove the first slash. We do this because some projects may deploy FPM package on a base URL eg, foo.com/bar/, so all reference to /x/ is actually a reference to foo.com/bar/x/. We also convert / to index.html for the same reason.

Sitemap $processor$
We have $processor$ called sitemap, which can be used to get sitemap data:

-- fpm.sitemap-data sitemap:
$processor$: sitemap
Consider a package contains the following sitemap
In FPM.ftd
-- fpm.sitemap:

# Section Title: /


## Subsection Title: /

- Toc Title: /
- Toc Title 1: /foo/
  - Bar Title: /bar/


## Subsection Title 2: subsection2/

- Other Toc Title: subsection2/foo/



# Second Section Title: section/

## Second Subsection Title: second-subsection/

- Second Toc Title: second-toc/
Now, for the sitemap processor in the document with id bar/ would return the value
{
  "sitemap": {
    "sections": [
      {
        "title": "Section Title",
        "url": "/",
        "is-active": true,
        "children": []
      }
    ],
    "subsections": [
      {
        "title": "Subsection Title",
        "url": "/",
        "is-active": true,
        "children": []
      },
      {
        "title": "Subsection Title 2",
        "url": "subsection2",
        "is-active": false,
        "children": []
      }
    ],
    "toc": [
      {
        "title": "Toc Title",
        "url": "",
        "is-active": false,
        "children": []
      },
      {
        "title": "Toc Title 1",
        "url": "foo/",
        "is-active": true,
        "children": [
          {
            "title": "Bar Title",
            "url": "bar/",
            "is-active": true,
            "children": []
          }
        ]
      }
    ]
  }
}

Missing Sub Section
If a TOC comes directly in a section the section would have a single anonymous sub-section, and this sub-section would not be shown in UI. In UI people will just see the section header and toc on left, no sub-section line.

# Section Title: section/

- Toc 1: toc/
  - Toc 2: toc2/

FPM Build Behaviour

If a document is not part of sitemap, it will be built as is built right now. All documents that are part of sitemap are built in a special way.

FPM build will first parse the sitemap and build all the URLs that are not part of it, and then in another pass build all the ones that are in it.

A document can appear in multiple places in sitemap, in that case FPM builds one HTML file for each time a ftd document appears in sitemap.

Note: A document can appear only once in a single TOC?

Canonical URL

If a ftd document appears multiple times in sitemap, one of them would be the canonical, the “main” URL.

Consider the following example:

Suppose foo.ftd has to be appeared more than once in the sitemap. The sitemap can include this document as foo/, this is the “main” URL. The other way to include it is by passing url something like this. foo/-/<something>/. The -/ is the pointer to get the document. Anything preceding -/ would be the document id. The generated html of this document will include the canonical url pointing to foo/.

document-metadata: Key Value Data in Sitemap

Document can use get-data processor to get value of any key specified in the sitemap. Since a document would get rendered once for each occurrence of the document in the sitemap, each occurrence can have different data and the occurrence specific data would be returned by get-data.

The document-metadata supports inheritance. This means that the document-metadata presents in section get passed to it’s subsection and TOCs. Similarly, subsection document-metadata get passed to TOCs. And also the parent TOC-item’s document-metadata get passed to its children TOC.

# name: section/url/
key1: value1
key2: value2

## sub name: subsection/url/
key3: value3

- toc/url/
  key4: value4
  - childtoc/url/
    key5: value5
In the above example, the section/url/ section have two document-metadata key1: value1 and key2: value2 The subsection/url/ subsection have three document-metadata where two are inherited from section. i.e. key1: value1, key2: value2 and key3: value3 The toc/url/ toc item have four document-metadata, where three are inherited from section and subsection. i.e. key1: value1, key2: value2, key3: value3 and key4: value4 The childtoc/url/ toc item have five document-metadata, where four are inherited from section, subsection and it’s parent TOC. i.e. key1: value1, key2: value2, key3: value3, key4: value4 and key5: value5

Variable can be changed based on document-metadata
Using the get-data, the title can be different:

-- boolean show-dev-info:
$processor$: get-data

-- string page-title: The Normal Title

-- page-title: The Dev Title
if: $show-dev-info

-- ft.h0: $page-title
sitemap
-- fpm.sitemap:

# Overview

- foo/

# Development

- foo/-/1
  show-dev-info: true

Including Documents From Other FPM Packages In Sitemap
A package foo.com can chose to include a document in bar.com by including it in sitemap.

