Typo patch (#19)

* mv md to adoc

* adoc format

* fix typo in fn doc

README.adoc

@@ -0,0 +1,65 @@
+= Waterfall
+
+image:https://github.com/robertluo/waterfall/actions/workflows/main.yml/badge.svg[CI,link=https://github.com/robertluo/waterfall/actions/workflows/main.yml]
+image:https://img.shields.io/clojars/v/io.github.robertluo/waterfall.svg[Clojars project,link=https://clojars.org/io.github.robertluo/waterfall]
+image:https://cljdoc.org/badge/io.github.robertluo/waterfall[cljdoc,link=https://cljdoc.org/d/io.github.robertluo/waterfall]
+
+== Rational
+
+Try to use Apache Kafka clients in idiomatic Clojure.
+
+Kafka has Java/Scala clients API, and programmers have no problem calling them directly in Clojure, so we do not need to "wrap" Java API.
+
+However, from a Clojure programmer's perspective, direct call Kafka Java API does not feel very good:
+
+ - It is complex, with many interfaces, classes, and methods, which is not a good experience for a dynamic language.
+ - It looks more like a framework or even frameworks. Users have to have a lot of background knowledge and implementation details.
+ - Clojure programmers could use transducers to deal with data and do not need to use Kafka streams. 
+
+Hence, Waterfall is an attempt at a minimalist library with additional optional batteries.
+
+ - Only two core functions: `producer` and `consumer`.
+ - Relying on the excellent https://github.com/clj-commons/manifold[manifold library], `producer`, and `consumer` is just the implementation of the Manifold stream.
+ - Minimum necessary dependencies. 
+ - Clojure version SERDES, in `robertluo.waterfall.shape` namespace. Underlying, give Kafka API byte array SERDES. 
+ - Optional features require optional dependencies. For example, if you want to use nippy as SERDES, you put it in your classpath.
+
+== Development experience
+
+Because all functions have their schemas incorporated, you can get the best development experience if dependency `metosin/malli` is in your classpath.
+
+Put `(malli.dev/start!)` in your `user.clj` will https://github.com/metosin/malli/blob/master/docs/function-schemas.md#tldr[enable clj-kondo] to use the schemas when editing.
+
+## Single API
+
+Powered by https://github.com/robertluo/fun-map[fun-map], you can use one single API for accessing a Kafka cluster without any further knowledge:
+
+ - https://cljdoc.org/d/io.github.robertluo/waterfall/CURRENT/api/robertluo.waterfall#kafka-cluster[`kafka-cluster`]
+
+You can see an example in link:notebook/easy.clj[this easy example notebook].
+
+## API namespaces
+
+[cols="1,1"]
+|===
+| namespace | Description
+
+| robertluo.waterfall | main API
+| robertluo.waterfall.shape | optional Clojure data shape transformation
+|===
+
+## Files Description
+
+[cols="1,1"]
+|===
+| Filename | Description
+
+| deps.edn  | Clojure tools.deps configuration
+| tests.edn | kaocha test runner configuration
+| build.clj | Clojure tools.build building script
+| notebook  | https://github.com/nextjournal/clerk[Clerk] notebooks to demostrate usage, use `clojure -M:dev:notebook` to use them
+|===
+
+== https://unlicense.org[Unlicense]
+
+2022, Robertluo

src/robertluo/waterfall.clj

@@ -3,10 +3,9 @@
   (:require 
    [robertluo.waterfall
     [core :as core]
-    [util :as util]]
-   [manifold.stream :as ms]
-   [robertluo.waterfall.shape :as shape]
-   [robertluo.waterfall.util :as util]))
+    [util :as util]
+    [shape :as shape]]
+   [manifold.stream :as ms]))
 
 (def schema
   "Schema for waterfall"
@@ -86,13 +85,13 @@
 (defn kafka-cluster
   "returns a convient life-cycle-map of a kafka cluster. It requires:
     - `::nodes` kafka bootstrap servers
-    - `::shapes` shape of data, example: [(shape/topic (constantly \"sentence\"))(shape/edn)(shape/value-only)]
+    - `::shapes` shape of data, example: `[(shape/topic (constantly \"sentence\"))(shape/edn)(shape/value-only)]`
    
    If you just use it to publish message,
     - optional `::producer-config` can specify additional kafka producer configuration.
    
    If you want to consume from topics:
-    - `::topics` the topic you want to subscribe to. example: [\"sentence\"]
+    - `::topics` the topic you want to subscribe to. example: `[\"sentence\"]`
     - `::group-id` the group id for the message consumer
     - optional `::source-xform` is a transducer to process message before consuming
     - optional `::consumer-config` can specify additional kafka consumer configuration. With additions:
@@ -101,10 +100,10 @@
    
    The returned map has different level of key-values let you use:
     - Highest level, no additional knowledge:
-      - For consumer: `::consume` a function, a one-arity (each message) function as its arg, returns nil.
+      - For consumer: `::consume` a function, a one-arity (each message) function as its arg, returns `nil`.
       - For producer: 
-        -`::put` a function with a message as its arg. e.g. ((::put return-map) {:a 3})
-        -`::put-all` a function with message sequence as its arg
+        -`::put!` a function with a message as its arg. e.g. `((::put return-map) {:a 3})`
+        -`::put-all!` a function with message sequence as its arg
     - Mid level, if you need access of underlying manifold sink/source.
       - `::source` a manifold source.
       - `::sink` a manifold sink.
@@ -139,7 +138,7 @@
                     (->> (comp (map (shape/deserializer shapes)) (or source-xform (map identity)))
                          (xform-source consumer)))
       ::put! (fnk [::sink] (partial ms/put! sink))
-      ::put-all (fnk [::sink] (partial ms/put-all! sink))
+      ::put-all! (fnk [::sink] (partial ms/put-all! sink))
       ::consume (fnk [::source] #(ms/consume % source))})
     kafka-conf-map)
    (throw (ClassNotFoundException. "Need io.github.robertluo/fun-map library in the classpath"))))