3 A minimal git‐based self‐hosted status publishing solution.
7 + A Python script, `post-receive`, suitable for use as a post‐receive
8 Git hook, which generates a number of `.jsonld` files from a
9 source repository of text posts.
11 + A small collection of H·T·M·L files designed to display the
12 generated `.jsonld` in a human‐readable fashion.
16 + A server or server configuration (you will need to supply these
19 + Suitable for deployment on a simple filesystem server like
20 NeoCities or GitHub Pages (it requires both Git hooks and H·T·T·P
23 + Interactive in any way (i·e “social media”).
25 + Compatible with the ActivityPub fediverse (or any other push‐based
26 social media platform).
28 ## Script Configuration
30 You will need to edit the `post-receive` script to adjust the following
31 constants to match your setup :—
33 + **`GIT_DIRECTORY`**:
34 The absolute path to this Git repository on your server.
35 It is expected that this will be a bare repository (ending in
38 + **`BUILD_DIRECTORY`**:
39 In order to access the files when you push, the repository will be
41 **This directory will be deleted on every push.**
43 + **`PUBLIC_DIRECTORY`**:
44 The directory that your server serves files from.
45 **This directory will be deleted on every push,** so if you need to
46 serve additional files (i·e those not generated by this script),
47 you should place those files in a different directory and adjust
48 your server configuration accordingly.
49 Note that the `post-receive` script and associted H·T·M·L files
50 provided by this repository expect a certain server configuration
54 The U·R·L that you are serving your site from (with no trailing
56 You cannot serve Status.git from a subdirectory.
58 A `Makefile` is provided to make installing the `post-receive` script
59 on your server easy (assuming you have `ssh` access).
60 You’ll need to supply some variables there, too :—
66 The domain or other address of your site.
68 + **`GIT_REPOSITORY`**:
69 The absolute path to this Git repository, as above.
71 ## Server Configuration
73 Your server should be configured to serve the following files from the
74 provided `PUBLIC_DIRECTORY` in response to the following requests :—
78 These responses **must** be served with a `Content-Type` of
79 `text/html;charset=UTF-8` (or equivalent).
80 Note that these paths **do not** have a trailing slash.
83 Serve the file at `/index.html`.
84 A `Link` header with the value
85 `</about.jsonld>;rel=meta;type="application/ld+json"` (or
86 equivalent) **must** be provided.
89 Serve the file at `/.about.html`.
90 A `Link` header with the value
91 `</about.jsonld>;rel=meta;type="application/ld+json"` (or
92 equivalent) **must** be provided.
94 + **`GET /statuses`**:
95 Serve the file at `/.statuses.html`.
96 A `Link` header with the value
97 `</statuses.jsonld>;rel=meta;type="application/ld+json"` (or
98 equivalent) **must** be provided.
100 + **`GET /$YYYY-MM`** (where `$YYYY-MM` is an `xsd:gYearMonth`):
101 Serve the file at `/.topic.html`.
102 A `Link` header with the value
103 `</$YYYY-MM.jsonld>;rel=meta;type="application/ld+json"` (or
104 equivalent) **must** be provided.
106 + **`GET /$YYYY-MM/*`** (where `$YYYY-MM` is an `xsd:gYearMonth`):
107 Serve the file at `/.status.html`.
108 A `Link` header with the value
109 `</$YYYY-MM.jsonld>;rel=meta;type="application/ld+json"` (or
110 equivalent) **must** be provided.
113 Serve the file at `/.topics.html`.
114 A `Link` header with the value
115 `</topics.jsonld>;rel=meta;type="application/ld+json"` (or
116 equivalent) **must** be provided.
118 + **`GET /topics/$TOPIC`** (where `$TOPIC` matches `[0-9A-Za-z_-]+`):
119 Serve the file at `/.topic.html`.
120 A `Link` header with the value
121 `</$TOPIC.jsonld>;rel=meta;type="application/ld+json"` (or
122 equivalent) **must** be provided.