-- fpm.sitemap:

- Intro: -/bar.com/intro/

In this case the file would get copied over, and the url of intro would be https://foo.com/-/bar.com/intro/. For dependent packages, the url should start with -/ and then the package name, following the document id, The canonical url for this would be the url of the document on the site of the package. i.e. The generated HTML, in this case, contains the canonical url as bar.com/intro

The document from dependent package can be included more than once. This can be achieved in the same manner as the document in the current package included more than once, which is mentioned earlier.

Consider the example below:

-- fpm.sitemap:

- Intro: -/bar.com/intro/-/main/
So the document intro.ftd in the package bar.com is included in the sitemap with the variant main. The generated HTML includes the canonical url with value as bar.com/intro

Skip Header
skip: true

Motivation Behind skip Header

If people want to draft something and don’t want to publish any section, sub section or toc, they can use skip in section, sub-section and toc header.

The skipped section, sub-section or toc would not be available in processor sitemap till it is not the active opened page.

skip in Section
We have header called skip(by default false), using this header we can skip the whole section.

-- fpm.sitemap:

# Section 1: /

## Section 1 Subsection 1: /subsection1

## Section 1 Subsection 2: /subsection2

# Section 2: /
skip: true

## Section 2 Subsection 1: /subsection1

## Section 2 Subsection 2: /subsection2
In this case, Whole Section 2 will be skipped and will not displayed.
Page without skip header
Page with skip header

skip in Subsection

We have header called skip(by default false), using this header we can skip the whole subsection.

In the below example Subsection 1 of Section 1 and Subsection 2 of Section 2 will be skipped

-- fpm.sitemap:

# Section 1: /

## Subsection 1: /subsection1
skip: true

## Subsection 2: /subsection2

# Section 2: /

## Subsection 1: /subsection1
skip: true

## Subsection 2: /subsection2

skip in ToC

We have header called skip(by default false), using this header we can skip the whole toc.

In the below example, ToC 3 and ToC 5 will be skipped.

-- fpm.sitemap:

# Section: /

## Subsection : /subsection

- ToC 1: /page1
- ToC 2: /page2
  - ToC 3: /page3
  skip: true
  - ToC 4: /page4
- ToC 5: /page5
skip: true
- ToC 6: /page6
skip ToC Header

Access Control Using Sitemap

Different parts of a package may have access control limitation, like who can read or edit a package. Read and write are only enforced when fpm serve is used for serving a fpm package, and do not work when using fpm build. fpm build ignores all documents that are not world readable.

By default if no access control is defined, the document is used to be readable by the world, and not writable by any.

Access control is specified in terms of user groups.

-- fpm.sitemap:

# name: section/url/
writers: write-group-id

## sub name: subsection/url/

- toc: /url/
  readers: reader-group-id
  - child toc: /url/
    key5: value5
If a readers is specified it is assumed that sub section of the site is no longer readable by the world. There exists a special group call “everyone”, which can be used to set a subtree readable by the world.

Access Control Inheritance
Both readers and writers are inherited from parent’s section, subsection and toc. And they get merged.

-- fpm.sitemap:

# name: /section/url/
readers: foo

## sub name: /subsection/url/

- U: /toc/url/
  readers: bar
  - U2: /child/toc/url/
    key5: value5


In this example, documents  `readers: foo` is specified on `/section/url/` and
`/subsection/url/` is a child, so `foo` has read access to `/subsection/url/`
as well.

`/toc/url/` is grand-child of `/section/url/`, and it has also specified an
extra reader `bar`, so both `foo` and `bar` have access to `/toc/url/` and it's
children, which is `/child/toc/url/`.

Resetting Access Control Inheritance

If you want to overwrite, you can say “readers: not-inherited” or “writers: not-inherited”, this is a special group which removes inheritance till now, and only other writers or readers defined at this level are used.

Eg

-- fpm.sitemap:

# name: /section/url/
readers: foo

## sub name: /subsection/url/

- U: /toc/url/
  readers: bar
  readers: not-inherited
  - U2: /child/toc/url/
    key5: value5
Now since /toc/url/ has specified both not-inherited and bar, foo will not have access to /toc/url/ and only bar will have access to it.

Global ACL
If you do not want to specify groups for each section, you can specify it at sitemap level as well:

-- fpm.sitemap:
readers: foo

# name: /section/url/

## sub name: /subsection/url/

- U: /toc/url/
  - U2: /child/toc/url/
    key5: value5

Here we have added readers key directly on fpm.sitemap itself, so entire site is only readable by foo.

You can still specify access control at any node, and regular inheritance rules specified above will apply.

How Is ACL Implemented
For HTTP request get doc-id and read. Based on this find the groups using get-readers or get-writers. Given these groups, total identities (traverse all group trees, and find all identities, remove the minus signs, and create a set of all identities). Pass total identities to get-identities API, it returns actual identities for the current user.