Merge pull request #24 from umccr/make-silent-night-distro

Made libumccr distribution light weight

README.md

@@ -8,8 +8,8 @@ UMCCR Reusable Python modules
 ## Usage
 
 - Install through ``pip`` like so
-```commandline
-pip install libumccr
+```
+pip install libumccr[aws]
 ```
 
 - Somewhere in your Python code
@@ -31,6 +31,32 @@ for obj in libs3.get_matching_s3_objects(bucket, prefix=key_prefix, suffix=key_s
     print(f"s3://{bucket}/{obj['Key']}")
 ```
 
+## Transitive Dependencies
+
+### Self Service
+
+- Starting from `v0.4` onwards, by default, the libumccr does not mandate any transitive dependencies installation for you.
+- That is, if you `pip install libumccr` then it just installs the libumccr package itself only. 
+- If you do this, you will have to manage transitive dependencies at your application as _end-of-chain_ requirement as you see fit. 
+- This depends on which libumccr modules you need to depend on at your code.
+- The _bare metal_ libumccr should only have some hundred KB (~180KB) by size. However, in this case, you can only use those modules that leverage built-in Python standard library (that does not require extra 3rd party library). Which might be perfect for use case like [libregex](libumccr/libregex.py) - well-known sample ID pattern parser contributed by team member ([Alexis](https://github.com/alexiswl)). You may contribute here similar and, reuse it at your end...
+
+### Feature Flag
+
+- If you would like to have all transitive dependencies requirement along with the libumccr, then do so `pip install libumccr[all]`.
+- For any aws related module usage with boto3 and botocore, then `pip install libumccr[aws]`.
+- Similarly, for gspread, gdrive and Pandas dataframe stuff, then `pip install libumccr[libgdrive]`.
+- Do check out [setup.py](setup.py) `extras_require` sections for all available feature flags. 
+
+_This arrangement is sometime also known as deferring feature flag toggle at runtime requirement. Google for more..._
+
+### AWS Lambda
+
+- The AWS Lambda runtime for Python environment already contain the boto3 and botocore.
+- Hence, if you are bundling your application for AWS Lambda Python runtime environment then you just need to `pip install libumccr`. 
+- The AWS Lambda Python runtime is already satisfied the `libumccr[aws]` feature requirement, and you can still leverage the [libumccr.aws](libumccr/aws) modules with the lowest asset bundle footprint. 
+- In this case, you just put `boto3` in your local development environment only; such as `reqirements-dev.txt` or equivalent.
+
 ## Development
 
 - Create Python virtual environment

setup.py

@@ -10,7 +10,7 @@
 
 setup(
     name="libumccr",
-    version="0.4.0rc1",
+    version="0.4.0rc2",
     author="UMCCR and Contributors",
     author_email="services@umccr.org",
     description="UMCCR Reusable Python modules",
@@ -49,18 +49,31 @@
             "nose2",
             "awscli-local",
         ],
+        "libgdrive": [
+            "requests",
+            "pandas",
+            "gspread",
+            "gspread-pandas",
+            "google-auth",
+        ],
+        "aws": [
+            "boto3",
+            "botocore",
+            "cachetools",
+        ],
         "all": [
             "Django",
+            "boto3",
+            "botocore",
+            "cachetools",
+            "requests",
+            "pandas",
+            "gspread",
+            "gspread-pandas",
+            "google-auth",
         ],
     },
     install_requires=[
-        "boto3",
-        "botocore",
-        "cachetools",
-        "requests",
-        "pandas",
-        "gspread",
-        "gspread-pandas",
-        "google-auth",
+        # silent night
     ],
 )

tox.ini

@@ -1,10 +1,10 @@
 [tox]
-envlist = py{36,37,38,39}
+envlist = py{36,37,38,39,310,311,312}
 skip_missing_interpreters=True
 
 [testenv]
 passenv = GITHUB_*
 deps =
-    .[test]
+    .[test,all]
 commands =
     pytest