diff --git a/doc/api/content/algo/getting-started/development.md b/doc/api/content/algo/getting-started/development.md index 1ffca3750..934d28424 100644 --- a/doc/api/content/algo/getting-started/development.md +++ b/doc/api/content/algo/getting-started/development.md @@ -133,5 +133,8 @@ Breakpoint 1 (algo_template.cpp:45) pending. Starting program: /sparkbin/sparkd run ``` -If your editor supports debugging over the ssh protocol you can configure and start sshd from inside the container. +If your editor supports debugging over the ssh protocol you can start sshd from inside the container. +## Next + +Continue to [Algo Lifecycle](/algo/ref/lifecycle) to learn how Algos are created and respond to external evens in Spark. diff --git a/doc/api/content/algo/getting-started/overview.md b/doc/api/content/algo/getting-started/overview.md index 87875f3a0..3b23e7f4d 100644 --- a/doc/api/content/algo/getting-started/overview.md +++ b/doc/api/content/algo/getting-started/overview.md @@ -5,7 +5,7 @@ weight = 0 Spark consists of a client UI ([Spark Client](https://www.bluetradingsystems.com/bts-spark#ladder-algo-screen)), a high performance professional futures front end, connected via our proprietary protocol over TCP to our Spark Server, which in turn connects to execution venues via their native protocols (typically FIX/TCP for execution and some proprietary protocol over UDP multicast for market data). - + ## What is the Spark Algo SDK? diff --git a/doc/api/content/algo/getting-started/testing.md b/doc/api/content/algo/getting-started/testing.md deleted file mode 100644 index 033fa787d..000000000 --- a/doc/api/content/algo/getting-started/testing.md +++ /dev/null @@ -1,6 +0,0 @@ -+++ -title = "Testing" -weight = 4 -+++ - -* Test your algo diff --git a/doc/api/content/algo/ref/lifecycle.md b/doc/api/content/algo/ref/lifecycle.md index 7b0b37456..9533aec5e 100644 --- a/doc/api/content/algo/ref/lifecycle.md +++ b/doc/api/content/algo/ref/lifecycle.md @@ -3,4 +3,30 @@ title = "Algo Lifecycle" weight = 0 +++ -todo +Spark Algos are registered when Spark Server starts and move though several states as they are activated and become live on the market. + +## Load and Register + +On startup Spark Server loads all plugins listed in the /sparkbin/plugins directory and calls an exported function named `init_spark_plugin` with the following signature. + +```cpp +extern "C" void init_spark_plugin(bts::SparkServices &serv) { } +``` + +Your Algo must export this function and use the bts::SparkServices instance to register your Algos. You can define more than one Algo type in a single plugin. In that case your init_spark_plugin function should register all available Algos. + +To register an Algo construct an instance of `DynamicOrderDescriptor` + +## Construction + + +## Before Launch +optional but suggested + +## Launch / Launch Attached +required, latency sensitive + +## Update + + +Every Spark Algo implements diff --git a/doc/api/static/images/API.png b/doc/api/static/images/API.png new file mode 100644 index 000000000..e53f95aaf Binary files /dev/null and b/doc/api/static/images/API.png differ diff --git a/doc/api/static/images/logo.svg b/doc/api/static/images/logo.svg new file mode 100644 index 000000000..d9d0f4b0c --- /dev/null +++ b/doc/api/static/images/logo.svg @@ -0,0 +1,2 @@ + + diff --git a/doc/api/templates/index.html b/doc/api/templates/index.html index 41d5a54c8..561dde3cd 100644 --- a/doc/api/templates/index.html +++ b/doc/api/templates/index.html @@ -20,7 +20,7 @@