124 + **`GET /topics/$TOPIC/*`** (where `$TOPIC` matches
126 Serve the file at `/.status.html`.
127 A `Link` header with the value
128 `</$TOPIC.jsonld>;rel=meta;type="application/ld+json"` (or
129 equivalent) **must** be provided.
131 ### Json‐L·D responses
133 These responses **should** be served with a `Content-Type` of
134 `application/ld+json`.
135 In all cases, for `/$PATH.jsonld`, this just serves the file at
136 `/$PATH/index.jsonld`.
138 + **`GET /about.jsonld`**:
139 Serve the file at `/about/index.jsonld`.
141 + **`GET /statuses.jsonld`**:
142 Serve the file at `/statuses/index.jsonld`.
144 + **`GET /$YYYY-MM.jsonld`** (where `$YYYY-MM` is an
146 Serve the file at `/$YYYY-MM/index.jsonld`.
148 + **`GET /topics.jsonld`**:
149 Serve the file at `/topics/index.jsonld`.
151 + **`GET /topics/$TOPIC.jsonld`** (where `$TOPIC` matches
153 Serve the file at `/topics/$TOPIC/index.jsonld`.
157 All responses **should** have a `Access-Control-Allow-Origin` header
158 with a value of `*` (assuming your server does not use credentials
159 and is not being served behind a firewall).
161 ## Committing Statuses To Git
163 The `post-receive` script will run whenever you make a commit to the
164 `live` branch, which should be set as your default.
165 Statuses are represented by a small collection of files committed to
166 particular locations in this repository :—
168 + Files committed to `/YYYY/MM/DD/HH.MM.SSZ/` (the final component
169 can actually take any form but a time is **recommended**) are
172 + Files committed to `/topic/TOPICNAME/a/b/c/d/xxxx` (where `a`, `b`,
173 `c`, and `d` are lowercase hexadecimal digits and `xxxx` can be
174 anything) is a status posted to a specific topic (`TOPICNAME`).
175 It is **recommended** that you make `a/b/c/d` the first four digits
176 of an MD5 hash of the status content and `xxxx` the remaining
177 digits (security is not an issue here, so MD5 is fine).
178 `TOPICNAME` should have the form `[0-9A-Za-z_-]+`.
180 The intent is that “ordinary statuses” are a bit more ephemeral whereas
181 “topic statuses” can serve as reference material.
182 Using an MD5 hash for topic statuses ensures you don’t post the same
185 The files which represent a status are as follows :—
188 The text of the status.
189 Blank lines separate paragraphs; linebreaks will be preserved.
190 There is a special markup for links: `<https://link.example>`, or
191 `<https://link.example>="link text"` if you want to supply link
193 At present, no other markup is supported.
195 + **`0=x_status_git_1.0`**:
196 This file is **optional** and not currently used for anything, but
197 indicates that the post follows the `1.0` format.
198 The contents of this file, if present, **must** be
199 `x_status_git_1.0`, optionally followed by a trailing newline.
201 + **`1=NAME`** (where `NAME` might be anything):
202 This file is **recommended** and indicates the author of the
204 Only one author is currently supported.
205 The value of `NAME` **must** give the name of the author of the
207 The contents of this file **must** give the author’s URL,
208 optionally followed by a trailing newline.
210 + **`3=YYYY-MM-DD`** (where `YYYY-MM-DD` is a date):
211 This file is **required** and indicates the date of the status.
212 Only one date is currently supported.
213 The value of `YYYY-MM-DD` **should** give the date of the status,
214 although this is not used (in favour of the full timestamp).
215 The contents of this file **must** give the full `xsd:dateTime`
216 timestamp of the status, optionally followed by a trailing
219 + **`4=IDENTIFIER`** (where `IDENTIFIER` might be anything):
220 This file is **required** and provides an identifier for the
222 Only one identifier is currently supported.
223 The contents of this file **must** be an I·R·I which uniquely
224 identifies the status (for example, a U·U·I·D or `tag:` URI).
