Add develop branch Dockerfile

- Update docs

docker/Dockerfile

@@ -3,7 +3,10 @@
 FROM alpine:3.6
 MAINTAINER Fred Park <https://github.com/Azure/blobxfer>
 
-RUN apk add --update --no-cache musl build-base python3 python3-dev libressl-dev libffi-dev ca-certificates \
+RUN apk update \
+    && apk add --update --no-cache \
+        musl build-base python3 python3-dev libressl-dev libffi-dev \
+        ca-certificates \
     && pip3 install --no-cache-dir --upgrade blobxfer \
     && apk del --purge build-base python3-dev libressl-dev libffi-dev \
     && rm /var/cache/apk/*

docker/Dockerfile.develop

@@ -0,0 +1,19 @@
+# Dockerfile for Azure/blobxfer develop branch
+
+FROM alpine:3.6
+MAINTAINER Fred Park <https://github.com/Azure/blobxfer>
+
+RUN apk update \
+    && apk add --update --no-cache musl \
+        musl build-base python3 python3-dev libressl-dev libffi-dev \
+        ca-certificates git \
+    && git clone -b develop --single-branch https://github.com/Azure/blobxfer.git /blobxfer \
+    && cd /blobxfer \
+    && pip3 install --no-cache-dir -e . \
+    && python3 setup.py install \
+    && cd / \
+    && rm -rf blobxfer \
+    && apk del --purge build-base python3-dev libressl-dev libffi-dev git \
+    && rm /var/cache/apk/*
+
+ENTRYPOINT ["blobxfer"]

docs/20-yaml-configuration.md

@@ -32,7 +32,9 @@ azure_storage:
 * `endpoint` specifies for which endpoint to connect to with Azure Storage.
 Generally this can be omitted if using Public Azure regions.
 * `accounts` is a dictionary of storage account names and either a
-storage account key or a shared access signature token.
+storage account key or a shared access signature token. Note that if you
+are downloading a striped blob (Vectored IO), then all storage accounts for
+which the blob is striped to must be populated in this list.
 
 ### <a name="options"></a>`options`
 The `options` section specifies general options that may be applied across

docs/30-vectored-io.md

@@ -92,4 +92,7 @@ keep this metadata in-tact or reconstruction will fail.
 ```
 
 In order to take advantage of `stripe` Vectored IO across multiple
-destinations, you must use a YAML configuration file.
+destinations, you must use a YAML configuration file. Additionally, when
+downloading a striped blob, you must specify all storage account locations
+of the striped blob in the `azure_storage` section of your YAML
+configuration file.