#12 introduce GATEWAY_PORT, rename GATEWAY_HOST and GATEWAY_PROTOCOL to EXTERNAL_HOST and EXTERNAL_PROTOCOL

Keep GATEWAY_HOST and GATEWAY_PROTOCOL for backwards compatibility (for now)

Readme.md

@@ -48,9 +48,6 @@ wget https://github.com/chanzuckerberg/cellxgene/raw/master/example-dataset/pbmc
 ```bash
 export CELLXGENE_DATA=../cellxgene_data  # change this directory if you put data in a different place.
 export CELLXGENE_LOCATION=`which cellxgene`
-export GATEWAY_HOST=localhost:5005
-export GATEWAY_PROTOCOL=http
-export GATEWAY_IP=127.0.0.1
 ```
 
 3. Now, execute the cellxgene gateway:
@@ -63,10 +60,11 @@ Here's what the environment variables mean:
 
 * `CELLXGENE_LOCATION` - the location of the cellxgene executable, e.g. `~/anaconda2/envs/cellxgene/bin/cellxgene`
 * `CELLXGENE_DATA` - a directory that can contain subdirectories with `.h5ad` data files, *without* trailing slash, e.g. `/mnt/cellxgene_data`
-* `GATEWAY_HOST` - the hostname and port that the gateway will run on, typically `localhost:5005` if running locally
-* `GATEWAY_PROTOCOL` - typically http when running locally, can be https when deployed if the gateway is behind a load balancer or reverse proxy.
-* `GATEWAY_IP` - ip addess of instance gateway is running on, mostly used to display SSH instructions
 Optional environment variables:
+* `EXTERNAL_HOST` - the hostname and port from the perspective of the web browser, typically `localhost:5005` if running locally. Defaults to "localhost:{GATEWAY_PORT}"
+* `EXTERNAL_PROTOCOL` - typically http when running locally, can be https when deployed if the gateway is behind a load balancer or reverse proxy that performs https termination. Default value "http"
+* `GATEWAY_IP` - ip addess of instance gateway is running on, mostly used to display SSH instructions. Defaults to `socket.gethostbyname(socket.gethostname())`
+* `GATEWAY_PORT` - local port that the gateway should bind to, defaults to 5005
 * `GATEWAY_EXTRA_SCRIPTS` - JSON array of script paths, will be embedded into each page and forwarded with `--scripts` to cellxgene server
 * `GATEWAY_ENABLE_UPLOAD` - Set to `true` or `1` to enable HTTP uploads. This is not recommended for a public server.
 

cellxgene_gateway/cache_entry.py

@@ -96,7 +96,7 @@ class CacheEntry:
         dataset = self.dataset
 
         gateway_basepath = (
-            f"{env.gateway_protocol}://{env.gateway_host}/view/{dataset}/"
+            f"{env.external_protocol}://{env.external_host}/view/{dataset}/"
         )
         subpath = path[len(dataset) :]  # noqa: E203
 

cellxgene_gateway/env.py

@@ -9,11 +9,13 @@
 
 import os
 import logging
+import socket
 
 cellxgene_location = os.environ.get("CELLXGENE_LOCATION")
 cellxgene_data = os.environ.get("CELLXGENE_DATA")
-gateway_host = os.environ.get("GATEWAY_HOST")
-gateway_protocol = os.environ.get("GATEWAY_PROTOCOL")
+gateway_port = int(os.environ.get("GATEWAY_PORT", "5005"))
+external_host = os.environ.get("EXTERNAL_HOST", os.environ.get("GATEWAY_HOST", f"localhost:{gateway_port}"))
+external_protocol = os.environ.get("EXTERNAL_PROTOCOL", os.environ.get("GATEWAY_PROTOCOL", "http"))
 ip = os.environ.get("GATEWAY_IP")
 extra_scripts = os.environ.get("GATEWAY_EXTRA_SCRIPTS")
 ttl = os.environ.get("GATEWAY_TTL")
@@ -22,12 +24,13 @@ enable_upload = os.environ.get("GATEWAY_ENABLE_UPLOAD", "").lower() in ['true',
 env_vars = {
     "CELLXGENE_LOCATION": cellxgene_location,
     "CELLXGENE_DATA": cellxgene_data,
-    "GATEWAY_HOST": gateway_host,
-    "GATEWAY_PROTOCOL": gateway_protocol,
     "GATEWAY_IP": ip,
 }
 
 optional_env_vars = {
+    "EXTERNAL_HOST": external_host,
+    "EXTERNAL_PROTOCOL": external_protocol,
+    "GATEWAY_PORT": gateway_port,
     "GATEWAY_EXTRA_SCRIPTS": extra_scripts,
     "GATEWAY_TTL": ttl,
     "GATEWAY_ENABLE_UPLOAD": enable_upload,
@@ -47,8 +50,6 @@ def validate():
 
         export CELLXGENE_LOCATION=~/anaconda/envs/cellxgene-dev/bin/cellxgene
         export CELLXGENE_DATA=../cellxgene_data
-        export GATEWAY_HOST=localhost:5005
-        export GATEWAY_PROTOCOL=http
         export GATEWAY_IP=127.0.0.1
     """
         )

cellxgene_gateway/extra_scripts.py

@@ -14,6 +14,6 @@ from json import loads
 def get_extra_scripts():
     # can be array of script tags to inject on every page, e.g. for google analytics could be
     # ['https://www.googletagmanager.com/gtag/js?id=UA-123456-2',
-    #  f"{env.gateway_protocol}://{env.gateway_host}/static/js/google_ua.js"]
+    #  f"{env.external_protocol}://{env.external_host}/static/js/google_ua.js"]
     # where google_ua.js is a script you add to the static/js folder prior to deployment.
     return [] if env.extra_scripts is None else loads(env.extra_scripts)

cellxgene_gateway/gateway.py

@@ -37,7 +37,7 @@ from cellxgene_gateway.util import current_time_stamp
 
 app = Flask(__name__)
 cache = BackendCache()
-location = f"{env.gateway_protocol}://{env.gateway_host}"
+location = f"{env.external_protocol}://{env.external_host}"
 
 
 @app.errorhandler(CellxgeneException)
@@ -232,7 +232,7 @@ def main():
     background_thread.start()
 
     app.launchtime = current_time_stamp()
-    app.run(host="0.0.0.0", port=5005, debug=False)
+    app.run(host="0.0.0.0", port=env.gateway_port, debug=False)
 
 
 if __name__ == "__main__":

run.sh.example

@@ -1,8 +1,5 @@
 export CELLXGENE_LOCATION=$(pwd)/.cellxgene-gateway/bin/cellxgene
 export CELLXGENE_DATA=../cellxgene_data
-export DEPLOYMENT_ENV=dev
-export GATEWAY_HOST=localhost:5005
-export GATEWAY_PROTOCOL=http
 export GATEWAY_IP=127.0.0.1
 
 #Once these are set, you run like a normal Flask app