Update Python READMEs

Python/Batch/README.md

@@ -1,9 +1,9 @@
 ##Azure Batch Python Samples
 
 ###Configuring the samples
-In order the run the samples, they must be configured with Azure Batch and
-Azure Storage credentials. The credentials for each sample are gathered from
-the common configuration located [here](./configuration.cfg). Once you
+In order to run these Python samples, they must be configured with Azure Batch
+and Azure Storage credentials. The credentials for each sample are gathered
+from the common configuration located [here](./configuration.cfg). Once you
 have configured your account credentials, you can run any of the samples and
 they will make use of the credentials provided in the common configuration
 file.
@@ -13,10 +13,11 @@ Each sample also has a configuration file specific to the individual sample
 
 ###Setting up the Python environment
 In order to run the samples, you will need a Python interpreter compatible
-with version 2.7, 3.3, 3.4 or 3.5. You will also need to install the Azure
-Batch and Azure Storage python packages.  This can be done using the
-[requirements.txt](./requirements.txt) file using
-`pip install -r requirements.txt`
+with version 2.7 or 3.3+. You will also need to install the
+[Azure Batch](https://pypi.python.org/pypi/azure-batch) and
+[Azure Storage](https://pypi.python.org/pypi/azure-storage) python packages.
+Installation can be performed using the [requirements.txt](./requirements.txt)
+file via the command `pip install -r requirements.txt`
 
 You can also optionally use the
 [Visual Studio project](./BatchSamples.pyproj) and the
@@ -41,9 +42,10 @@ sku ids, selecting a publisher, offer and sku for the Linux VM gallery image.
 ####[sample3\_encrypted\_resourcefiles.py](./sample3_encrypted_resourcefiles.py)
 This sample shows how to generate on-demand encryption keys in conjunction with
 [blobxfer](../Storage) to encrypt local files into Azure Storage which will
-then be decrypted for the task when it executes. This sample showcases a
-variety of Azure Batch interaction including: adding certificates to an
-account, creating a pool with a certificate reference, and accessing
+then be decrypted for the task when it executes which ensures encrypted files
+not only in transit but encrypted in storage immediately. This sample
+showcases a variety of Azure Batch interaction including: adding certificates
+to an account, creating a pool with a certificate reference, and accessing
 certificates on a Linux Batch compute node. This sample is geared towards
 Linux with the assumption of a locally available OpenSSL and blobxfer
 installation that is accessible from the sample path invocation. This sample
@@ -64,37 +66,38 @@ tunnel script to interact with the batch pool after the sample is run. When
 using this option, you will also need to disable the delete pool option as
 well.
 
-####[sample5\_docker\_batch\_task.py](./sample4\_docker\_batch\_task.py)
-This sample shows how to schedule tasks that run a docker container. 
+####[sample5\_docker\_batch\_task.py](./sample5\_docker\_batch\_task.py)
+This sample shows how to schedule tasks that run a docker container.
 Specifically, this sample uses pool start up task to install docker
 on Batch VMs which is Ubuntu based. It then uses job preparation tasks to
 pull the application image from docker hub and finally submit a set of simple
 tasks. Each task will launch a docker container to perform ffmpeg transcoding
 on a downloaded mp4 file and upload the result to blob storage. The docker
 image is CentOS based with ffmpeg and blobxfer preinstalled. The docker image
-(yidingz/ffmpeg:v3) is available on docker hub. 
+(yidingz/ffmpeg:v3) is available on docker hub.
 
-###Azure Batch on Linux Best Practices
+##Azure Batch on Linux Best Practices
 
 Although some of the Python samples are not specific to Linux, the Azure Batch
 team would like to provide guidance on best practices for hosting your Linux
 workloads on Azure Batch.
 
-* _Wrap your command(s) in a shell or provide a shell script_
+####Wrap your command(s) in a shell or provide a shell script
 
 Unless you have a single program you wish to execute that is resolvable in the
-default `PATH` specified by the distribution (i.e., `/bin`, `/usr/bin`), then
-it is advisable to wrap your command in a shell. For instance,
+default `PATH` specified by the distribution (i.e., `/bin`, `/usr/bin`) or
+the `$AZ_BATCH_TASK_WORKING_DIR`, then it is advisable to wrap your command
+in a shell. For instance,
 
     /bin/bash -c "command1 && command2"
 
-would execute `command1` followed by `command2`, if command1 was successful
+would execute `command1` followed by `command2`, if `command1` was successful
 inside a bash shell.
 
 Alternatively, upload a shell script as part of your resource files for
 your task that encompasses your program execution workflow.
 