225 The value of `IDENTIFIER` **must** be a locally‐unique identifier
226 for the status and **should** resemble the contents where
228 (Note, however, that `IDENTIFIER` cannot contain slashes and need
229 not be a valid I·R·I.)
231 Files with names that begin with the strings `2=` or `x_status_git_`
232 are reserved for backwards‐compatible extensions.
234 Status.git has no opinion on how these files make their way into the
235 Git repository, except that all the files for a single status should
236 be added in the same commit.
237 The intention is that the simple nature of these files will make them
240 ## I Am Computer, How Do I Get Status?
242 Assume you are given a U·R·L `resource_url` which you think points to
243 some kind of Status.git resource.
244 Start by resolving it as follows :—
246 01. Make a `HEAD` request to `resource_url`.
248 02. If there is a `Link` header with a `rel` of `meta` and a `type` of
249 `application/ld+json`, set `resource_url` to the URL provided in
250 that header and restart these steps from step 1.
252 03. Make a `GET` request to `resource_url` and let `response` be the
255 04. Set `document` as follows :—
257 01. If the `Content-Type` header of `response` has a type of `text`
258 and subtype of `html`, let `document` be the result of
259 processing the body of `response` into a D·O·M tree as an
261 It is an error if this process fails.
263 02. If the `Content-Type` header of `response` has a type of
264 `application` and a subtype which is `xml` or which ends in
265 `+xml`, let `document` be the result of processing the body
266 of `response` into a D·O·M tree as an X·M·L document .
267 It is an error if this process fails.
269 03. Otherwise, let `document` be null.
271 05. If `document` is not null :—
273 01. If there is a `<link>` element in either the H·T·M·L namespace
274 or the Atom namespace in `document` with a `rel` of `meta`
275 and a `type` of `application/ld+json`, set `resource_url` to
276 the `href` of that `<link>` element and restart these steps
278 If multiple such elements exist, choose the first one.
280 02. Otherwise, it is an error.
282 06. If the body of `response` is not a Json document, it is an error.
284 Assuming the U·R·L you were given was valid, you will end this
285 algorithm with a Json‐L·D response, and you can use the `@type`
286 attribute to determine the response type.
287 `@type` will be either a string or an array.
289 + If the `@type` is or contains `Forum`, the resource is a collection
292 + If the `@type` is or contains `Thread`, the resource is a
293 collection of statuses.
295 + If the `@type` is or contains `Microblog`, the resource describes
296 this site as a whole.
297 The `streams` property will contain a list of available `Forum`s
298 and `Thread`s, as objects with an `@id` and `@type`.
300 The items in the collection (`Forum` or `Thread`) may be determined
301 through one of the following methods :—
303 + If the `@type` is or contains `CollectionPage` or
304 `OrderedCollectionPage`, then its `items` will contain resources.
305 This is a partial collection, and the `prev` and `next` properties
306 can be used to access further items from the parent collection
307 (indicated by `partOf`).
308 `first` and `current`, in this scenario, point “horizontally” to
309 the first and latest pages of items, not to subpages.
311 + If the `@type` is or contains `Collection` or `OrderedCollection`
312 and the resource has `first` and/or `current` properties, then
313 the `items` property will not be present.
314 `first` and `current` can be accessed to provide
315 `OrderedCollectionPage`s listing the items of the collection.
317 + Otherwise, the `items` property will contain every item in the
320 Statuses themselves have the following properties :—
323 The identifier of the status.
326 The value `MicroblogPost`.
328 + **`created`** [`dcterms:created`]:
329 The creation date for the status, as an `xsd:dateTime`.
331 + **`creator`** [`dcterms:creator`] (optional):
332 The author of the status, as an object with an `@id` and `name`
335 + **`identifier`** [`dcterms:identifier`]:
336 An I·R·I which uniquely identifies the status.
337 This differs from the `@id` in that it is not expected to be
340 + **`subject`** [`dcterms:subject`] (optional):
341 The topic of the status, for topic statuses.
343 + **`content`** [`sioc:content`]:
344 The content of the status, as an `rdf:XMLLiteral`.