git ssb

0+

mixmix / ssb-client-basic-tutorial



Commit 732c846359bc5138a27a082a1fa03a3d8ba0bf85

Browse Files

v01 - a pull-stream query

mixmix committed on 10/17/2018, 7:14:35 AM
Parent: 6a6bdd02661ff78d984eb5b4df7130ae3ab01135

Files changed

README.mdchanged
v01.jsadded
README.mdView
@@ -16,4 +16,33 @@
1616 … `whoami` is an asynchronous method which calls back with the details of the feed your scuttlebot is currently running. It's the basic "hello world" of scuttlebutt.
1717 …
1818 … We close the connection to the server using `server.close()` otherwise the connection stays open forever!
1919 …
20 …+
21 …+## `v01` - a pull-stream query!
22 …+
23 …+We introduce [**pull-stream**](https://github.com/pull-stream/pull-stream), which is a really common way to handle data in scuttlebutt.
24 …+The basic idea is a every complete pull-stream connects a source of data and runs that into a sink (some output).
25 …+Along the way, your data might go _through_ some steps which filter or modify the data
26 …+
27 …+Have a read of `v01.js` and see if you can guess what it does.
28 …+Run it by running `node v01.js` in the terminal and seeing what comes out.
29 …+_Kick the tyres_ by modifying the code and running it again to see what happens!
30 …+
31 …+```js
32 …+pull(
33 …+ server.query.read(opts), // the source
34 …+ pull.filter(msg => msg.value.content.type === 'post'), // filter 'through'
35 …+ pull.collect(onDone) // the sink
36 …+)
37 …+```
38 …+
39 …+The `pull` function wrapping the source, through, and sink connects these into a complete stream which data will immediately flow through.
40 …+
41 …+The source is provided by [**ssb-query**](https://github.com/dominictarr/ssb-query) which is super fancy, but we'll get to that later. All you need to know now is that opts says "gimme the last 100 messages going _backwards_ from right now".
42 …+
43 …+The `pull.filter` gets passed each one of the results that the source spits out, and we've set it up only to let `post` type messages continue on.
44 …+
45 …+The sink is a `pull.collect`, which waits until the stream is finished (here when we've pulled 100 messages), collecting all the results then passing them as an Array to the callback `onDone`.
46 …+
47 …+
48 …+NOTE - you need to be using a server with the ssb-query plugin installed for this to work (most have this!)
v01.jsView
@@ -1,0 +1,42 @@
1 …+const Connection = require('ssb-client')
2 …+const pull = require('pull-stream')
3 …+
4 …+console.log('Connecting')
5 …+
6 …+Connection((err, server) => {
7 …+ if (err) throw err
8 …+ console.log('Connection established')
9 …+
10 …+ const opts = {
11 …+ limit: 100,
12 …+ reverse: true
13 …+ }
14 …+
15 …+ pull(
16 …+ server.query.read(opts),
17 …+ pull.filter(msg => msg.value.content.type === 'post'),
18 …+ pull.collect(onDone)
19 …+ )
20 …+
21 …+ function onDone (err, msgs) {
22 …+ if (err) {
23 …+ console.error(err)
24 …+ server.close()
25 …+ return
26 …+ }
27 …+
28 …+ msgs.forEach(msg => {
29 …+ prettyPrint(msg)
30 …+ console.log('------')
31 …+ })
32 …+
33 …+ console.log(`${msgs.length} messages`)
34 …+ server.close()
35 …+ }
36 …+})
37 …+
38 …+function prettyPrint (msg) {
39 …+ console.log(JSON.stringify(msg, null, 2))
40 …+ // this just print the full object out as a string that's been nicely indented
41 …+ // with each level of nesting
42 …+}

Built with git-ssb-web