-* _Check for exit codes for each command in a shell script_
+####Check for exit codes for each command in a shell script
 
 You should check for exit codes within your shell script for each command
 invocation in a series if you depend on successful program execution for
@@ -120,31 +123,36 @@ modified to:
 If `command2` fails, then the entire script will exit with the proper
 return code of the failing command.
 
-* _Wait for your background commands_
+####Wait for your background commands
 
 If you require executing multiple programs at the same time and cannot split
 the invocation across multiple tasks, ensure you wrap your execution flow in
 a shell and provide the appropriate wait command for all child processes. For
 instance,
 
-    /bin/bash -c "command1 &; command2 &; command3 &; wait"
+    #!/usr/bin/env bash
+    set -e
+    command1 &
+    command2 &
+    command3 &
+    wait
 
 This would ensure that all child processes exit before the parent exits.
 Without the `wait` command, the Azure Batch service will not be able to
 properly track when the compute node has completed execution of the
 backgrounded tasks.
 
-* _Set preferred locale and encoding_
+####Set preferred locale and encoding
 
 Linux shell scripts or program invocations via Azure Batch tasks will execute
 under the `POSIX` locale. If your programs require a specific locale and
 encoding, e.g., to encode Unicode characters, then you will need to set the
 locale via an environment variable. For example,
 
     # set environment variables on job: applies to all tasks in job
-    job = batchserviceclient.models.CloudJob(
+    job = azure.batch.models.CloudJob(
         common_environment_settings=[
-            batchserviceclient.models.EnvironmentSettings('LC_ALL', 'en_US.UTF-8')
+            azure.batch.models.EnvironmentSetting('LC_ALL', 'en_US.UTF-8')
         ],
         # ... more args
     )
@@ -154,16 +162,17 @@ encoding for all tasks added to the job. Alternatively you can set environment
 variables for each individual task:
 
     # set environment variables on single task
-    task = batchserviceclient.models.CloudTask(
+    task = azure.batch.models.CloudTask(
         environment_settings=[
-            batchserviceclient.models.EnvironmentSettings('LC_ALL', 'en_US.UTF-8')
+            azure.batch.models.EnvironmentSetting('LC_ALL', 'en_US.UTF-8')
         ],
         # ... more args
     )
 
 There are similar environment settings arguments for start task, job
 preparation task, and job release task. Although we recommend using the
-built-in environment variable control provided by the Azure Batch API, you
+built-in environment variable control provided by the Azure Batch API as
+these environment variables will be set on execution of the task, you
 can, as always, directly set shell environment variables in the shell
 invocation for your task command(s):
 
@@ -173,13 +182,14 @@ A final note: not all locales may be present and installed on the compute node
 and may require a start task or job preparation task for installation of the
 desired locale.
 
-* _stdout.txt and stderr.txt encoding_
+####stdout.txt and stderr.txt encoding
 
-On Linux compute nodes, these files are encoded with UTF-8. If your program
-generates Unicode characters, ensure that the file is interpreted with UTF-8
-encoding. Please see above related note regarding locale and encoding.
+On Linux compute nodes, task `stdout.txt` and `stderr.txt` files are encoded
+with UTF-8. If your program generates Unicode characters, ensure that the file
+is interpreted with UTF-8 encoding. Please see above related note regarding
+locale and encoding.
 
-* _Do not perform release upgrades on compute nodes_
+####Do not perform release upgrades on compute nodes
 
 Many distributions offer the ability to perform a release upgrade. By
 "release upgrade," we refer to major version upgrades such as from Ubuntu
@@ -195,7 +205,7 @@ new pool.
 Note that we are evaluating automating os and security updates as a possible
 future enhancement.
 
-* _Consider asyncio for blocking Azure Batch calls_
+####Consider asyncio for blocking Azure Batch calls
 
 With Python [3.4](https://docs.python.org/3.4/library/asyncio.html),
 [3.5+ (async/await)](https://docs.python.org/3.5/library/asyncio.html), or

Python/Storage/README.rst

@@ -224,7 +224,10 @@ specify the share name as the second positional argument.
 
 The above example would upload all files in the ``localfiles`` directory to
 the share named ``myshare``. Encryption/decryption options are compatible with
-Azure Files as the destination or source.
+Azure Files as the destination or source. Please refer to this `MSDN article_`
+for features not supported by the Azure File Service.
+
+.. _MSDN article: https://msdn.microsoft.com/en-us/library/azure/dn744326.aspx
 
 General Notes
 -------------